Talk:Periodic table/Archive 9

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

Archive 5

←

Archive 7

→

Change location of the f-block

Opening statement

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

~~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

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

. Right above the asterisks, it says "3". -DePiep (talk) 22:11, 8 October 2015 (UTC)

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/*/**"

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
	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)

v t e 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	Hydrogen1H1.0080																	Helium2He4.0026
2	Lithium3Li6.94	Beryllium4Be9.0122											Boron5B10.81	Carbon6C12.011	Nitrogen7N14.007	Oxygen8O15.999	Fluorine9F18.998	Neon10Ne20.180
3	Sodium11Na22.990	Magnesium12Mg24.305											Aluminium13Al26.982	Silicon14Si28.085	Phosphorus15P30.974	Sulfur16S32.06	Chlorine17Cl35.45	Argon18Ar39.95
4	Potassium19K39.098	Calcium20Ca40.078	Scandium21Sc44.956	Titanium22Ti47.867	Vanadium23V50.942	Chromium24Cr51.996	Manganese25Mn54.938	Iron26Fe55.845	Cobalt27Co58.933	Nickel28Ni58.693	Copper29Cu63.546	Zinc30Zn65.38	Gallium31Ga69.723	Germanium32Ge72.630	Arsenic33As74.922	Selenium34Se78.971	Bromine35Br79.904	Krypton36Kr83.798
5	Rubidium37Rb85.468	Strontium38Sr87.62	Yttrium39Y88.906	Zirconium40Zr91.224	Niobium41Nb92.906	Molybdenum42Mo95.95	Technetium43Tc	Ruthenium44Ru101.07	Rhodium45Rh102.91	Palladium46Pd106.42	Silver47Ag107.87	Cadmium48Cd112.41	Indium49In114.82	Tin50Sn118.71	Antimony51Sb121.76	Tellurium52Te127.60	53 I	Xenon54Xe131.29
6	Caesium55Cs132.91	Barium56Ba137.33		Hafnium72Hf178.49	Tantalum73Ta180.95	Tungsten74W183.84	Rhenium75Re186.21	Osmium76Os190.23	Iridium77Ir192.22	Platinum78Pt195.08	Gold79Au196.97	Mercury80Hg200.59	Thallium81Tl204.38	Lead82Pb207.2	Bismuth83Bi208.98	Polonium84Po	Astatine85At	Radon86Rn
7	Francium87Fr	Radium88Ra		Rutherfordium104Rf	Dubnium105Db	Seaborgium106Sg	Bohrium107Bh	Hassium108Hs	Meitnerium109Mt	Darmstadtium110Ds	Roentgenium111Rg	Copernicium112Cn	113Uut	Flerovium114Fl	115Uup	Livermorium116Lv	117Uus	118Uuo

			Lanthanum57La138.91	Cerium58Ce140.12	Praseodymium59Pr140.91	Neodymium60Nd144.24	Promethium61Pm	Samarium62Sm150.36	Europium63Eu151.96	Gadolinium64Gd157.25	Terbium65Tb158.93	Dysprosium66Dy162.50	Holmium67Ho164.93	Erbium68Er167.26	Thulium69Tm168.93	Ytterbium70Yb173.05	Lutetium71Lu174.97
			Actinium89Ac	Thorium90Th232.04	Protactinium91Pa231.04	Uranium92U238.03	Neptunium93Np	Plutonium94Pu	Americium95Am	Curium96Cm	Berkelium97Bk	Californium98Cf	Einsteinium99Es	Fermium100Fm	Mendelevium101Md	Nobelium102No	Lawrencium103Lr
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:

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

Nh

Fl

Mc

Lv

Ts

Og

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

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

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

In svg overview form, they are:

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 course~~In 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

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)

File:15LaAc with LuLr in group 3.jpg

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

@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

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 ~~too~~either. 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

Graphic variants for the gap (as aside)

@R8R Gtrs:@DePiep: Would there be any appetite for this, presuming it could be done in html? Sandbh (talk) 06:14, 21 November 2015 (UTC)

Nice twist, using completely different symbols from the traditional * and **. Another idea would be to use something like
_* and ^*
_* – using asterisks provides a tip-of-the-hat to the historical precedent, but placing the two asterisks vertically provides a hint that something is different, and at the same time, conserves horizontal real estate. YBG (talk) 08:10, 21 November 2015 (UTC)

