Talk:Periodic table

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Featured article Periodic table is a featured article; it (or a previous version of it) has been identified as one of the best articles produced by the Wikipedia community. Even so, if you can update or improve it, please do so.
Main Page trophy This article appeared on Wikipedia's Main Page as Today's featured article on February 28, 2004.
          This article is of interest to the following WikiProjects:
WikiProject Chemistry (Rated FA-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Chemistry, a collaborative effort to improve the coverage of chemistry on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Featured article FA  This article has been rated as FA-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
 
WikiProject Elements (Rated FA-class, Top-importance)
WikiProject icon This article 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.
Featured article FA  This article has been rated as FA-Class on the quality scale.
 Top  This article has been rated as Top-importance on the importance scale.
 
WikiProject Physics (Rated FA-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Physics, a collaborative effort to improve the coverage of Physics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Featured article FA  This article has been rated as FA-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
 
WikiProject Russia / Science & education (Rated FA-class, High-importance)
WikiProject icon This article is within the scope of WikiProject Russia, a WikiProject dedicated to coverage of Russia on Wikipedia.
To participate: Feel free to edit the article attached to this page, join up at the project page, or contribute to the project discussion.
Featured article FA  This article has been rated as FA-Class on the project's quality scale.
 High  This article has been rated as High-importance on the project's importance scale.
Taskforce icon
This article is supported by the science and education in Russia task force.
 
Wikipedia Version 1.0 Editorial Team / v0.5 / Vital (Rated FA-class)
WikiProject icon This article has been reviewed by the Version 1.0 Editorial Team.
Featured article FA  This article has been rated as FA-Class on the quality scale.
 ???  This article has not yet received a rating on the importance scale.


Change location of the f-block[edit]

Opening statement[edit]

As the article currently appears, the f-block (57-71 and 89-103) is placed underneath the table, marked by two asterisks to the left of Lu and Lr, creating a column in-between 2 and 3. Lu and Lr are both members of the f-block, which is why I propose that they be moved into the disjoined rows, and the asterisks take their place. This would shorten the table and make the grouping of the f-block more easily recognizable.

G31r0d (talk) 00:01, 10 September 2015 (UTC)

Initial responses[edit]

Hydrogen (diatomic nonmetal)
Helium (noble gas)
Lithium (alkali metal)
Beryllium (alkaline earth metal)
Boron (metalloid)
Carbon (polyatomic nonmetal)
Nitrogen (diatomic nonmetal)
Oxygen (diatomic nonmetal)
Fluorine (diatomic nonmetal)
Neon (noble gas)
Sodium (alkali metal)
Magnesium (alkaline earth metal)
Aluminium (post-transition metal)
Silicon (metalloid)
Phosphorus (polyatomic nonmetal)
Sulfur (polyatomic nonmetal)
Chlorine (diatomic nonmetal)
Argon (noble gas)
Potassium (alkali metal)
Calcium (alkaline earth metal)
Scandium (transition metal)
Titanium (transition metal)
Vanadium (transition metal)
Chromium (transition metal)
Manganese (transition metal)
Iron (transition metal)
Cobalt (transition metal)
Nickel (transition metal)
Copper (transition metal)
Zinc (transition metal)
Gallium (post-transition metal)
Germanium (metalloid)
Arsenic (metalloid)
Selenium (polyatomic nonmetal)
Bromine (diatomic nonmetal)
Krypton (noble gas)
Rubidium (alkali metal)
Strontium (alkaline earth metal)
Yttrium (transition metal)
Zirconium (transition metal)
Niobium (transition metal)
Molybdenum (transition metal)
Technetium (transition metal)
Ruthenium (transition metal)
Rhodium (transition metal)
Palladium (transition metal)
Silver (transition metal)
Cadmium (transition metal)
Indium (post-transition metal)
Tin (post-transition metal)
Antimony (metalloid)
Tellurium (metalloid)
Iodine (diatomic nonmetal)
Xenon (noble gas)
Caesium (alkali metal)
Barium (alkaline earth metal)
Lanthanum (lanthanide)
Cerium (lanthanide)
Praseodymium (lanthanide)
Neodymium (lanthanide)
Promethium (lanthanide)
Samarium (lanthanide)
Europium (lanthanide)
Gadolinium (lanthanide)
Terbium (lanthanide)
Dysprosium (lanthanide)
Holmium (lanthanide)
Erbium (lanthanide)
Thulium (lanthanide)
Ytterbium (lanthanide)
Lutetium (lanthanide)
Hafnium (transition metal)
Tantalum (transition metal)
Tungsten (transition metal)
Rhenium (transition metal)
Osmium (transition metal)
Iridium (transition metal)
Platinum (transition metal)
Gold (transition metal)
Mercury (transition metal)
Thallium (post-transition metal)
Lead (post-transition metal)
Bismuth (post-transition metal)
Polonium (post-transition metal)
Astatine (metalloid)
Radon (noble gas)
Francium (alkali metal)
Radium (alkaline earth metal)
Actinium (actinide)
Thorium (actinide)
Protactinium (actinide)
Uranium (actinide)
Neptunium (actinide)
Plutonium (actinide)
Americium (actinide)
Curium (actinide)
Berkelium (actinide)
Californium (actinide)
Einsteinium (actinide)
Fermium (actinide)
Mendelevium (actinide)
Nobelium (actinide)
Lawrencium (actinide)
Rutherfordium (transition metal)
Dubnium (transition metal)
Seaborgium (transition metal)
Bohrium (transition metal)
Hassium (transition metal)
Meitnerium (unknown chemical properties)
Darmstadtium (unknown chemical properties)
Roentgenium (unknown chemical properties)
Copernicium (transition metal)
Ununtrium (unknown chemical properties)
Flerovium (post-transition metal)
Ununpentium (unknown chemical properties)
Livermorium (unknown chemical properties)
Ununseptium (unknown chemical properties)
Ununoctium (unknown chemical properties)
Talk from 32-column PT. 18-column PT discussions are astrologic to me.Please talk from the 32-column form. 18-column PT with that rectangle below introduces description difficulties, unnecessary (It's like an IKEA cupboard unboxed, deerly needing a manual ;-) ). Now, you mention "f-block", but there also exists the "lanthanides", which does not equals f-block. It also involves the description of "group 3". -DePiep (talk) 00:13, 10 September 2015 (UTC)
(Don't mind DePiep's strong words on the 18-column PT: sure the 32-column one is more accurate, but unfortunately it isn't actually used seriously by anyone to a first approximation yet.) Now, to answer your question: Lu and Lr are not f-block elements - they are d-block elements. (Also how can the f-block have 15 elements per period when an f-subshell can only hold 14 electrons?) The term you are looking for is "inner transition metal". Double sharp (talk) 17:20, 10 September 2015 (UTC)
Your're right, I rephrased. -DePiep (talk) 17:44, 10 September 2015 (UTC)
Thank you! ^_^ Yes, 18-column needs a manual, especially for the version the OP advocates (Sc/Y/*/**) - what is under Y, and how does the f-block fit in? Sc/Y/La/Ac or Sc/Y/Lu/Lr 18-columns are at least unambiguous. Double sharp (talk) 17:58, 10 September 2015 (UTC)

G31r0d, I agree with your suggestion even though the main reason you gave isn't valid, and therefore, the title of this talk page section also isn't valid. I'll be creating a new talk page section below with a valid title to discuss the same proposal for different reasons. As pointed out above, the f-block must have 14 elements per period (and the d-block must have 10). That's a consequence of solutions to the Schrödinger equation when the azimuthal quantum number is three for f and two for d. This means that, if emphasizing the f-block is the goal, then elements (not asterisks) must be placed below Sc and Y. However, if blocks are going to be emphasized then the reader is best served if they are fully emphasized for all blocks, including the s-block. A periodic table layout that does this is already present at Wikipedia at Block_(periodic_table). Different periodic tables in different articles ought to serve the reader in different ways as discussed at Talk:Block_(periodic_table)#Pedagogical_goals.2C_empiricism_and_theory. Flying Jazz (talk) 00:54, 11 September 2015 (UTC)

On second thought, after reminding myself of other talk page discussions on similar periodic table topics in recent years, I changed my mind about creating a new talk page section about this matter. If a larger community of knowledgeable, focused editors with a reader-focused, pedagogical mindset were present then a focused discussion about this topic might be doable here. Flying Jazz (talk) 02:22, 11 September 2015 (UTC)
My main gripe with Sc/Y/*/** is that there is no obvious way to translate it to 32-column, as shown in the header and footer PTs on each element article. IMHO it ought to be either Sc/Y/La/Ac (following Greenwood and many esteemed textbooks) or Sc/Y/Lu/Lr (according to the arguments by Scerri and Jensen, among others), if you want to keep those tables 32-column. I do admit that Sc/Y/La/Ac would seem preferred according to an RS count today. It is not my favourite, but my personal preference should not come into this for WP. Double sharp (talk) 12:11, 11 September 2015 (UTC)
P.S. Yes I know IUPAC uses Sc/Y/*/** in their PT, but they don't think the question is their business IIRC. Double sharp (talk) 12:14, 11 September 2015 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────A focused discussion would involve goals that editors are attempting to achieve for the sake of a general reader in the broad context of an article about the table and in the even broader context of many other articles where some "generic Wikipedia table" is used. What layout reinforces the most important chemical knowledge at an appropriate level for that hypothetical general reader? Your view seems to be that one particular correct, true, and best identity of group 3 in the 32-column table is so important that group 3 must be presented in the 18-column table in such a way as to maintain that identity. This might be why you frame the discussion as a decision among "Sc/Y/*/**," "Sc/Y/La/Ac," and "Sc/Y/Lu/Lr." Someone with my preference would frame the discussion as a decision among "unify and label the lanthanides/actinides" "separate La/Ac from the other lanthanides/actinides," and "separate Lu/Lr from the other lanthanides/actinides." My view is that the group 3 issue, the 18-versus-32 column issue, and the relationship of both issues to each other are all relatively unimportant matters in introductory chemistry pedagogy when compared to the more central role of the table as an arrangement to spatially unite elements with particular chemical identities. In addition to having united background colors that match united spatial locations, the general reader is also served in my view by having correct and fully meaningful labels (i.e., * = lanthanide and ** = actinide). The discussion among the authors that we've both read is interesting, and I'm familiar with it. But a correspondence between location, chemical identity, and meaningful labels are more important to the general reader than a layout that depends upon our editorial judgement about the outcome of that discussion. Flying Jazz (talk) 20:21, 11 September 2015 (UTC)

I don't really care which is the best anymore: I just think that all those PTs on WP are meant to be generic ones, not really illustrating any particular point: so they should probably all be consistent with each other, as otherwise the difference raises questions that really are unimportant in context. It's just that if you use "keep Ln and An together", there's no way to make a consistent 32-column table as we already use. Which is perhaps an argument to get rid of the 32-column table in favour of the 18-column ones in our element articles. A much stronger argument for that is that 18-column is what you will find just about everywhere when nobody's trying to prove a point.
Once you stop proudly displaying 32-column everywhere, there are no further obstacles to keeping Ln and An together – which is also the form supported by the most RSs, save perhaps only separating La and Ac. And I agree that for the beginning reader, the f-block elements are not a big deal, and the most salient point is showing how similar they all are - which Sc/Y/*/** does best. (Even Greenwood, for example, which discusses La and Ac along with Sc and Y and restricts the Ln and An chapters to Ce-Lu and Th-Lr respectively, admits in those chapters that it is very useful to include La and Ac IIRC. I don't have it with me right now so if I'm wrong, please correct me.)
tl;dr: use 18-column throughout whenevera standard PT is needed, and use Sc/Y/*/**, as used by many RSs and fits the chemistry the best. Sc/Y/La/Ac is really the only possible alternative by RSs and it gives the false impression that La and Ac act noticeably differently from the rest of the lanthanides and actinides. Double sharp (talk) 08:18, 12 September 2015 (UTC)
It's been stated before that any of the three layouts under discussion for the 18-column table used in this article can be found in excellent reliable sources. Therefore, separating La/Ac from the other lanthanides/actinides is not the only possible alternative by RSs. At some point in the past, someone may have argued that huge literature surveys will result in a judgement that a particular table is the one supported by sources. In an actual community of science editors at an encyclopedia, an inappropriate literature survey that doesn't serve the reader would be treated with derision. Also, at some point in the past, someone may have argued that our judgement should be based on how to "make a consistent 32-column table." Again, in an actual community of science editors, the idea that the reader is served when editors make decisions based on how one table can make another would be treated with derision. As for displaying the 32-column table everywhere, that is a separate issue for a different discussion. If you've reached a decision about what serves the general reader of an encyclopedia for the template table in the Overview section of this particular article, then you've made a focused editorial judgement about a particular content issue, and that's what we're here to do. Flying Jazz (talk) 13:55, 12 September 2015 (UTC)
As I said above: for this table, Sc/Y/*/** is my personal editorial judgement, based on what I think would serve the general reader. I do wonder what is so wrong with wanting PTs that serve the same purpose to look the same, but that is a different issue: now you have an answer for what you state we are here to do. Double sharp (talk) 22:06, 12 September 2015 (UTC)
I'll refactor the talk page in a day or two, wait a few weeks to see if anyone else chimes in, and then make the change. Flying Jazz (talk) 23:24, 14 September 2015 (UTC)
Although a note: while I think Sc/Y/*/** would be OK in isolation, I still think the consistency argument (if we have both an 18-column table and a 32-column table, and both are meant as simple illustrations, they should show the same layout, having the same purpose) would trump it as long as we have 32-column tables displayed proudly in WP for general purposes. Double sharp (talk) 09:54, 21 September 2015 (UTC)
re Double sharp's "I don't really care which ...." full post. Sorry I'm this late, must have been distracted in between. Now your general line, to me, seems to be that an 18-column PT everywhere (as our 'general presentation form', except for special issues of course) is better for the reader. First I note that an 18-column PT does not solve the group-3 question. 18-col form still has to make a point in this (just as it does with the position of He). We'd still need to decide on how to present group 3 even if we abolish and delete every 32-col form! Namely: group 3 = Sc/Y/Lu/Lr OR group 3 = Sc/Y/La/Ac. The suggestion that "group 3 = Sc/Y/*/**" solves it is wrong: it just buries the issue in ambivalence (See my #Sc/Y/*/** in 32-column form below). Then, have we solved & made this unhidable group 3 point, there is the question on why we would prefer to put those 14 or 15 elements (times 2) below. I only hear this argued by people educated before 1990 or so. [warning: I'm charging, to make this clear] That is: those who have internalised the 18-col, and saying "I understand it all right this way, so young students don't need a new form". (Disclosure: I was educated back then too). -DePiep (talk) 22:00, 8 October 2015 (UTC)

Sc/Y*/** once more[edit]

──────────────────────────────────────────────────────────────────────────────────────────────────── Sc/Y*/** should appear in the overview section for the reasons given by Flying Jazz, including general reader interests, and pedagogy. This would be consistent with the 32-column table appearing in the lede and footer, since both tables (18-column and 32-column) are premised on spatially uniting, to the greatest extent practical, chemically similar elements. Unresolved arguments on which elements go with Sc and Y in group 3 are appropriately and well enough summarized in the open questions and controversies section of the article and don't need to spill over into the lede, the overview section, or the foooter. I have argued Sc/Y/Lu/Lr for WP in the past and have now changed my mind in light of discussions on WP and in other forums. Sandbh (talk) 04:23, 5 October 2015 (UTC)

But again: what or which "Sc/Y/*/**" presentation do you mean? Given the topic, I request that the group number "3" is present (something even Scerri does not do consistantly). I can think of multiple graphical variants (seriously) that would comply. You can link to a 18-, 32-, or x-column variant as you like, as long as it is not ambiguous and not ambivalent wrt this issue. (Is why I discard the IUPAC drawing). btw, I read Sc/Y/*/**. -DePiep (talk) 21:46, 6 October 2015 (UTC)
I missed what FJ said about pedagogy. But re Sandbh in this, I disagree. There is no reason why the 18-col and 32-col form should be structurally different. Any pupil or student should be able to zoom in on a more detailed PT and discover, but this requires that these are the same. One does not start telling that the earth is flat, and then leave it to the pupil to discover something different. 18-col PT and 32-col PT must represent the same. (See #Sc/Y/*/** in 32-column form). And btw, in PT overview the f-block (two rows and then a step) is more convincing that the Ln, An categories. Categories are loosing their PT-behaviour in the p-block anyway. -DePiep (talk) 22:54, 8 October 2015 (UTC)
The version appearing in the lede of the periodic table article, top right. Like the template was originally, here. But I would like to hear your view and the views of other editors about this. Sandbh (talk) 09:13, 7 October 2015 (UTC)
That's the undecided one (bad). Gap column missing. Does not show whether Sc/Y is glued to group 2 or to group 4. Not a new PT. -DePiep (talk) 22:24, 7 October 2015 (UTC)
Sandbh: I am not a specialist in elements. I am a technician. Now it occurs to me time and time again that scientist in PT (physicists, chemicists) can not reflect. Can not convey. It is a "we all know" attitude --as in "we in the lab"--. Each and every time I ask for a graphical representation here, I get these variants/not-a-variant/whatever/youknowwhatimean things, but *not* an answer. From Sandbh, from IUPAC, that sort of people ;-). -DePiep (talk) 22:41, 7 October 2015 (UTC)
Yes, the gap column is missing but I think it is more imporant that an encyclopedia aimed at the general reader emphasizes the most important chemical similarities, and reinforces this with joined up background colours. Hence I support showing 15 lanthanides and actinides at the foot of the table. If the curious reader wonders what is going on with the gap in group 3, and just how the lanthanides and actinides fit in the gap they will be able to work this out from looking at the 32-column table in the article footer. If they get really interested there is also the accompanying text in the article on which elements are the period 6 and 7 members of group 3.
I further submit that Lu shares more in common with the rest of the lanthanides than it does with Sc and Y, or the other period 6 transition metals. If we accept that the purpose of our 18-column table is to (as best we can) group like with like then this requires showing 15 lanthanides at the foot of the table. Sandbh (talk) 11:52, 8 October 2015 (UTC)
...I think I'm convinced. Sc/Y/*/** it is as default. 32-column will stay as it is showing Sc/Y/Lu/Lr. Double sharp (talk) 15:12, 8 October 2015 (UTC)
re Sandbh above: "If the curious reader wonders what is going on with the gap in group 3, and just how the lanthanides and actinides fit in the gap" -- uh, there is no gap in the Sc/Y/*/** form (eg the IUPAC graph). The reader is lead into thinking that the PT, asterisks expanded, looks like #Sc/Y/*/** in 32-column form Red XN. Right above the asterisks, it says "3". -DePiep (talk) 22:11, 8 October 2015 (UTC)
Yes check.svg Done: changed {{Periodic table}} and {{Periodic table (18 columns, large cells)}} to show Sc/Y/*/**. Double sharp (talk) 15:16, 8 October 2015 (UTC)
I reverted. This "Sc/Y/*/**" is not the result of this discussion. And I noted elsewhere and often, it introduces an ambiguous form for group 3 and the f-block. As explained (e.g. in #Graphic_presentations), the 18-column form should follow from a 32-column form, and so represent it (the gap column does that). If the general 32-column shows group 3 being Sc/Y/La/Ac: fine, and we'll move the gap column to the right (btw, this is how dewiki has their general PT). Now this is what happens if you read that "Sc/Y/*/**" Red XN presentation made into a 32-column: -DePiep (talk) 20:31, 8 October 2015 (UTC)
Source: Sc/Y/*/** (IUPAC)

Periodic Table, showing Sc/Y/*/** in 32-column form   (view)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
1 H He
2 Li Be B C N O F Ne
3 Na Mg Al Si P S Cl Ar
4 K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
5 Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
6 Cs Ba La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
7 Fr Ra Ac Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr Rf Db Sg Bh Hs Mt Ds Rg Cn 113 Fl 115 Lv 117 118
-DePiep (talk) 20:58, 8 October 2015 (UTC)
We know the gap is made ambiguous by Sc/Y/*/**. We know it results in an inconsistency with the Sc/Y/Lu/Lr 32-column periodic table, but the main argument for Sc/Y/*/** is that it emphasizes the chemical similarities of the lanthanides and actinides, as well as Sc and Y. Then both the 18- and 32-column periodic tables unite similar elements as much as possible (e.g. Yb is next to Lu in Sc/Y/Lu/Lr 32-column, but not in Sc/Y/Lu/Lr 18-column, whereas it is in Sc/Y/*/** 18-column), given the universally accepted constraint that one element takes up exactly one cell. Double sharp (talk) 05:36, 9 October 2015 (UTC)
P.S. It's not that I don't think your argument regarding 18/32-column consistency has merit. I guess Sandbh+FJ+my position of yesterday is that we should have 18- and 32-column tables constructed on the same principles, whereas yours is that the end result should look the same (i.e. both should have Sc/Y/Lu/Lr, or Sc/Y/La/Ac). I guess it depends on which you think is more likely to confuse a reader: two slightly differently-looking periodic tables, or two periodic tables that appear to imply different chemical similarities. It may perhaps be arguable that the latter is a subjective interpretation, while the former is a fact about how the tables look. But I'm not going to take a side on this any longer. Whatever a majority consensus thinks is fine, I will accept. Double sharp (talk) 05:50, 9 October 2015 (UTC)
Double sharp: with all due respect, times two: no. That is: no. I don't have the time, nor the plan, nor the patience, to explain the same thing again to you. Must say I am quite astonished that you keep opposing (even by edits) without discussing. -DePiep (talk) 23:05, 10 October 2015 (UTC)
What is this if not a discussion? What is this if not a 3:1 consensus? Double sharp (talk) 04:55, 11 October 2015 (UTC)
C'mon. You claim 'discussion' and you do 'vote-counting'? -DePiep (talk) 22:16, 11 October 2015 (UTC)
I see three people (me, Sandbh, and FJ) who have discussed it with arguments and came to a conclusion, and one (you) who opposes that conclusion. Looks like a discussion and consensus to me. The counting is just a rough estimate to get a sense of how clear the consensus is, but it seems pretty clear that there is one. Double sharp (talk) 04:27, 12 October 2015 (UTC)

DePiep, there is no need to stretch the Sc and Y cells. This is not done for the 32 column table in the footer. A simple note added to the Sc|Y|*|** table will point the reader to an explanation (with pictures) on how the lanthanides and actindes fit under Y (see below). I agree Sc|Y|*|**, at face value, is ambiguous. However, I think this is outweighed for reasons others have noted previously. And Sc|Y|La|Ac or Sc|Y|Lu|Lr have their own problem. Sandbh (talk) 02:09, 11 October 2015 (UTC)

Periodic table
Group 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Alkali metals Alkaline earth metals Pnicto­gens Chal­co­gens Halo­gens Noble gases
Period

1

Hydro­gen
1
He­lium
2
2
Lith­ium
3
Beryl­lium
4
Boron
5
Carbon
6
Nitro­gen
7
Oxy­gen
8
Fluor­ine
9
Neon
10
3
So­dium
11
Magne­sium
12
Alumin­ium
13
Sili­con
14
Phos­phorus
15
Sulfur
16
Chlor­ine
17
Argon
18
4
Potas­sium
19
Cal­cium
20
Scan­dium
21
Tita­nium
22
Vana­dium
23
Chrom­ium
24
Manga­nese
25
Iron
26
Cobalt
27
Nickel
28
Copper
29
Zinc
30
Gallium
31
Germa­nium
32
Arsenic
33
Sele­nium
34
Bromine
35
Kryp­ton
36
5
Rubid­ium
37
Stront­ium
38
Yttrium
39
Zirco­nium
40
Nio­bium
41
Molyb­denum
42
Tech­netium
43
Ruthe­nium
44
Rho­dium
45
Pallad­ium
46
Silver
47
Cad­mium
48
Indium
49
Tin
50
Anti­mony
51
Tellur­ium
52
Iodine
53
Xenon
54
6
Cae­sium
55
Ba­rium
56
1 asterisk
Haf­nium
72
Tanta­lum
73
Tung­sten
74
Rhe­nium
75
Os­mium
76
Iridium
77
Plat­inum
78
Gold
79
Mer­cury
80
Thallium
81
Lead
82
Bis­muth
83
Polo­nium
84
Asta­tine
85
Radon
86
7
Fran­cium
87
Ra­dium
88
2 asterisks
Ruther­fordium
104
Dub­nium
105
Sea­borgium
106
Bohr­ium
107
Has­sium
108
Meit­nerium
109
Darm­stadtium
110
Roent­genium
111
Coper­nicium
112
Unun­trium
113
Flerov­ium
114
Unun­pentium
115
Liver­morium
116
Unun­septium
117
Unun­octium
118
1 asterisk
Lan­thanum
57
Cerium
58
Praseo­dymium
59
Neo­dymium
60
Prome­thium
61
Sama­rium
62
Europ­ium
63
Gadolin­ium
64
Ter­bium
65
Dyspro­sium
66
Hol­mium
67
Erbium
68
Thulium
69
Ytter­bium
70
Lute­tium
71
2 asterisks
Actin­ium
89
Thor­ium
90
Protac­tinium
91
Ura­nium
92
Neptu­nium
93
Pluto­nium
94
Ameri­cium
95
Curium
96
Berkel­ium
97
Califor­nium
98
Einstei­nium
99
Fer­mium
100
Mende­levium
101
Nobel­ium
102
Lawren­cium
103
 N.B. The occupancy of the group 3 positions below yttrium is discussed in the article.

P.S. If you look at some historical periodic tables, you will find that the "*" under Y is sometimes explained as La and the following elements: so you can even argue that the form suggested is in fact a Sc/Y/La/Ac 32-column table, but with La and Ac put together with the (other) lanthanides and actinides, which are similar. No need to stretch cells. Similarly it can also be explained as Lu and the preceding elements. But since there is a disagreement it is best to stay neutral on the matter unless we are trying to illustrate a point – and here we are not. Double sharp (talk) 05:02, 11 October 2015 (UTC)

re Double sharp there is no need to stretch the Sc and Y cells - Yes there is, because that is what the graph says. The IUPAC Sc/Y/*/** graph has "3" (i.e. group 3) over the Sc an Y, and over the asterisks. So your IUPAC form says it. It is not even a "need", it is a fact. -DePiep (talk) 21:32, 11 October 2015 (UTC)
Double sharp. To cut things short: how does the 32-column PT look like for your Sc/Y/*/** 18-column version? Graphs only. -DePiep (talk) 21:38, 11 October 2015 (UTC)
Like this:
Double sharp (talk) 04:26, 12 October 2015 (UTC)
DePiep, Unless we find a WP:RS that shows a 32-column PT with stretched cells, attempting to show it on WP would cross the line into WP:OR. Even if we find such a RS, we should be careful that we are not violating WP:UNDUE. YBG (talk) 03:32, 12 October 2015 (UTC)
It is the other way around, YBG. The 18-col version Double sharp is promoting itself has "3" above the asterisks, i.e. all lanthanides and actinides are in group 3, by drawing. Not me. Also, it has Sc and Y glued to both group 2 and group 4 (no space --gap-- between). So when the reader expands it to 32-col form, these properties are to be maintained with the result. It is not me who drew them. In general, the two forms should be representing the same PT. As for shortcut arguments: the 18-col form is OR or FRINGE for putting those elements in group 3. -DePiep (talk) 08:18, 14 October 2015 (UTC)
What? If IUPAC is fringe, then what is Jensen?
And to be clear, I am assuredly not the only one here promoting this. And how do you know the first asterisk means "all the lanthanides" and not "La and the following elements" (creating Sc/Y/La/Ac) or "Lu and the preceding elements" (creating Sc/Y/Lu/Lr)? Greenwood and Earnshaw doesn't treat La as a lanthanide, incidentally. Double sharp (talk) 12:16, 14 October 2015 (UTC)
re "What? If IUPAC is fringe, ...": Ask IUPAC why they draw group 3 this way. Not me. -DePiep (talk) 19:48, 17 October 2015 (UTC)
So re Double sharp: your 32-col differs by taking f-block (14 columns of Ln andAN). out of group 3. Also it states that SAC/Y are to the right hand, while that does not follow from the 18-col form. It is not the same PT, you have added statements that are not in the 18-col form. -DePiep (talk) 08:21, 14 October 2015 (UTC)
I prefer to think of it this way: Sc/Y/*/** cannot expand to your form, as an element cannot take up more than one cell: the only PTs that have done this are pretty fringe/specific-in-intention (with the exception of duplicating H in group 1 and 17) and not meant for a beginner's overview (unlike Sc/Y/*/**). The answer is that Sc/Y/*/** is a form to use for beginning, when the f-block elements don't really matter. This keeps similar elements together. When you put them in the main body of the table, though, you have to stop equivocating over La and Lu, so the best way to preserve the table's intentions (keeping similar elements together) would be Sc/Y/Lu/Lr if you want to keep the blocks together. But doing this in the 18-column table tears Lu away from its fellow lanthanides. Double sharp (talk) 12:16, 14 October 2015 (UTC)
If IUPAC is fringe - WP putting all Ln and An in group 3 is OR or FRINGE. IUPAC doing that is wrong - they have drawn the PT wrong. -DePiep (talk) 19:32, 15 October 2015 (UTC)
If we're going to throw around acronyms: you are moving into WP:IDIDNTHEARTHAT mode. Double sharp (talk) 02:15, 16 October 2015 (UTC)
  • Sandbh, I add a new argument here, so as not to disrupt your summary.
I note that, say, elegance of the result is mentioned (i.e., for cutting or not cutting same colored groups). Of course, if factual things are the same, we are free to choose an elegant result from options. However, elegance should not be an argument to form a drawing some way, into making it wrong/distorted/ambiguous.
I add that another point of elegance is to be mentioned: showing the steps clearly (two-period steps). Some weeks ago it looked like the User:Sandbh/sandbox seemed to end up preferring Sc/Y/Lu/La, which would nicely show the steps, and keep colors & blocks together (as opposed to the alternative there Sc/Y/La/Ac). Again, this does not add to the conclusion, but is says "nice if our preference has this too". -DePiep (talk) 15:12, 16 October 2015 (UTC)

Summary of arguments[edit]

Sc|Y|*|**

Pros Cons
Spatially unites elements with similar chemical identities At face value, membership of group 3 is uncertain
Unites background colors that match united spatial locations Not clear how the Ln and An fit in
Does not take sides in the inconclusive debate on which element (La or Lu) occupies the position below Y At face value, implies that "*|**" is the correct notation rather than La and Ac, or Lu and Lr

Sc|Y|Lu|Lr

Pros Cons
Membership of group 3 is clear Contradicts other authors who assign La and Ac to group 3
Clear how the Ln and An fit in Spatially separates elements with similar chemical identities
Spatially separates elements with similar background colours

Is this a fair summary? Sandbh (talk) 10:49, 13 October 2015 (UTC)

Thanks for the summary of a very long set of discussions. I think that an encyclopedia article on the periodic table should show the various presentations which are common in books and on websites, whether or not they seem logical to the editors of the article. That includes at least three 18-column formats: Sc|Y|La|Ac, Sc|Y|Lu|Lr and Sc|Y|*|**. The first (Sc|Y|La|Ac) seems to have disappeared from the article, although it is the most common in textbooks and even on classroom walls. I do not agree that Sc|Y|*|** preserves neutrality - it implies that * and ** are the correct notation rather than La and Ac or Lu and Lr. Dirac66 (talk) 14:03, 13 October 2015 (UTC)
So here's a summary for Sc/Y/La/Ac (which I would like, if Sc/Y/*/** continues to be unacceptable for whatever reason, because this way we have Greenwood & Earnshaw and Holleman & Wiberg etc. behind us):
I added "Contradicts other authors who assign Lu and Lr to group 3" and moved the other two con entries down one row to be more consistent with the second table. Sandbh (talk) 10:54, 14 October 2015 (UTC)
I amended the neutrality "pro" to make its meaning clearer. Sandbh (talk) 10:59, 14 October 2015 (UTC)

Sc|Y|La|Ac

Pros Cons
Membership of group 3 is clear Contradicts other authors who assign Lu and Lr to group 3
Clear how the Ln and An fit in Spatially separates elements with similar chemical identities
Many authors use this arrangement Spatially separates elements with similar background colours
(I still like it myself, as Sc and Y are rather peripheral transition metals, almost main-group-ish, and La would fit a main-group-style trend better than Lu. But the key point is that this way we don't equivocate on the composition of group 3, if you don't like that, and we would have more support from authorities using Sc/Y/La/Ac than if we chose to use Sc/Y/Lu/Lr.) Double sharp (talk) 14:25, 13 October 2015 (UTC)
The current PT in the lead shouldn't be considered to be the standard to comply with; it can be altered as well. I agree we should show the three in this article, but since we will use just one in most circumstances, I would advocate using the /Lu/Lr one. If the group 3 topic is not all that important for most readers (it is not), it shouldn't be a problem to have the /Lu/Lr version. Except can we (DePiep (talk · contribs)) make the cell between groups 2 and 3 thinner (use thinner cells?), so it's not too intense at attracting attention?--R8R (talk) 09:00, 14 October 2015 (UTC)
Thinner would be great. I tried that a few times, but could not get it done in the fixed-colwidth setting. (btw, it is that column that I call "gap", for disconnected cells). -DePiep (talk) 19:36, 15 October 2015 (UTC)
I removed the "consistent with/contradicts table in the lede" arguments as I agree these arguments are neither here nor there. Sandbh (talk) 10:48, 14 October 2015 (UTC)

Since Sc/Y/*/** continues to be unacceptable for DePiep, my second choice is Sc/Y/La/Ac along with those famous textbooks that I mentioned. With /Lu/Lr, I think we're going to have questions. I learnt the table with Sc/Y/La/Ac (which I still see a lot) and have seen Sc/Y/*/** many times too, but I never once saw Sc/Y/Lu/Lr before encountering WebElements (and then got interested in the issue). Double sharp (talk) 11:56, 14 October 2015 (UTC)

Analysis of arguments[edit]

  • I agree with User:Dirac66, who said, "I think that an encyclopedia article on the periodic table should show the various presentations which are common in books and on websites, whether or not they seem logical to the editors of the article." This is a separate issue that can be addressed in the Layout variants section of the article.
  • I agree with Flying Jazz (retired), who said, "At some point in the past, someone may have argued that huge literature surveys will result in a judgement that a particular table is the one supported by sources. In an actual community of science editors at an encyclopedia, an inappropriate literature survey [i.e. one] that doesn't serve the reader would be treated with derision." For that reason I have struck out the pro and con arguments about consistency with the literature.
  • I agree with Flying Jazz, who said, "My view is that the group 3 issue, the 18-versus-32 column issue, and the relationship of both issues to each other are all relatively unimportant matters in introductory chemistry pedagogy when compared to the more central role of the table as an arrangement to spatially unite elements with particular chemical identities…a correspondence between location [and] chemical identity are more important to the general reader than a layout that depends upon our editorial judgement about the outcome of that discussion." For this reason I have struck out the pros or cons to do with the group 3 issue.
  • I agree with User:Double sharp's observations that Sc|Y|*|** tables (imagined or real) that widen the Sc and Y cells to span the 15 Ln and Ac elements, "are pretty fringe/specific-in-intention (with the exception of duplicating H in group 1 and 17) and not meant for a beginner's overview (unlike Sc/Y/*/**)" and "The answer is that Sc/Y/*/** is a form to use for beginning, when the f-block elements don't really matter."
  • Looking back on the tables 1–3 as they appear now it seems to me that the general reader's interests would be best served by featuring Sc|Y|*|** in the article.
  • And I think the Layout variants section would be more clear if it made mention of Sc|Y|La|Ac and Sc|Y|Lu|Lr.
  • User:DePiep, I respect your right to disagree and ask you to place the interests of the general reader ahead of your concern about graphical integrity in the region of group 3. My preference is to instead make mention of this concern in the Layout variants section. Sandbh (talk) 10:57, 15 October 2015 (UTC)

RBR Gtrs comment: For me, one point is not clear. So if the f block does not matter, why does it help anyone advocate the /*/** position? I think if it is not, we would be fine with any of the three. And if I am correct and it actually is not all that important for a newcomer, we could also aid a more educated person by not choosing the ambiguous variant? We're an encyclopedia, after all, we may be user-friendly (personally, I am doing my best to ensure user-friendliness in my writing), but not on expense of anyone else.--R8R (talk) 11:20, 15 October 2015 (UTC)

I'll try and address your comments as best as I can. All the relevant considerations for a general reader are set out in tables 1–3. Yes, you are right about the f block. Catering for a more educated reader is not really on the agenda of this particular discussion. That is why I struck out the reference in table 1 to the membership of group 3 being unclear, as this is not an issue for the general reader. I was almost going to say the same thing about User:Dirac66's mention of having the article show the various presentations which are common in books and on websites etc as this was somewhat of another distraction from the core discussion but I capitulated because I thought it could be quickly addressed. Now a couple of things about the more educated reader. I shouldn't be saying anything about this because it's another distraction from the central theme of the discussion. Anyway, 1. A more educated reader would likely be comfortable with a degree of ambiguity. And if this causes some wonderment they can read more about it in the article. 2. If we choose a less ambiguous variant then this results in spatially separating elements with similar chemical identities and spatially separating elements with similar background colours which, for the general reader, is bad. I don't want to go on about these side issues because---in good faith, and trying to be civil---we end up going down energy-sucking and argument-distracting rabbit holes. Sandbh (talk) 10:41, 16 October 2015 (UTC)

Dirac66 comment: I agree with Sandbh that the Layout variants section would be a good place to present the different versions used today: Sc/Y/La/Ac, Sc/Y/*/**, Sc/Y/Lu/Lr and perhaps 32-column and Janet too. Dirac66 (talk) 11:49, 15 October 2015 (UTC)

YBG comment: I think ambiguity is one of the strongest arguments for */**. Experts seem to disagree as to the correct membership of group 3; the /*/** form, by its very ambiguity, neatly avoids this issue. YBG (talk) 02:18, 16 October 2015 (UTC)

With respect, YBG, ambiguity is a side issue that is not relevant at the general reader level. Sandbh (talk) 10:41, 16 October 2015 (UTC)

Double sharp comment: I think the idea is that with Sc/Y/*/**, all the lanthanides (which are similar) are together, while Sc/Y/La/Ac tears La away from Ce and Sc/Y/Lu/Lr tears Lu away from Yb in their 18-column forms. In a 32-column form, there is no obstacle to Sc/Y/Lu/Lr, as Yb and Lu remain together. Double sharp (talk) 02:22, 16 October 2015 (UTC)

DePiep comment. There is a flaw in this setup. It suggests there is an equal 3-way choice at hand. However, this is not the case. There is the choice between group 3 = Sc/Y/Lu/Lr and group 3 = Sc/Y/La/Ac. Both have a scientific base, and the choice would only be for all of our general PT's being presented in the same way (as a preference). Of course, our general reader is best-served if our PT's are consistently congruent in this. The non-chose option would be explained in article group 3 (most likely). The text is being prepared by Sandbh in User:Sandbh/sandbox. I have no preference between these two (except that we should pick just one for that purpose). And I thought it was converging to one of these.

Important to note is that these two variants are based on scientific (chemical, physical) arguments, both options being well-based. Once a preference is chosen, we can draw whichever one (see here). They are two, say, structural variants. This has a big big analogy with the positioning of He in the PT: multiple options are well-based in science, but we use only one in our general PT's. For this, the group 3 = Sc/Y/Lu/Lr or Sc/Y/La/Ac variants do not belong in the Periodic_table#Layout_variants section.

However. The third option under discussion, named Sc/Y/*/** here, is *not* about a scientific third variant. It is a presentation variant. Now this could be an improvement, given that we know what structural PT we want to present. And to cut this short: up until now I have not been able to get the structural variant it represents (seeing the IUPAC 18-col drawing). I repeat: it does not show which group 3 variant it uses. Even worse: it has two graphic features that are undeniably wrong or ambivalent: "group 3" is said to contain 32 elements, and it says that Sc and Y are spanning 15 columns. This is unacceptable. Another sign of bad drawing is that no one has been able to show how its 32-col drawing looks like. This goes against the second argument mentioned: the Good of our Reader. Well, inconsistent PT's are helping no one. (At this point, I see no need to evaluate 18-col vs 32-col. Useless if they are inconsistent).

I think the order of choices is 1. which structural variant do we want to use in general; 2. By what layout (graphic features) do we draw that. -DePiep (talk) 15:45, 16 October 2015 (UTC)

DePiep, as I see it you are re-presenting an argument that has been previously discussed and set aside i.e. re the composition of group 3. Your attempted distinction between "scientific" and "presentation" variants is a dressed-up version of the same theme. All three variants are based in science; they just differ in which scientific aspects are emphasized. Concerns about the composition of group 3 are relatively unimportant in introductory chemistry pedagogy. A correspondence between location and chemical identity is more important for the general reader. Sandbh (talk) 00:01, 18 October 2015 (UTC)

Sandh comment: I'll see if I can draft a Sc|Y|*|** version of the periodic table article in my sandbox. This would include Dirac66's suggestion to say some more about the other major variants. I'll try and do justice to DePiep's comments by seeing if I can refer to the graphical ambiguity issue, without blowing the general reader's mind. I'm not sure yet but this may mean moving the Layout alternatives section further down the article, nearer to the Open questions and controversies section, possibly amalgamating it with the Alternative structures section. Sandbh (talk) 10:30, 21 October 2015 (UTC)

I'll await this result before making any analysis or argument wrt these replies. -DePiep (talk) 10:14, 23 October 2015 (UTC)
I'm sorry. I can't hold my breath any longer. So Sandbh is stating that "Sc/Y/*/**" represents a third scientifically based PT structure? How and when did that #3 creep into the User:Sandbh/sandbox? I must say, today I do not understand the Sandbh sandbox any more. Up to you, Sandbh. -DePiep (talk) 22:37, 23 October 2015 (UTC)
You did well holding your breathe for so long. The sandbox version is still evolving. Things may come and go. I don't know yet what will happen to the draft Sc/Y/*/** content. I'm trying to say what needs to be said in the "Layout variants subsection" without duplicating what's in the "Period 6 and 7 elements in group 3 subsection." This is hard, so bear with me. Sandbh (talk) 06:09, 24 October 2015 (UTC)
OK, and it feels good to be breathing again. You know, if you want me to sketch weird graphical things I'll loyally make views I might not yet agree with. -DePiep (talk) 19:56, 25 October 2015 (UTC)
Thank you DePiep. Are you able to make 15LaAC, 14CeTh and 14LaAc forms, to go into the periodic table article draft Layout subsection? I had in mind the same style as the two 32-column tables in the Period 6 and 7 elements in group 3 subsection, except that these need to be separate 18-column tables, like the existing svg image in the Layout subsection. Sandbh (talk) 10:58, 27 October 2015 (UTC)

Three forms[edit]

────────────────────────────────────────────────────────────────────────────────────────────────────

In svg overview form, they are:
15LaAc
('Sc/Y/*/**')
14LaAc (group 3=Sc/Y/Lu/Lr)
14CeTh (group 3=Sc/Y/La/Ac)
You want them in micro-form?
Does anyone else think it would look better if the "*/**" in the main table were aligned vertically with the "*/**" in bottom? Of course, what is actually used in the published literature is infinitely more important that what I think would be visually appealing. YBG (talk) 04:04, 28 October 2015 (UTC)
re YBG. Of courseIn the processs I considered that, but this is my outcome so far. The job of the 'asterisks' (big dots here) is to explain the displacement. Nothing more, nothing less. Now to make the re-gathering of the 'footnote' (as the bottom set is sometimes called) the simplest mental step possible, I prefer to hint only a vertical displacement of the leftmost footnote elements (is what you see). Note that we tend to read left-to-right. Then, it is easily to grasp that all the elements aright of the asterisks need a vertical shift to allow the footnote in. Aligning the 'asterisks' vertically would imply that the footnote is shifted one column to the right too (a horizontal displacement). Needless complication in the mental step.
Even better, consider this. Since the one and only connection between the footnote and the main table are the asterisks (and specifically not the downward extended column/group suggestion), we should stress the disconnection even further. OrWe must prevent that the reader, in their mental re-gathering step, shifts the whole right-of-asterisks bunch of elements one row up to accommodate the footnote in (period 4 Sc or Ti ending up in period 2!). This bad suggestion/possibility we must prevent. I even plan to draw the gap-column (i.e., the column with only the asterisks) smaller, intentionally breaking the suggestion that the footnote-groups (columns) are a continuation of the true group columns above. Also R8R asked about this.
Finally, this is not about "visually appealing", I am not lead by aesthetics (iirc, above such arguments were mentioned like keeping colors together). I only use aesthetics after the What of the PT is in, and after I have a How-to-graph-it. Next, to use "what is actually used in the published literature" as you suggest, we have problems in there. 1. As Jensen (2008) notes, he sees that quite often textbook text does not represent that textbook's PT graph. (I claimed that the IUPAC 18-column has this error). 2. I note too that any google search for an 18-col image shows dozens of graphical variants of the same PT (times the two or three structural variants we are talking about here). In any PT you see out there (books, internet), just look at exactly how the footnote is graphically linked to its place the main table. Quite often graphically ambivalent or wrong. These are unfit for a vote count on 'usage in literature'. 3. We are talking about a graphical structure, not a Van Gogh painting. The graphics must reflect the underlying structure. We can not describe a certain PT structure in the texts, and then pick a 'I like this face' PT from internet. -DePiep (talk) 11:06, 28 October 2015 (UTC)
Points well taken. Thanks for bearing with me. YBG (talk) 17:11, 28 October 2015 (UTC)
Yes, agree. Could you please go ahead DePiep. Sandbh (talk) 20:57, 28 October 2015 (UTC)
Edited and corrected my text for clarity. -DePiep (talk) 16:47, 29 October 2015 (UTC)
DePiep, could you also do a micro-form version of the 32-column table as per Jensen i.e. with the cells for Sc and Y stretched across La-Lu? Thank you. Sandbh (talk) 23:27, 29 October 2015 (UTC)
You mean to show the PT I wrote here at "20:58" [search term]? -DePiep (talk) 01:30, 31 October 2015 (UTC)
Does not Jensen (2008) say that this form was abandoned after 1946 (admitting that Seaborg wrote it that way back then?). -DePiep (talk) 02:03, 31 October 2015 (UTC)
────────────────────────────────────────────────────────────────────────────────────────────────────Yes, 20:58 is the one. As I read Jensen, he refers to it as antiquated, "chemical nonsense", rather than saying it was abandoned. It was still used by, for example, Housecroft and Sharpe (2008), Inorganic Chemistry, 3rd ed. They show "La-Lu" and "Ac-Lr" under Y. As well, the IUPAC table, which shows "57-71 lanthanoids [sic]" and "89-103 actinoids" in the two positions under Y is essentially the same. I've included a place for a micro-form 20:58 table in my sandbox, to illustrate what Jensen was describing, and to do justice to your concerns about the graphical ambiguity issue. PS: Could we use a dagger rather than two asterisks to refer to the actinides? Sandbh (talk) 04:11, 31 October 2015 (UTC)
re Sandbh This reply may be telegraph-style. May sound like making claims - so be it. No obstruction intended in there, but these days I don't have the time or patience to reply carefully & with due care, as the topics & you fellow editors deserve.
- dagger - no, not now. It's complicated enough. January.
- this edit by Double sharp, read the es. Pending this discussion, I'll leave it alone, and not discuss it here, again for reason of unneeded complication. However, once we have an outcome, it might be reversed. (I'd happily do that. Double sharp says that cerium is in group 4).
- Sandbh asks for micro-PT's to have 18-col variants. Today, I don't think that shows well in a webpage. For example, take a look at the extended PT in micro form, which uses asterisks for for footnote elements (predicted elements 118+, this instance). It is way too tiny and scribbled to explain something to the reader. Also, the template does not scale at all (one cannot set the size). Let's consider this: the three svg-forms above are fine and can show anything we want. Now they don't have hyperlinks to articles (cannot click on Hg to get to mercury). But that can be added to the svg (I'm experimenting). So we can have scalable, clickable PT images. Great IMO, even and especially in this whole topic.
- Sandbh also asks for PT's that are "chemical nonsense" (dixit Jensen), and not to be used on the PT history page (fun section)! I don't get the proposal (by Sandbh/sandbox). This is a tough not nut to crack. I need time, and a clear & fresh mind to understand what this is leading to.
- Concluding. Technically (as in: graphs & webpage-smart) things are limited and can be handled, when isolated. But there are content issues too. I have not enough energy to solve the questions, nor to respond fresh and sound. Is why I can not promise to deliver all tries, for now. Nor can I make serious replies to content issues. OTOH, most of my arguments are on this wiki already. -DePiep (talk) 19:19, 1 November 2015 (UTC)
Best way forward is when Sandbh drops the graph issue, and firstly has the group 3 scientific aspects into articles. (IOW, do-not-touch the 18-col issue). -DePiep (talk) 22:34, 3 November 2015 (UTC)

Draft Sc|Y|*|** version ready[edit]

I think the draft in my sandbox has got to a more or less stable state (aside from persnickety formatting and linking fussiness). The main changes appear in section 5 "Alternative structures", which is now called "Different periodic tables", and has two subsections, 5.1 "Layout variations", which summaries the three main forms of 18 column table; and 5.2 "Alternative structures" (no change). I've added a paragraph to section 6.6 "Period 6 and 7 elements in group 3" about the attempted use of the 32 column table to address this question. I've tried to do justice to DePiep's graphical concerns by including content about the apparent implications of some 15LaAc tables. Comments welcome. Sandbh (talk) 11:23, 5 November 2015 (UTC)

I'll go and take a look. Must say, after what I met last weeks and you throwing together, as I read here, in #5 "Layout variations" and "Alternative structures" (in that order!!!), I'll prepare for a disappointment. -DePiep (talk) 21:51, 5 November 2015 (UTC)
No. -DePiep (talk) 23:04, 5 November 2015 (UTC)
  • Nature gave us elements in a 32-column PT structure. Nature did not mind classroom size, book page format or as-I-learned-it habits. That PT was gradually and gently unfolded (discovered) by Mendeleev's "8-col" etcetera. Group 0 (=18) could be added without being disruptive. Ln's: more unfolding. Seaborg: more of the same (though he didn't mind the graphic back in '46). Some stuff is still scientifically ambivalent: position of He, categorisation of metallishness, and group 3 composition. Fine. We should aim to present this PT, in its forms & variants as based, and by detail when & where required, to our readers. And, by the way, as far as I know a classroom wall has the right format for a 32-column PT (longitude:height), much better than for an 18-column.
However. Proposal User:Sandbh/sandbox#Different_periodic_tables has this: It describes graphical variants and structural variants of the PT as one sort of. It says: "5. Different_periodic_tables" (and, sic, in the reversed order of relevance at that). We must recognise: structural variants like 14CeTh and 14LaAc are scientifically based, and throw a nice new light on the PT. As do Janet's Left Step and ADOMAH. A graphical variant on the other hand, is just an illustrators choice: pic font and background colors, choose decomposition into footnote elements maybe. By definition, graphical variants do not alter the structure being presented.
In short: that is what the "Sc|Y|*|**" (or "15LaAc") graph version does/so badly. It does not state which structure it represents. Or, even worse: it claims/suggests that group 3 consists of all 32 rare earth elements. And now Sandbh proposes this as some "3rd variant". But I see no source for such a structural variant (no Jensen, no Scerri). The check questions are: A. So what is in group 3 then?, and B. how do Sc and Y position in your 32-col form? (and C one can check the accompanying textr for discrepancies. e.g., the published IUPAC graph says group 3 = REM, but the text never says so).
On top of this: OR. The linked paragraph opens like this: "The 18-column periodic table featured in this article [and elsewhere in this wiki, I add - DePiep] has been referred to as the 15LaAc form.[79] It shows 15 lanthanides and 15 actinides at the foot of the table". Sure it does, but why was this form chosen to represent it at all? Nowhere its 32-column equivalent is shown or referenced (of course not: it does not exist, in science). The only reference, #79, is jolly titled "Flyleaf [i.e. of textbooks] periodic table" (btw, I hope I understand the frivolity right. How is this a scientific ref?). So the source still has book formatting for base. The proposal also writes "other common forms": 'common' is not a source. Really, not.
I propose to remove the Sc/Y/*/** form from the proposal altogether, and from all our enwiki pages. It can stay in the curiosities section of History of the periodic table. For our kids to laugh at their parents: why did you have to learn the IKEA version of the PT?
-DePiep (talk) 23:04, 5 November 2015 (UTC)

Mind the gap (It lives again)[edit]

A. Medium long form of the periodic table with a gap to accommodate the lanthanides and actinides, and showing Lu and Lr as belonging to group 3
B. Medium long form of the periodic table with a gap to accommodate the lanthanides and actinides, and showing La and Ac as belonging to group 3

@DePiep: I suspect your concerns would be addressed by either option A (left) or option B (right)? If I recall correctly you have no particular preference for either Lu|Lr or La|Ac lining up under Y in group 3. (talk) 03:18, 7 November 2015 (UTC)

DePiep response: re Sandbh, "De Piep, I suspect your concerns would be addressed by ..." - indeed. So please throw the 15LaAc out. My fly over: my response above ("Nature gave us elements in a 32-column PT structure", [23:04]) was against current sandbox proposal [1] that treats three forms alike (the 18-cols: 15LaAc, 14CeTh and 14LaAc). I repeated my note (from "There is a flaw in this setup", [15:45]) that 14CeTh and 14LaAc are sound scientific variants, but 15LaAc is not a new variant, e.g. per my "The check questions are", above; it is a historic curiosity. So I concluded that 15LaAc should be removed from the proposal.
R8R Gtrs comment: I am observing this discussion, rather than actively participating in it; but one thing I want to be clear is that we can't throw away the 15LaAc option. We may say it is not as natural in the sense of how it places 15 elements under one, and describe it as an unnatural construct (except not using these terms) used for particular reasons, but we can't ignore if we think it's wrong. People would use a description of PT variants in an encyclopedia, and we should provide one, since we happen to be editors of an encyclopedia.--R8R (talk) 15:41, 7 November 2015 (UTC)
Double sharp comment: We cannot throw out 15LaAc; it has been used by enough reliable sources that it has to be mentioned. We can of course point out its problems, but it has to stay as a prominent mention. The most we can do is not use it as our main format.(talk) 16:21, 7 November 2015 (UTC)

DePiep, 15LaAc should not be removed for the reasons mentioned above but we can use its unambiguous form (option A or B above) instead of the confusing version currently used in the lede. Sandbh (talk) 22:12, 7 November 2015 (UTC)

Then, which RS states that group 3 contains 32 elements? (don't forget what Jensen wrote: text may not correspond with the accompanying graph [2]). We also do so in this file (I disapprove). -DePiep (talk) 22:47, 7 November 2015 (UTC)
Let me be more specific, re Double sharp: "We cannot throw out 15LaAc; it has been used by enough reliable sources that it has to be mentioned" and Sandbh: "should not be removed for the reasons mentioned above". I propose to throw it out of the topic/sandbox, and move it to the history department because sure it is/was used widely. The question is: which RS uses this graph variant while claiming this describes group 3 correctly? Of course, we should be alerted that the group 3 issue often was not considered at all (e.g., a sort of carelessness or sloppyness in drawing), and that the text might be inconsistent with the graph (IUPAC has, I claim). I also know that the number "3" is missing often, which reduces the preciseness of a PT. The notion "commonly used" may be true, but does not qualify as an RS. -DePiep (talk) 03:11, 8 November 2015 (UTC)
re Sandbh we can use its unambiguous form (option A or B above) instead of the confusing version currently used in the lede (this one). - Yess, this is progress! Glad to have your signature on this. Because the lede PT is an 18-column PT saying group 3=Sc/Y/*/**. It is this 'ambiguity' (or worse) why I want that version out, out of the whole sandbox. (Now after that removal, we should talk about best replacement: one of A/B or those. That's a n independent topic. Really). -DePiep (talk) 03:29, 8 November 2015 (UTC)
YES!!! You have it! High five time! Dip me in molasses, cover me in feathers and call me a lucky duck! Happy days are here again. I'll see if I can refactor the A or B discussion into a new section or subsection, unless someone else beats me to it. Sandbh (talk) 04:06, 8 November 2015 (UTC)
Comment on the new graphs A and B (top of this section). Sandbh is mixing up graphic presentation and scientific base. 1. Quite disturbingly, in here the notion "15LaAc" is used in a new definition, thereby introducing mistakes and mixing up two opposite statements. As Jensen wrote about 15LaAc (as opposed to 14LaAc & 14CeTh): In the case of the 15LaAc form, however, an entirely different interpretation is placed on these elements. The 30 elements are treated not as a separate independent electronic block but rather as degenerate members of group 3 of the d-block. 2. A and B still do not explain any scientific base for the "Sc/Y/*/**" variant. 3. There is an inconsistency wanting to group all lanthanides and actanides together in the footnote. Because this form in general of the PT has blocks not categories together. Categories are not the structure basics in here. It is an addition of complexity to start using categories as drawing criteria. 3 (minor at the moment but still): the re-introduction of "minor long form" terminology does not help, it is ambiguous.
More worrying is that lot of scientists around here seem unable to keep independent issues independent. Instead, even more variants by different issues are created by mixing up. -DePiep (talk) 23:11, 7 November 2015 (UTC)
I confess I don't understand Sandbh's PTs here. If you want to show Sc/Y/La/Ac, and yet use an 18-column table, then doesn't the gap have to be between groups 3 and 4, and La and Ac put under Sc and Y? That's how Greenwood and Earnshaw does it. Double sharp (talk) 02:39, 8 November 2015 (UTC)
(I understand, after some studying, that they are both graphic variants of our familiar scientific group-3 variants: Sc/Y/La/Ac=14CeTh and Sc/Y/Lu/Lr=14LaAc. In essence, the variants have the Ln and An together. Earlier graphs of the same, you might recongnise more easily, are shown in top of #Graphic presentations below). -DePiep (talk) 02:59, 8 November 2015 (UTC)
@Double sharp: DePiep's explanation is correct. In option B the lanthanides and actinides can only be moved into the main body of the table if La and Ac end up under Y in group 3 (and the end result would be a 32 column table). Does this help? Sandbh (talk) 03:33, 8 November 2015 (UTC)
It does, but I'm still concerned about it, because I've never seen this form of Sc/Y/La/Ac 18-column before. Can you point me to a reliable source that uses it? In my experience, the most common 18-column table that seeks to put La and Ac under Y simply does so and has a an asterisk inserted in the cells of La and Ac, obviously intending to signify a gap between La/Ac and Hf/Rf. Double sharp (talk) 04:19, 8 November 2015 (UTC)
No, I can't Double sharp, as I wasn't specifically looking for such an example when I was working on the problem. My focus was instead on seeing if I could eliminate ambiguity. The best I can offer is an example of its 32-column equivalent in: MacKay, MacKay & Henderson (2002, p. 196), Introduction to Modern Inorganic Chemistry, 6th ed., Nelson Thornes, Cheltenham, p. 196. Could you elaborate your concern? Not that I've ever looked closely, but it seems to me that there are effectively innumerable ways of representing a Sc/Y/La/Ac 18-column table, involving (either singly or in combination) asterisks in various positions, daggers, arrows, └> { , vertical pill-box slits, dotted lines, bars, connectors, and sundry other graphical artifacts and doodads, or sometimes even nothing. The tables involved are the equivalent of written forms of communication (are they not?) that all attempt to effectively say the same thing, but in different ways, presumably in an effort to be as clear as possible in their graphical representations. Hence option B. Sandbh (talk) 11:02, 8 November 2015 (UTC)
re Double sharp "It does, but I'm still". My concern is this. At hand at this point are two pairs of graphs:
one pair says "group 3= Sc/Y/Lu/Lr" (graphs 14CeTh and Sandbh's 'A')
other pair says "group 3= Sc/Y/La/Ac" (graphs 14LaAc and Sandbh's 'B')
not at hand now is: any "Sc/Y/*/**" form.
The surpising new A and B forms keep Ln and Ac together by putting exactly all of them in the footnote. Were those form an improvement, it could be worth introducing them to the world through our readers, and having scientists frown & smile. But I see two flaws. Flaw 1: it breaks group 3. The core periodic table structure is columns & rows first, with reasons. A and B forms break this base: group 3(!) is split over two places: two elements are displaced into the footnote. Only after reconstruction (into 32-col form) are they visibly in the place -- while exactly that is what we want to show in this detail. IOW, we want to state "group 3 = ./././.", and then we cut that group in half. It is this extra mental step added that imo does not help the reader. Flaw 2: introducing cuts by category. The Ln and Ac together looks very nice at 1st sight because of their categorycolors being together. But categories are not the structuring base of the PT. Of the 12 categories we show, only one falls together with column(s). Categories are great to show the trend, but they should not be used to make a cut (for practical reason only) in the PT. -DePiep (talk) 06:32, 12 November 2015 (UTC)
Exactly. If we want to show Sc/Y/La/Ac, then we had better see La and Ac in the positions right under Y, and not force the reader to mentally fit them in. Furthermore, I don't see anything wrong with splitting La and Ac from the other lanthanides and actinides if they are coloured similarly. Nobody complains that H is very far away from the other diatomic nonmetals. Double sharp (talk) 06:46, 12 November 2015 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────

Let us proceed with Sc|Y|La|Ac[edit]

A 14CeTh periodic table.jpg

Sc|Y|La|Ac periodic table

Let us proceed with Sc|Y|La|Ac. The lanthanides and actinides are reasonably proximate to one another, spatially and chromatically. Membership of group 3 is clear. How the Ln and An fit into the main body of the table is straightforward, and this yields a 32-column form with no IKEA tears (the crying type). Sc|Y|La|Ac is the most common form. The gap between groups 3 and 4 is concordant with the group 3 metals behaving chemically more like the alkaline earths, and the s-block metals generally. Sc and Y remain coloured as transition metals in homage to their d1s2 electron configurations. The footnoted Ln and An line up nicely under the main body group numbers: Ce under group 4; the transition-metal-like earlier actinides---Th, Pa, U, Np, Pu---under groups 4, 5, 6, 7 and 8. Lanthanum and actinium are coloured as, respectively, a lanthanide and an actinide, seemingly out of whack with their d1s2 elecron configuations but consistent with their stature as progenitors of each of their own series of elements. An elegant package all round. And the asterisks line up nicely, too. Sandbh (talk) 00:28, 15 November 2015 (UTC)

I would support this, as it seems to be the best option. After making this change we can think about the group-12 change later. I'll accept keeping Sc and Y as transition metals as this seems to be the majority opinion, although we should note that they are rather borderline. Double sharp (talk) 04:33, 15 November 2015 (UTC)
Please give me a few days to write my opinion in detail before anything happens.--R8R (talk) 11:20, 15 November 2015 (UTC)

There's a big version of the table in my sandbox, if you want see what it'd look like (before you craft your opinion). The rest of the sandbox article has largely been updated accordingly. Sandbh (talk) 11:31, 15 November 2015 (UTC)

Before I start to write the comment itself, I must say, I'm glad this long discussion has taken place. I have abstained from taking part because it seemed (and, in fact, still seems) to be very inefficient in consuming editor energy. I have limited time available for Wiki, and most of it goes into improving lead. Nonetheless, I am glad; I did follow the discussion, and took my time to reconsider the group 3 and 12 issues.
If everyone suddenly does decide to color group 12 as non-transition, I'll be fine with it. I get the logic how group 12 is not the last step of the transition, but the first step of post-transition... something. That still doesn't make it a main group, and it leaves group 12 somewhat puzzling, but this is minor compared to the whole question of whether it is a part of transition. Actually, the whole issue is quite minor: "transition" is a human-created term. (The question is, is it worth spending our time? I finally understand Stone and his remarks about efficiency of time spending.)
The group 3 issue is more extensive in that respect. It does change the structure of the PT, moving two cells and either splitting the d block, or leaving it a rectangle. I no longer blindly follow the argument it should be a rectangle; but still, it does seem to still be important. So basically, we have a choice between aligning group 3 with the s block, or the rest of the d block. I have previously assumed the former option just steamed from an electron configuration false data obtained in the '50s or something; I have changed my opinion.
I see little point in discussing specific differences between the Sc-Y-La and Sc-Y-Lu trends (although it would indeed make the argument look more detailed and reasonable; but that would be just an illusion); we all know singular, or even structural deviations within the PT are possible (starting with the electron configuration of Cr through d-block and lanthanide contractions and everything they cause). But a comparison of the general trends is indeed useful. So clear trends in Sc-Y-La resemble the s block, and Sc-Y-Lu trends resemble the unclear trends elsewhere in d block.
There are a few places in the PT that could also cause similar uncertainties; for one, if group 3 is not so transitional, then group 4 is not very transitional as well, as group 4 element tend to lose all four electrons. Exceptions occur, and they happen more and more commonly as we go further into the d block, but they again start a trend that I would say begins with group 3. It is the best example of how group 3 is a rightful member of the set of transition metal groups, and not just a poor neighbor. Then we have another problematic point, p block; if the d block is ripped, why is the p block not? And I think there are great similarities between gallium and lutetium (a new block just ended, and it is followed by a trivalent element, in which the contraction coming from it sets a great difference between it and a previous element in its group.
So, again, there is no correct answer on whether group 3 is -Lu-Lr or -La-Ac (I think we all know that, it's said to ensure the comment is complete); the nature did not have a "group 3" in mind when creating atoms. We humans choose. This problem—apparently—has no easy solution, because if there was, we would already know it (which is also why I sincerely believe a majority of sources is not so important in this issue: it's not that overwhelming). So it's putting us in a difficult situation. I have re-examined the problem (I am afraid to not know where the end would be if I went on and on; so I emphasized the points I consider most important, doing my best in generalization), and I believe that in a situation so questionable we should stay with -Lu-Lr. I think I have enough arguments to say this proposal does have a strong basis, see above, and not just the lesser of two evils. In short, my idea could be simplified to "if we have a choice between breaking a block in pieces and not doing it, and we would still have question if we did break it, we shouldn't" (but it's not the statement to oppose, as it would be oversimplification).
As for the sandbox, I think it's great structurally; two minor things I would fix are this—"It has been claimed that such arguments are proof that, 'it is a mistake to break the [periodic] system into sharply delimited blocks.'"— quote and the 14LaAc notation. The quote is fine by itself; but it opposes a statement saying the PT is divided into sharply delimited blocks, and we lack such a statement just before it (or elsewhere). The 14LaAc notation is somewhat user-not-so-friendly; as an alternative, I propose simple naming them options A, B, and C; they are clearly generic and easy to use, so they shouldn't cause problems. (Although the latter may be just me overdoing simplifying things so they are most accessible to anyone.) -- R8R (talk) 09:36, 17 November 2015‎

@R8R Gtrs: Could you clarify why, if there is a gap between the s block and the d block, there would need to be a gap in the p block too? I'm not following your reasoning here. Thank you. Sandbh (talk) 07:29, 18 November 2015 (UTC)

My idea is, why should there be a gap in the d block (since we're debating over whether we would get one or not), when there is no such gap elsewhere, particularly in the p block? (which was outlined because it continues the trend of how groups become taller as you go to the right after the s block. The f block couldn't be divided now even if we wanted to divide it because a g block isn't there yet; and since the s block is so small, the p block would be the only possible candidate for a gap in it.) I don't include the gap between the s and d blocks in my reasoning. Hope this makes my idea clear.--R8R (talk) 07:50, 18 November 2015 (UTC)

Well, split blocks are nothing new. Many tables, including the one appearing (for years now) in the lede, have a split d block in which two d block elements are instead collocated with the footnoted lanthanides and actinides. For the most common table, the one with La and Ac in group 3, when this is assembled in its 32-column form the result is a split d block. Nearly all periodic tables split the s block by showing He in group 18 rather in group 2, and effectively nobody complains about that. So split blocks are nothing new. And nobody has ever suggested that the presence of a split s or d block implies that the p block should be split.