I think the triangle and square may be a little too difficult. I am certain a majority of people will get the idea, but even the original 57 La* and 89 Ac** were not so obvious when I saw the PT for the first time in my life, and this does not promise to be much easier (squeeze in between?). I think a gap would be the most accessible version for everyone out there, ideally a gap of non-standard width. The table in the section "Let us proceed with Sc|Y|La|Ac" is just great (except I would make gap between groups 2 and 3, rather than 3 and 4, but it's a separate issue.).--R8R (talk) 08:24, 21 November 2015 (UTC)

As for horizontal and vertical double asterisks, I think it's a matter of taste rather than changing anything; I'll be fine with either version.--R8R (talk) 08:26, 21 November 2015 (UTC)

57 La	*	72 Hf
89 Ac	* *	104 Rf

Glad you understood that I wasn't actually proposing that we use {{chem}} to display the asterisks, just that they be spaced vertically. That chem template (even with <big>) makes the two asterisks too small and much too far apart. Even using *<br/>* as in the sidebar to the right puts the asterisks vertically too far apart for my taste. Ideally they'd be occupy the same vertical space as regular character, so that (†) is to (‡) as (*) is to (what I want). If there's no way to provide this close vertical spacing, I'll withdraw my proposal. I was so hoping for something that would allow us to use not just a non-standard width, but a minimal non-standard width. YBG (talk) 08:51, 21 November 2015 (UTC)

57 La	_*	72 Hf		57 La	_*	72 Hf		57 La	_*	72 Hf
89 Ac	^* _*	104 Rf		89 Ac	^* _*	104 Rf		89 Ac	^* _*	104 Rf

I found {{su}} which with |lh=0.6 is pretty much what I was looking for. — Preceding unsigned comment added by YBG (talk • contribs) 01:10, 21 November 2015‎

I've added two more versions, surrounded once and twice by {{big}} YBG (talk) 09:24, 21 November 2015 (UTC)

So FWIW, my favorites for */** are the ones in the rightmost pic:

_* created by {{big | {{big | {{su |w=f |lh=0.6 |b=* }}}}}}

^*
_* created by {{big | {{big | {{su |w=f |lh=0.6 |p=* |b=* }}}}}}

But I don't feel super strong about it. YBG (talk) 09:33, 21 November 2015 (UTC)

If we have asterisks I like the ones on the far right too. Sandbh (talk) 10:49, 21 November 2015 (UTC)

Not enthousiastic about the first one (square+triangle), because is has no visible gap (wider space with no elements). While we surely intend to state that there is a series of groups (columns) in between, the unnumbered ones. The "3" is supposed to appear in the PT, I assume. Also, I don't think the placeholders are clear enough, they do not show their job at first sight (with me).

I still prefer asterisks, also over dagger/cross variants. My opinion is that asterisks are the most recognised typography to refer to a footnote, as we do here. (Maybe, Sandbh, can you expand on your choice, instead, for the dagger?). With this, the vertical arrangement of the pair looks like an improvement. re vertical positioning: The pair should be tight, and both symbols (1 and 2 *) best be centered vertically in their row, as they replace a whole row (not just element symbols). (pls don't worry too much about the technique to draw them. In most of our PT's this is a graph not the type character, so we can change them). -DePiep (talk) 11:49, 21 November 2015 (UTC)

See User:Sandbh/sandbox#Overview where I've inserted vertical asterisks. -DePiep (talk) 13:14, 21 November 2015 (UTC)

Please note my post above. Page-search '19:06' or 'perverted'. -DePiep (talk) 19:12, 21 November 2015 (UTC)

I like the vertically aligned asterisks and don't intend to further contribute to this subsection. Sandbh (talk) 23:02, 21 November 2015 (UTC)

La|Ac and Lu|Lr

---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

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

---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

---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?

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

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

---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

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

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."

Russell HN & Megger WF 1932, An analysis of lanthanum spectra (La I, La II, La III), Bureau of Standards, Journal of Research, vol. 9, no. 5, pp. 625--668

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

. -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+

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.

Leal W, Restrepo G & Bernal. A 2012, "A network study of chemical elements: From binary compounds to chemical trends", Communications in Mathematical and in Computer Chemistry, vol. 68, pp. 417–442, http://match.pmf.kg.ac.rs/electronic_versions/Match68/n2/match68n2_417-442.pdf

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)₅^2–complexes…In Ce(NO₃)₆^3–, 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)₅^2– 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

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

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)f^x-1(n-1)d, unlike the current data, which indicates most lanthanides have ground-state configurations of (n-2)f^x. 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)