More generally, the periodic table in the lede has never emphasized blocks. It has always been about seeking to categorise and collocate (spatially and chromatically) elements with similar chemical identities—never mind that this has resulted in some block splitting.

In light of the above could you reconsider your argument about block splitting, as there doesn't seem to be anything in it. I have some comments about your other arguments but I'd like to see if we can square away this one first. And I appreciate the thought that went into writing your opinion. Sandbh (talk) 11:31, 19 November 2015 (UTC)

Indeed, many tables break the d block into parts; I don't mean to question that. However, some don't. And we are discussing whether we should pick a PT variety that does make the break or not. There are no blocks (other than the block in question) that are broken into parts. The helium analogy is clearly not applicable here, because helium is a gas with no chemistry, and is moved from above a pronounced metal to another gas with no chemistry, with this anomaly coming from an indisputable physics principle; scandium and yttrium, trivalent metals, would in both cases be located above a trivalent metal.
The emphasis statement is broad, possibly broader than intended, and therefore easily arguable with. (One example obvious when we discuss group three: aluminum is a group III metal that does not experience the d block contraction; the same is true for scandium, but not gallium, its actual neighbor. This causes some anomalies in group 13 at gallium; there would be no such anomalies in the B-Al-Sc trend.) But it must be noted that while chemical similarities were the property the PT was originally built on, nowadays it is commonly described as coming from physics, with chemistry coming from physics as well, in most context beyond school chemistry even possibly even in it.
I, however, can't deny that the original argument about block splitting was not a universal one; there are no breaks in other blocks, but alone it does not mean one can't be found in the d block. I never intended to use this statement. However, in an arguable position when we can't determine any truth none would argue with, and therefore, any solution would be arbitrary (like this one, since blocks are human constructs), I find it most appealing to choose the option that breaks fewer rules; in the context of blocks, a block with no breaks, thus -Lu-Lr.
Like any argument, it can be argued with; it is not a piece of non-disputable truth; but I wouldn't call it null, either.--R8R (talk) 05:48, 20 November 2015 (UTC)

OK I think this is progress. You are saying that as there is otherwise no reasonable way to decide if -La-Ac or -Lu-Lr would be better placed in group 3 under Y, then -Lu-Lr is preferred as it gives a d block with no break. Is that right? Sandbh (talk) 10:29, 21 November 2015 (UTC)

Long story short, yes. There are reasonable arguments for either side, and it depends on you which ones are the more reasonable ones.--R8R (talk) 10:32, 21 November 2015 (UTC)

Good. Are Jensen's "rules" for assigning an element to a position in the periodic table, as set out below, reasonable(?):

The following steps are applied, in order.

1. Assignment to a major block based on the kinds of available valence electrons (i.e., s, p, d, f, etc.).
2. Assignment of the elements within each block to groups based on the total number of available valence electrons.
3. Verification of the validity of the resulting block and group assignments through the establishment of consistent patterns in overall block, group, and period property trends.
4. Verification that the elements are arranged in order of increasing atomic number as required by the periodic law
Unfortunately criteria 1 and 2 do not always lead to an unambiguous assignment and in those few cases where they fail one must resort instead to criterion 3 to help resolve the impasse.

According to Jensen, application of the above rules strongly suggests -Lu-Lr. Sandbh (talk) 11:47, 21 November 2015 (UTC)

Well, given that group 3 behaves more like an s-block group than a d-block group, it seems that Sc-Y-La-Ac is better in showing consistent patterns with Ca-Sr-Ba-Ra, for example. Double sharp (talk) 14:12, 22 November 2015 (UTC)
If I was to create a list of rules listed by importance (not sure if I actually would get into this, but suppose I would), I think that I would come up with something similar. While I understand the desire to break the d block and reasons behind it, to me, this makes most sense. Since Hf is so chemically similar to Zr, and Ta is so close to Nb, and so on, I expect there is no clear chemical trend in group 3, another d block group—so I would jump to that same conclusion.
I do not agree that "group 3 behaves more like an s-block group than a d-block group." Group 3 does remind me of groups 1 and 2, although not exactly; but again, group 4 does do that, too, although to an even smaller extent. Sc and Y almost always lose their three electrons? Ti and Hf tend to lose their four, etc. Sc has hardness between those of Ca and Ti (little data available to make other comparisons), Ca-Sc-Ti from a trend in densities (1.5, 3, 4.5 g/cm3), etc. These metals are called "transition metals" because they represent a transition from the s block to the p block. That's how a transition works: you would expect the group closest to the s block to have something similar with it (although even if it was not for the name, you would still not expect group 3 to be completely different from group 2 (which it isn't), as you wouldn't expect group 4 to be completely different from group 3 (which it isn't), and so on).--R8R (talk) 16:44, 22 November 2015 (UTC)
Alright. Jensen's rules explain why, for example, B and Al sit over Ga rather than Sc. So I don't think we need to worry about peculiarities in periodic trends such as these. The resolution of the group 3 issue rests on Jensen's third rule. I'll post and analyse Jensen's arguments for -Lu-Lr here and I'll further post the arguments that I'm aware of for -La-Ac. Double sharp did some earlier work on this which I'll also add. Ideally, a side-by-side comparison will result in a reasonably clear result. If not, i.e. the difference is not clear enough, we would probably have the flexibility to go either way. Sandbh (talk) 02:02, 25 November 2015 (UTC)
  • re Sandbh "Let us proceed with Sc|Y|La|Ac": I like the reasoning, and so I accept the result (I support). I hope this reflects the text (proposed article text on this). However, pls leave out argument "the most common form", as this is not a sound base (and I dare claiming that there is a more common form: the ambiguous and wrong graph Sc/Y/*/**).
Thank you for your support. The main issue for me is sorting out which 18-column we show. Finalizing the accompanying text will be easy, in comparison. Sandbh (talk) 23:58, 21 November 2015 (UTC)
Is this also a firm rejection of the Sc/Y/*/** form? Or can we expect it to pop up again in the future? -DePiep (talk) 12:12, 21 November 2015 (UTC)
It is one of the common forms mentioned by Clark & White and, regardless of its faults, has become well known thanks or no thanks to IUPAC. Pick a selection of chemistry text books and, chances are, one of them will feature it. It seems to me we should mention it but point out its faults (as rightly criticized by you), including those given by Jensen. Sandbh (talk) 23:58, 21 November 2015 (UTC)
  • re R8R: thx for the careful & patient response. I think we can decide on group 3 without interference of the group 12 issue (that is, a group 12 discussion can continue as it is independent). And I think the 'gap' topic as you describe is superfluous (I can skip it).
One major question: can you push it over the hill, and support the implicit statement wrt group 3 (or do you think am I wrong in this): there are two compositions possible for group 3, both scientifically based. The composition proposed by Sandbh here (group 3 = Sc|Y|La|Ac) is the preferred grouping, as chosen by consensus here. That is, in general we show and describe our PT's with this group 3 constitution. In relevant places (like article group 3, periodic table, ...) the two options may be/will be described. -DePiep (talk) 12:12, 21 November 2015 (UTC)
No support (for the part positioning Ce under group 4). This morning I completely missed this Sandbh point: "The footnoted Ln and An line up nicely under the main body group numbers: Ce under group 4; ..". Well, that would also definitely imply that neptunium is in group 7/VII, and Yb in group 16/VI, etcetera. No way shall we suggest or promote or imply that these footnote elements have group numbers (by this positioning). Again (sigh), in a 32-col PT we do not add group numbers there tooeither. Or, said this way: the footnote is graphically unconnected to the main graph, except for the placeholders (asterisks). The graphic position of the footnote is unrelated to the main graph. Those 14CeTh might as well be on the opposite wall in the classroom, or on the other bookpage -- that's how footnotes work. (And really, I am stunned that this perversion keeps creeping up. I'll have to ask my lawyer to read our talks). Graphically we should prevent this error by shifting the footnote by an irregular col width. -DePiep (talk) 18:48, 21 November 2015 (UTC)
The actinides do not have group numbers but lining up Th, Pa, U etc with Hf, Ta, W etc has a strong historical basis and is consistent with the early actinides showing some similarities to transition metals. There was a periodic table in a recent issue of the Journal of Chemical Education that highlighted this relationship. See also list of oxidation states of the elements and compare Hf, Ta, W etc with Th, Pa, U etc. Sandbh (talk) 23:58, 21 November 2015 (UTC)
Exactly. And furthermore, if it truly does not matter where the footnote is, then they can very well be aligned this way in the 18-column table and not in the 32-column table. Double sharp (talk) 14:29, 22 November 2015 (UTC)
Worth a separate, distinctive topic: continued at #Suggestion that PT grouping applies to footnote elements [3]. -DePiep (talk) 21:31, 24 November 2015 (UTC)
Actually, I'm getting enough of this. Why do these perverted unrelated side-topics keep creeping up in the main topic (ie, group 3 & its presentation)? Why is it proposed as a package-deal, 'all or nothing'? (even mixing up graph and science variants in the process, time and time again). Why did not we decide already crisp & clear that the 3rd thing "Sc/Y/*/**" is to be rejected always everywhere? Every time we are near an outcome, some other ingredient is added to make it a soup. -DePiep (talk) 19:06, 21 November 2015 (UTC)
The side topics are largely a result, as I understand it, of me trying to get a solution that addresses the major concerns. I thought we had solution in the Sc|Y|*|** form however I could not get you over the line on ambiguity. I thought we had a solution in Sc|Y|La|Ac however R8R Gtrs has concerns about a split d-block. I thought I had a solution in Sc|Y|La|Ac with a square and a triangle however other editors prefer asterisks. So now I'm back to either -La-Ac or -Lu-Lr, and on this choice I'm waiting to hear from R8R Gtrs re Jensen's "rules" (see above or search for my 11:47) so I can progress the discussion. Sandbh (talk) 23:58, 21 November 2015 (UTC)

Graphical interlude[edit]

La|Ac and Lu|Lr[edit]

---Extracted by me from the preceding section: "Mind the gap (It lives again)"--- Sandbh (talk) 11:43, 8 November 2015 (UTC)

I was an Lu|Lr in group 3 supporter, having regard to Jensen. Having looked closely into this question again—including doubts raised by Scerri on some of Jensen's arguments—I now (tentatively) support La|Ac in group 3 i.e. option B. As I believe User:Double sharp intimated, the chemistry of Sc, Y, La and Ac has much more in common with the alkaline earth metals in group 2, and the s-block metals more generally, than it does with that of the transition metals proper in groups 4 to 11.

If that is the case, then slightly "gapping" the d-block into portions of one and nine groups would be a more natural grouping of like with like.

Some of Jensen’s arguments for Sc|Y|Lu versus Sc|Y|La are highly questionable e.g. his comparison of periodic trends in atomic radii; sum of the first two ionization potentials (why these two given we are talking about group 3 elements?); melting point; and electronegativity. Here, Sc|Y|La clearly resembles Ca|Sr|Ba rather than the pattern seen in the transition metals.

I further like the fact that in the set {Sc, Y, La and the rest of the lanthanides}, the elements are in atomic number order, whereas this is not the case for the set {Sc, Y, Lu and the rest of the lanthanides}.

I also like an argument I stumbled upon in the 4th edition of Shriver & Atkins, here, re ionic radius and its influence on chemical properties. Here, Sc|Y|La is a better fit than Sc|Y|Lu.

I do agree that Sc, Y, La and Ac are physically more like transition metals (and I’m OK, I think, with Jensen’s arguments in this regard); however it seems to me that---these days---chemical properties trump physical properties when it comes to organising the periodic table.

Hence, to my surprise, I now find myself thinking that the composition of group 3 appears to indeed be better regarded as Sc, Y, La and Ac. I also like the fact that in option B, La and Ac are lined up under group 3. And, most of all, there is nil ambiguity as to the composition of group 3. Sandbh (talk) 03:18, 7 November 2015 (UTC)

A bit strange that Sandbh now dives into the 14CeTh/14LaAc dispute without addressing this [the above] 15LaAc issue first. I'd expect arguments on why this third "variant" should be in there -- or not. Once this is solved, we can talk about the remaining choice. Note that in this I avoid the question about "it should be in xx-column form", as this complicates the topic with irrelevant arguments (cluttering). -DePiep (talk) 13:35, 7 November 2015 (UTC)
And yes, Sandbh, I think it was I who said that! Even physically, though, we have Ti, Zr and Hf being similar hard refractory metals with high melting points, while Sc, Y, and La show a clearer trend (decresing melting point). And unlike Ti, Zr, and Hf being similar in reactivity, we have a clear trend going down from Sc to Ac of increasing reactivity. My favourite comparison is to see how Be/Mg/Zn/Cd/Hg looks like a p-block group, while Be/Mg/Ca/Sr/Ba/(Ra) looks like an s-block group, looking at densities and melting points. If you consider Sc and Y to be more like transition metals than main-group elements, then Lu makes for nicer-looking trends; but as you say, Sc, Y, La, and Ac have a chemistry more like their s-block neighbours, so it makes sense to choose the trends for Sc/Y/La resembling those of Ca/Sr/Ba. Double sharp (talk) 16:21, 7 November 2015 (UTC
Hydrogen (diatomic nonmetal)
Helium (noble gas)
Lithium (alkali metal)
Beryllium (alkaline earth metal)
Boron (metalloid)
Carbon (polyatomic nonmetal)
Nitrogen (diatomic nonmetal)
Oxygen (diatomic nonmetal)
Fluorine (diatomic nonmetal)
Neon (noble gas)
Sodium (alkali metal)
Magnesium (alkaline earth metal)
Aluminium (post-transition metal)
Silicon (metalloid)
Phosphorus (polyatomic nonmetal)
Sulfur (polyatomic nonmetal)
Chlorine (diatomic nonmetal)
Argon (noble gas)
Potassium (alkali metal)
Calcium (alkaline earth metal)
Scandium (transition metal)
Titanium (transition metal)
Vanadium (transition metal)
Chromium (transition metal)
Manganese (transition metal)
Iron (transition metal)
Cobalt (transition metal)
Nickel (transition metal)
Copper (transition metal)
Zinc (transition metal)
Gallium (post-transition metal)
Germanium (metalloid)
Arsenic (metalloid)
Selenium (polyatomic nonmetal)
Bromine (diatomic nonmetal)
Krypton (noble gas)
Rubidium (alkali metal)
Strontium (alkaline earth metal)
Yttrium (transition metal)
Zirconium (transition metal)
Niobium (transition metal)
Molybdenum (transition metal)
Technetium (transition metal)
Ruthenium (transition metal)
Rhodium (transition metal)
Palladium (transition metal)
Silver (transition metal)
Cadmium (transition metal)
Indium (post-transition metal)
Tin (post-transition metal)
Antimony (metalloid)
Tellurium (metalloid)
Iodine (diatomic nonmetal)
Xenon (noble gas)
Caesium (alkali metal)
Barium (alkaline earth metal)
Lanthanum (lanthanide)
Cerium (lanthanide)
Praseodymium (lanthanide)
Neodymium (lanthanide)
Promethium (lanthanide)
Samarium (lanthanide)
Europium (lanthanide)
Gadolinium (lanthanide)
Terbium (lanthanide)
Dysprosium (lanthanide)
Holmium (lanthanide)
Erbium (lanthanide)
Thulium (lanthanide)
Ytterbium (lanthanide)
Lutetium (lanthanide)
Hafnium (transition metal)
Tantalum (transition metal)
Tungsten (transition metal)
Rhenium (transition metal)
Osmium (transition metal)
Iridium (transition metal)
Platinum (transition metal)
Gold (transition metal)
Mercury (transition metal)
Thallium (post-transition metal)
Lead (post-transition metal)
Bismuth (post-transition metal)
Polonium (post-transition metal)
Astatine (metalloid)
Radon (noble gas)
Francium (alkali metal)
Radium (alkaline earth metal)
Actinium (actinide)
Thorium (actinide)
Protactinium (actinide)
Uranium (actinide)
Neptunium (actinide)
Plutonium (actinide)
Americium (actinide)
Curium (actinide)
Berkelium (actinide)
Californium (actinide)
Einsteinium (actinide)
Fermium (actinide)
Mendelevium (actinide)
Nobelium (actinide)
Lawrencium (actinide)
Rutherfordium (transition metal)
Dubnium (transition metal)
Seaborgium (transition metal)
Bohrium (transition metal)
Hassium (transition metal)
Meitnerium (unknown chemical properties)
Darmstadtium (unknown chemical properties)
Roentgenium (unknown chemical properties)
Copernicium (transition metal)
Ununtrium (unknown chemical properties)
Flerovium (post-transition metal)
Ununpentium (unknown chemical properties)
Livermorium (unknown chemical properties)
Ununseptium (unknown chemical properties)
Ununoctium (unknown chemical properties)
(1) Be/Mg + Ca/Sr/Ba/(Ra) v. Zn/Cd/Hg

Be/Mg/Zn/Cd/Hg looks like a p-block group;
Be/Mg/Ca/Sr/Ba/(Ra) like an s-block group,
looking at densities and melting points.
Hydrogen (diatomic nonmetal)
Helium (noble gas)
Lithium (alkali metal)
Beryllium (alkaline earth metal)
Boron (metalloid)
Carbon (polyatomic nonmetal)
Nitrogen (diatomic nonmetal)
Oxygen (diatomic nonmetal)
Fluorine (diatomic nonmetal)
Neon (noble gas)
Sodium (alkali metal)
Magnesium (alkaline earth metal)
Aluminium (post-transition metal)
Silicon (metalloid)
Phosphorus (polyatomic nonmetal)
Sulfur (polyatomic nonmetal)
Chlorine (diatomic nonmetal)
Argon (noble gas)
Potassium (alkali metal)
Calcium (alkaline earth metal)
Scandium (transition metal)
Titanium (transition metal)
Vanadium (transition metal)
Chromium (transition metal)
Manganese (transition metal)
Iron (transition metal)
Cobalt (transition metal)
Nickel (transition metal)
Copper (transition metal)
Zinc (transition metal)
Gallium (post-transition metal)
Germanium (metalloid)
Arsenic (metalloid)
Selenium (polyatomic nonmetal)
Bromine (diatomic nonmetal)
Krypton (noble gas)
Rubidium (alkali metal)
Strontium (alkaline earth metal)
Yttrium (transition metal)
Zirconium (transition metal)
Niobium (transition metal)
Molybdenum (transition metal)
Technetium (transition metal)
Ruthenium (transition metal)
Rhodium (transition metal)
Palladium (transition metal)
Silver (transition metal)
Cadmium (transition metal)
Indium (post-transition metal)
Tin (post-transition metal)
Antimony (metalloid)
Tellurium (metalloid)
Iodine (diatomic nonmetal)
Xenon (noble gas)
Caesium (alkali metal)
Barium (alkaline earth metal)
Lanthanum (lanthanide)
Cerium (lanthanide)
Praseodymium (lanthanide)
Neodymium (lanthanide)
Promethium (lanthanide)
Samarium (lanthanide)
Europium (lanthanide)
Gadolinium (lanthanide)
Terbium (lanthanide)
Dysprosium (lanthanide)
Holmium (lanthanide)
Erbium (lanthanide)
Thulium (lanthanide)
Ytterbium (lanthanide)
Lutetium (lanthanide)
Hafnium (transition metal)
Tantalum (transition metal)
Tungsten (transition metal)
Rhenium (transition metal)
Osmium (transition metal)
Iridium (transition metal)
Platinum (transition metal)
Gold (transition metal)
Mercury (transition metal)
Thallium (post-transition metal)
Lead (post-transition metal)
Bismuth (post-transition metal)
Polonium (post-transition metal)
Astatine (metalloid)
Radon (noble gas)
Francium (alkali metal)
Radium (alkaline earth metal)
Actinium (actinide)
Thorium (actinide)
Protactinium (actinide)
Uranium (actinide)
Neptunium (actinide)
Plutonium (actinide)
Americium (actinide)
Curium (actinide)
Berkelium (actinide)
Californium (actinide)
Einsteinium (actinide)
Fermium (actinide)
Mendelevium (actinide)
Nobelium (actinide)
Lawrencium (actinide)
Rutherfordium (transition metal)
Dubnium (transition metal)
Seaborgium (transition metal)
Bohrium (transition metal)
Hassium (transition metal)
Meitnerium (unknown chemical properties)
Darmstadtium (unknown chemical properties)
Roentgenium (unknown chemical properties)
Copernicium (transition metal)
Ununtrium (unknown chemical properties)
Flerovium (post-transition metal)
Ununpentium (unknown chemical properties)
Livermorium (unknown chemical properties)
Ununseptium (unknown chemical properties)
Ununoctium (unknown chemical properties)
(2) Sc/Y/La/Ac v. Ti/Zr/Hf/Rf
Ti/Zr/Hf: similar hard refractory metals
with high melting points;
Sc/Y/La: Clearer trend (decreasing MP).
Unlike Ti/Zr/Hf being similar in reactivity;
ScAc Clear trend (increasing reactivity).
Hydrogen (diatomic nonmetal)
Helium (noble gas)
Lithium (alkali metal)
Beryllium (alkaline earth metal)
Boron (metalloid)
Carbon (polyatomic nonmetal)
Nitrogen (diatomic nonmetal)
Oxygen (diatomic nonmetal)
Fluorine (diatomic nonmetal)
Neon (noble gas)
Sodium (alkali metal)
Magnesium (alkaline earth metal)
Aluminium (post-transition metal)
Silicon (metalloid)
Phosphorus (polyatomic nonmetal)
Sulfur (polyatomic nonmetal)
Chlorine (diatomic nonmetal)
Argon (noble gas)
Potassium (alkali metal)
Calcium (alkaline earth metal)
Scandium (transition metal)
Titanium (transition metal)
Vanadium (transition metal)
Chromium (transition metal)
Manganese (transition metal)
Iron (transition metal)
Cobalt (transition metal)
Nickel (transition metal)
Copper (transition metal)
Zinc (transition metal)
Gallium (post-transition metal)
Germanium (metalloid)
Arsenic (metalloid)
Selenium (polyatomic nonmetal)
Bromine (diatomic nonmetal)
Krypton (noble gas)
Rubidium (alkali metal)
Strontium (alkaline earth metal)
Yttrium (transition metal)
Zirconium (transition metal)
Niobium (transition metal)
Molybdenum (transition metal)
Technetium (transition metal)
Ruthenium (transition metal)
Rhodium (transition metal)
Palladium (transition metal)
Silver (transition metal)
Cadmium (transition metal)
Indium (post-transition metal)
Tin (post-transition metal)
Antimony (metalloid)
Tellurium (metalloid)
Iodine (diatomic nonmetal)
Xenon (noble gas)
Caesium (alkali metal)
Barium (alkaline earth metal)
Lanthanum (lanthanide)
Cerium (lanthanide)
Praseodymium (lanthanide)
Neodymium (lanthanide)
Promethium (lanthanide)
Samarium (lanthanide)
Europium (lanthanide)
Gadolinium (lanthanide)
Terbium (lanthanide)
Dysprosium (lanthanide)
Holmium (lanthanide)
Erbium (lanthanide)
Thulium (lanthanide)
Ytterbium (lanthanide)
Lutetium (lanthanide)
Hafnium (transition metal)
Tantalum (transition metal)
Tungsten (transition metal)
Rhenium (transition metal)
Osmium (transition metal)
Iridium (transition metal)
Platinum (transition metal)
Gold (transition metal)
Mercury (transition metal)
Thallium (post-transition metal)
Lead (post-transition metal)
Bismuth (post-transition metal)
Polonium (post-transition metal)
Astatine (metalloid)
Radon (noble gas)
Francium (alkali metal)
Radium (alkaline earth metal)
Actinium (actinide)
Thorium (actinide)
Protactinium (actinide)
Uranium (actinide)
Neptunium (actinide)
Plutonium (actinide)
Americium (actinide)
Curium (actinide)
Berkelium (actinide)
Californium (actinide)
Einsteinium (actinide)
Fermium (actinide)
Mendelevium (actinide)
Nobelium (actinide)
Lawrencium (actinide)
Rutherfordium (transition metal)
Dubnium (transition metal)
Seaborgium (transition metal)
Bohrium (transition metal)
Hassium (transition metal)
Meitnerium (unknown chemical properties)
Darmstadtium (unknown chemical properties)
Roentgenium (unknown chemical properties)
Copernicium (transition metal)
Ununtrium (unknown chemical properties)
Flerovium (post-transition metal)
Ununpentium (unknown chemical properties)
Livermorium (unknown chemical properties)
Ununseptium (unknown chemical properties)
Ununoctium (unknown chemical properties)
(3) ... we perhaps would be better off
leaving a note that Sc and Y
are marginal transition metals,
like what we do for group 12
(treating them otherwise
leads to logical problems with
K, Rb, Cs, predicted E113 and Fl
I have added these micro PTs to help me follow DS's thought processes. Perhaps others might find them helpful.
(Side note: the above illustrations might have been improved if {{periodic table (micro)}} not only had |mark= but also |mark2= to provide two contrasting markings.)
YBG (talk) 22:17, 7 November 2015 (UTC)

Rare earths[edit]

---Extracted by me from the La|Ac and Lu|Lr section---Sandbh (talk) 07:19, 13 November 2015 (UTC)

Actually this is making me think of using rare-earth metals as a classification again, given how Cotton and Wilkinson group Sc and Y with the lanthanides and say "the properties of Y are extremely similar to, and those of Sc mainly like, those of the lanthanide elements proper, and quite different from those of the regular d-block elements.", corroborating what has been said above. But this may be too hard to explain, so we perhaps would be better off leaving a note that Sc and Y are marginal transition metals, like what we do for group 12 (treating them otherwise leads to logical problems with K, Rb, Cs, predicted E113 and Fl, etc.) [see Group 12 section, below, re logical problems]. Double sharp (talk) 16:21, 7 November 2015 (UTC)

Using "rare earth metals" (REM) as a category to encompass Sc, Y and the lanthanide elements (La to Lu) is a potentially neat and interesting option for carving up that part of the periodic table. But it means the lanthanides would not show as a separate colour category and I don't know if that would be a good or bad thing, or neither. I recall R8R Gtrs was not in favour of using a REM category since only older Russian professors apparently still use this term. Certainly, the general reader is more likely to have heard of the rare earths due to the amount of popular press coverage this term tends to attract ("China, a Rare Earths Giant, Set to Start Importing the Elements"; "Once scarce the world is suddenly awash in rare earths"; "Rare earth demand to increase 50% by 2020" etc). Sandbh (talk) 23:06, 11 November 2015 (UTC)
I wouldn't be too sad about losing lanthanides as a separate category, as Sc and Y are so similar to the lanthanides (and are often covered together). The press coverage can't hurt. But getting rid of one borderline group (group 3) and not the other (group 12) does bother me a little. Double sharp (talk) 06:56, 12 November 2015 (UTC)
re YBG "And yes, Sandbh, I think it was I ..." [in this section, above] About showing REMs as a separate category(-color). I do not doubt its scientific base and encyclopedic relevance. But this is off topic, because recoloring (i.e., using different categorisation-criteria) does not relate to the core question: what constitutes group 3 and how do we present that? Recoloring does not state or clarify anything with this. In general, groups and categories are per se not interchangeable (only once our PT says "group=category"!). Illustrative, we solved a lot by saying "halogens is a group name, but not a category name" (At). And remember, this is independent of the exact memberlist of any category, usually not definitive. Categories do not follow the periodic table structure (cols & rows), except for showing a definitive trend in periods (same colors are always adjacent in a period!). That said, this topic started by YBG may very well be a sound encyclopedic article or section. For sanity if the main topic discussion, I ask us all to keep this subtopic isolated & separate. Better not add another opion-multiplier that comes from the wrong preposition. -DePiep (talk) 05:59, 12 November 2015 (UTC)

re Double sharp: "using rare-earth metals as a classification again", Sandbh: "Using "rare earth metals" (REM) as a category ... is a potentially neat and interesting option for carving up that part of the periodic table". First of all, this issue is only a consequence for trying to write an 18-column PT. In 32-col format this does not exist. (I maintain that they should be interchangeable, not stating different things). Then, the REM-grouping was done this way by Seaborg 1946 (says Jensen 2008). It is still present in the "Sc/Y/*/**" presentation form used today. So I can be repetitive about this: 1. approach by category introduces an extra criteria, thereby adding complexity to the graph. Except for the tetris-like color effect, we should not try to add more 'explanation' (incidental at that, b/c not done in the main group) to the basic PT. Just as we don't use natural occurrence (cell border) ofor state (Z font color) to reorganise the PT graph. 2. It has this huge, huge disadvantage that it is stating group 3 wrongly. 3. I see no advantages, except for elegance of grouped colors. That's not enough. -DePiep (talk) 08:15, 13 November 2015 (UTC)
Huh? In a 32-column Sc/Y/La/Ac PT, it is quite easy to use REM as a category: it would encompass Sc, Y, La, and Ce-Lu, without disturbing the alignment of La below Y. Maybe I am misunderstanding you, but using REM would require no more than taking Template:Periodic table/Sc-Y-La-Ac/sandbox and colouring Sc and Y pink as well (and changing the legend from "Ln" to "REM"). This doesn't imply that we're putting all the lanthanides and actinides into group 3; the fact that we currently colour all the actinides the same way in a Sc/Y/Lu/Lr periodic table (where only Lr is in group 3) does not suddenly make Ac-No group 3 elements too. The only thing Sc/Y/La/Ac with REM would change is that La and Ac would be the group 3 elements of period 6 and 7, not Lu and Lr. There would still be only one element in each position below Y. Now [Sandbh] for group 3: do you think we should take Sc and Y out of the TMs and lump them with the lanthanides into REMs? I'm leaning towards yes, as they are also quite borderline as TMs. Double sharp (talk) 15:20, 13 November 2015 (UTC)
re "it is quite easy to use REM as a category" -what? WHAT? are we drawing the PT by reason of easyness? -DePiep (talk) 21:45, 13 November 2015 (UTC)
What we should be doing according to Wikipedia policy is following reliable sources. So how about someone doing a survey of recent (say 0-10 years) textbooks and review articles to see how many use Rare earths as a category? Dirac66 (talk) 22:20, 13 November 2015 (UTC)
REM is an existing category, RS & no dispute. But to use introduce it as a graph organiser for the 18-col only, I oppose. (graphs please) -DePiep (talk)
I don't think it should be used in 18-col only. If we end up using it, it ought to be in both 18-col and 32-col. Double sharp (talk) 11:11, 14 November 2015 (UTC)
So, Double sharp, you really maintain that we should drop Ln and An categories completely, and in the process declare Sc and Y non-TM because they are --undisputed-- REM's? That is bad categorization because it mixes up category schemes (as if we'd recolor the PT by occurrence for only some elements). But could you be clear: do you propose & support & push each and every PT to be "Sc/Y/*/**" still? (Would save me a lot of time, not having to repeat my arguments). -DePiep (talk) 23:59, 14 November 2015 (UTC)
I think you've misunderstood me. In such a categorisation, An would stay, and Sc+Y+Ln would together form a larger REM category. There are valid scientific reasons to take Sc and Y out of the transition metals: see Periodic table#Groups included in the transition metals. And no, I am not now proposing Sc/Y/*/**; instead I'm supporting Sc/Y/La/Ac with Sandbh. Double sharp (talk) 04:29, 15 November 2015 (UTC)
Admit, I missed that An are not in REM. So that leaves Ln being invisible in your PT proposal. Next: how does your proposal help the topic of group 3 clarification/description? Whether we switch to "REM" of keep "Ln" (-colors), the core issue is the same and unanswered. And I add that the group 12-issue you introduce is not related. Instead, it adds to confusion by introducing an unrelated issue. Sorry, group 12 = wrong place, let's forget this here -DePiep (talk) 17:48, 15 November 2015 (UTC)

I went ahead and made a graph of usage of the phrases "lanthanide," "lanthanoid," and "rare earrh element" (2000–2008, since the service doesn't scan in later books). Lanthanides undisputably won.

Well, in compounds we get terms like "rare earth mineral" and "rare earth magnet", so it may be fairer to use just "rare earth", in which case "rare earth" actually wins. Double sharp (talk) 11:13, 14 November 2015 (UTC)
Hmm. Indeed. .0000256% vs. .0000155% is a larger share.
However, this can't be called an indisputable victory: I have the feeling "lanthanide" is better in the context of chemistry, and "rare earth" is now a geology-related word. And both terms you suggest are not chemistry really: they're geology and electromagnetism, correspondingly (which was how I came to this thinking). The whole term "rare earth" is geology. We are concerned with chemistry for now, so I would pick the good old "lanthanide" for element categorization.--R8R (talk) 12:51, 14 November 2015 (UTC)

Maybe I was too emotional against the REM as a term, but I still stand against it being used here. Not just because it is used so infrequently—we have overlapping categories here, we have to make our editorial choices—it removes scandium and yttrium from the TM category, a rare approach to the problem these days as well. Two elements carved from TMs is sort of the least bad solution we have, in part because it is so unnatural. Four is a certain structural decision, and a really questionable one. (I would say d electron being involved in basically every group 3 compound is more than enough to call them TMs.)--R8R (talk) 07:13, 14 November 2015 (UTC)

I don't believe nor think this occurrence statitics says anything about how we should present our PT. What's wrong with thinking & arguments? -DePiep (talk) 20:02, 14 November 2015 (UTC)

"Common" or "standard form" of PT[edit]

---Extracted by me from the Let us proceed with Sc|Y|La|Ac section--- Sandbh (talk) 06:54, 25 November 2015 (UTC)

I add: So I prefer not to use common form wording. For similar reasons, let's not use standard form. This "long form/extremely long form/standard/medium/short" relative wording is historical, and tied to the development of the PT (ie, discovery of more PT structure, and sequential at that). But today they are confusing and even mixing up. Jensen and Scerri avoid these terms. And again, what we are talking about, there are multiple forms in writing group 3 so there is no single standard. Yes we at enwiki, on this very page, aim to use this as standard PT (ie our standard), but that word should not be used for the reader. -DePiep (talk) 13:25, 21 November 2015 (UTC)

We can certainly look at this and see what the sources say/ask Scerri. Sandbh (talk) 23:58, 21 November 2015 (UTC)
thanks. -DePiep (talk) 22:48, 26 November 2015 (UTC)
  • Today [4], Sandbh uses "common" to push some PT presentation. Even worse: Sandbh writes: "Medium-long form of periodic table" File:14LaAc periodic table IIa.jpg etc. etc. Bad habit. -DePiep (talk) 23:43, 8 December 2015 (UTC)
    I've made (and then reverted) what I think is a middle ground that saves horizontal real estate while graphically showing that footnote elements belong to periods 6+7 and do not belong to any group. Comments? YBG (talk) 01:26, 9 December 2015 (UTC)
Whatever. As long as Sandbh keeps pushing that mid-20th century habit, I won't win the word. The only thing I know: both me and Mendeleev did not refer to 'common' or 'longer than yesterday'. -DePiep (talk) 01:36, 9 December 2015 (UTC)
Not sure what you mean by "win the word". YBG (talk) 01:45, 9 December 2015 (UTC)
Sandbh is better in PT science than I am. Still, hammering on some 'common' PT form though is bad (I dare saying: I can beat Sandbh). I trust Sandbh as a top PT editor, but he needs to accept criticism in his 1945 PT side-views. -DePiep (talk) 02:21, 9 December 2015 (UTC)

A 'medium length' PT?[edit]

Today, Sandbh re-introduces PT descriptions like 'Medium-long form of periodic table' here. This description is to be deprecated for multiple reasons. First: it is unclear. Second: it is relative to some previous PT. Third: it is out of date. Sooo 19th-century. (Why not use col-count, as Scerri does?). Really, I am confused by Sandbh prefering to re-use these terms. To be clear: I state that this should never be our basic PT descriptive form.-DePiep (talk) 21:33, 9 December 2015 (UTC)

This is not something Sandbh came up with. And it is still in use today. If I thought it was wrong in some way, I would say that aloud; but I don't. The idea of using column count is worth discussing, but for the sake of objectivity, it doesn't make a clear well-established nomenclature unclear. Also, old != bad. The relativity argument is the most prominent one here; yes, it does seem reasonable enough to use the "xx-column" notation, as it is pretty self-descriptive and does not rely on anything. Still, let's keep calm about the issue.
(Also, I feel the two paras you wrote don't match up; could you please clarify that to me? In this case, are you bothered by the terms used or by the 18-column layout of the PT?)
One thing that bothers me much more is that group 2 shares its color with the transition metals.--R8R (talk) 01:06, 10 December 2015 (UTC)
Re notation: I agree, using "xx-column" notation is preferable because it is self-describing.
Re colors: When I look carefully at the colors in the jpg used in Sandbh's sandbox, File:14LaAc periodic table IIa.jpg, it appears that the color in group 2 and in transition metals are different, but much too similar. I presume this is an artifact of the way the jpg was created, and that eventually this jpg will be replaced by one with the colors used in the svg currently used, File:Periodic table (polyatomic).svg
YBG (talk) 14:19, 10 December 2015 (UTC)
re R8R: "And it is still in use today" -- do you really mean to say that is an argument? Any source for that statement?
re YBG: "colors" --???
Really, I claim that the wording is for historical sections only. Someone serious? -DePiep (talk) 22:38, 11 December 2015 (UTC)
The colour of the group 2 metals is my bad. Will fix. Sandbh (talk) 03:51, 12 December 2015 (UTC)
You uploaded to IIb: File:14LaAc periodic table IIb.jpg. A mistake? -DePiep (talk) 17:56, 13 December 2015 (UTC)

Priorities & preferences[edit]

Today, the sandbox says:

==Different periodic tables==
===18-column forms===
The three common 18-column forms of the periodic table differ in which elements appear in the group 3 column.
====Type I====
====Type II====
====Type III====
===Other arrangements===
(Benfey's, Janet's Left Step)
  • I note that the 32-column graph is elimintated completely.
    As it ought to, since it is nowhere near as common as any of the 18-column formats. Look, I get that you don't like the 18-column form for splitting out part of periods 6 and 7 unnecessarily. The fact remains that it is the most common form to be found now. If that changes, we can talk about changing the accent. But right now mentioning 32-column here puts too much weight on it. Double sharp (talk) 21:31, 24 December 2015 (UTC)
"as it ought to": OR. And remember that truly the most "common" form is the crippled wrong 'Type III' version (32 elements in group 3). -DePiep (talk) 00:20, 25 December 2015 (UTC)
Err, no. Type I is the most common type as per the citation. Sandbh (talk) 00:41, 25 December 2015 (UTC)
See also the postscript to argument 28, above. Sandbh (talk) 00:43, 25 December 2015 (UTC)
Alas, I am at least partly responsible for the surprisingness lack of a 32-column PT in this section. I changed the header from "Common forms" to "18-column forms" on the idea that I didn't think "common" was a good cover term to describe all three types (I/II/III). When I made this change, I was reading the level ==2== header "Different periodic tables" as meaning "Here is a description of periodic tables different from the 32-column one discussed above." But I now realize that an equally valid understanding of "Different periodic tables" is "Here is a description of all of the different kinds of periodic tables". With the former reading, there is no need for any mention of the 32-column form, but with the latter reading, it is certainly needed. So what to do? Maybe change the level ===3=== header back? Or describe the 32-column table in a lede paragraph before the first level ===3=== section? Or add a new level ===3=== section describing the 32-column version? Something should be done, but I'm not sure what. YBG (talk) 08:39, 25 December 2015 (UTC)
I'll have a closer look at this one. Sandbh (talk) 09:51, 25 December 2015 (UTC)
  • The composition of group 3 is tied to the graph form (eg, see the opening line as quoted here)
    But group 3 is exactly the one group where the formats differ. All three choose to keep the elements in order of increasing atomic number. The difference is: shall we put La under Y, Lu under Y, or compromise and appear to put all 15 lanthanides there? So I do not see a problem with this. Double sharp (talk) 21:31, 24 December 2015 (UTC)
Let me rephrase: it is written saying (witout an other option) that the 18-col form is the one that describes group 3. This is incorrect in two ppoints: A. Type III is nonsense (although, the nonsense is hard of at all to repeat in 32-col format), and the same group3 variants can also be written in 32-col format. -DePiep (talk) 00:20, 25 December 2015 (UTC)
This is why I added the word 'column' at the end of the sentence, in an attempt to emphasize the graph rather than group membership. YBG (talk) 08:39, 25 December 2015 (UTC)
  • Type III does not represent a different group 3, but is wrong, misleading and wrongfooting.
    Yes, I think we all know you don't like it. We are not here to change the fact that it is used seriously by many people, including IUPAC. It simply arises from a wish to emphasise the chemical similarities among the lanthanides by keeping them together. This is rather wrong-headed as if these are truly group 3 elements, it does not make sense to include Ce which can easily get to the +4 oxidation state, or the actinides from Th to Cf which have accessible states from +4 to +8. It also violates the principle of putting one element in one space. So we can criticise it. But for all that, it is still used, and we have to describe its peculiarities, such as treating the f-block as degenerate members of the d-block, even as we criticise it as Jensen does. Double sharp (talk) 21:31, 24 December 2015 (UTC)
It is not about 'I don't like it' (no need to distort my point). It is: there is no source for that. Not one single Type III PT (source) is accompanied by text that states & convinces that this describes group 3. -DePiep (talk) 00:20, 25 December 2015 (UTC)
Observation: Fine (1978, pp. 702–707) has a six page section called THE 32 ELEMENTS IN III B [his formatting, not mine] which includes a Type III periodic table with the 32 elements in question shaded.
  • Fine LW 1978, Chemistry, 2nd ed., The Williams & Wilkins Company, Baltimore
Sandbh (talk) 07:09, 25 December 2015 (UTC)

To be sure, I'll remove the FA star from the sandbox. -DePiep (talk) 16:52, 24 December 2015 (UTC)

Group 3 composition debate[edit]

---This where I'll list the arguments for -Lu-Lr and arguments that I'm aware of for -La-Ac, as flagged in my 02:02 of 25 Nov re Let us proceed with Sc|Y|La|Ac --- Sandbh (talk) 10:37, 26 November 2015 (UTC)

Arguments 1–10[edit]

REFERENCE: Jensen WB 1982, 'The Positions of Lanthanum (Actinium] and Lutetium (Lawrencium] in the Periodic Table,' Journal of Chemical Education, vol. 59, no. 8, pp. 634–636, doi:10.1021/ed059p634


1. Separation groups: "For quite some time it has been known that Y, and, to a lesser degree, Sc are closer in their chemical properties to Lu and the other heavy rare earths than they are to lanthanum (1,2)…

(1) In the classical chemical methods for separating the rare earths Sc, Y, and Lu occur together in what is called the Y group, whereas La and Ac occur together in what is called the Ce group. See, for example, Levi, S.I.,"The Rare Earths," Longmans, Green and Co, London, 1915, Chap. IX.
(2) Moeller, T., "The Rare Earths", Spedding, F. H., and Daane, A. H., (Editors), Wiley, New York. 1961. Chap. 2."

Analysis: Sc, Y and Lu do occur in the so called yttrium group; La and Ac do occur in the "cerium" group. This does not imply anything particularly significant. It is simply a reflection of the increasing basicity of these elements as atomic radius increases. Thus, for another example, among the alkaline earth metals, Mg (less basic) belongs in the "soluble group" and Ca, Sr and Ba (more basic) occur in the "ammonium carbonate group" (Moeller et al. 1989, pp. 955–956, 958). Arguing that Lu should go under Y simply because they occur in the same chemical separation group ignores periodic trends.

"ignores periodic trends": the latter points cause some doubt on this. This could be justified if specified, but now it's too vague as a statement, I think.--R8R (talk) 17:45, 27 November 2015 (UTC)

Moeller (1961, p. 10) gives Sc, Y, La and Ac as the first members of the four "transition" series. I couldn't find anything supporting argument J1 aside from a comment (p. 21) that the chemisty of Sc differs significantly from that of the "rare-earth elements" (La–Lu).

Isn't that to be expected? It is a lot smaller! Double sharp (talk) 15:04, 29 November 2015 (UTC)
  • Moeller et al. 1989, Chemistry with Inorganic Qualitiative Analysis, 3rd ed., Harcourt Brace Jovanovich Publishers, San Diego, p. 955–956, 958

2. Older tables: [fits on to end of J1 quote] "…and on this basis alone a number of chemists in the 1920's and 1930's assigned Lu rather than La to group IIIB (3).

(3) See, for example, Shemyakin, F. M., Zh. Obshch. Khim., 2, 62 (1932) and Bury, C. R., J. Amer. Chem. Soc., 43, 1602 (1921). Further examples can be found in reference (14): Mazurs, E. G., "Graphic Representations of the Periodic System During One Hundred Years" Univ. Alabama Press, University, Alabama, 1974"

Analysis: Bury shows Sc-Y-Lu on the basis of chemical properties, but does not elaborate which properties he had in mind. He draws an analogy to Be and Mg resembling Zn better than Ca. Mazurs has plenty of tables with Lu in group 3 as well as -La-Ac tables pre-dating, dated during, and post-dating the 20s and 30s. You have to be careful with Mazurs as some of his renderings of tables appearing in the literature are dodgy but there is no doubting that both kinds of tables have been around for quite a while. @R8R Gtrs: When you get time could you look up Shemyakin, and see if there is anything in there worth noting? Sandbh (talk) 01:32, 27 November 2015 (UTC)

I've checked the source. It's a short article about placing lanthanides into the 8-group table (since as you may remember, Russian chemists are still keen on the Mendeleev's original format, not to say back in the 30s). It is mostly interesting from a historical perspective; it does not specifically discuss the group 3 problem, although it does very briefly mention "close similarities between Jt and Lu," without adding a word about it. The author suggests one would use, apart from the regular group 6a/6b notation, also 6a2/6a1 for lanthanides (Lu-Eu and Gd-Yb, accordingly), and nc and nd for what we would call group 9 and group 10.
It won't help us much. :( --R8R (talk) 17:45, 27 November 2015 (UTC)

3. Electron configurations: Jensen says that La and Lu have equal claims to the position under Y, based on their differentiating d electrons. He then asserts that nobody doubts Th is an f block element with an irregular electron configuration i.e. [Rn]6d27s2 and that this therefore "strongly" supports treating La i.e. [Xe]5d16s2 and Ac i.e. [Rn] 6d17s2 as f block elements with irregular electron configurations. Thus Lu i.e. Xe4f145d16s2 and Lr i.e. Rn5f147s27p1 fit under Y, with the result that each element in periods 6 and 7 of the d block has either a completed 4f14 or 5f14 shell.

Analysis: I don't agree that this is a strong argument. It seems to me to be only a "tipping point" argument. That is to say, if the merits of -La-Ac and -Lu-Lr are otherwise similar in terms of which one is placed under Y then, of course, -Lu-Lr would be the one given this would result in completed 4f14 or 5f14 shells across periods 6 and 7 of the d block. Sandbh (talk) 05:14, 27 November 2015 (UTC)

Postscript: The Russian authors, Landau and Lifshitz (1977, p. 273–274) write: "The filling up of the 3d, 4d, and 5d shells…[has] a characteristic feature of [of] "competition" between the s and d states…The filling up of the 4 fshell also occurs in a slightly irregular manner characterized by the competition between 4f, 5d, and 6s states…In books of chemistry, lutetium is usually placed in the rare-earth elements. This, however, is incorrect, since the 4f shell is complete in lutetium; it must therefore be place in the the platinum group…"[They refer to Sc–Ni as the Iron group; Y–Pd as the Palladium group; and Lu–Pt as the Platinum group]. They first wrote these words in 1956, which Scerri refers to as "one of the oldest categorical statements in favor of Sc Y Lu Lr" (pers. comm). I like the 'competition' metaphor and the simplicity of their assertion.
  • Landau LD & Lifshitz EM 1977, Quantum Mechanics (Non-relativistic Theory), 3rd ed., Pergamon Press, Oxford
Sandbh (talk) 10:42, 14 December 2015 (UTC)

4. Ionization potentials: Citing the Russian chemist Chistyakov, Jensen argues that the trend in the sum of the first two ionization potentials going down Sc-Y-Lu is similar to that occurring in groups 4 to 8, and unlike that of Sc-Y-La.

Analysis: True, but not the full story, and of questionable relevancy. Why the sum of the first two ionization potentials? In any event, the trend in this case (for Sc-Y-La) is similar to the trend seen in the group 2 and 1 metals, -Ca-Sr-Ba- and -K-Rb-Cs-. As well, the trend in the sum of the first three ionization energies for Sc-Y-La is a better fit with the trend occurring in groups 4 to 8, than is the case for Sc-Y-Lu.

Postscript: The Russian source (English translation) is: Chistyakov VM 1968, "Biron's secondary periodicity of the side d-subgroups of Mendeleev's short table", Journal of General Chemistry of the USSR, vol. 38, no. 2, pp. 213–214. The author compares the atomic radii and sum of the first two ionization energies (IE) for groups 3–8 and 11-12 and finds that Sc-Y-Lu is a better fit than -Y-La. He says this is probably due to the impact of the lanthanide contraction on the periodi 6 transition metals. For Group 11 he only uses the first IE's. In his conclusion he writes, "The radii of the free atoms, recently calculated from Dirac's equation, and the sums of the first two ionization potentials, i.e., parameters of the external ns-electrons of the d-elements, are periodic functions of the atomic numbers in each side d-subgroup." I thought his reference to "ns-electrons" explained why he was using the sum of the first two IE's rather than the first three (since it was group 3 that was the anomaly) but this doesn't work since, for example, Nb, Cr and Mo only have one s electron, so he isn't really comparing like with like as, to get the sum of the first two IE's for these metals he has to include one d electron each. And comparing the first two IE's of the group 11 metals fails to produce the pattern seen in the other groups. Explains why Jensen left out the group 11 and 12 metals in his article. Sandbh (talk) 10:21, 14 December 2015 (UTC)
@R8R Gtrs: Could you look up this reference(?): Shemyakin FM, Zh. Obschch. Kim., 2, 62 (1932). Chistyakov mentions this source as one consideration which requires -Y-Lu rather than -Y-La, but does not clearly elaborate. It may be something to do with atomic spectra judging by some of the other references he cites, but as he doesn't list the article title there's no way of knowing without checking it out. Thank you. Sandbh (talk) 01:51, 15 December 2015 (UTC)


I looked it up some time ago; see my re under point 2. The article is titled, "To the question of including the rare earths"; here's a photo of the table described in the article: http://s4.postimg.org/n22tahdbx/image.jpg --R8R (talk) 19:10, 15 December 2015 (UTC)
Picture is great. I see that he lists what we call group 3 as Sc-Y-Lu-Ac and that he then lists La to Eu and Gd to Yb as two separate but parallel group 3 subgroups. Pretty good effort for 1932, particularly for an 8-column table. Sandbh (talk) 10:10, 16 December 2015 (UTC)

5. Atomic radii: Citing Chistyakov, Jensen argues that the trend in atomic radii going down Sc-Y-Lu is similar to that occurring in groups 4 to 8, and unlike that of Sc-Y-La.

Analysis: True, but not the full story. The trend in atomic radii going down Sc-Y-La is instead similar to the trend seen in the group 2 and 1 metals, -Ca-Sr-Ba- and -K-Rb-Cs-.

6. Ionic radii: Jensen argues that comparisons in periodic trends favour Sc-Y-Lu.

Analysis: No data nor a reference is given. In contrast, Shriver and Atkins in their 4th edition, here, discuss ionic radius and its influence on chemical properties. Based on a data comparison, they say Sc-Y-La is a better fit than Sc-Y-Lu. Sandbh (talk) 10:01, 27 November 2015 (UTC)

7. Redox potentials: Jensen argues that comparisons in periodic trends favour Sc-Y-Lu.

Analysis: No data nor a reference is given. I had a look at the NIST standard electrode potential data here but wasn't able to discern any meaningful differences in periodic trends between Sc-Y-La and Sc-Y-Lu. Sandbh (talk) 10:41, 27 November 2015 (UTC)

8. Electronegativities: Jensen argues that comparisons in periodic trends for Allred-Rochow electronegativity favour Sc-Y-Lu.

Analysis: True, but not the full story. The trend in going down Sc-Y-La is instead similar to the trend seen in the group 2 and 1 metals. Sandbh (talk) 11:15, 27 November 2015 (UTC)

9. Melting points: Jensen argues that the trend in melting points going down Sc-Y-Lu is a better fit with groups 4 to 10, than is the case with Sc-Y-La.

Analysis: True, but not the full story. The trend in melting points going down Sc-Y-La is instead similar to the trend seen in the group 2 and 1 metals, -Ca-Sr-Ba- and -K-Rb-Cs-.

10. Crystal structures (elements): Jensen contends that the most compelling evidence for Sc-Y-Lu comes from the physicists. He says that, as a first example, the crystalline structures for Sc, Y, Lu are all hexagonal close packed (HCP) whereas that of La is double hexagonal close packed.

Analysis: The example given is true, but its relevance is questionable. For example, the structures of the group two metals Be, Mg, Ca, Sr, Ba, and Ra are HCP; HCP; face centred cubic; face centred cubic; body centred cubic; and body centred cubic. Groups 7, 8, 9, and 10 also show inconsistencies in crystalline structures. Sandbh (talk) 23:42, 27 November 2015 (UTC)

One should also note, however, that in groups 7-9 we have a first-row anomaly between the 3d metal and the heavier two members. Sc-Y-La-Ac has the heavier elements be inconsistent. Double sharp (talk) 15:04, 29 November 2015 (UTC)

Arguments 11–20[edit]

11. Crystal structures (oxides, chlorides, various intermetallics): Jensen says that the crystalline structures of the oxides X2O3 for Sc, Y and Lu are the same whereas that of La is different. The same pattern occurs with the chlorides -Cl3 and various intermetallic compounds.

Analysis: True, but not relevant at least for oxides and chlorides. Different structures for homologous ionic compounds are unremarkable. Consider, for example, NaCl vs. CsCl ("and this is in the alkali metals, the model example of great group trends"---to quote Double sharp). NaCl has the sodium chloride (rocksalt) structure; CsCl has a different (primitive) cubic structure, as shared with caesium bromide and caesium iodide, and many binary metallic alloys. When both ions are similar in size (Cs+ ionic radius 174 pm for this coordination number, Cl 181 pm) the CsCl structure is adopted, when they are different (Na+ ionic radius 102 pm, Cl 181 pm) the sodium chloride structure is adopted. Sandbh (talk) 03:31, 28 November 2015 (UTC)

The reference to intermetallic compounds is from Hamilton (1952) who says, "There are at least several intermetallic compounds where the compound with La has a different crystal structure from the corresponding compounds with Sc, Y, and Lu." Sandbh (talk) 04:32, 28 November 2015 (UTC)

  • Hamilton DC 1965, 'Position of Lanthanum in the Periodic Table', American Journal of Physics, vol. 33, pp. 637–640
Well, slap me with a wet fish and blow me down. Hamilton shows a periodic table extract (groups 1 to 11, plus footnoted Ln and An, showing Ce, Pr…Lu; and Th, Pa…Lw) with a split d block (the gap is between groups 3 and 4) and, get this, he says this is, "the periodic table as it is usually presented."(!) :) Now, back to our scheduled presentation. Sandbh (talk) 04:53, 28 November 2015 (UTC)

Yes, this shows again that everybody should be very careful claiming something like "this is the common way of presenting the PT". Must add, such a wrong statement made in 1965 is more forgiveable then one done 50 years later. In general, it is correcter to say that most PT drawings are wrong or ambiguous, even in RS. -DePiep (talk) 15:57, 16 December 2015 (UTC)

12. Excited state spectra: Citing Hamilton, Jensen says that the atomic spectra for Sc, Y, and Lu differ from La, in that for La, "excited energy levels have been observed which can be attributed to an electron in an f orbit" (Hamilton 1965 p. 637) whereas this is not the case for Sc, Y, or Lu thereby indicating, "that the 4 f wave function in La differs from the 4 f wave function in Sc and Y or the 5 f wave function in Lu; this causes the various line strengths to be different."

Analysis: OK, but incomplete. I have no primary grounds to query Hamilton's discussion about Sc, Y, Lu and La. However, I am troubled by Myers (1997, p. 201–202) who says that, "Normally we define transition metals as only those containing incompletely filled d shells but the empty d bands in the alkaline earths (Ca, Sr and Ba) and the filled d bands in Cu, Ag, and Au lie sufficiently close to the Fermi energy to produce significant effects, Because of this, the former group may be termed incipient transition metals, whereas the latter are immediate post-transion metals." So, I tend to find Hamilton inconclusive as he does not say anything about the spectra of Ca, Sr, and Ba. Sandbh (talk) 05:33, 28 November 2015 (UTC)

  • Myers HP 1997, Introductory Solid State Physics, 2nd ed., CRC Press, Boca Raton, Florida

Sandbh (talk) 05:33, 28 November 2015 (UTC)

Postscript: From Russell & Megger (1932, p. 625): "All the available data (wave-length measurements and intensity estimates, temperature classes, Zeeman effects) on the lanthanum lines have been correlated and interpreted in an analysis of the successive optical spectra. The total number of lines classified is 540 in the La I spectrum, 728 in the La II spectrum, and 10 on the La III spectrum.
Series-forming terms have been identified in each spectrum and from these the ionization potentials of 5.59 volts for neutral La atoms, 11.38 volts for La+ atoms and 19.1 volts for La++ have been deduced.
Lanthanum is a chemical analogue of scandium and yttrium but, although the corresponding spectra are strikingly similar, some interesting differences are noted...the... s2d configuration describes the normal state of the neutral atoms in each case...In addition, the first two spectra of La exhibit a large number of (odd) middle-set terms ascribed to the binding of an f electron."
Sandbh (talk) 21:27, 14 December 2015 (UTC)

13. Superconductivity: Jensen says the Sc, Y, and Lu are not capable of superconductivity in bulk form at normal pressure whereas La is.

Analysis: Incorrect. Lu is capable of such conductivity. See here. Sandbh (talk) 03:56, 28 November 2015 (UTC)

14. Conduction band structures. Citing Merz and Ulmer (1967) Jensen says that Sc, Y and Lu have conductivity bands with a d block like structure whereas La does not.

Analysis: Correct.

  • Merz H & Ulmer K 1967, 'Position of Lanthanum and Lutetium in the Periodic Table', Physics Letters,, vol. 26A, no. 1, pp. 6–7

Sandbh (talk) 05:52, 28 November 2015 (UTC)


REFERENCE: Jensen WB 2015, 'The Positions of Lanthanum (Actinium) and Lutetium (Lawrencium) in the Periodic Table: An Update,' Foundations of Chemistry, vol. 17, no. 1, pp. 23-31, doi:10.1007/s10698-015-9216-1


15. Dimer spectroscopy: Jensen cites Fang et al. (2000) who, in discussing the spectra of Sc, Y La and Lu X2 dimers, and those of some other period 6 transition metals, conclude that lutetium is more like the other transition metals and is therefore a better fit under Y than is the case for La.

Analysis: Correct, but the story is incomplete as Fang et al. do not say anything about the spectra of the group 2 or 1 metals.

  • Fang L, Chen X, Shen X & Lombardi JR 2000, 'Raman and absorption spectrum of mass-selected lutetium dimers in argon matrices,' Journal of Chemical Physics, vol. 113, no. 22, pp. 10202–10206, doi:10.1063/1.1322635

Sandbh (talk) 23:44, 28 November 2015 (UTC)

16. Relativistic contraction of 6s shell: Jensen further cites Fang et al. who note that the relativistic contraction of the 6s shell falls on the same trend line as that applying to the periodic 6 transition metals Hf to Ir whereas the contraction for La is more consistent with the trend line for Ce to Yb.

Analysis: Correct at face value, but not the full story since Fang et al. do not extend their trend line into the period 6 metals Ba and Cs. Looking up the applicable data tables cited by Fang et al. (Desclaux 1973) shows that La falls on the Ba and Cs trendline.

  • Desclaux JP 1973, 'Relativistic Dirac-Fock Expectation Values for Atoms with Z = 1 to 120, Atomic Data and Nuclear Data Tables, vol. 12, pp. 311–406 (370–371), doi:10.1016/0092-640X(73)90020-X

Sandbh (talk) 00:22, 29 November 2015 (UTC)

17. Aluminide dimers: Jensen cites Ouyang et al. (2008) who note that the "AlLa dimer has a different chemical bond compared with its congeners AlSc, AlY and AlLu. This discrepancy raises the question as to whether it would be more suitable to replace La with Lu in the periodic table."

Analysis : Fine, but not the whole story as, unfortunately (once again) the authors fail to say anything about Ba and Cs.

You know, the fact that such arguments are raised speaks volumes about how close La and Lu are. Does anyone ever complain about the bonding differences in the chalcogen dioxides?! This thus strikes me as a lame argument for placing elements in the periodic table when presented alone. Double sharp (talk) 15:04, 29 November 2015 (UTC)
  • Ouyang Y, Wang J, Hou Y, Zhong X, Du Y and Feng Y 2008, 'First principle study of AlX (X=3d, 4d, 5d elements and Lu) dimer', Journal of Chemical Physics,, vol. 128, no. 7, pp. 074305-1–074305-6, doi:10.1063/1.2831506

Sandbh (talk) 10:54, 29 November 2015 (UTC)

18. Heat of vapourization: Jensen refers to "trends in the [heat] of vaporization for the alternative group sequences Sc-Y-La versus Sc-Y-Lu" and goes on to say, "the latter, rather than the former, corresponds most closely to the group tends observed for this property for the other elements in the early part of the d block."

Analysis: Unclear. The values for La (402 kJ/mol) and Lu (415) are quite close and in comparing these to values for other nearby elements I wasn’t able to discern anything favouring the placement of either La or Lu under Y. Sandbh (talk) 00:56, 29 November 2015 (UTC)


REFERENCE: Scerri E 2012, 'Mendeleev's Periodic Table Is Finally Completed and What To Do about Group 3?', Chemistry International, vol. 34, no. 4; Scerri 2015, 'Five ideas in chemical education that must die - part five', Education in Chemistry blog


19. Split d block: Scerri supports –Lu-Lr on the grounds that, essentially, -La-Ac results in a split d-block, with a gap between groups 3 (Sc-Y-La-Ac) and 4 (Ti-Zr-Hf-Rf) caused by the insertion of the lanthanides (Ce-Lu) and actinides (Th-Lr) between La and Hf, and between Ac and Rf.

Analysis : Scerri notes that, "some textbook authors have taken –Lu-Lr up, but the majority seem reluctant." Indeed, La-Ac is still the most common form although very few authors show it with a split d block. The vast majority use alternative graphical solutions. For example, the Sargent-Welch table shows a single star above the 'a' in La and a double star above the 'c' in Ac, to denote the footnoted lanthanides and actindes. Sandbh (talk) 10:24, 29 November 2015 (UTC)

Furthermore, what's wrong with a split d-block? We have a split s-block in most tables since He (and sometimes H) are graphically disconnected from groups 1 and 2. Double sharp (talk) 15:04, 29 November 2015 (UTC)
The word "graphically" is the answer. Even as helium or even hydrogen float away, they still remain s block members, just for obvious reasons very different than the others. Splitting the d block means there is a d block group, then 14 f block columns, and then the d block again. Nothing like that is present in the PT elsewhere. (As I argued above, this does not unquestionably state the d block just can't be split, but this does reduce the likeliness of such a split if we assume we don't actually know the correct answer.)--R8R (talk) 09:06, 30 November 2015 (UTC)
Actually, I think introducing the word "graphically" is the problem. One can not reason: 'that's a unique graphic thing in the PT, so that "unlikelyness" is an argument against it'. Obviously, the He positioning is another break of pattern example. If one concludes that group 3 is Sc-Y-La-Ac (or that this is a well-based variant, given the criteria applied), then this ends up with a split d block, full stop. Unless you want to break the rule of ascending atomic numbers in the PT.
What we want to show is for the sources, how we show it is just an editorial choice. As we are free to pick out background category colors. And the graph is just a how thing. Now those many authors who show La-Ac with footnote-placeholders in the very same element cell are just sloppy or bad graphicists. Their PT does not show what they state. (For example, here is another(!) IUPAC PT that 'says' there are 32 elements in group 3). I am not worrying about the IUPAC, they'll learn to be consistent in the end. But I am worried about fellow-editors here who keep creating arguments from a proven misformed PT. It is simple: we reject each and every PT presentation that has, clearly or by suggestion, by graphical intention or bad form, 30 (or 32) elements in group 3 Red XN. -DePiep (talk) 15:51, 16 December 2015 (UTC)
  • 1. re Sandbh: "Scerri supports –Lu-Lr on the grounds that, essentially, -La-Ac results in a split d-block". This is a stronger conclusion that Scerri makes. He just calls it "a very asymmetrical possibility" to split the d block, and nothing more. So (how strange it is to foul Scerri for this), this is not enough a source to dismiss one option. It would refute all previous 18 more thorough (and non-graphical!) sources.
And from your 2nd ink, Scerri again:

For me, this [increasing atomic numbers] is a virtually conclusive argument in favour of group 3 consisting of Sc, Y, Lu and Lr. The only fly in the ointment is a third possibility, but this involves an awkward sub-division of the d-block elements (Figure 3). As such, it is not a fatal objection to the group 3 assignment that is being proposed in this article.

So nothing 'essentially' or 'grouds' in a split d block'.
  • 2. re "Sargent-Welch" PTs you introduce(!): [5] (images) shows they all have 32 elements in group 3. To be binned. Zooming out: Sandbh, why do you keep pushing & abusing the Sc/Y/*/** form? Some times, like here, even trying to construct a source-argument from that graph? Thought we (Scerri and I ;-) ) had mentioned the arguments. -DePiep (talk) 16:29, 16 December 2015 (UTC)

20. Name matches position: Putting La and Ac in the f block is pleasingly consistent with their names. This is a comment to Scerri's blog by McCaw (who, according to Scerri, has expertise in f-block chemistry).

Analysis : Yes, good point. Sandbh (talk) 10:32, 29 November 2015 (UTC)

I dunno. It strikes me as a little lame. One could argue just as well that "lanthanide/oid" implies "lanthanum-like" and thus cannot include lanthanum itself. And where does that leave lutetium, surely a lanthanide in chemistry? Double sharp (talk) 15:04, 29 November 2015 (UTC)
I think, used either way, the argument can't be used to even partially defend either version.--R8R (talk) 09:06, 30 November 2015 (UTC)

Arguments 21+[edit]

REFERENCE: Trifonov DN 1970, Rare-earth elements and their position in the periodic system, translated from the 1966 Russian edition, Academy of Sciences of the USSR Institute of the History of Natural Sciences and Technology, Moscow, published for the Atomic Energy Commission and the National Science Foundation, Washington, by the Indian National Scientific Documentation Centre


21. Comparison of six properties: Trifonov compares La and Lu across electronic structure; atomic volume, radius, ionisation energy, and density; and basicity.

Electronically he says (a) Sc (2, 8, 9, 2), Y (2, 8, 18, 9, 2) and Lu (2, 8, 18, 32, 9, 2) each have only two incomplete shells and that this pattern holds true for the rest of the transition metals proper for periods 4 to 6, whereas La (2, 8, 18, 18, 9, 2) has three incomplete shells. Placing Lu in group 3 also means (b) that there is a consistent difference in atomic numbers of 32 between the period 5 and 6 transition metals, whereas this is not the case for La in group 3. He further notes (c) that, "…in the spectrum of La the configuration levels containing 4f-electrons are extremely deep—already there is a tendency to strenghten the bonds of 4f-electrons" but that "this can hardly serve as a sufficient basis for considering La as the first element of 4f-family."

For atomic volume, radius, ionisation energy and density he says vertical trends going down groups 3 to 7 favour Sc-Y-Lu but that horizontal trends in periods 4 to 6 for groups 1 to 3 favour Sc-Y-La.

On basic character he says that increasing basicity with increasing atomic number is a general principle for the entire periodic system and since Sc-Y-La follows this pattern, whereas Sc-Y-Lu does not, he overall favours La in group 3.

Analysis: Trifonov appears to discount his 2½ electronic arguments in favour of his single basicity argument. The latter argument is flawed since basicity does not always increase with increasing atomic number. The scope of his arguments on vertical and horizontal trends is too narrow and overlooks, for example, anomalous trends in melting points and Young's modulus, if La is placed in group 3, as discussed elsewhere on this page.


REFERENCES: Matthias BT, Zachariasen WH, Webb GW & Engelhardt JJ 1967, 'Melting point anomalies', Physical Review Letters, vol. 18, no. 19, pp. 781–784; Kmetko EA & Hill HH 1976, 'Anomalous melting of f electron metals (with attention to Pu)', Journal of Physics F: Metal Physics, vol. 6, no. 6, pp. 1025–1037


22. f character of La: Matthias et al. attribute the melting point of La, which is much lower than would be expected from its periodic table position, to the presence of some f character, "in the hybridized wave functions describing the band structure for the valence electrons." Kmetko and Hill offer a theoretical explanation for this anomaly. They calculate a value of 0.6–0.7 for the amount of f-like charge in La (which agrees with Matthias et al.). Sandbh (talk) 21:34, 1 December 2015 (UTC)

Analysis: Electron configurations are normally determined for the gas phase. The two works cited above refer to the condensed phase, which is much more relevant to chemistry. I've seen other references to some suspected f character in La but never anything this robust. I'm not expecting anything dramatic from Trifonov (#21) so, especially on the basis of this argument, plus #'s 3, 14, 16, and 20 (weak as it is), I intend to change my allegiance to Sc-Y-Lu-Lr. Any resulting kinks in periodic trends, such as ionic radius (#6) are then explained by the lanthanide and actinide contractions, just as occurs in going down group 3 (which is also disturbed by the d-block contraction).

Браво to User: R8R Gtrs for motivating me to compile this whole section, and thank you to User: Double sharp for reasonably long ago discussions about this topic. There's one more argument from Jensen that I'll post, re lawrencium and the kerfuffle about the relevance of its ionization energy. Sandbh (talk) 06:00, 2 December 2015 (UTC)

Thanks; but then bravo tenfold to you for actually compiling it.--R8R (talk) 12:35, 2 December 2015 (UTC)

Postscript: A couple more refs on the f character of La:

"Lanthanum has a very high density of states at the Fermi level and a much higher superconducting critical temperature compared with other trivalent transition metals. These have led to speculation that the conduction bands at the Ferm level contain the f character. The band calculation of Glötzel and Fritsche(14) has shown that the f bands of La are above the Fermi level by about 2eV, but because of hybridization the occupied states contain 0.3 electrons of the f character for each atom." (p. 125)

  • Liu SH 1980 'Electronic Structure of Rare Earth Systems', in Science and Technology of Rare Earth Materials, EC Subbarao & WE Wallace (eds), Academic Press, New York, pp. 121–142

"The existence of f states at or near the Fermi level is demonstrated by depressed melting points, by high low-temperature capacities, by the ability to make the solids either superconducting or magnetic in various intermetallic compounds, by the exterme sensititvity of most physical properties to pressure, and by the occurrence of certain unique crystal structures." (p. 39)

"…lanthanum, actinium, thorium, and americum…have some f character…These elements have f bands above the Fermi energy…The existence of these unfilled bands is shown, for example, by depressed melting points…by the double hexagonal close-packed (dhcp) structure of lanthanum and americium…and by the need for including some f character in the calculation of the Fermi surface of thorium." (p. 42)

"More specifically, the evidence for several pressure-induced phase transitions [in La] shows the f character in the bonding." (p. 43)

  • Smith JL 1980, 'Superconductivity in the Actinides', in Superconductivity in d- and f-band metals, H Suhl and MB Maple (eds), Academic Press, New York
  • Wow, thank you for the info. I was aware there might be something quite similar might be present at superheavy elements, in part element 116 (but also surely some others to some extent as well), for which some theoretic data for atomic-scale properties is available, but I wasn't aware this might happen with not so heavy elements to such a degree. Very interesting.--R8R (talk) 15:48, 4 December 2015 (UTC)



REFERENCE: Jensen WB 2015, Some Comments on the Position of Lawrencium in the Periodic Table


23. Ionization energy of Lr: Jensen discusses, as reported in the 9 April 2015 edition of Nature, the experimental confirmation of the ionization energy (IE) of Lr and the implications of this for the composition of group 3. The latter question dominated all subsequent news stories on the Nature paper. He finds that no conclusions can be drawn on this question.

Analysis: Agree.


REFERENCE: Settouti N & Aouragi H 2014, 'A Study of the Physical and Mechanical Properties of Lutetium Compared with Those of Transition Metals: A Data Mining Approach', JOM, vol. 67. no. 1, pp. 87–93, doi:10.1007/s11837-014-1247-x


24. Lu as a transition metal: The physical and mechanical properties of Lu are compared with those of Cs, Ba, Hf, Ta, W, Re, Os, Ir, Pt, Au, Tl, Pb, and Bi, using mathematical analysis. The authors conclude that Lu shares many properties and similarities with period six transition metals and, "and can be well described as a transition metal."

Analysis: The authors go too far in saying Lu may well be described as a transition metal. To substantiate this claim they'd have to show that the properties of Lu in question are closer to e.g. Hf than they are to the other heavy lanthanides, and they did not do that. Other references have different observations. Spedding and Beadry (1968, p. 377), wrote, "Since metallic lutetium resembles closely erbium and holmium, except that it melts at a slightly higher temperature and is essentially non-magnetic, the details of producing, purifying and fabricating it are almost identical with those described under Holmium." Leal, Restrepo and Bernal (2012) compared 4,700 binary compounds of 94 elements. Sc and Y ended up in their own cluster; Lu ended up in a cluster with Er, Ho and Gd.

  • Spedding FH & Beadry BJ 1968, "Lutetium", in CA Hampel (ed.), The encyclopedia of the chemical elements, Reinhold Book Corporation, New York, pp. 374–378

Sandbh (talk) 04:20, 5 December 2015 (UTC)


REFERENCE: Nelson PG 2012, 'Periodicity in the formulae of carbonyls and the electronic basis of the Periodic Table,' Foundations of Chemistry, vol. 15, no. 2, pp. 199–208


25. Carbonyls: Nelson argues that the number of outer electrons possessed by an atom, and the number required for it to achieve an inert gas configuration exhibit an almost exact periodicity. Further, these two numbers correlate almost exactly with the highest conventional valency and the highest carbonyl valency exhibited by an element. For example in iron carbonyl, Fe(CO)5, the carbonyl valency is taken to be 10 whereas Fe has a highest conventional valency of 6. Now, while Y, La and Lu all have a highest conventional valency of 3, Y and Lu require only 15 electrons to achieve an inert gas configuration whereas La would need 29. On this basis Nelson perfers Y-Lu rather than Y-La.

Analysis: I can see periodicity in carbonyl valencies but the values of the rare earths carbonyl valencies are tentative, incomplete and based on matrix isolation studies. Nelson however reckons errors in the available data will not significantly affect the argument. Sandbh (talk) 09:33, 12 December 2015 (UTC)


REFERENCE: Wulfsberg G 2000, Inorganic Chemistry, University Science Books, Sausalito, CA


26. Metallurgy and symmetry: Wulsberg reckons the chemical and electronic properties of La and Lu (and Ac and Lr) are too close to make a call. He cites Jensen's 1982 arguments saying that the "metallurgical" resemblance is much stronger for Lu than La, so has adopted Lu (and by extension, Lr) below Y. He goes on to note that "an important additional advantage is that the periodic table becomes more symmetrical, and it becomes easier to predict electronic configurations." (p. 53)

Analysis: Like he says. I will caveat however that symmetry does not necessarily presuppose truth—does it? Sandbh (talk) 09:58, 12 December 2015 (UTC)


REFERENCES: MacKay KM, MacKay RA & Henderson W 2002, Introduction to Modern Inorganic Chemistry, 6th ed., Nelson Thornes, Cheltenham, p. 256 • Zhang et al., 2014, 'Preparation, characterization, and photocatalytic activity of boron and lanthanum co-doped TiO2', Catalysis Communications vol. 45, pp. 144–147 (144)


27. f orbital participation in bonding?

From MacKay et al. (in their chapter on the Scandium Group and the Lanthanides): "The one case in which contributions to the bonding from the f orbitals is possible is in complexes of the heavier elements in which the coordination is high. Use of the s orbital, together with all of the p and d orbitals of one valency shell, permit a coordination number of nine in a covalent species. Thus, higher coordination numbers imply either bond orders less than unity or else use of the f orbitals. In addition, certain shapes (such as a regular cube) of lower coordination number also demand the use of f orbitals on symmetry grounds. These higher coordination numbers have only become clearly established recently, but their occurrence in lanthanide or actinide element complexes suggest the possibility of f orbital participation. Example include the 10-coordinate complexes mentioned above ["Yttrium, and the other lanthanides investigated, also form 10-coordinate M(NO)3)52– complexes…In Ce(NO3)63–, the coordination number is twelve…In the ion La(C6H9N3)43+, the twelve nitrogen atoms form an almost regular icosahedron around the La atom."], LaEDTA(H2O)4 and Ce(NO3)52– or 10-coordinate La2 (CO3)3.8H2O; 11-coordinate Th(NO3)4.5H2O (coordination by four bidentate nitrate groups and three of the water molecules; and the 12-coordinate lanthanum atoms in La2 (SO4)3.9H2O—with twelve sulfate O atoms around one type of La atom position."

From Zhang et al: "Lanthanide ions have a strong ability to form complexes by the coordination bonding between their f orbitals and lone electron pairs of various Lewis bases. Therefore, TiO2 doping with lanthanide ions can concentrate the organic pollutants at the semiconductor surface."

Analysis: MacKay et al. are plausible but not definitive. The only implication I can draw from Zhang et al is that the lanthanum ion (La3+?) uses its empty f orbitals in coordination bonding (as suggested by MacKay et al., and consistent with Nelson in #25) but why it would do this in the substance being studied (B and La co-doped TiO2) isn't clear to me. Sandbh (talk) 22:07, 13 December 2015 (UTC)

Postscript

"...the relatively small value of 0.04 Å for the lanthanide monoxides results from f-orbital participation in LaO, whereas the 4f shell has core character in LuO." (p. 626)

  • Dolg M & Stoll H 1996, "Electronic structure calculations for molecules containing lanthanide atoms", Handbook on the Physics and Chemistry of Rare Earths, vol. 22, KA Gschneider Jr. & L Eyring (eds), Elsevier, Amsterdam, pp. 607–729

"Two other polyhedra possible for 8-coordination are hexagonal bipryamid and cube...Cubic coordination requires f orbital participation and is a possible configuration for some lanthanide compounds. Some structures of lanthanide compounds with octa-coordination are given in Table 5.10." The example is given of LaTaO4, with a cubic structure, although the column heading says nearest [italics added] regular shape. (pp. 397, 399)

  • Sastri VS, Bünzli J, Rao V, Rayudu GVS & Perumareddi JR 2003, Modern Aspects of Rare Earths and Their Complexes, Elsevier, Amsterdam

Sandbh (talk) 03:36, 18 December 2015 (UTC)


REFERENCES: Lavelle L 2008, "Lanthanum (La) and Actinium (Ac) should remain in the d-block", Journal of Chemical Education, vol. 85, no. 11, pp. 1482–1483 • Jensen WB 2009, "Misapplying the periodic law", Journal of Chemical Education, vol. 86, no. 10, p. 1186 • Lavelle L 2009, "Response to Misapplying the periodic law", Journal of Chemical Education, vol. 86, no. 10, p. 1187 • Clark RW & White GD 2008, "The flyleaf periodic table", Journal of Chemical Education, vol. 85, no. 4, p. 497.


28. Block consistency and literature record: Lavelle (2008) says that, "the entire modern basis of the periodic table is the grouping of elements by occupied outer orbital type giving rise to the s-block (two outer electrons in an s-orbital and two groups), the p-block (six outer electrons in three p-orbitals and six groups); the d-block (ten outer electrons in five d-orbitals and ten groups), and the f-block (14 outer electrons in seven f-orbitals and 14 groups)." He says that placing Lu and Lr in the d-block, and La and Ac in the f-block leads to a worse outcome than leaving La and Ac in the d-block since this would represent, "the only case where a pair of elements [i.e. La-Ac] is placed such that they are part of block [i.e., the f- block] with no outer electrons in common with that block." He also relies on the fact that several well-known reference books show La and Ac in the d-block. His position is that, "we [should] use well-established forms of the periodic table…and that, "to suggest otherwise may result in a Pandora's box of a never-ending mutltitude of different periodic tables" (Lavelle 2009).

Analysis: Lavelle's argument about La-Ac being the only example of "a pair out of place", if they were in the f-block is right. And I agree with him that Sc-Y-La-Ac is well established in the literature. I think the counter-argument—drawing partly from Jensen (2009)—would be: (1) the modern periodic table is based on idealized electronic configurations rather than actual configurations "and in this fashion functions in chemistry much as the ideal gas law or the concept of ideal crystals and ideal solutions" (Jensen 2009); (2) there are many exceptions to idealized electronic configurations; (3) an 18-column table with Sc-Y-Lu-Lr is closer overall to a periodic table based on idealized configurations than is the case with Sc-Y-La-Ac. [Check me on this. If you say that the f-block starts with Ce-Th then all 28 of the f-block elements are out by one space from their idealized electron configurations. I just had a look at three textbooks with Sc-Y-La-Ac or Sc-Y-*-** and it is a bit of a sorry story to see how they depict their f-blocks. According to one there are 30 f-block elements; a second shows group 3 as Sc-Y-La-Ac but their 32-column table showing the relationship between orbital filling and the periodic table has group 3 as Sc-Y-Lu-Lr, with Lu and Lr coloured as f-block elements, and La-Ac as d-block elements at the the start of the f-block; the third has the blocks coloured right but if you try and compare the f-electron filling sequence in the footnoted Ce-Lu series with the number of f-electons you'll see it's largely out of sync: Ce +1 f1; Pr +2 f3; Nd +3 f4; Pm +4 f5; Sm +5 f6; Eu +6 f7; Gd +7 f7; Tb +8 f9; Dy +9 f10; Ho +10 f11; Er +11 f12; Tm +12 f13; Yb +13 f14; Lu +14 f14. Whereas if you start the f filling sequence at La, you get the right synchronization, with only a few irregularities].

Now, since the Sc-Y-Lu-Lr table was identified by Clark and White (2008) as one of the three common forms of 18-column table, in addition to Sc-Y-La-Ac and Sc-Y-*-**, and since a Sc-Y-Lu-Lr periodic table is closer, overall, to the idealized electron configuration periodic table upon which the actual periodic table is based, it follows that this is the "better" (more scientific?) form. Sandbh (talk) 12:15, 17 December 2015 (UTC)

Postscript: Clark and White (2008, above) pooled their general chemistry text collections to survey trends in flyleaf periodic tables from 1948 to 2008. From 35 texts they found 9 × Sc-Y-*-**; 9 × Sc-Y-Lu-Lr and 11 × Sc-Y-La-Ac. Over the last 20 years of their survey period the count was 2 × Sc-Y-*-**; 6 × Sc-Y-Lu-Lr and 6 × Sc-Y-La-Ac. Sandbh (talk) 21:45, 17 December 2015 (UTC)

Further to my above observation about idealized electron configurations, it can be seen from tables 1 and 2 hereunder that 20 of 28 f-block electron configurations in a Sc-Y-Lu-Lr table match the idealized electron configurations underlying the modern periodic table whereas this is the case for only 9 of 28 f-block elements in a Sc-Y-La-Ac table, a disparity of over 2:1 Sandbh (talk) 02:12, 19 December 2015 (UTC)
TABLE 1: Sc-Y-Lu-Lr periodic table f-block showing electron configurations (light grey shading = match with idealized configuration; dark grey shading = irregularity)
Period 6 La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb
Idealized no. of f-electrons 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Actual configuration 5d16s2 4f15d16s2 4f36s2 4f46s2 4f56s2 4f66s2 4f76s2 4f75d16s2 4f96s2 4f106s2 4f116s2 4f126s2 4f136s2 4f146s2
Period 7 Ac Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No
Idealized no. of f-electrons 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Actual configuration 6d17s2 6d27s2 5f26d17s2 5f36d17s2 5f46d17s2 5f67s2 5f77s2 5f76d17s2 5f97s2 5f107s2 5f11s72 5f127s2 5f137s2 5f147s2
TABLE 2: Sc-Y-La-Ac periodic table f-block showing electron configurations (light grey shading = match with idealized configuration; dark grey shading = irregularity)
Period 6 Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
Idealized no. of f-electrons 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Actual configuration 4f15d16s2 4f36s2 4f46s2 4f56s2 4f66s2 4f76s2 4f75d16s2 4f96s2 4f106s2 4f116s2 4f126s2 4f136s2 4f146s2 4f145d16s2
Period 7 Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr
Idealized no. of f-electrons 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Actual configuration 6d27s2 5f26d17s2 5f36d17s2 5f46d17s2 5f67s2 5f77s2 5f76d17s2 5f97s2 5f107s2 5f11s72 5f127s2 5f137s2 5f147s2 5f147s27p1

General comments[edit]

I hope I'll get to the pre-WWII book today (can't find it online, but let's see). Meanwhile, could you explain the "true, but misleading" labels? In what way are they misleading? Is stating the -La trends fall in line not misleading as well? If not, why? What is the difference?--R8R (talk) 09:57, 27 November 2015 (UTC)

That's great, tx. 'Misleading' is a word used by Jensen to refer to comparisons based on inter-group trends without also considering inter-group trends. He does this himself by only looking to the right of group 3 and not to the left. I may change the label to 'True, but not the full story' since that is less emotive. 124.171.67.54 (talk) 11:39, 27 November 2015 (UTC)
One question: since sources just judge raw data, would it be wise if we just list it here, put in a proper format for comparison, and make our decisions based on that and not on someone else's judgments, or would it not?--R8R (talk) 17:45, 27 November 2015 (UTC)
Yes, thank you, most definitely. At the moment I just happen to be more focused on listing all of the arguments. Sandbh (talk) 23:16, 27 November 2015 (UTC)
Property
1 2 3 4 5
K Ca Sc Ti V
Rb Sr Y Zr Nb
Cs Ba La Lu Hf Ta

for example

Electronegativity
1.00 1.36 1.54
0.95 1.22 1.33
0.89 1.1 1.27 1.3
Test: please add group numbers to columns. -DePiep (talk) 23:37, 27 November 2015 (UTC)
I had you asking for this in mind when adding the numbers; here you go.--R8R (talk) 09:15, 30 November 2015 (UTC)

@Sandbh: Thank you for taking the time to list all these! I notice that several (4, 5, 8, 9, and maybe 15) seem to be amount to the following

  • 3=Sc-Y-La follows the pattern established in groups 1 and 2
  • 3=Sc-Y-Lu follows the pattern established in groups 4 and beyond

What would be really telling would be cases where one or the other follows the patterns established both before and after group 3. YBG (talk) 22:54, 29 November 2015 (UTC)

I feel a little sorry for not being able to get to the question at the moment. I have updated the base table format, in which data for primary analysis will be stored. Groups 1 and 5 are important because they show steps in the s and d block, rather than just s to 3 and 3 to rest of d we had prior to now. In general, I expect and I think it is to be expected (and I remember a little data as well) the -La trend will resemble groups 1 and 2, and the -Lu trend will fall in line with the rest of the d block. (Just what YBG said, I noticed as I hit "show preview.") I believe the major factor in this is the f block contraction, present for Lu, Hf, Ta, and so on, but not La, Ba, or Cs. But I will jump to any colclusions only after I present the numeric data.--R8R (talk) 08:54, 30 November 2015 (UTC)

@Double sharp: Does Cotton's Lanthanide and Actinide Chemistry throw any light onto the question? Sandbh (talk) 22:34, 30 November 2015 (UTC)

@Double sharp: SAme question entered my mind one hour ago. -DePiep (talk) 16:47, 16 December 2015 (UTC)

I didn't see this the first time! Here's what he has to say:

"The Periodic Table places the elements in atomic number order, with the lanthanides falling between barium (56) and hafnium (72). For reasons of space, most present-day Periodic Tables are presented with Groups IIA and IVB (2 and 4) separated only by the group IIIB (3) elements. Normally La (and Ac) are grouped with Sc and Y, but arguments have been advanced for an alternative format, in which Lu (and Lr) are grouped with Sc and Y (see e.g. [Jensen]) on the grounds that trends in properties (e.g. atomic radius, IE, melting point) in the block[?!] Sc-Y-Lu parallel those in the Group Ti-Zr-Hf rather closely, and there are resemblances in the structures of certain binary compounds. Certainly on size grounds, Lu resembles Y and Sc (it is intermediate in size between them) rather more than does La, owing to the effects of the 'lanthanide contraction'. The resemblances between Sc and Lu are, however, by no means complete." (p. 7)

"Scandium and yttrium are elements in group IIIA (3) of the Periodic Table, usually placed above La (or Lu). Their treatment is frequently grouped with the lanthanides in textbooks (often for reasons of convenience). ... purely on size grounds, it would be predicted that yttrium would resemble the later lanthanides but that scandium would exhibit considerable differences, and this expectation is largely borne out in practice." (p. 107)

"As expected on steric grounds, the smaller scandium generally exhibits lower coordination numbers than the lanthanides, although sometimes the value is the same as for lutetium, the smallest lanthanide." (pp. 112–4)

And yttrium is such a good pseudolanthanide that he always includes it for comparison with the other lanthanides. Scandium, on the other hand, is a weird mix of a 3d transition metal (in size leading to lower coordination numbers) and a lanthanide. We can also see that Sc/Y/Lu/Lr is gaining enough respectability to be acknowledged by him. The only periodic table picture he gives, believe it or not, is Seaborg's eight-period "future" periodic table on p. 235, which of course gives Sc/Y/La/Ac. This is so unexpected that I confess I'm not quite sure if I didn't miss one. Double sharp (talk) 01:29, 17 December 2015 (UTC)

Draft Sc|Y|Lu|Lr version ready[edit]

The draft in my sandbox has once again got to a more or less stable state. Changes are:

  • Section 1 Overview. I turned User:YBG's mention of element 0 periodic tables into a note.
  • Section 5 Alternative structures. This section is now called "Different periodic tables", and has two subsections, 5.1 "Common forms", which summarizes the three main forms of 18-column table, and mentions why Sc|Y|Lu|Lr is in the lede (see note for some further detail); and 5.2 "Other arrangements" (no change, aside from title). I kept the secondary 14CeTh etc labels since these have been used by other authors in the literature.
  • Section 6.6 Period 6 and 7 elements in group 3. I added a paragraph about the attempted use of the 32-column table to address this question.

I'm not convinced there is a significant issue with positioning the footnoted f-block elements in such a way that La and Ac are aligned with, but separate from, the group 3 elements, so have left that as is.

If I've forgotten anything please let me know. Comments welcome. Sandbh (talk) 11:26, 20 December 2015 (UTC)

I haven't looked closely at the sandbox, but for now, I have a question re note 14: I remember once reading it was thought sometime in the past (the '50s, the '60s, or maybe the '70s) the then-current data indicated most lanthanides had ground-state configurations of (n-2)fx-1(n-1)d, unlike the current data, which indicates most lanthanides have ground-state configurations of (n-2)fx. Is it worth mentioning?
Is there an easy way to collect this kind of content? If we add it and then decide it's an overkill for the main PT article, it'll still fit nicely into group 3 element.--R8R (talk) 17:04, 20 December 2015 (UTC)
Jensen mention this in his 1982 paper. The first ground state config works only for La, Gd and Lu whereas the second one works for all the rest of the rare earths in period 6. Adding this to note 14 is an excellent suggestion. I'll see what I can do. Sandbh (talk) 04:02, 21 December 2015 (UTC)
I've added four citations re note 14, one at the end of the parent paragraph, and the other three in the note itself, including one mentioning the ideal ground state configuration of the f-block elements (as determined by Jensen). Sandbh (talk) 05:01, 23 December 2015 (UTC)
  • Comments by YBG:
    1. The neutronium info isn't mine -- I just moved it from the first sentence of a paragraph in Periodic table to a separate, 1-sentence 2nd paragraph, as I thought the information was really secondary (or tertiary or less) and certainly didn't belong at the beginning of a paragraph. Putting it is a note is perfectly acceptable. I'd be fine with eliminating it altogether.
      I don't mind it in a note, which is where it is now. Sandbh (talk) 05:01, 23 December 2015 (UTC)
    2. What about repeating the period 6/7 numbers in the rows with the footnoted elements, to show that those elements are members of periods 6/7 also?
    3. What about putting a horizontal rule between the main table body and footnote elements, to show that footnoted elements don't belong to PT groups?
    I included changes 2 & 3 in an edit which I self-reverted.
    I remember seeing this suggestion and not liking the way it implied that the f-block elements were somehow separated from the rest of the elements by being "below the line". I don't think there's a need to repeat the period 6/7 numbers. Sandbh (talk) 05:01, 23 December 2015 (UTC)
    YBG (talk) 04:30, 22 December 2015 (UTC)
Another thing - regarding the nomenclature for the three types of 18 column periodic tables. It seems we've been using a couple of different ways of identifying them:
Option Type I Type II Type III Naming method
Option 1 14CeTh 14LaAc 15LaAc Number of footnote elements per row + 1st element in each footnote row
Option 2a Sc|Y|La|Ac Sc|Y|Lu|Lr Sc|Y|*|** Symbols in group 3 column (separated by pipe symbol)
Option 2b Sc/Y/La/Ac Sc/Y/Lu/Lr Sc/Y/*/** Symbols in group 3 column (separated by virgule)
I tend to think that options 2a and 2b are more intuitive, but that's just my 2¢ worth. YBG (talk) 05:35, 22 December 2015 (UTC)
I'll have another look at this. Sandbh (talk) 05:01, 23 December 2015 (UTC)
Done. Sandbh (talk) 10:24, 24 December 2015 (UTC)

I went ahead and updated the lede with an Sc|Y|Lu|Lr image, and the rest of the article with the associated and other changes that were in my sandbox. Merry Xmas Sc|Y|Lu|Lr. Sandbh (talk) 00:25, 25 December 2015 (UTC)

Data for group 3 trends[edit]

I have extracted this data from List of elements by atomic properties. Not sure how helpful this will be. Feel free to delete it if it isn't useful. YBG (talk) 06:22, 1 December 2015 (UTC)

Atomic number & symbol
1 2 3 4 5
19 K 20 Ca 21 Sc 22 Ti 23 V
37 Rb 38 Sr 39 Y 40 Zr 41 Nb
55 Cs 56 Ba 57 La 71 Lu 72 Hf 73 Ta
Atomic Weight
1 2 3 4 5
39.10 40.08 44.96 47.87 50.94
85.47 87.62 88.91 91.22 92.91
132.9 137.3 138.9 175.0 178.5 181.0
Electronegativity (Pauling)
1 2 3 4 5
0.82 1 1.36 1.54 1.63
0.82 0.95 1.22 1.33 1.6
0.79 0.89 1.1 1.27 1.3 1.5
1st ionization energy
1 2 3 4 5
4.34066 6.11316 6.5615 6.8281 6.7462
4.17713 5.6949 6.2171 6.6339 6.75885
3.8939 5.2117 5.5769 5.4259 6.82507 7.5496
Atomic radii
1 2 3 4 5
220 180 160 140 135
235 200 180 155 145
260 215 195 175 155 145
Covalent radii
1 2 3 4 5
196 174 144 136 125
211 192 162 148 137
225 198 169 160 150 138
Ionic radii
1 2 3 4 5
152 114 88.5 74.5 68
166 132 104 86 78
181 149 117.2 100.1 85 78
Melting and boiling points, degrees Celsius
1 2 3 4 5
63.5
759
842
1484
1451
2836
1668
3287
1910
3407
39.3
688
777
1382
1526
2930
1855
4377
2477
4744
28.5
671
727
1637
920
3464
1652
3402
2233
4603
3017
5458
27
677
700
1140
1051
3197
1627
 
2100
5500

Based on vertical and horizontal trends, note the anomalously low melting points for La and Ac in group 3, which has been attributed to these elements having some f character. See "Group 3 composition debate" above, argument #22. Sandbh (talk) 11:18, 3 December 2015 (UTC)

Redox potentials
1 2 3 4 5
-2.936 -2.868 -2.09 - -1.380
-2.943 -2.899 -2.38 -1.45 -0.8
-3.026 -2.906 -2.379 -2.28 -1.55 -0.6

(Data in this table given for M+ to M for group 1, M2+ to M for group 2, M3+ to M for groups 3 and 5, and M4+ to M for group 4)

Yes, that's great; thank you very much! I have a few other properties to list, but in general, I will do just the same thing.--R8R (talk) 14:59, 1 December 2015 (UTC)

List of properties we could list and have not listed yet (additions by anyone would be very welcome):

  • Ionization energies (1st for group 1, 1st+2nd for group 2, etc.)
  • Ionic radii
  • Redox
  • mp/bp
Any effect for our group 3 presentation from this? -DePiep (talk) 02:09, 2 December 2015 (UTC)

Calculating ionization energies would surely take much time, so I skipped it. I don't think this would change the whole picture.

Of these tables, we may analyze electronegativity, atomic/covalent/ionic radii, mp and bp, and possibly the redox potentials. (I am ignoring 1st IE because I believe it plays a small role in an element's character, if that element is not univalent, and it is not even a complete picture, but rather a part of the whole IE energy picture taken out of context, when it comes to different groups.) There might be other properties, but I am not aware of any property that would change throughout the elements in a way different than theses ones (please let me know if there is such a property). Data listed by Sandbh above follows the same pattern.

I have joined the discussion without knowing the answer, coming open-minded and unbiased, mostly thanks to having come so late. Analyzing the data (which I have done a few years ago; back then, however, I looked to support my then-current bias, and this time, I was looking for the truth); by now, I have formed an opinion. The Sc-Y-La trend closely resembles those of Ca-Sr-Ba and K-Rb-Cs, while the Sc-Y-Lu is much closer to Ti-Zr-Hf and V-Nb-Ta instead. That could be said in every single case (except the redox part; we can't say it strongly advocates for either version). Which one would be right to be named as group 3? I strongly believe (and I think everyone would agree) group 3 is a d block group. As such, it would make most sense to group it with the other d block groups, and one would expect it to behave similarly to the other groups. And since we humans decide what a "periodic table group" is in first place, and we have to assign the name of "group 3" to either Sc-Y-La or Sc-Y-Lu, I think picking Sc-Y-Lu would make most sense. The numeric data is a great illustration for my words, and data Sandbh quotes is even greater, for the most part. (I am okay with singular deviations, as they are not nearly close to being a counter-argument as strong as what they oppose to, and as the PT does have such deviations elsewhere, even when it doesn't yet come to relativity.)

The numeric data makes me draw conclusions that generally match Sandbh's, in that the optimal composition of the group 3 would be Sc-Y-Lu-Lr.--R8R (talk) 12:32, 2 December 2015 (UTC)

This is convincing - the data, with 3 and 14 and 22. We can see d and f contributions starting at Sc/Y and La/Ac respectively, so that physically they are certainly from those blocks. An instructive comparison is Ti-Zr-Ce-Th vs Ti-Zr-Hf-Rf, where no one advocates the former. Also note extentuating circumstances, like why Be-Mg-Zn works since Be and Mg are small. Likewise, Sc and Y are slightly degenerate as having only one d electron, they have to lose it and form ions without any, but excluding them as TMs is a bit like moving Rb to period 4 as it forms ions without any 5s electrons. Hence they probably belong with groups 4 to 11, and not so much 1 to 2. Double sharp (talk) 14:22, 3 December 2015 (UTC)
Young's modulus, GPa
1 2 3 4 5
3.1 2.0 74 116 128
2.4 15.7 64 68 105
1.7 13 37 69 78 186
13.2 25

Based on vertical and horizontal trends, note the anomalously low Young's modulus (a measure of rigidity) figures for La and Ac in group 3, a phenomenon which has also been attributed to these elements having some f character. Sandbh (talk) 21:36, 3 December 2015 (UTC)

Group 12: Post-transition metals[edit]

---Extracted by me from the Rare earths section--- Sandbh (talk) 23:24, 14 November 2015 (UTC)
Promoted into toplevel topic (== header). Not related to discussion about f-block placement, group 3, or Sc/Y/*/**. -DePiep (talk) 21:17, 16 November 2015 (UTC)

The group 12 metals (Zn, Cd and Hg) have effectively zero transition metal (TM) properties. They would be better categorised as post transition metals (PTM), but see below. E113 and Fl would count as TMs given the predicted involvement of their d electrons in compound formation.

In the transition metal article I would probably include the group 3 metals (Sc, Y, La, and possibly even Ac), and the group 12 metals (Zn, Cd, Hg) for comparative purposes but show them coloured as , respectively, rare earth metals and post transition metals.

What logical problems did you foresee with K, Rb and Cs? Sandbh (talk) 23:06, 11 November 2015 (UTC)

See this article. "At high pressure the alkali metals potassium, rubidium, and cesium transform to metals that have a d1 electron configuration, becoming transition metal-like." Actually, I think you linked it first! This is why I'm not so keen on the d-electron argument: there isn't an obvious place to draw the line (except saying "any d-electron involvement at all counts", which makes Hg a transition metal but Cd not, which is rather silly). If you say Hg doesn't pass it, how do we know if E113 and Fl are going to?
I wouldn't say Zn, Cd, and Hg have no TM properties. Zn forms stable complexes with O-, N-, and S-donor ligands as well as halides and CN, like a transition metal, and Cd is rather similar. Additionally, many compounds of HgII (and CdII to a lesser extent) are highly coloured, a characteristic of transition metals. It seems like the easiest way to go is to call everything from groups 3 to 12 TMs, but call Sc/Y/La/Ac and Zn/Cd/Hg borderline TMs.

Double sharp, thank you. Your reference to Hg as a transition metal lacks any rigour. In this case you've quoted me out of context. The full story is that Nergaal regarded mercury as a transition metal on the basis that it could form HgF4 in a cryogenic matrix at a few degrees above absolute zero. I responded by saying that, ergo, K was a transition metal since it becomes one at high pressure i.e. greater than 30 Gpa. These are both ridiculous statements given periodic table categorisations are self-evidently based on the properties of the elements in ambient conditions. Regarding element 113 and flerovium, these may well qualify as transition metals based on their predicted chemistry, as I understand from reading their articles. These two elements are distractions in any event---we could easily leave them as post-transition metals until stronger data comes in.

Your argument about the properties of group 12 metals i.e. complex formation and some highly coloured compounds is lame. As I understand it, all or nearly all metals and metalloids can form stable complexes. And while many transition compounds are strongly coloured, some main group compounds are too e.g. SbI3 (ruby red); cesium oxide and sub-oxides ("brightly coloured"); and lets not forget brightly coloured Zintl phases The best that can be said about the group 12 metals is that they are capable of forming some stable complexes, just like e.g. the alkaline earth metals and group 13 metals can, and they are also capable of forming some brightly coloured compounds, just like some main group metals can. To suggest that these properties warrant calling them transition metals is like calling an orange and apple. Greenwood and Earnshaw, whom you were rightly quoting from, say these elements show few of the characteristic properties of transition metals; the only ones they discuss are complexes and highly coloured compounds and they then go on to note the drawbacks of having d10 configurations on complex formation. Cotton and Wilkinson go further and say that the group 12 elements are not classified as transition metals and that while their capacity to form complexes is "reminiscent" of transition metals, it is significantly qualified. Group 12 as transition metals on par with groups 4–11? Phooey! Sandbh (talk) 10:43, 13 November 2015 (UTC)

Thank you for stating where the line is drawn: does it happen at standard conditions, or not? Now we have a completely sound basis to exclude group 12, so now I agree with you. (Really: apart from not being sure where to draw the line, I don't really have any objections to taking out group 12).
Now how do we colour copernicium? Minor element, I know, but it's going to show up on all our periodic tables. Physically it is known to behave like a typical group 12 metal, except that it is predicted to have Cn(IV) as a significant part of its chemistry. The trouble with flerovium is that Fricke's old predictions mention sd hybridisation as a possibility, but the recent sources all assume Fl(IV) would go for sp3; this seems to imply that new predictions have the 6d energy levels too low. E113 OTOH is still predicted to have 113(V) AFAIK. But these are all still predictions, so maybe we ought to just colour them like their groups (post-transition metals) until we get evidence otherwise. After all, the only thing we know about them so far is that physically, they behave like their congeners in their groups. As such we would colour Cn and Fl as post-transition metals (and E113 as a predicted PTM) until evidence comes in for Cn(IV), 113(V), and Fl(VI). This is what Japanese Wikipedia does (or at least it will until Dlgkstmf ar1103 insists on restoring his/her favourite classification from his/her favourite science magazine, which apparently is not aware of the existence of chemical studies on At, Fr, and Fl but spontaneously invented studies on Mt, Ds, and Rg out of thin air, and has some penchant for keeping Ge and Sb out of the metalloid club while putting Se in it, and before I revert him/her again). In fact, they're very similar to your current proposal for the right side of the periodic table (but calls those just "other metals", and puts Be and Mg in that club; is that traditional there?). (talk) 15:20, 13 November 2015 (UTC)
So it got protected at The Wrong Version. Oh well. Time to write more screeds on the talk page (alas in English). (I do find the classification of Be and Mg outside the alkaline earth metals interesting. I will ask about that.) Double sharp (talk) 06:04, 16 November 2015 (UTC)

Group 3 in detail[edit]

A few hours ago, much prose on what elements should be included in group 3 was removed; this was soon undone, and it was proposed the topic should be raised here, which I am doing. Personally, I believe the material in question is going somewhat into detail; compare with, say, the part on placement of hydrogen and helium. Moreover, I think it is not FA-worthy in general, because it gives a somewhat one-sided description of the alternatives; i've had this feeling for a while, I was the one to write that part (for group 3 element, originally, where it did make more sense than it does here).

So yes, I also believe we'll be better off without it. Does anyone else want to drop a comment?--R8R (talk) 08:37, 18 September 2015 (UTC)

I think this material should be included because it concerns the classification of an important part of the periodic table, and it is an open question in the literature. If it is one-sided, perhaps it can be improved and made more neutral, but that is not a reason to delete it entirely. Dirac66 (talk) 11:13, 18 September 2015 (UTC)
I like that brief overview of group 3, giving a few reasons for each, but I've got two objections: firstly, it's biased towards Sc/Y/Lu/Lr (while it may be the best, it is not obviously better than Sc/Y/La/Ac and Sc/Y/*/**), and secondly, the referencing is terrible for an FA. So I would keep it, but address these problems. It's not obviously worse than the H/He section, which also states all alternatives and gives reasons for each. Double sharp (talk) 12:23, 18 September 2015 (UTC)
That said, all of this is true and correct, yet should be discussed into the main articles (period 1 and group 3). Here, we should just outline the problem and give a proper Main article link.--R8R (talk) 19:52, 18 September 2015 (UTC)

I was the one who removed the prose. I did so because (1) the content wasn't up to FA standard; (2) the use of a bullet list wasn't justified for an FA article; (3) it should never have been added into the article in this form—it came from a discontinued article called 'Placement of lanthanides and actinides in the periodic table'; and (4) getting it to FA standard would require a complete rewrite. Of course we encourage anybody to make contributions but in this case I felt that the importance of maintaining the quality of the FA was more important than keeping the prose in question. I've now started drafting a replacement in my sandbox. Given this I'll delete the prose in question again, unless there are any further objections.Sandbh (talk) 03:50, 19 September 2015 (UTC)

@Double sharp: Sandbh is cleaning up the prose in userspace (see above), so I have no objections anymore to the material being removed. Karl Dickman talk 05:49, 19 September 2015 (UTC)
OK. Yes check.svg Done. Double sharp (talk) 09:25, 19 September 2015 (UTC)
I love the development now visible in User:Sandbh/sandbox. The group 3 issues described well. I want to add some notes:
The topic & the prose is quite complicated. I had to make notes to get it. Can it be lighter?
Also, I think we could add two images: a PT with group 3 being Sc/Y/La/Ac and Sc/Y/Lu/Lr respectively. This could be parial only, say groups 1-4 and periods 4-5-6-7 only, f-block included. Shall I make some sketches, Sandbh?
The IUPAC PT is a horror [6]. Not about its status (being "not the only official one"), but because it is ambiguous graphically. It suggests states that, by substituting (inserting) the two rows from below, the whole f-block is group 3 and Sc, Y elements somehow span 15 columns (R8R wrote: Sc/Y/*/**). I am not convinced this is what it purports to say. We do this in our first periodic table image too. A second ambiguous hiccup are the two dashed lines that vertically make a connection between group 3 and La, Ac (it is screaming: but what about the other 14+14 elements?).
-DePiep (talk) 11:22, 20 September 2015 (UTC)
Thank you DePiep. I will try and make the prose clearer. Pls go ahead with the partial images and let's see how they look. We don't necessarily have to link to the IUPAC table. Sandbh (talk) 12:03, 20 September 2015 (UTC)
Maybe they were thinking of it meaning Sc/Y/La*/Ac**, i.e. Ln = La and the following elements? How very obfuscated indeed. It seems like an attempt to offend no one, with it possibly implying both Sc/Y/La/Ac (where the f-block is Ce–Lu) and Sc/Y/*/** (where somehow the f-block elements have become degenerate d-block elements).
Of course Sc/Y/*/** is pretty good for an introduction, where the f-block is unimportant. Unfortunately it also implies that since the lanthanides and actinides are in group 3, they won't exhibit common higher oxidation states, which is not true (e.g. UF6). Oh, the urge to make everything show Sc/Y/La/Ac like Greenwood and Holleman-Wiberg. Might annoy some people, but given how common this layout is no one can say it does not serve the beginner well (if not, why is it in very many introductory textbooks?) And it would score one over Sc/Y/*/** for the consistency argument (18- and 32-column tables that are both for general purposes should look the same), given that I'm not sure we can easily get rid of 32-column here thanks to the width issue. Double sharp (talk) 13:36, 20 September 2015 (UTC)
...sigh. No, as much as that would work, I don't like Sc/Y/La/Ac anymore (I don't feel like repeating all the arguments), and IUPAC gives me a perfect excuse not to use it. Sc/Y/*/** it will have to remain, then. Double sharp (talk) 13:52, 20 September 2015 (UTC)
I think the idea of partial PT's to illustrate this section is a great one. Originally, I was thinking of suggesting a side-by-side comparison akin to Periodic table § Layout variants, but having seen the sandbox, I now think having a separate PT for each paragraph would be best. YBG (talk) 14:12, 20 September 2015 (UTC)
About the IUPAC PT. as DS writes: Maybe they were thinking of it meaning .... Says it all: a useless ambivalent presentation by IUPAC. Since Sandbh suggests it can be left out, I'm convinced it does not offer a new or different variant (next to the two discussed). So we can leave it out without loss of information (instead, it's cleaning up this neighborhood!). I support it being removed from the sandbhox.
DS, Sc/Y/*/** is pretty good for an introduction - I strongly disagree, because it has the IUPAC ambiguity. IMO, we must show a gap. Compare: this (bad) vs. this (good). The variant Sc/Y/La/Ac should have the gap to the right of group 3 (good), but will not be our general presentation. -DePiep (talk) 09:15, 21 September 2015 (UTC)
Yes, if we need to be consistent (and I still think that's a good argument, despite FJ comments), Sc/Y/*/** is ruled out. (I say it's an OK introduction as some have used it that way, and because when you first learn chemistry the f-block is really unimportant. But once consistency with 32-column is needed, it shatters.) I mention Sc/Y/La/Ac as a possibility for general presentation only because (1) it ought to keep the FJ-ish derision towards Sc/Y/Lu/Lr out, as it's been used as a general-purpose PT by such esteemed textbooks as Greenwood & Earnshaw and Holleman & Wiberg, and (2) it allows for a table with a clear gap between groups 3 and 4, as you say. Double sharp (talk) 09:56, 21 September 2015 (UTC)
Yes. Sc/Y/La/Ac will be described in detail in the section Sandbh is making. There will be two equal graphs available. After that, we can choose one variant to be the general one for out PTs. (Can you give article(s) that have the two sources?). -DePiep (talk) 10:40, 21 September 2015 (UTC)
@Sandbh, a text suggestion. Maybe leave out "rare earth metals" completely, even making the quote like: ... "closer in their chemical properties to lutetium [...] than they are to lanthanum." REM is another new term to be looked up when reading, adding to the complications. It should be defined elsewhere in the article sure. -DePiep (talk) 10:40, 21 September 2015 (UTC)
What's wrong with "closer in their chemical properties to lutetium and the other heavy [lanthanides] than they are to lanthanum."? Heavy REM = heavy Ln, as the only REMs that are not Lns are Sc and Y, which are light (though Y acts as a heavy one). Double sharp (talk) 10:51, 21 September 2015 (UTC)
Nicer reading indeed. It is a literal quote from a source, so maybe remove the "'s before rephrasing. -DePiep (talk) 11:16, 21 September 2015 (UTC)
So Double sharp, do you know one or two articles that have the links to Greenwood & Earnshaw and Holleman & Wiberg sources you mentioned? Would be great. -DePiep (talk) 23:16, 21 September 2015 (UTC)
Try fluorine or yttrium. (The latter perhaps more relevant.) I have no doubt that there are many more. Double sharp (talk) 01:32, 22 September 2015 (UTC)

Graphic presentations[edit]

The essential two structures are:

  • Group 3 = Sc/Y/Lu/Lr
Periodic table overview (18-columns, group 3 = Sc-Y-Lu-Lr).svg
Hydrogen (diatomic nonmetal)
Helium (noble gas)
Lithium (alkali metal)
Beryllium (alkaline earth metal)
Boron (metalloid)
Carbon (polyatomic nonmetal)
Nitrogen (diatomic nonmetal)
Oxygen (diatomic nonmetal)
Fluorine (diatomic nonmetal)
Neon (noble gas)
Sodium (alkali metal)
Magnesium (alkaline earth metal)
Aluminium (post-transition metal)
Silicon (metalloid)
Phosphorus (polyatomic nonmetal)
Sulfur (polyatomic nonmetal)
Chlorine (diatomic nonmetal)
Argon (noble gas)
Potassium (alkali metal)
Calcium (alkaline earth metal)
Scandium (transition metal)
Titanium (transition metal)
Vanadium (transition metal)
Chromium (transition metal)
Manganese (transition metal)
Iron (transition metal)
Cobalt (transition metal)
Nickel (transition metal)
Copper (transition metal)
Zinc (transition metal)
Gallium (post-transition metal)
Germanium (metalloid)
Arsenic (metalloid)
Selenium (polyatomic nonmetal)
Bromine (diatomic nonmetal)
Krypton (noble gas)
Rubidium (alkali metal)
Strontium (alkaline earth metal)
Yttrium (transition metal)
Zirconium (transition metal)
Niobium (transition metal)
Molybdenum (transition metal)
Technetium (transition metal)
Ruthenium (transition metal)
Rhodium (transition metal)
Palladium (transition metal)
Silver (transition metal)
Cadmium (transition metal)
Indium (post-transition metal)
Tin (post-transition metal)
Antimony (metalloid)
Tellurium (metalloid)
Iodine (diatomic nonmetal)
Xenon (noble gas)
Caesium (alkali metal)
Barium (alkaline earth metal)
Lanthanum (lanthanide)
Cerium (lanthanide)
Praseodymium (lanthanide)
Neodymium (lanthanide)
Promethium (lanthanide)
Samarium (lanthanide)
Europium (lanthanide)
Gadolinium (lanthanide)
Terbium (lanthanide)
Dysprosium (lanthanide)
Holmium (lanthanide)
Erbium (lanthanide)
Thulium (lanthanide)
Ytterbium (lanthanide)
Lutetium (lanthanide)
Hafnium (transition metal)
Tantalum (transition metal)
Tungsten (transition metal)
Rhenium (transition metal)
Osmium (transition metal)
Iridium (transition metal)
Platinum (transition metal)
Gold (transition metal)
Mercury (transition metal)
Thallium (post-transition metal)
Lead (post-transition metal)
Bismuth (post-transition metal)
Polonium (post-transition metal)
Astatine (metalloid)
Radon (noble gas)
Francium (alkali metal)
Radium (alkaline earth metal)
Actinium (actinide)
Thorium (actinide)
Protactinium (actinide)
Uranium (actinide)
Neptunium (actinide)
Plutonium (actinide)
Americium (actinide)
Curium (actinide)
Berkelium (actinide)
Californium (actinide)
Einsteinium (actinide)
Fermium (actinide)
Mendelevium (actinide)
Nobelium (actinide)
Lawrencium (actinide)
Rutherfordium (transition metal)
Dubnium (transition metal)
Seaborgium (transition metal)
Bohrium (transition metal)
Hassium (transition metal)
Meitnerium (unknown chemical properties)
Darmstadtium (unknown chemical properties)
Roentgenium (unknown chemical properties)
Copernicium (transition metal)
Ununtrium (unknown chemical properties)
Flerovium (post-transition metal)
Ununpentium (unknown chemical properties)
Livermorium (unknown chemical properties)
Ununseptium (unknown chemical properties)
Ununoctium (unknown chemical properties)
Corresponding 18-column graphs: designated 14LaAc (right, and {{this}})
  • Group 3 = Sc/Y/La/Ac
Periodic table overview (18-columns, group 3 = Sc-Y-La-Ac).svg
Hydrogen (diatomic nonmetal)
Helium (noble gas)
Lithium (alkali metal)
Beryllium (alkaline earth metal)
Boron (metalloid)
Carbon (polyatomic nonmetal)
Nitrogen (diatomic nonmetal)
Oxygen (diatomic nonmetal)
Fluorine (diatomic nonmetal)
Neon (noble gas)
Sodium (alkali metal)
Magnesium (alkaline earth metal)
Aluminium (post-transition metal)
Silicon (metalloid)
Phosphorus (polyatomic nonmetal)
Sulfur (polyatomic nonmetal)
Chlorine (diatomic nonmetal)
Argon (noble gas)
Potassium (alkali metal)
Calcium (alkaline earth metal)
Scandium (transition metal)
Titanium (transition metal)
Vanadium (transition metal)
Chromium (transition metal)
Manganese (transition metal)
Iron (transition metal)
Cobalt (transition metal)
Nickel (transition metal)
Copper (transition metal)
Zinc (transition metal)
Gallium (post-transition metal)
Germanium (metalloid)
Arsenic (metalloid)
Selenium (polyatomic nonmetal)
Bromine (diatomic nonmetal)
Krypton (noble gas)
Rubidium (alkali metal)
Strontium (alkaline earth metal)
Yttrium (transition metal)
Zirconium (transition metal)
Niobium (transition metal)
Molybdenum (transition metal)
Technetium (transition metal)
Ruthenium (transition metal)
Rhodium (transition metal)
Palladium (transition metal)
Silver (transition metal)
Cadmium (transition metal)
Indium (post-transition metal)
Tin (post-transition metal)
Antimony (metalloid)
Tellurium (metalloid)
Iodine (diatomic nonmetal)
Xenon (noble gas)
Caesium (alkali metal)
Barium (alkaline earth metal)
Lanthanum (lanthanide)
Cerium (lanthanide)
Praseodymium (lanthanide)
Neodymium (lanthanide)
Promethium (lanthanide)
Samarium (lanthanide)
Europium (lanthanide)
Gadolinium (lanthanide)
Terbium (lanthanide)
Dysprosium (lanthanide)
Holmium (lanthanide)
Erbium (lanthanide)
Thulium (lanthanide)
Ytterbium (lanthanide)
Lutetium (lanthanide)
Hafnium (transition metal)
Tantalum (transition metal)
Tungsten (transition metal)
Rhenium (transition metal)
Osmium (transition metal)
Iridium (transition metal)
Platinum (transition metal)
Gold (transition metal)
Mercury (transition metal)
Thallium (post-transition metal)
Lead (post-transition metal)
Bismuth (post-transition metal)
Polonium (post-transition metal)
Astatine (metalloid)
Radon (noble gas)
Francium (alkali metal)
Radium (alkaline earth metal)
Actinium (actinide)
Thorium (actinide)
Protactinium (actinide)
Uranium (actinide)
Neptunium (actinide)
Plutonium (actinide)
Americium (actinide)
Curium (actinide)
Berkelium (actinide)
Californium (actinide)
Einsteinium (actinide)
Fermium (actinide)
Mendelevium (actinide)
Nobelium (actinide)
Lawrencium (actinide)
Rutherfordium (transition metal)
Dubnium (transition metal)
Seaborgium (transition metal)
Bohrium (transition metal)
Hassium (transition metal)
Meitnerium (unknown chemical properties)
Darmstadtium (unknown chemical properties)
Roentgenium (unknown chemical properties)
Copernicium (transition metal)
Ununtrium (unknown chemical properties)
Flerovium (post-transition metal)
Ununpentium (unknown chemical properties)
Livermorium (unknown chemical properties)
Ununseptium (unknown chemical properties)
Ununoctium (unknown chemical properties)
Corresponding 18-column graphs: designated 14CeTh (right, and {{this}})

These are the two PT structures for the current development (User:Sandbh/sandbox). I'll create two large-cell cutouts with more info per cell: name, Z, symbol and electron configuration (useful?). -DePiep (talk) 10:07, 21 September 2015 (UTC)

Yeah, electron configuration is kind of the point here. Sc = [Ar]3d14s2; Y = [Kr]4d15s2; La = [Xe]5d16s2; Ac = [Rn]6d17s2; Lu = [Xe]4f145d16s2; Lr = [Rn]5f147s27p1. Double sharp (talk) 10:11, 21 September 2015 (UTC)
OK. -DePiep (talk) 10:32, 21 September 2015 (UTC)
  • Recap (check me): A. for the topic at hand, we have the two PT structure variants as shown here. No other variants are involved. The grapic look will change, but not the PT structures. B. Later on, we can decide on which variant shall be our main general presentation (today it is Sc/Y/Lu/Lr; de:wiki has the other one). C. Even later, when this is stable content, we can consider inserting gaps consistently in our PT's, like changing this one. -DePiep (talk) 10:32, 21 September 2015 (UTC)
    • A. Yes. We need to mention the confusing IUPAC one as well (Sc/Y/*/**), as it has been used (unfortunately), but I believe we can just show it in 18-column and note that it does not make clear how the lanthanides and actinides are meant to fit in. We will not be using that one generally, then.
    • B. Agree. Yes Sc/Y/Lu/Lr is for now the main if we judge by the contents of {{Periodic table}}: de.wiki consistently applies Sc/Y/La/Ac. (I imagine because that's what Holleman and Wiberg do.)
    • C. Agree of course. Double sharp (talk) 10:43, 21 September 2015 (UTC)
Interesting, even today. -DePiep (talk) 22:52, 26 November 2015 (UTC)
Yep. -DePiep (talk) 21:21, 9 December 2015 (UTC)

The IUPAC immature drawing[edit]

New subsection for this. I don't want this IUPAC PT variant, fringe option, hash to interfere. -DePiep (talk) 22:01, 21 September 2015 (UTC)

"We need to mention the confusing IUPAC one as well" -- noooo, please kill that gem from now on. It is not even one of the 700 true PT variants that exist, not even fringe opinion, just a very sloppy one. Where its it mentioned that we would have to mention it at all? -14:48, 21 September 2015 (UTC)
We are here to represent the current science, not to build a new one. That's the main idea of Wiki.--R8R (talk) 14:55, 21 September 2015 (UTC)
We have to mention it, because it is in use by a significant number of people. You don't have to like it – I don't either – but that cannot get in the way of documenting what has been used. Double sharp (talk) 15:06, 21 September 2015 (UTC)
DS: "in use by a significant number of people" -- I object, and will do so by capslock. The IUPAC PT image is not part of the group 3-issue at hand. At all. It should not be in the Sandbh/sandbox prose. Doing the two variants correct, we are fine here at wiki. (And then again and again I ask, Double sharp: actually WHAT FORM or INTERPRETATION of that iupac PT you refer to?). -DePiep (talk) 21:50, 21 September 2015 (UTC)
And yet it is in Sandbh's sandbox, in the last paragraph, as it should be. I refer to the interpretation of it as Sc/Y/*/**, i.e. with every lanthanide and actinide being somehow a group 3 element; there does not seem to be any other common one. This is mentioned by Jensen, who calls it "15LaAc", and of course criticizes it roundly (as 32-columnizing it means that the Sc and Y cells have to stretch 15 columns). (He calls Sc/Y/Lu/Lr "14LaAc" and Sc/Y/La/Ac "14CeTh", referring to the number of columns in the resultant f-block portrayal, and then its first column.) Double sharp (talk) 01:23, 22 September 2015 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────(outdent) These are issues of interpretation and how to frame the issue. Jensen's letter frames the issue as "the representation of the f-block elements." With that framing, placing 15 elements in the bottom section is ridiculous. You and many others frame the issue as "every lanthanide and actinide being somehow a group 3 element." And Jensen, Scerri, and many others have framed the issue in terms of either uncertainty or the need to "stretch boxes" to make a 32 column table. My view is that the table used at IUPAC exists as a graphic on its own merits with no explicit interpretation about blocks, groups, adding columns, or anything else. Its lexical utilitarianism (* = lanthanides, ** = actinides) suits me, so it's my preference, but that's not a strong preference, and I never meant to express derision toward Sc/Y/Lu/Lr ("...it ought to keep the FJ-ish derision towards Sc/Y/Lu/Lr out..."). I did mean to express derision about talk page discussions and literature survey usage that isn't reader-focused. An example would be this current talk page subsection. An editor created a talk page subsection in order to declare that it is not part of the issue at hand. Knowledgeable people looking for volunteer opportunities avoid communities when that sort of thing happens. Flying Jazz (talk) 23:41, 3 October 2015 (UTC)

re "These are issues of interpretation and how to frame the issue". What 'framing' are you talking about? We already have the issues "scientific PT" (like: Mendeleev, Mendeleev +, Jane, Adomah) and the graphic forms like 18-column or 32-column. Now you really want to introduce a third dimension of confusion called "frame"? Please explain & convince. Oh, and if you care to reply: stop using words like 'ridiculous'. That's not an argument. -DePiep (talk) 00:36, 4 October 2015 (UTC)
@DePiep:, I think that by "framing", Flying Jazz means that there are different ways of presenting a problem (how the PT should look) that are "framed" or presented by what question you ask. YBG (talk) 02:16, 4 October 2015 (UTC)
If so, he should have said so. -DePiep (talk) 22:28, 7 October 2015 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── The IUPAC PT expanded into 32-col form looks like #Sc/Y/*/** in 32-column form. It is undeniable that the asterisks for the 2 × 14/15 elements are under the header of "(group) 3". That is incorrect or fringe. -DePiep (talk) 22:19, 8 October 2015 (UTC) Note to myself: keep in mind and bring as argument when needed: 3 = Sc/Y/Lu/Lr emphasizes the f-block structure in the whole, great. 3 = Sc/Y/La/Ac makes f-bock splitting d-block? How unelegant to misform the PT! btw, how would 3 = Sc/Y/La/Ac look in a Left Step PT? -DePiep (talk) 22:32, 8 October 2015 (UTC)

If I wanted a left-step Sc/Y/La/Ac PT, I'd take the left-step Sc/Y/Lu/Lr PT and cut out Sc, Y, La, and Ac. Then I'd shift them all over to the right, with Sc/Y/La/Ac/E121 to the right of Ca/Sr/Ba/Ra/E120. Probably I would also move Ce and Th to the right as well, so that La/Ac/E121 would neighbour Ce/Th/E122. Double sharp (talk) 14:30, 13 October 2015 (UTC)

Suggestion that PT grouping applies to footnote elements[edit]

New section, discussion started in previous thread
A 14CeTh periodic table.jpg

So. This PT graph (see right) is a recognisable presentation of the periodic table (18-col form, 14CeTh for group 3 constitution). All fine for this.

Now Sandbh and Double sharp earlier promoted to:

position Ce right under Rf, to state and show that Ce is in group 4 too.

I oppose. My point is: if you state that, you state too the PT-groups for all the footnote elements:

PT claiming groups for f-block elements
Periodic table
Group 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
5
Rubid­ium
37
Stront­ium
38
Yttrium
39
Zirco­nium
40
Nio­bium
41
Molyb­denum
42
Tech­netium
43
Ruthe­nium
44
Rho­dium
45
Pallad­ium
46
Silver
47
Cad­mium
48
Indium
49
Tin
50
Anti­mony
51
Tellur­ium
52
Iodine
53
Xenon
54
6
Cae­sium
55
Ba­rium
56
Lan­thanum
57
1 asterisk
Haf­nium
72
Tanta­lum
73
Tung­sten
74
Rhe­nium
75
Os­mium
76
Iridium
77
Plat­inum
78
Gold
79
Mer­cury
80
Thallium
81
Lead
82
Bis­muth
83
Polo­nium
84
Asta­tine
85
Radon
86
7
Fran­cium
87
Ra­dium
88
Actin­ium
89
1 asterisk
Ruther­fordium
104
Dub­nium
105
Sea­borgium
106
Bohr­ium
107
Has­sium
108
Meit­nerium
109
Darm­stadtium
110
Roent­genium
111
Coper­nicium
112
Unun­trium
113
Flerov­ium
114
Unun­pentium
115
Liver­morium
116
Unun­septium
117
Unun­octium
118
group, stated (Red XN) 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 asterisk
Cerium
58
Praseo­dymium
59
Neo­dymium
60
Prome­thium
61
Sama­rium
62
Europ­ium
63
Gadolin­ium
64
Ter­bium
65
Dyspro­sium
66
Hol­mium
67
Erbium
68
Thulium
69
Ytter­bium
70
Lute­tium
71
1 asterisk
Thor­ium
90
Protac­tinium
91
Ura­nium
92
Neptu­nium
93
Pluto­nium
94
Ameri­cium
95
Curium
96
Berkel­ium
97
Califor­nium
98
Einstei­nium
99
Fer­mium
100
Mende­levium
101
Nobel­ium
102
Lawren­cium
103

Undisputed in this:

  • group 3 = Sc/Y/La/Ac

Questions about the PT-grouping claims for the f-block:

  • By what RS?
  • Why should enwiki make this a general feature (a common presentation)?
  • How are these group numbers/statements in any 32-col PT? (RS PT's, enwiki PTs, ..)
  • Does the switchover by Lu/Lr from group 17 to group 3 make sense?
Better: If I pick the (scientifically correct) group 3=Sc/Y/Lu/Lr structure, where does that leave Ce, re group number & footnote graphical position? And all the others in the footnote?
-DePiep (talk) 22:52, 24 November 2015 (UTC); (better:) -DePiep (talk) 20:54, 25 November 2015 (UTC)

(New section, discussion started in previous thread: as a side note, I'd like to say I don't like the desire to categorize a discussion into sharply outlined topics because it splits them and many questions are interrelated, and setting them apart makes maintaining/using this connection more difficult.)

From what I've read, I believe (if I am wrong, I'd like to be told that) Sandbh does not want to state cerium is group 4; rather he just wants to softly show their similarity by simply putting Ce and Th right under group 4, etc. (Not in a way that would unquestionably tell us something, but it would be a nice thing for an attentive audience.) Here's what it would look like for -La-Ac and -Lu-Lr:

Ba   La   *    Hf   Ta
Ra   Ac   **   Rf   Db

          *    Ce   ...
          **   Th   ...

and

Ba   *    Lu   Hf   Ta
Ra   **   Lr   Rf   Db

     *    La   Ce   ...
     **   Ac   Th   ...

--R8R (talk) 07:19, 26 November 2015 (UTC)

re "I'd like to say I don't like ...": well I do. I am academic in sciences, and I know in science &tc we treat independent topics independently. That is how we solve & explain things in science. If you can point out why & where these topics are related: please do so. But until today, the group 3-composition has nothing to do with graphing Ce in a group. They are independent. (And thank nature & our mind: that is the way we can understand those things!). No use or need to throw all topics into one soup. -DePiep (talk) 21:15, 26 November 2015 (UTC)
@R8R Gtrs: That is my (YBG's) understanding, also: it is how I understood what Sandbh (and Double sharp) wrote, which I have copied here from above.
  • @Sandbh: Please correct me if I have misuderstood you. You wrote on 00:28, 15 November 2015 (UTC) "The footnoted Ln and An line up nicely under the main body group numbers: Ce under group 4; the transition-metal-like earlier actinides---Th, Pa, U, Np, Pu---under groups 4, 5, 6, 7 and 8." (emphasis added) You later added on 23:58, 21 November 2015 (UTC) "The actinides do not have group numbers but lining up Th, Pa, U etc with Hf, Ta, W etc has a strong historical basis and is consistent with the early actinides showing some similarities to transition metals." In the first quotation, I believe that by under, you did NOT mean under in a membership sense, but rather in a spatial sense. Please correct me if I am wrong.
  • @Double sharp: Please correct me if I have misunderstood you. You answered Sandbh on 14:29, 22 November 2015 (UTC) "Exactly. And furthermore, if it truly does not matter where the footnote is, then they can very well be aligned this way in the 18-column table and not in the 32-column table." So I believe you also agree that the 'footnote elements' do not belong to any of the groups 1-18.
So, Double sharp & Sandbh, please correct me if I am wrong in my understanding that both of you believe that 'footnote elements' do not belong to any of the groups 1-18, but that it would be beneficial, all else being equal, if those elements could be placed physically below non-footnote elements with which they share some chemical similarities.
Again, please, if I have misunderstood what anyone has stated, please correct me. YBG (talk) 08:34, 26 November 2015 (UTC)
Yes, thank you, both of your understandings are correct. Sandbh (talk) 09:10, 26 November 2015 (UTC)
Yes, you're absolutely right. Double sharp (talk) 13:02, 26 November 2015 (UTC)
Nicely shows the issue at hand. YBG is rephrasing: "is group 4" into "but ... they share some chemical similarities", with Sandbh and Double sharp confirming. This introduces a new, secondary-group-thing. I don't deny those similarities exist, duh. Still, none of my questions are answered by now. Not one of four. To check our thinking process, I add: why would we add this information to the graph, while not actually confirming it (eg by repeating group number?). Only suggesting (I must assume it extends to all 14 footnote columns)? There are more stable facts that a general PT does not or can not show. -DePiep (talk) 20:57, 26 November 2015 (UTC)

This POV seems to be the same as R8R's and DePiep's, so, we add to our list of undisputed/resolved issues:

  1. group 3 = Sc/Y/La/Ac
  2. elements 58(Ce)–71(Lu) and 90(Th)-103(Lr) do not belong to any periodic table groups

YBG (talk) 16:13, 26 November 2015 (UTC)

I have struck out my post above as it seems to have caused some misunderstanding. Let me try to restate:
So it seems that we all agree that
  • elements 58(Ce)–71(Lu) and 90(Th)-103(Lr) do not belong to any periodic table group.
YBG (talk) 21:44, 26 November 2015 (UTC)

Why is this undisputed; I dispute and I have my arguments (see above on this page, although I can't give an exact link at the moment). Sandbh wants to summarize arguments for either version and analyze them; I intend to join him in this as I find time for this. My desire is not in pushing for -Lu-Lr, and I can imagine myself agreeing with -La-Ac; moreover, I abstained from the discussion until very late because I was rethinking the arguments, having no desire to advocate for either version. But I want to be confident this would be the right decision, and I'm leaning towards the idea it wouldn't be; but I'm open for any new ideas. I must admit, I am disappointed by how Sandbh adds a post (a reasonable one, of course; Sandbh's usual), and everyone just agrees with him, with no discussion taking place. I perfectly understand now why WP:Wikipedia is not a democracy. We all can take another breath and actually analyze the arguments.

No need to rush.--R8R (talk) 18:01, 26 November 2015 (UTC)

"Why is this undisputed;" - No, bad quote. I wrote: "Undisputed in this: "group 3=Sc/Y/La/Ac". That is: For this topic, we can assume that (sake of reasoning). Also, in my objection #4 I mentioned the alternative (whith possibly more devastating effects). How wold your reasoning be, with this? (actually, I can't follow it because of this). -DePiep (talk) 21:05, 26 November 2015 (UTC)
@DePiep: Thank you for clarifying. I had completely overlooked your phrase "in this", which I now understand better. Thanks! YBG (talk) 21:44, 26 November 2015 (UTC)
Thanks. m. You just won one bonus point: 24h of patience from me ;-). -DePiep (talk) 22:44, 26 November 2015 (UTC)

I must've made my thought not clear enough. "So it seems that we all agree that elements 58(Ce)–71(Lu) and 90(Th)-103(Lr) do not belong to any periodic table group." No, I don't agree. I have provided my rationale above, section "Let us proceed with Sc|Y|La|Ac." I have finally indulged myself into the discussion in great part because I was led by the famous saying, "From the clash of opinions emerges the truth." Sandbh did try to understand my position, and now the question is being revisited in detail. That said, I am open for new ideas, and Sandbh seems to be open for new ideas as well; that's great. That's what I would expect from an important discussion.

We may avoid this question in this section, but not yet say the matter is settled in general. I don't understand why you are in such a rush on this.--R8R (talk) 00:07, 27 November 2015 (UTC)

Misunderstanding, R8R Gtrs? This section is *not* about group 3 at all. It is about: do we show & state that "all Ln/An are in a group" (eg Ho is in group 13)? -DePiep (talk) 00:29, 27 November 2015 (UTC)
@R8R Gtrs: No, the problem is not that you didn't make yourself clear, but rather that I was too myopic to remember and not thoughtful enough to craft my words carefully. Let me try again to express more carefully what I believe we all agree on.
In the 3=Sc/Y/La/Ac PT arrangement being considered here, the footnoted elements 58(Ce)–71(Lu) and 90(Th)-103(Lr) are not considered to be a part of PT groups.
To this it seems one could add:
  • (a) nevertheless, all else being equal, it would be nice to place the footnoted elements physically underneath chemically similar ones.
  • (b) consequently, it is best to place the footnoted elements in such a way as to make it obvious that no group membership is implied.
In this, it seems that Sandbh and Double sharp would agree with (a) and DePiep would agree with (b). Am I right in this? I'm not quite sure which of these R8R Gtrs would agree with. YBG (talk) 02:15, 27 November 2015 (UTC)
Oh. I didn't realize that at first (nor at second). In that case, if I was to decide, I would choose the former option for the reasons already said, as I have little to add to what I already posted on the matter. The fact no membership is implied is best shown by some vertical space between period seven and the lanthanides.--R8R (talk) 10:07, 27 November 2015 (UTC)
Please, can we come home to my original Four Questions? (Fm is in group 14 then?) -DePiep (talk) 22:35, 27 November 2015 (UTC)
@DePiep: your questions are based on the following statement that you posted above:
Now Sandbh and Double sharp earlier promoted to:
position Ce right under Rf, to state and show that Ce is in group 4 too.
@Sandbh and Double sharp: does this statement accurately reflect your POV? YBG (talk) 05:34, 28 November 2015 (UTC)
No. Ce is not in group 4, although it shows some similarities to the group 4 elements. Sandbh (talk) 05:45, 28 November 2015 (UTC)
No. Okay...the main reason I support this, with Sandbh, is that while Ce is not in group 4 (nor indeed any group), it is more similar to the elements of group 4 than those of any other group. Same for Th, Pa, U, Np, and Pu in groups 4 to 8 inclusive. This also complements the similarity of the lanthanides and actinides to group 3 in general, as shown by the position of the asterisks (they come right after group 3). You cannot show both similarities in a 32-column table, but I think that if it is possible to show something, we should – especially since displacing these elements by a column or two aligns them under elements they are not closely related to. Fm in group 14 isn't true, but this is the only possible location for the f-block where at least some of these relationships are true. So it is a nice touch for the reader who knows this, and for the reader who doesn't, it doesn't really matter where the footnote is.
But TBH, this is a minor issue. I don't mind if it doesn't go through. (In fact I'm getting rather close to not really caring about any of this, save the question of group 12 as transition metals – which is a separate topic that I will not go into further here.) Double sharp (talk) 12:02, 28 November 2015 (UTC)
re Double sharp saying "in groups 4 to 8 inclusive": so the statement is explicitly not valid for all footnoted elements. The thing that is being illustrated is not to be read for elements Eu to Lu (under group 9 and up+). I'd say that is a misleading presentation then. -DePiep (talk) 18:39, 28 November 2015 (UTC)
Ce "shows some similarities to the group 4" - being true, isn't that a weak statement to be made in a general PT? In parallel, there are other statements like "Xx is considered a metalloid by multiple sources" is correct too, but we don't cram that in either. We'd have to add extra explaining legends or, er, footnotes. -DePiep (talk) 18:49, 28 November 2015 (UTC)
  • Apart from this all, there is a graphic bad issue with this group 4 (etc) suggestion. That is: this way, once the 18-col 14CeTh from is drawn, the CeTh column is tied to group four! One can not freely remove it (the whole footnote rectangle 30 elements), freely, from its position without breaking the original '(bad) intent'. Not into a 32-col PT! What is it: are the asterisks to be followed, or is the "4, ..." column numbering to be followed? My mind hurts. -DePiep (talk) 02:19, 2 December 2015 (UTC)
Extract of periodic table with lanthanides and actinides swapped to emphasize similarity in chemical behaviour of the early actinides with their period 6 transition metal cousins

Hey DePiep, here's a sketch (see right) of an 18 column periodic table, with Ac lined up with group 3, thorium and cerium with group 4 etc, that recently appeared in the Journal of Chemical Education (Hoffman 2009, p. 1125). The author notes, "Figure 5 from our 1999 paper shows a periodic table in which the actinides are pictured in a stairstep arrangement leading from Rf down to the trivalent actinides Am and Cm to portray graphically the similarities in chemical behavior between the early actinides and possible pseudo-homologues in groups 4 through 8—Th and Pa and to a lesser degree, U, Np, and Pu." In the 1999 paper in the same journal Hoffman (and Lee, p. 334) wrote: "Recently, periodic tables similar to the one shown in Figure 4 [i.e. figure 5 in Hoffman's 2009 paper] have been proposed (11–13)." These earlier reference are 11. Herrmann, G. Nucl. Phys. News 1998, 8, 7; 12. Schädel, M.; Brüchle, W.; Dressler, R.; Eichler, B.; Gäggeler, H. W.; Günther, R.; Gregorich, K. E.; Hoffman, D. C.; Hübener, S.; Jost, D. T.; Kratz, J. V.; Paulus, W.; Schumann, D.; Timokhin, S.; Trautmann, N.; Türler, A.; Wirth G.; Yakuschev, A. Nature 1997, 388, 55; and 13. Schädel, M.; Brüchle, W.; Schausten, B.; Schimpf, E.; Jaeger, E.; Wirth, G.; Guenther, R.; Kratz, J. V.; Paulus, W.; Seibert, A.; Thoerle, P.; Trautmann, N.; Zauner, S.; Schumann, D.; Andrassy, M.; Misiak, R.; Gregorich, K. E.; Hoffman, D. C.; Lee, D. M.; Sylwester, E. R.; Nagame, Y.; Oura, Y. Radiochim. Acta 1997, 77, 149. Lining up actinides with their transition metal psuedo-homologues is non-controversial. Nobody is taking this to mean that the lanthanides or the actinides have individual group numbers. Sandbh (talk) 10:14, 10 December 2015 (UTC)

  • Hoffman DC 2009, 'The Periodic Table Key to Past "Elemental" Discoveries—A New Role in the Future?', Journal of Chemical Education, vol. 86, no. 10, pp. 1122–1128
  • Hoffman DC & Lee DM 1999, 'Chemistry of the Heaviest Elements—One Atom at a Time', Journal of Chemical Education, vol. 76, no. 3, pp. 331–347
Great! Thanks for this Hey DePiep, here's a sketch. Still, what you claim is OR and not RS. Could be Fringe. Bye. -DePiep (talk) 22:52, 11 December 2015 (UTC)
This is not OR as it has been sourced from reliable, peer reviewed sources. Positioning the Ln and Ac so that Ce-Th are aligned with, but separate from, Ti-Zr-Hf-Rf is quite common and clearly not fringe. Even the dreaded table appearing on the IUPAC site features such an alignment. Sandbh (talk) 03:42, 12 December 2015 (UTC)
I did not contest or deny (ever) the claims Hoffman (2009) makes wrt these similarities in behaviour. And she even has them graphically supported correctly (text and graph correspond, great!).
I do claim that (1) adding this information to the graph is way too detailed (unimportant) for our general purpose PT. Of course they can be described in some dedicated article or section, maybe we'll see Hoffman there too. (2) If we want to add this info, it must be done graphically correct. i.e., the graph must explicitly show the data correct. The Hoffman graph does so. Hoffman also has the feature that it does not use the asterisks, so it does not point to replacement (footnote elements in the main body ie into 32-col). This way she prevents the bad issue that the graph would have contradicting features: does Ce belong below group 4 by position or somewhere between group 2 and 4 by asterisks? But the proposed PT does have this ambivalence and there for is graphically incorrect (elements are taken out into a footnote, and then the footnote is tied to a different position wrt the main table). Also, the proposed PT does not have the stairs. So it does not differentiate between Ln's, nor between Ln-An in a column (Hoffman does). About the IUPAC PT: why do you conclude that they want to convey this information? How is that put into their PT? What does it say about the gradual change of fact both in period and in group (column) for these? So far, IUPAC only happens to show it this way. And, not unimportant, for other reasons other topics their PT already is dismissed. -DePiep (talk) 12:35, 14 December 2015 (UTC)
TL;DR Given the undisputed scientific facts, the Hoffman graph has everything right. Transplanting any of her graphic features into the our general (18-column only) PT is wrong always. -DePiep (talk) 09:42, 19 December 2015 (UTC)

Group 3 as (possibly borderline) transition metals[edit]

Can we make a case for this to rescue transition metal from its current problems, to get it behind the reasonable IUPAC definition (like Cotton and Wilkinson) of groups 3-11? Double sharp (talk) 03:59, 24 December 2015 (UTC)

On a quick look, the article seems to give a more or less balanced overview; what problems do you have in mind?--R8R (talk) 04:58, 24 December 2015 (UTC)
It keeps using the 4-11 definition as default, like in the first section ("eight groups") and its criticism of including group 3 (either Sc/Y/La/Ac or Sc/Y/Lu/Lr - I prefer the latter now) based on their behaviour as catalysts. All fine and well, but IUPAC demands you include group 3, so is there a way to refute the arguments against it? One of them is that the +3 state dominates their chemistry (um, yeah, but it's also true for Zr, Hf, Nb, and Ta with their group oxidation states, and lower states are shown at standard conditions - IIRC, CsScCl3 is an example), and another is their behaviour as catalysts. Double sharp (talk) 14:53, 24 December 2015 (UTC)

P.S. If you ask me about the ion thing, BTW, I'm uncomfortable with it. If you look at Li, for example, its chemistry is utterly dominated by the Li+ ion where it gives up its only 2s electron to get the helium configuration of 1s2. That does not make it a period 1 element because its ions have no electrons with n = 2. The electron that participates in reactions is a 2s electron! So for Sc, the fact that Sc3+ dominates its chemistry just means that its 3d electron has participated in its chemistry. It's not like Zn or Ga where nothing is going to let the d-electrons out. So if our idea of a transition element is one that uses electrons that are not from the outermost shell, then we have the 3-11 definition, where the f-block become inner transition elements as they give up their s-electrons and at least a single d- or f-electron.

So a reasonable definition might be that an element is a transition element iff it is in the d-block and has a chemically contributing inner shell (e.g. Sc turnings combusting to form the sesquioxide), and an element is an inner transition element iff it is in the f-block and has a chemically contributing inner shell (yes, I phrased it to take care of annoying cases like La, Gd, Ac, and Th, so that the d-electrons would count). This would mean that Lu is a lanthanide, but not an inner transition metal, but La is both (my excuse is that 4f is playing a role in La, whereas in Lu it is part of the core). There are thus 30 lanthanides and actinides (La-Lu, Ac-Lr) but only 28 inner transition elements (La-Yb, Ac-No).

All of this must be applied at standard conditions, of course, or we get absurdities like claiming K, Rb, and Cs to be transition metals.

The only trouble is that this bars Lr from TM status, because 7p is not an inner shell then. Troublesome! A possible way out is if the 6d configuration is a very low excited state: given the uncertainty in determining the correct configuration, this is probably the case. Then we do indeed have a chemically contributing 6d subshell that just happens to be empty in the ground state, like the case of 4f in La and 5f in Ac and Th.

The superheavy elements Cn, E113 (predicted), and Fl are PTMs until someone actually sees their 6d subshells breached. Until then, all we have are physical properties, which are in line with their groups and thus give a verdict of PTM.

Hence suggested colouring: same as now, but with Sc/Y/Lu/Lr as group 3, and with all of group 12 (Zn, Cd, Hg, and Cn) listed as PTMs. (Since we colour Ln/An and not ITM, and this is my OR definition after all, Lu and Lr are still coloured as a lanthanide and actinide respectively, but they would also be considered d-block elements.)

We would consider Lu and Lr to be the heavier members of group 3 by default, i.e. in all contexts except different ways to draw the periodic table. This is fine as all three versions are supported by numerous reliable sources, and we have to pick one. We might as well pick the best of them! ^_^

We also ought to do this sort of thing for the placement of H and He in another thread to determine our defaults. Although I suppose I can take the opportunity to say that because we are looking at standard conditions, putting He anywhere except over Ne is ludicrous for a default. Double sharp (talk) 15:19, 24 December 2015 (UTC)

P.S. I can't believe I forgot to define the blocks. Okay, here goes: an element is in the x-block (x = s, p, d, f, or g) if the outermost x-orbitals are (expected to be) chemically active in it. The precedence rules are: p, g, f, d, s. Hence La is in the f-block since 4f and 5d are both active and f precedes d. E113 is in the p-block since p beats everything: the same is true for C with both 2s and 2p. That covers the majority of cases, but falls down for group 12 as well as He and Ne. We presumably want group 12 in the d-block, as well as He in the s-block and Ne in the p-block; this puts group 12 in the s-block to get a bifurcating group 2 past Be, and there is no way to assign He and Ne to blocks until their nobility collapses. Double sharp (talk) 15:27, 24 December 2015 (UTC)

P.P.S. Oops. Lr. Actually it's not possible to create a precedence rule such that both Lr and E113 work because you need p to both be before d (for E113) and after d (for Lr). Obviously this is impossible, so here goes half my idea. I think my definitions of TM and ITM work, though. Double sharp (talk) 15:44, 24 December 2015 (UTC)

Scratch that: we are fine since 6d contributes to Lr's chemistry. Now only group 12, He, and Ne really create problems. Double sharp (talk) 21:18, 24 December 2015 (UTC)
Thanks for the Xmas gift. I'm slowly writing a response to your concerns about the transition metal article's focus on groups 4–11. There is more to this than what may be apparent. In the literature, the group 3 elements—and whether to treat them as d-block elements, transition metals or the other kind of transition metals—are the source of much categorical hand waving and contortion. Hope to post something soon. Sandbh (talk) 07:01, 5 January 2016 (UTC)

The group 4–11 focus[edit]

Abstract
In response to Double sharp's concerns, this post discusses the focus of the transition metal article on groups 4–11, and why this is not at odds with IUPAC or the literature. I conclude that colouring Sc and Y as transition metals is misguided since this is based on electron configurations rather than overall chemical, metallurgical and physical properties, and it Is the latter which informs how we colour categorise each element. (I count myself as having been misguided on this point until I thought at length about the issues raised by Double sharp.) I contend that Sc, Y and the lanthanides would be more relevantly colour coded as either "rare earths", "rare earth metals", "rare earths (lanthanides 57–71)" or "rare earth metals (lanthanides 57–71)".

Our transition metal article
I think the article uses the 4–11 definition because (a) Wikipedia categorizes only half of group 3 as transition metals; and (b) IUPAC says that group 12 elements are not always counted as transition metals. In these two senses the 4–11 definition is safe. The other sentence that uses the 4–11 definition is the second one in the Classification section, the one that says: "The elements of groups 4–11 are now generally [italics added] recognized as transition metals, justified by their typical chemistry, ie large range of complex ions in various oxidation states, coloured complexes and catalytic properties either as the element or as ions (or both)." Given the "generally" qualifier, this is a reasonably accurate statement.

A group 4–11 focus is not necessarily at odds with IUPAC as they are only definitive on d-block elements. The Red Book (2005 p. 51) says: "The elements of groups 3–12 are the d-block elements." That's definitive. And then it goes on to say, "These elements are also commonly referred to as the transition elements, though the elements of group 12 are not always included…". That's not definitive.† The d-block elements certainly are commonly referred to as transition elements but as well as group 12 not always being included, neither are group 3 or group 11 always counted as transition metals.

† IUPAC had earlier (in 1990) defined a transition element as "An element whose atom has an incomplete d sub-shell, or which can give rise to cations with an incomplete d sub-shell" i.e. groups 3–11. This guidance was superseded by the 2005 recommendation although it still appears in IUPAC's online Compendium of Chemical Terminology (The Gold Book).

Group 3 in the literature
The literature has a hard time dealing with group 3 as transition metals. More often than not authors have to either [1] mention that the group 3 elements don't characteristically behave like transition metals; or [2] treat them in the same vein as the lanthanides; or [3] deny they are transition metals in the first place.

Earnshaw and Harrington (1973, p. 52), say that it is group 4, rather than group 3, "in which the really characteristic transitional properties of variable oxidation state, color and paramagnetism are encountered."

Gschneidner (1975, p. 76) says of scandium that it…

"…is the lightest of the rare earth group of elements and is the first transition metal in the periodic table [i.e. in the electronic sense]…In its general chemical, metallurgical and physical behaviours it is similar to those observed for other rare earth metals, but there are a number of significant differences. These are in part due to its smaller size (5.4% smaller than the next larger rare earth, lutetium) and its larger electronegativity value (1.28 compared to 1.12–1.22, for the other trivalent rare earths)."

There is nothing particularly surprising about the fact that scandium exhibits some differences when compared to its heavier congeners. This is a reflection of what Jensen (1986, p. 506) describes as…

"…a systematic variation in the periodic table which shows that the elements in the first row of any new electronic block tend to show abnormalities relative to the elements in later rows of the same block, and that the degree of divergence decreases in the order s-block>> p-block > d-block > f-block…C–F are quite distinct compared to other p-block elements, and the same is true to a lesser degree for Sc–Zn relative to the heavier d-block elements."

Neither scandium nor yttrium meet the profile given by F. A. Cotton (2000, p. 1961):

"The chemistry of the transition elements is differentiated from that of the other (so-called main group) elements in several ways, of which the following three are perhaps most important.

(1) The transition elements typically form compounds in two or more oxidation states, and redox chemistry, including electrochemistry, is of major importance.

(2) The majority of transition element compounds have visible spectra (which are why they are colored) and the interpretation of these spectra provides a wealth of information concerning their electronic structures. A classic example is provided by the spectra of tetrahedral and octahedral complexes of cobalt(II), as shown in Fig. 2.

(3) A great many transition element compounds have one or more unpaired electrons and therefore have interesting and often useful magnetic properties. These magnetic properties range from simple Curie paramagnetism to those associated with high-temperature superconductivity."

Greenwood and Earnshaw (2002, p. 946), refer to the group 3 elements (Sc, Y, La, Ac) as having properties that might be expected for elements immediately following the strongly electropositive alkaline-earth metals and preceding the transition elements proper [italics added]". That is, they don't regard the group 3 elements as "proper" transition metals. A little earlier (p. 958) they say that, "whatever arguments may be advanced against the description to Sc, there is no doubt that Ti is a "transition metal".

In their methodical survey of structure-property relations in nonferrous metals, Russell and Lee (2005) proceed from left to right across groups 1–2 and 4–16 of the periodic table. They skip the group 3 metals as these are included with the lanthanides.

Simon Cotton, a noted author on lanthanide chemistry, has a few words to say about scandium as a supposed transition metal:

(1) "The first 3d metal, scandium is not a transition metal, but has significantly different properties, partly on account of its slightly greater size, forming complexes with higher coordination numbers (e.g. the aqua ion [Sc(H2O)7]3+). Its developing chemistry is that of a slightly smaller version of lutetium." (2011)

(2) With a colleague, he again criticizes its treatment as a transition metal: "More still needs to be known about scandium chemistry; it is still too often regarded as a 3d transition metal even though it is now clear that it is quite unlike 3+ [sic] transition metal species." (Cotton & Harrowfield 2012).

(3) He observes that the Group 3 elements are "frequently" treated together with the lanthanides (2006, p. 107).

Rayner-Canham and Overton (2006, p. 484–485) write that, "Although some people use the terms d-block elements and transition metals interchangeably, this is not strictly correct. Inorganic chemists generally restrict the term transition metal to an element that has at least one simple ion with an incomplete outer set of d electrons." According to these authors, the elements commonly considered as transition metals are found in groups 4–11. They discuss the group 3 elements and lanthanides together in another chapter, given resemblances in the chemistry of the two sets of elements.

In the case of our nearest "competitor", The Encylopædia Britannica Online, their entry for transition metals (F. A. Cotton 2014) says:

"Because scandium, yttrium, and lanthanum actually do not form compounds analogous to those of the other transition elements and because their chemistry is quite homologous to that of the lanthanoids, they are excluded from the present discussion of the main transition elements. Similarly, because zinc, cadmium, and mercury exhibit few of the properties characteristic of the other transition elements, they are treated separately (see zinc group element)."

Britannica's own 18-column colour-coded periodic table (2010) categorizes the group 3 metals as rare earth elements [!] while still noting that the lanthanides run from 57–71.

The 48 volumes of The Handbook on the Physics and Chemistry of Rare Earths (1978–2015+), which has Sc, Y and the lanthanides within its scope (Elsevier 2016), runs to more than 25,000 pages and is the longest continuously running series of books on any periodic table category that I'm aware of.

Conclusion
Of the d-block elements, only those in groups 4–11 have a well established reputation for exhibiting the important characteristics of transition metals. In the literature, the group 3 metals Sc and Y are routinely compared or treated with the lanthanides in terms of their chemical, metallurgical and physical properties.

Sure, when scandium is counted as a transition metal, this is rightly done in light of its electron configuration. And this needs to be noted in our transition metal article.

But the way we categorise the elements in our periodic table is not primarily based on electron configurations. Rather, we have a categorization scheme that is informed by overall chemical, metallurgical and physical properties. This needs to made clear in the caption to the periodic table in the lede, as per the example.

18-column periodic table, with the elements color coded according to their overall chemical, metallurgical and physical properties [Sc, Y and lanthanides shown as rare earth metals]

Viewed this way, the transition metal colouring of scandium and yttrium in our periodic table is damned incongruous. Scandium is more accurately coloured as a rare earth metal (which is an IUPAC approved category), along with yttrium and the lanthanides. This would not mean that scandium and yttrium are not transition metals; it only means that in the case of our colour-coded periodic table we choose to show the more representative category.

Similar to The Encyclopedia Britannica Online table, our rare earth metals category could read, "Rare earth metal (lanthanides 57–71)" or "Rare earths (lanthanides 57–71)" as I think it's worth retaining some mention of the lanthanides.

Actinium should remain coloured as an actinide. I mention this in the event that you're wondering why it shouldn't be coloured as a rare earth metal given it's similar to lanthanum in its comportment. Actinium and the actinides are worth retaining as a separate category in recognition of the radioactivity that characterizes the entire series, and the chemical effects this can induce in solids and solutions (Nash & Braley 2001, p. 13). Actinium, in particular, is an intensely radioactive metal, so much so that this contributes to its reactivity and makes it difficult to study in even milligram amounts (Katz & Seaborg 1957, p. 11; Cotton et al. 1999, p. 1142). The relativistic effects that become increasingly prominent in this part of the periodic table (Dholabhai 2008, p. 2) are another consideration.

References

  • Connelly NG, Damhus T, Hartshorn RM & Hutton AT 2005, Nomenclature of Inorganic Chemistry: IUPAC Recommendations 2005, RSC Publishing, Cambridge
  • Cotton FA, Wilkinson G, Murillo CA & Bochmann M 1999, Advanced Inorganic Chemistry, 6th ed., John Wiley & Sons, New York
  • Cotton FA 2000, "A millennial overview of transition metal chemistry", Journal of the Chemical Society, Dalton Transactions, pp. 1961–1968
  • Cotton FA 2014, Transition element, in Encyclopædia Britannica Online,
  • Cotton S 2006, Lanthanide and Actinide Chemistry, John Wiley & Sons, New York
  • Cotton S 2011, "Scandium, Yttrium & the Lanthanides: Inorganic & Coordination Chemistry", in Encyclopedia of Inorganic and Bioinorganic Chemistry, John Wiley & Sons, New York
  • Cotton S & Harrowfield JM 2012, "Lanthanides: Coordination chemistry", in DA Attwood, The Rare Earth Elements: Fundamentals and Applications, John Wiley & Sons, Chichester
  • Dholabhai PP 2008, "Probing the 5f electrons: A relativistic DFT study of americium surfaces", PhD thesis, University of Texas at Arlington, Proquest, Ann Arbor
  • Earnshaw A & Harrington TJ 1973, The Chemistry of the Transition elements, Clarendon Press, Oxford
  • Elsevier BV 2016, "Publisher Summary", Handbook on the Physics and Chemistry of Rare Earths, vol. 1
  • Greenwood NN & Earnshaw A 2002, Chemistry of the Elements, 2nd ed., Butterworth Heinemann, Oxford
  • Gschneidner Jr. KA 1975, "Physical metallurgy", in CT Horovitz et al. (eds), Scandium: Its Occurrence, Chemistry, Physics, Metallurgy, Biology and Technology, Academic Press, London, pp. 76–110
  • Jensen WB 1986, "Classification, Symmetry and the Periodic Table", Computers & Mathematics with Applications, vol. 12B, no. I/2, pp. 487–510
  • Katz JJ & Seaborg GT 1957, The Chemistry of the Actinide Elements", Methuen, London
  • Nash KL & Braley JC 2001, "Chemistry of radioactive materials in the nuclear fuel cycle", in KL Nash & GJ Lumetta (eds) 0000, Advanced Separation Techniques for Nuclear Fuel Reprocessing and Radioactive Waste Treatment, Woodhead Publishing, Oxford, pp. 3‒22
  • Rayner-Canham G & Overton T 2006, Descriptive Inorganic Chemistry, 4th ed., WH Freeman, New York, pp. 484–485
  • Russell AM & Lee KL 2005, Structure Property Relations in Nonferrous Metals, John Wiley & Sons, New York
  • The Red Book—see Connelly et al.

Sandbh (talk) 02:20, 13 January 2016 (UTC)


This is an interesting posting, indeed. I want to note a couple of points re group 3 (i.e., not a substantial re):

  • From Greenwood, another quote would be this: "Because of this, although each member of this group is the first member of a transition series, its chemistry is largely atypical of the transition elements." So, they do not question, while mentioning that some others do, that group 3 also belongs to the transition metals set. I think the word "proper" mostly relied not on properties alone, but also on group 3 being split from the rest of the d block, as they use the Sc-Y-La-Ac group 3 layout.
  • Cotton's arguments are not criteria; that is concluded from words "typically" and so on that he uses. As they stand, they're not universal: the multiple OS criterion does not apply to neighboring zirconium, and zirconium also commonly forms colorless/white compounds; both Y and Zr are paramagnetic; and this falls in line with the idea of how transition elements represent an actual transition, with group 3 standing at the very beginning of it.
  • Viewed this way, the current coloring matches the profile of the elements in question. Another point, which plays a greater role to me, is, the proposed scheme is not as user-friendly, as I will describe below. It must be remembered that Wikipedia is oriented towards a wide audience, far wider than the G&E book, for example. Since we (suppose) have two possible variants, the current scheme and the proposed scheme, and accuracy is considered fulfilled for both, it may be a better reason to stay with the current coloring because it is easier to understand. "This would not mean that scandium and yttrium are not transition metals" -- for many readers, it would mean exactly this. Many won't read; some will, and won't understand it. The current scheme has a more subtle problem, but the fact two group 3 elements are TMs and two aren't may ignite some thinking. We still have a very common category for the 4f elements (lanthanides). That said, the current scheme is okay. The proposed scheme, apart from its advantages, also has its great disadvantage: it visually unquestionably excludes group 3 from the TM series. Wider audience does not mean we have to simplify our material; but we need to make an editorial decision either way and both alternatives are correct. In such a situation, I think the current scheme is preferred; but I'd like to hear your reception to this thesis.--R8R (talk) 12:04, 14 January 2016 (UTC)
    • Actually I think I'm with Sandbh this time. Scandium and yttrium do indeed appear to be much closer to the lanthanides than the rest of the transition metals, so I think it makes more sense to group them together to form the rare earth metals – a term that I think more people know than "lanthanides" today, judging by how often it occurs in the news. (Yes, I know the lanthanides probably are named as such in the periodic table in your first chemistry textbook, but I really doubt they will be mentioned for a long, long while. Poor lanthanides. They're so pretty!) Such a classification makes it easier to write about chemistry. If you look at Cotton's book, for instance, he covers Y explicitly together with all the lanthanides, as it is so close in behaviour to a heavy lanthanide. He covers Sc as a special case in its own chapter, but then he does the same for Pm, whose only problem is its radioactivity; so I think this shows that Sc's specialness mostly comes from it being so small, and that indeed it simply acts like a mini-Lu. (OTOH, he covers the actinides separately, because the radioactivity common to all of them does make a difference, and the increased relativistic effects play a huge role in the chemistry of these elements.)
    • I think it is not too much of a demand on our categories that if we place element X into category Y, category Y must be the best possible fit for element X's properties. In particular, if there is a category Z that fits the properties of element X better, then it should be moved to category Z. Thus, the way I see it, Sc and Y are much closer to the lanthanides than the transition metals they currently are grouped with, so they should be moved to be categorised together with the lanthanides. This is even better supported by the fact that there is a standard name for scandium, yttrium, and the lanthanides: the "rare earth metals". (It's even IUPAC-approved!) Clearly, this categorisation has been thought to be needed by many reliable sources.
    • I think getting the reader to follow the natural overlapping of these categories is a lost cause. Is lutetium a transition metal? If you follow the table explicitly, it isn't. If you follow the electron-configuration definition, it most certainly is. So already we have this problem. When faced with this, we should pick the most representative categories. According to the IUPAC 2005 Red Book, astatine is a halogen. Yet this is a poor fit for it, which is why we moved the halogens to simply being a group and not a category. According to the Red Book again, N is a pnictogen and O is a chalcogen; but you can bet that if I ever get around to rewriting those two articles, I am going to treat them separately as special cases and refer to the pnictogens and chalcogens as simply P–Bi and S–Po respectively (following what appears to be more common among chemists), because these (as specialist categories) are poor fits for N and O respectively. Their behaviour is too different!
    • There are so many possible categories not shown on our table. We have chosen to limit ourselves to just ten, with one more that basically is "don't know". We had better then carve up the table to keep the most similar elements together.
    • For the same reason, group 12 should be placed with the post-transition metals. Actually I might even advocate colouring Cn as a post-transition metal for now, as we know that physically acts like its lighter congeners. It is only the chemistry that is expected to be that of a transition metal, and we do not know that yet. All the experimental evidence we have is on physical properties, which fall on the side of PTM.
    • (This is also why I am fine with keeping the status quo on the nonmetals; the polyatomic/diatomic/monatomic distinction does indeed correlate very well with many other properties, so that each element fits better in its own category than any other. I would also think that separating iodine from the other three halogens would be a grave mistake. This is one of the model examples of great group trends, at least if you ask astatine to politely decay before you begin your lecture. And I think dumping them all together results in a category that is much more diverse than the others: look at the dramatic difference between fluorine and selenium, two extreme cases. If you compared the extreme members of any other category, it wouldn't be that drastic a difference.)
    • Naturally we should still retain mention of the lanthanides as being a thing, but increasingly I am thinking that the category (and main flagship article for it, with the detailed chemistry) should be the rare earths. Lanthanides could even be simply a redirect to a section of the rare earths, as they overlap so much. You'll notice that even in Greenwood and Earnshaw's attempt to cover group 3 separately from the lanthanides and actinides, he kind of cheats by looking at La in both the group 3 and lanthanide chapters. (He can get away with this more easily for Ac, covering it mostly only in the group 3 chapter, as that group's chemistry is more diverse.) Double sharp (talk) 14:07, 14 January 2016 (UTC)
      • P.S. I'm sorry for always referring to Greenwood and Earnshaw 1st edition! The reason is that I just can't seem to find my copy of the 2nd edition anywhere, and I've been looking for a while. It'll probably turn up eventually, but for now I'm using this. If my memory serves me right, details have changed on just about every page, but the general structure is intact, so it should be safe to use this on categorisation topics. Double sharp (talk) 14:09, 14 January 2016 (UTC)

Downwards and backwards[edit]

Today Sandbh pushed their /sandboox version of this article into live. Repeatedly I have pointed to flaws and errors in this version, but for some months the editor did not bother to reply or base their changes on sound arguments. In short, and as an incomplete list, this is wrong in the article (all topics + arguments can be found above).

  • Using the wording "Standard form" and "Long form" (eg in top image): not sourced (=OR), not correct, and misleading. On top of this, there are two viable forms for group 3, and nowhere is explained why this version would be "standard". Actually, the current "standard form" (ie mostly used in serious sources) is an other one (namely, putting 32 elements in group 3—incorrectly, see below).
Good point. Scerri (The Periodic Table, 2007, p. 21) says, "The standard form of the periodic table has also undergone some minor changes regarding the elements that mark the beginning of the third and fourth rows of the transition elements. Whereas older periodic tables show these elements to be lanthanum (57) and actinium (89), more recent experimental evidence and analysis have put lutetium (71) and lawrencium (103) in their places. It is also interesting to note that some even older periodic tables based on macroscopic properties had anticipated these changes." How does that look? Sandbh (talk) 09:26, 25 December 2015 (UTC)
I've edited the article to emphasise that the concept of a "standard" form refers to the 18-column form, of which there are three main variants. Does this help? Sandbh (talk) 23:24, 25 December 2015 (UTC)
"Does this help" - No, no 'help'. "Standard form" is a useless, misguiding old concept. There is no "standard" from, and no authority can claim it to be (Which is why you can not source it). The reader is helped into a trainaccident this way, because you introduce the subliminal point: what is not standard, then? Just leave out those relative, indecisive and outdated judgemental wording. Don't use = no need to explain. Oh and claiming a Standard unbsourced is OR/POV. -DePiep (talk) 08:23, 5 January 2016 (UTC)
The article sources Eric Scerri on this point and his book on the history and significance of the periodic table. Sandbh (talk) 22:34, 5 January 2016 (UTC)
Another OR. Scerri does not define "the standard", let alone in a way you do here. The word "standard" form is used for many variants (historical ones included), even covering the wrong "third" variant. In situations it also refers to the scientific form (be it 18- or 32-colomn) to differentiate it from scientific variants like Janet's Left Step. As such it is useless as an identifier for a particular form. Proof: the Scerri source you point to says: "The standard form of the PT has also undergone some minor changes re the elements re.." (i.e., not stable in history), all this in an larger description of (1) the single law of periodicity, 2. graphic variants, and 3. scientific variants (together ~700 he says). So Scerri does not define this "standard". And as we know, Scerri does not so consistently over his publishings. -DePiep (talk) 08:53, 8 January 2016 (UTC)
  • In section "#Different periodic tables": introducing "Type I, II, III" denominations. These do not exist as such in literature (unsourced, so OR). There is no need to deviate from descriptive wording like "Group 3 is ...".
I added a sentence to say that these labels (of convenience) are used only for the purposes of the article. Sandbh (talk) 11:18, 25 December 2015 (UTC)
  • Wrong classification setup. After the two well sourced, structural variants having Group 3 is SC, Y, La, Ac en Group 3 = Sc, Y, Lu, Lr, a third from appears that is not sourced and is not a new structural variant. Tellingly, it is labeled "[Group 3 =] Sc, Y, and markers": markers are elements? At best, the third form can be described as a sloppy, careless version. But actually it is a wrong version (nowhere in the sources it is claimed that group 3 has 32 elements; and so of course nowhere in the article can this be referenced or described). This is more like dumb, uncritically copying the sloppyness of a source into prominence. I checked two sources Housecroft and (not in the article but used on this talkpage) the IUPAC Red Book: none states that that is a description of group 3. Tellingly, Housecroft chapters skip groups 3–12. Mixing up two scientific variants with a third graphic droodle is amateuristic approach.
You may have missed my previous post to this page. Fine (1978, pp. 702–707) has a six page section called THE 32 ELEMENTS IN III B [his formatting, not mine] which includes a Type III periodic table with the 32 elements in question shaded.
  • Fine LW 1978, Chemistry, 2nd ed., The Williams & Wilkins Company, Baltimore [post by Sandbh, I assume]
And this singles source [fringe] you [POV] promote [OR] into mainstage science, even transposing Fine's statement to be the base for all other such PTs like IUPAC? -DePiep (talk) 08:13, 5 January 2016 (UTC)
No, the practice of extending Group 3 membership to rope in the lanthanides and actinides is not fringe, although Jensen disparages it as noted in the article per your concerns. Here some further examples from the literature: "The lanthanides and actinide elements are all metallic and are formally members of Group 3 of the Periodic Table." Barrett J 2002, Atomic structure and periodicity, The Royal Society of Chemistry, Hoboken, New Jersey, p. 80 • "The carbides of Group 3, ie, Sc, Y, the lanthanides, and the actinides, are opaque." Stoll WM & Santhanam AT 1992, 'Carbides (Industrial Hard)', in Kirk-Othmer Encyclopedia of Chemical Technology, 4th ed., vol. 4, Bearing Materials to Carbon, John Wiley & Sons, New York, p. 844 Sandbh (talk) 22:19, 5 January 2016 (UTC)
Wait wait. So you have this one (one) source that actually does describe group 3 being 32-element, and it is not in the article? At least, do you admit/agree that the current sources (including Housecroft I could check) do not support the claim? Those sources must be removed then. -DePiep (talk) 08:39, 5 January 2016 (UTC)
The current sources are examples of type III Sc|Y|*|** tables. Whether or not they address or imply that group 3 is comprised of 32 elements is irrelevant since they are cited only to illustrate examples of the type III table. As explained in the article, Jensen has criticised some presentations of type III tables for implying that all the Ln fit in the single box below Y. Sandbh (talk) 23:55, 5 January 2016 (UTC)
This describes your sin. Their graphs state group 3 = 32 elements, and their text does not source it. From this, you conclude these 'sources' "define" a new variant. Qoud non. -DePiep (talk) 08:40, 13 January 2016 (UTC)
Their graphs can be read to imply that group 3 has 32 elements or they can be read to imply that 57–71 and 89–103 fit between the s-block and the rest of the d-block, in an ambiguous manner. Either way I'm only describing a variant, warts and all, that is widely used in the literature, as observed by Clark and White (2008) in "The Flyleaf Periodic Table", Journal of Chemical Education 85 (4): 497. This article prompted a flurry of responses from other authors including Jensen, Scerri and Phillip Stewart (of Chemical Galaxy fame) none of whom criticized Clark and White's finding that there were three main types of periodic table, including Sc|Y|*|**. Sandbh (talk) 06:20, 16 January 2016 (UTC)
"Their graphs can be read to imply ... or" - no, not 'can be'. 'Must be' is the right wording. These graphs undisputedly, unambiguously state column3/group3=32 elements. You can not talk some out of it.
"none of whom criticized" - that is not enough. Even stronger, it supports the claim that group3=32 elements is not sourced. -DePiep (talk) 10:13, 17 January 2016 (UTC)
Here are two more examples of authors who say there are 32 elements in group 3: Enghag P 2008, Encyclopedia of the Elements, Wiley-VCH, Weinheim, p. 65: He uses a Sc|Y|La|Ac table! Quadbeck-Seeger HJ 2007, World of the Elements: Elements of the World, Wiley-VCH, Weinheim, inside cover: he shows a 32-column table, and says in a footnote, "all lanthanides and actinides belong to group 3". Those two are in addition to the previously mentioned examples (Fine 1978; Barrett 2002; Stoll and Santhanam 1992). I pass no judgement as to the merits of saying group 3 has 32 elements. I simply note this is done by some authors. Sandbh (talk) 11:32, 17 January 2016 (UTC)
About 'Quarkonium' News from Eric Scerri - Sandbh
Great. [...]
Second: Scerri announces a IUPAC taskforce (his, so to speak) into the constitution of Group 3. That is good news. I also note that with this, he does not even mention Sc/Y/*/** being group 3. Well, if it takes IUPAC to get rid of it as a serious PT construction, so be it. But we can shaft it on sound grounds today. -DePiep (talk) 02:36, 8 January 2016 (UTC)
The main problem with Sc/Y/*/** is that while you will see people using it, they don't seem to agree what it means. As a result it is difficult to critique it because you cannot start critiquing a form until you are sure what it is supposed to mean. Thus we use Jensen's assumption of its meaning (though you will hear people saying "oh, it's to keep Ln/An together" or "oh, obviously La and Ac are primary since they are the first" or any number of strange things to justify it). Nonetheless, I would still say that from personal experience, the unequivocal Sc/Y/La/Ac beats it in popularity.
As popular as it is, though, we seem to be stuck with it for now. If IUPAC actually makes a decision, though (hopefully on Sc/Y/Lu/Lr), then we could probably get away with downgrading Sc/Y/*/** from "Type III" to an unnumbered form, mentioned due to its history of usage, but as a side thing to be critiqued and immediately thrown out as a thing. Double sharp (talk) 03:30, 8 January 2016 (UTC)
"... what it is supposed to mean". No, there is no question about it: it graphically says that group 3 has 32 elements and that Sc, Y border both their period elements left and right. And 99/100 sources who publish a Sc/Y/*/** do not base that statement on scince but bad editorship (it is not described as such in those same sources that draw Sc/Y/*/**). In this, these are an unRS. The real problem with Sc/Y/*/** is that people here keep using & pushing that drawing as a third, scientific variant for group 3. I oppose, and Scerri does not even mention it (also not in his earlier papers). -DePiep (talk) 08:40, 8 January 2016 (UTC)
  • The single usage of the 18-column form, omitting the 32-column form completely, is an editorial choice and a POV. I agree with Scerri: as a modest question, please keep atomic numbers in order (done) and give us the 32-column form (not done).
I'll have a look at this. It seems odd not to say anything about the 32-column form in the Different periodic tables section. Sandbh (talk) 10:26, 25 December 2015 (UTC)
32-column PT is not different from its 18-column form. It is the same PT, presented different form. The scientific statements are the same, and must be. Examples of a different PT are : Janet's Lewft Step, ADOMAH, and hundreds of other structural variants. -DePiep (talk) 08:07, 5 January 2016 (UTC)
I've added a 32 column table and accompanying text. Sandbh (talk) 23:34, 26 December 2015 (UTC)
In Russian chemical literature, the statement that all the lanthanides and actinides belong to the "transitional subgroup of group III" (=group 3, as opposed to the "main subgroup of group III"=group 13) is frequent; the Great Soviet Encyclopedia and the Chemical Encyclopedia state so. In Russia, the terms "lanthanides" and "actinides" (or, more frequently, "lanthanoids" and "actinoids") are usually understood to mean Ce to Lu and Th to Lr; but the arrangement of the footnotes in the 8-column table (as R8R Gtrs has already said in some talk pages, it is still common in Russia, slowly being superseded by the 18-column table) may be either 14CeTh or 15LaAc. Hydrogen in Russian 8-column tables is normally shown in groups I and VII; either with full information in group I, parenthesized and without information in group VII, or vice versa.Burzuchius (talk) 20:05, 26 December 2015 (UTC)
Very interesting, Burzuchius. Do these two sources give a base or primary source for this statement? Scientific reasoning? Sandbh is dearly in need of a second source support for these fringe graphics. So far, all sources but one (one) do not support their own wrongful group-3=32-elements presentation.
... but the second part of your answer suggests this: coming from the 8-column PT, adding f-block columns 19–32 may not be that accurately done. All in all, we are not interested in how the elements are drawn into a source's PT, but whether the drawing is conform its textual statements. If not, the drawing is a misleading fraud, and can not count as a 'source' for a 32-element group 3. -DePiep (talk) 08:32, 5 January 2016 (UTC)
I mean the explicit statement, not just the drawings. For example: "The rare earth elements are a family of 17 elements of group III of the periodic system, comprising scandium, yttrium, lanthanum and the lanthanoids..." (Chemical Encyclopedia, 1994 [7]). I think the reasoning is the following: the lanthanides are very similar in their properties, and they all have the common oxidation state +3. Some of them can have +4, especially cerium, but they are not very much like group IV. After all, the rule "maximal oxidation state is equal to the group number" already has exceptions in the 8-column table: Cu, Ag, Au are in group I, but they can have oxidation states more than +1. And the principle "one cell, one element" in the 8-column PT is violated by the triads in group VIII. The actinides are placed in group III by analogy with the lanthanides, although the early actinides (Th, Pa, U, Np, Pu) are not very much like lanthanides.Burzuchius (talk) 13:33, 5 January 2016 (UTC)
For sake of the argument, I assume your 'I think the reasoning is ...' (OR) can be replaced by the original sources. That says by now we have two sources that claim so explicitly (that is: sourced as a scientific statement). btw, the Ce exception and others is weighed quite heavy elsewhere on this page. This assumption does not validate presenting the group 3 = Sc/Y/*/** variant as mainstream PT. Already a dozen sources (Scerri, Jensen come to mind) do explicitly not mention this one, which can be counted as a counter-source (so sources: going below zero this way!).
I state (working hypothesis) that 99/100 PT's that show Sc/Y/*/** do so out of sloppyness/unknowingly. To make matters worse, these PT's do not re-graph into 32-column form without exposing the fraud. This is bad for every reader and scolar in the field. -DePiep (talk) 09:06, 8 January 2016 (UTC)
  • The article keeps pushing (silently, but predictably it will be added later) that group 4–group 17 continue below into the footnoted elements.
DePiep (talk) 09:00, 25 December 2015 (UTC)
I don't think f-block group numbers (however these look) will ever happen unless elements occupying the next row of the f-block are synthesised i.e. in the vicinity of 141 onwards. Sandbh (talk) 11:49, 25 December 2015 (UTC)

It disturbs me that the group 3 issue is covered twice: once in "Common form variants" and again in "Period 6 and 7 elements in group 3". Double sharp (talk) 08:17, 5 January 2016 (UTC)

This may be unavoidable since the reason why there are three common types of 18-column table arises from the group 3 issue hence it is mentioned in the first section. The controversy is then explored in the second section. I'll have another look at this. Sandbh (talk) 00:00, 6 January 2016 (UTC)
Sandbh there are three common types of 18-column table arises from the group 3 issue - Nonsense. Unsourced, synthesis, incorrect synthesis (because you are mixing up two grouping 'rules' into one set), no consensus, mixing up editorial chooices with scientific bases, introducing fringe for mainstream, bad and judgemental qualifying (eg pushing 'common'), and so bad loads onto the reader. -DePiep (talk) 09:36, 6 January 2016 (UTC)

What changes would you like to see in the article? Sandbh (talk) 10:29, 6 January 2016 (UTC)

1. Remove the third 'variant' (Sc/Y/*/**) completely from the article. Not a mainsteam scientific variant, more a bad graphic inherited habit. Lacks sources. Could be a section in PT History, adding: Why does your classroom PT not look like WP's?
DePiep, is this still your position? Sandbh (talk) 10:53, 2 February 2016 (UTC)
2. Throughout the article(s), maintain a split within all 'variants' wrt scientific variant (ie structurally different: like Janet's), and graphically different (like 18/32-col for the same group3=Sc/Y/Lu/La PT). Those graph variants should be reduced to a minimum, basically saying in this article: "these two graphs (18, 32) are the same PT".
I believe this has been done? Sandbh (talk) 10:53, 2 February 2016 (UTC)
3. (Have not checked whether this is done bad today, but we should aim for it:) clearly explain that & why we choose to show the Sc/Y/Lu/La variant as the standard group3 variant (enwiki wide). In a section, not in the lede. Group 3 can have more on this all.
Done Sandbh (talk) 10:53, 2 February 2016 (UTC)
4. Remove "Type X" naming. Is not mainstream, and not needed because the description "group 3 = ..." already says it, and more directly at that. The description (definition) is required anyway.
Done Sandbh (talk) 10:53, 2 February 2016 (UTC)
5. Remove all historical/indecisive/ambiguous/relative graph definitions like long form, standard form, medium-long form, etc. Sources are inconsistent between them, relative naming adds complexity in understanding, it is loaded with historical changes. In short: there is no "standard form". -DePiep (talk) 10:47, 8 January 2016 (UTC)

-DePiep (talk) 10:40, 8 January 2016 (UTC)

In general, the literature refers to 8-column tables as the short form, 18-column tables as the modern, standard, common, popular, medium, or medium long form and 32-column tables as the long or expanded form. I propose to explain this in the article rather than not saying anything at all, given how frequently these terms are encountered. Sandbh (talk) 10:53, 2 February 2016 (UTC)

Thank you DePiep; I'll respond to each of the above items. Sandbh (talk) 11:16, 8 January 2016 (UTC)

Thx. I request we pause this thread for some weeks, because I don't have the time to add serious background for this (it is just telegraph-style now). For more thorough argumentation, I want to re-read the article, the whole thread and use/read linked sources etc. But these weeks I'm bizzy in RL. You are free to ponder the seeds I planted, of course ;-). -DePiep (talk) 11:30, 8 January 2016 (UTC)
Sandbh, another (similar) question here you did not answer. Your pattern? -DePiep (talk) 01:27, 31 January 2016 (UTC)
DiPiep, have I addressed all of your requests/question?
No you have not. Need to ask? Shortcut to Q2: just read & react for starters (I was about to repeat my boilerplate fact but hey). -DePiep (talk) 22:50, 2 February 2016 (UTC)

Four new elements[edit]

7th periode is now completed. Someone should add info... --Obsuser (talk) 21:13, 5 January 2016 (UTC)

It's mentioned several times, earliest is the last paragraph of the lead section (including mentioning the recnet-news aspect) and then with more detail at the end of the History section. What additional info would you like to see? Alternately, you should be able to edit the article if you have some ideas. DMacks (talk) 21:33, 5 January 2016 (UTC)
I got news on 5th January 2016 and IUPAC formally recognized them about 5 days earlier. My bad. --Obsuser (talk) 22:55, 15 January 2016 (UTC)

Coming changes in visual presentation[edit]

The wiki PT is currently undergoing a great discussion about an update, both in colors and possibly in some categorization and structure issues. Everyone would be welcome to join the discussion on the issues related to the table to present itself, see here, and potentially the discussion focused on technical details of presentation, here.--R8R (talk) 22:51, 10 January 2016 (UTC)

Two separate topics are opened:
WT:ELEMENTS Metallicity categorization review (scientific side, category definitions)
WT:ELEMENTS Category Color Set review (presentation side, colors to be used)
-DePiep (talk) 11:16, 11 January 2016 (UTC)

Origin of the Sc|Y|*|** table[edit]

Any idea who originated this form?

I was looking at The Internet Database of Periodic Tables today, to so how far this version went back. The first tables that caught my eye were Seaborg's of 1944 and 1945. I mention him because his actinide hypothesis was quite influential. Before that there is Deming's table of 1923 which popularized the 18-column form. Earlier yet there is Meyer's Periodisches System der Elemente of 1918 but this is an 8-/18-column hybrid. It may be that although there were predecessors, Deming's 1923 table was the candle that lit the fire. Or will my question be similar to attempting to find the source of the Nile/Amazon? Sandbh (talk) 11:37, 21 January 2016 (UTC)

I added a note (#12) to the article that addresses this question. Sandbh (talk) 12:04, 22 January 2016 (UTC)
Not read the sources yet. Anyway: Graphs-By-Heros are not defining:
- Mendeleev put uranium in group VI (Reihe 12). Th in Reihe 12 (group IV). You maintain?
Mendeleev's assignment of Th to group IV and U to group VI was a reasonable decision in its time. But effectively no one has been doing this since Seaborg's actinide hypothesis became universally adopted just after World War Two. Sandbh (talk) 10:25, 27 January 2016 (UTC)
- Seaborg 1946 did not bother about the graphic issue. Nor about group 3 composition.
Agree. Neither has IUPAC bothered with the graphical issue. Nevertheless, the Sc|Y|*|** table is one of the three most common forms. Sandbh (talk) 10:25, 27 January 2016 (UTC)
S, you float from "PT in RS x shows it [so proof]"to "[Fringe I found] defines it" (quite a point here: what are actually the sources for all those wallpapers?).
Again, Sandbh, you abuse editors energy to make your moving fog point: that a graph in an (otherwise) RS proves "group3=32 elements". That is your core thinking error. Eric Scerri does not even mention your OR assumption.
Please take note of the 'abuse of editors energy' thng. -DePiep (talk) 23:18, 23 January 2016 (UTC)

DePiep, let me be clear on what you are saying. I will do this in small questions so that there is no confusion:

Q1. I think you're saying that a Sc|Y|*|** table, from a graphical point of view, implies there are 32 elements in group 3. Is that right? Sandbh (talk) 10:23, 24 January 2016 (UTC)
A1: re "from a graphical point of view" No, not by some "view". BY FACTUAL PRESENTATION. There is no choice for the reader. The graph says: "column header 3 = 32 elements". -DePiep (talk) 22:14, 24 January 2016 (UTC)
Q2. Do you think that showing 32 elements in group 3 is factually wrong? Sandbh (talk) 02:18, 25 January 2016 (UTC)
A2: It was great, in 1946! How could I disagree with Seaborg?
Now back to your point: when do you digest & apply my answer? -DePiep (talk) 22:09, 25 January 2016 (UTC)
I'm trying to digest your answer by seeing if I can follow your logic. I'm doing this by asking more questions. It's a slow process but will be worth it if both of us can be 100% clear on each other's position. Sandbh (talk) 23:13, 25 January 2016 (UTC)
I've answered dozens of times, including A1 here: A Sc|Y|*|** table states (not just suggests or optionises) that there are 32 elements in group 3. Irrespective on whether drawn in 32-col or 18-col format. It was introduced ca 1946 by Seaborg to allow for actinides/f-block etc. By then is was great, by today it is historical. You (Sandbh) may have found two or three sources that actually base that group 3 has 32 elements (after I pushed you, remember), but that only moves the 3=32 fact from "wrong" into "fringe". When you sourced the other two group-3-options (20 sources), you yourslef did not even find or noted that 3=32 is a viable third statement.
Confusingly enough (for you that is) you keep mixing this topic with the "18- or 32-col PT" thing. A3: Find a wall, bang your head 3 times against it, then leave this mixing behind. -DePiep (talk) 23:32, 25 January 2016 (UTC)
Q3. Are the following 14 sources, which refer to the Ln and An being part of group 3, reliable(?):
  • As noted by User:Burzuchius, "In Russian chemical literature, the statement that all the lanthanides and actinides belong to the "transitional subgroup of group III" (=group 3, as opposed to the "main subgroup of group III"=group 13) is frequent; the Great Soviet Encyclopedia and the Chemical Encyclopedia state so."
  • Fine (1978, pp. 702–707) has a six page section called THE 32 ELEMENTS IN III B [his formatting, not mine] which includes a Sc|Y|*|** periodic table with the 32 elements in question shaded. Fine LW 1978, Chemistry, 2nd ed., The Williams & Wilkins Company, Baltimore
  • "Scandium (Sc), yttrium (Y), the lanthanides and actinides constitute subgroup B of group III in the periodic table."—Luckey TD & Venugopal B 1978, Metal toxicity in mammals: Chemical toxicity of metals and metalloids, Plenum Press, p. 101
  • "Both the lanthanoids and actinoids are considered to be part of group IIIB, the scandium subgroup." ; "The lanthanoids and actinoids…are often counted with group IIIB."—Russell JB 1980 General chemistry, McGraw-Hill Higher Education, pp. 161, 641
  • "The carbides of Group 3, ie, Sc, Y, the lanthanides, and the actinides, are opaque."—Stoll WM & Santhanam AT 1992, "Carbides (Industrial Hard)", in Kirk-Othmer Encyclopedia of Chemical Technology, 4th ed., vol. 4, Bearing Materials to Carbon, John Wiley & Sons, New York, p. 844
  • Williams et al. (1996, p. 10) features a periodic table with the f block elements marked as "Lanthanides (Group 3)" and "Actinides (Group 3)".—Williams RJP et al. 1996, The natural selection of the chemical elements: The environment and life's chemistry, Clarendon Press
  • "The 28 lanthanides and actinides together make up about one quarter of the periodic system as it is now known, with 109+ elements. Together with the four related elements Sc, Y, La and Ac, they all belong to transition group III, which is thus the largest subgroup in the period table."—Wiberg N 2001, Inorganic chemistry, Academic Press, p. 1645
  • "The lanthanides and actinide elements are all metallic and are formally members of Group 3 of the Periodic Table."—Barrett J 2002, Atomic structure and periodicity,' The Royal Society of Chemistry, Hoboken, New Jersey, p. 80
  • "Group 3 elements…include the lanthanides and actinides."—Byrne CH 2002, "Speciation in seawater", in AM Ure & CM Davidson (eds), Chemical speciation in the environment, Blackwell Science, pp. 322–357 (332)
  • "Modern group numbering runs from 1 to 18, with the f blocks being subsumed into group 3."—Cox PA 2002, Inorganic Chemistry, 2nd ed., Bios Scientific Publishers, p. 11
  • Quadbeck-Seeger (2007) shows a 32-column table, and says in a footnote, "all lanthanides and actinides belong to group 3."—Quadbeck-Seeger HJ 2007, World of the Elements: Elements of the World, Wiley-VCH, Weinheim, inside cover
  • Enghag (2008, p. 65) uses a Sc|Y|La|Ac table and above the lanthanides he says, "Lanthanides (the 14 elements between lanthanum and hafnium (period 6, group 3)". Above the actinides he says, "Actinides (the 14 elements after actinium, thorium-lawrencium (period 7, group 3)."—Enghag P 2008, Encyclopedia of the Elements, Wiley-VCH, Weinheim
  • "The 30…elements…at the bottom of the periodic table belong to the lanthanide and actinide series found in group 3 of the periodic table."—Ganfalvi G 2011, "Heavy metals, trace elements and their cellular effects", in G Ganfalvi (ed.), Cellular effects of heavy metals, Springer-Science+Business Media, pp. 3–28 (7)

Sandbh (talk) 05:51, 26 January 2016 (UTC)

You keep jumping from one foot on the other. Have you decided on whether this graph defines group 3 or just suggests it (see Q1-A1)? At the moment, the "suggestion" is relegated to reference #13 only. (Background of this daily alternating is that you keep relating the two issues. If you want to get things clear, it's your turn to conclude). -DePiep (talk) 11:27, 27 January 2016 (UTC)
I would personally say that drawing Sc/Y/*/** absolutely implies that all 15 lanthanides and all 15 actinides are group 3. (Is this a valid characterisation of your stand, DePiep?) I think that the above is evidence that 12 of these 14 sources (a rather significant number) intentionally wanted to show 32 elements in group 3, so that this implication of Sc/Y/*/** would have to be taken seriously as a legitimate variant (albeit not a good one). However, if one looks at the periodic tables in Holleman and Wiberg, as well as Enghag, we get a strange inconsistency. They use a Sc/Y/La/Ac table, and yet treat group 3 as having 32 elements. One can't have it both ways. (My favourite source, Greenwood and Earnshaw, is consistently Sc/Y/La/Ac. It disappoints me somewhat that the German standard text contains this inconsistency...) Double sharp (talk) 11:45, 27 January 2016 (UTC)
A group does not actually exist; it's a theoretical construct invented by us humans for classification purposes. As such, we humans can define the term to be whatever we want.
 2              3               4
|Ca| |Sc                    |  |Ti|
|Sr| |Y                     |  |Zr|
|Ba| |La Ce Pr ....... Yb Lu|  |Hf|

An example of how this could be. This is a self-consistent model.--R8R (talk) 13:36, 27 January 2016 (UTC)

Yes, that is consistent. What is not consistent is putting La alone under Y in group 3, and then proclaiming that the other 14 lanthanides are also members of group 3. Double sharp (talk) 14:31, 27 January 2016 (UTC)
As R8R is drawing the group 3 here [completed into a PT], it defines 32 elements in there. If R8R wants to convey some other 'theoretical construct', he should draw a different graphical construct. -DePiep (talk) 16:07, 27 January 2016 (UTC)
Well, the two Russian sources I have mentioned (the Great Soviet Encyclopedia and the Chemical Encyclopedia) are also somewhat inconsistent: they consider the lanthanides and actinides to be group III elements in the text and in the 8-column table, but they also show the 32-column Sc-Y-La-Ac form, and in that table, the label "IIIb" is placed above the Sc column only.Burzuchius (talk) 21:31, 27 January 2016 (UTC)
Thanks for detailing this. Not devastating for these encyclopedias (it happens 'most commonly, for dozens of years, re group 3). But in this thread it is core: text-does-not-match-graph. -DePiep (talk) 00:23, 28 January 2016 (UTC)

If I was a general reader I would agree with Double sharp. Drawing Sc/Y/*/** will nearly always imply that all 15 lanthanides and all 15 actinides are in group 3. Greenwood & Earnshaw (2nd ed.) struggle a bit with this. In their periodic table each element box is 11 mm × 11 mm. In the two boxes under Y they show La* and Ac†. But each of these boxes is only 10mm wide. There is are two empty pillboxes between La and Hf and between Ac and Rf that are each 1 mm wide and 11 mm high. *Ce to Lu, and †Th to Lr are shown at the foot of their table. Thus it looks like they say 32 elements in group 3 or do they? In their mighty text they imply that only Sc, Y, La and Ac are in group 3. Sandbh (talk) 10:39, 28 January 2016 (UTC)

Saw this only just now. Sandbh, don't get distracted into a mud. That is IUPAC behaviour. -DePiep (talk) 20:51, 28 January 2016 (UTC)
Q4. Placeholder for a question I haven't worked out yet
  • "The lanthanides and actinides have been of so little relative importance that they have not been given group numbers."—Dickerson et al. 1979, Chemical principles, Benjamin/Cummings Publishing Company, p. 268
  • "There are no group numbers assigned to this portion of the table, because similarities within each period are more important than any vertical relationships."—Siebring BR & Schaff ME 1980, General chemistry, Wadsworth Publishing Company, p. 128
  • "The lanthanides and actinides are metallic, and they give up a number of electrons equal to their group number, IIIB."—Becker RS & Wentworth WE 1989, General chemistry, Houghton Mifflin, p. 202
  • "The f block does not have group numbers."—Olmsted J & Williams G 2002, Chemistry: A molecular science, 3rd ed., John Wiley & Sons, p. 292
  • "The inner transition elements (also called lanthanides and actinides) are metals that have no group numbers."—Hughes KJ & Kelter PB 2002, Student study guide to accompany Chemistry: A world of choices, McGraw-Hill, pp. 28–29
  • "We'd like to resume with a discussion of the lanthanide and actinide elements. These elements, while assigned no group numbers, are…shown below the main body of the periodic table."—Krishnamurthy N & Gupta CK 2004, Extractive metallurgy of rare earths, CRC Press, p. 460
  • "In the case of the 15LaAc form…the 30 elements are treated not as a separate independent electronic block but rather as degenerate members of group 3 of the d-block. The two boxes below Sc and Y…contain either the atomic numbers 57–71 and 89–103 or the symbols La–Lu and Ac–Lr, respectively, thus indicating that all 30 of the elements in the footnote belong in just those two boxes. Expanding such a table into a 32 column table would require one to stretch the boxes for Sc and Y so that they span all 15 of the inserted columns. This interpretation goes back to the 1920s and the original electronic interpretation of the so-called rare earth elements, as shown in the 8-column table in Figure 1 taken from Ephraim’s textbook of 1929 (5), and was later reformatted in terms of an 18-column table and extended to the actinoids by Seaborg (6). In the case of the lanthanoids it was based on the assumption that all of them had, as per Sc and Y, a common (n – 1)d1ns2 valence configuration and a maximum possible oxidation state of 3+, their only differences being the presence of a variable, but chemically insignificant, [noble gas](n – 2)fxcore. As shown in my original article, all of these assumptions are now known to be false (7). Many of these elements have (n – 2)fxns2 valence configurations and exhibit maximum oxidation states greater than 3+, thus making their assignment to group 3 of the d-block chemical nonsense. IUPAC or not, I can hardly believe that a modern inorganic chemist would advocate such an antiquated interpretation of these elements, unless, as noted above, they have lost all contact between the underlying premises of their periodic table and the facts of chemistry."—Jensen WB 2008, "The Periodic Table: Facts or Committees?", Journal of Chemical Education, vol. 85, no. 11, 1491–1492

Sandbh (talk) 22:30, 27 January 2016 (UTC)

Sandbh, you're expected to respond to my 11:27, 27 January 2016 post. (If you got distracted by intermediate posts possibly OT: don't let yourself be). DePiep -00:14, 28 January 2016 (UTC)

Yep. I expect to be able to do this later today. Sandbh (talk) 03:41, 28 January 2016 (UTC)

rm Type I-II-III references[edit]

I have started to remove all Type I-II-III references (namings for group 3 variants):

- They are OR (no source lists them as such)
- Their categorisation base is broken (like: Type a = by color, type z = by length)
- There is no need or usefulness for any reader to shortify these concepts into a local code. Essentially saying "Group 3 = ..." does the job.
- More cleanup=clarification can be done.
Must say, very disappointing having to enforce this upon Sandbh's edits. -DePiep (talk) 00:20, 24 January 2016 (UTC)

PT graph versus structure[edit]

Won't explain it too much. A diff graphic PT is NOT the same as a diff structure PT. (eg bg colors, font type, 18/32 cols versus Janet's Left Step, spiral, ADOMAH). Ask me if you do not get this. (I edited it this way, btw) -DePiep (talk) 00:38, 24 January 2016 (UTC)

DePiep, in general I like your edits. Will have a closer look later. Sandbh (talk) 02:50, 24 January 2016 (UTC)
I still like them. I tidied the Group 3 constitution variants section, removed some redundant text and fixed some mistakes. How does it look now? Sandbh (talk) 11:41, 24 January 2016 (UTC)
Like what you like. Won't help our Reader. Just say: "Group 3 = ...". Now repeat. -DePiep (talk) 22:11, 24 January 2016 (UTC)