Talk:Periodic table/Archive 8

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 6

Archive 7

→

Typo

Typo in "Placement of Hydrogen and Helium: "...both corresponding tot hose...". Please fix. Pity it's blocked, otherwise I could have fixed it myself.

Sure, done; thanks for the signal.--R8R (talk) 21:05, 6 October 2015 (UTC)

Moseley

where is Henry Moseley who is actually a real father of periodic table???????? periodicity was discovered not by mendeleev, but by other scientists before him, so I do not understand why he is being mentioned at all?? — Preceding unsigned comment added by 134.7.190.150 (talk) 13:28, 7 August 2015 (UTC)

I agree with your first point that Moseley should be included, and I have now added mentions of his measuring atomic number, showing that Mendeleev's element order actually corresponds to atomic number, and predicting missing elements.

For your second point however, I think the history section adequately considers scientists before Mendeleev and explains how Mendeleev's work was original and surpassed what went before. Dirac66 (talk) 01:25, 8 August 2015 (UTC)

See Henry Moseley. -DePiep (talk) 23:09, 21 September 2015 (UTC)

Observable francium

The article says "No element heavier than einsteinium (element 99) has ever been observed in macroscopic quantities in its pure form, nor has astatine (element 85)." This implies that Fr has been observed in macroscopic quantities. The francium article, with two pictures, says, "Enough francium is trapped that a video camera can capture the light given off by the atoms as they fluoresce. The atoms appear as a glowing sphere about 1 millimeter in diameter. This was the first time that anyone had ever seen francium."

I was going to say that this counts as seeing Fr. But now I am not so sure. I think that to qualify as a "macroscopic" quantity of Fr you ought to be able to see whether or not it has a metallic appearance. The experiment referred to in the francium article, while genuinely impressive in its own right, has not done this. They only show the emitted light or heat, not the actual visible appearance of the surface of Fr. So, I'm inclined to agree with User:Double sharp on this point rather than with User:Materialscientist. Although I don't want to make a big song and dance about it. Sandbh (talk) 11:21, 21 October 2015 (UTC)

Where does the periodic table end?

A better place to delete, I think, is the purely hypothetical discussion of whether the periodic table will "end" someday at element 137 or 155 or 173. In contrast, the lanthanides and actinides are known elements of some real interest. Dirac66 (talk) 11:13, 18 September 2015 (UTC)

As for when the periodic table ends: I think we can ditch 128 and 155, as nobody seems to believe those but their authors. Additionally 137 was an important historical proposal, but nobody believes it anymore. AFAICS now there are two camps: (1) it ends ≤ 173 and (2) 173 is not a limit. (One of them writes "it will never end": I think we must take that with a few moles of salt, as even if the electron-energy thing isn't a factor, I find it hard to believe that the nucleon drip line will not intervene sooner or later.) Double sharp (talk) 12:23, 18 September 2015 (UTC)

It is a fair enough question to ask, and well answered. OK, there is no definitive answer but I find it interesting to read about the range of possible answers. A better heading might be "Where might the periodic table end?" Sandbh (talk) 03:59, 19 September 2015 (UTC)

Personally, I prefer an even more generalized "Is there an end of the periodic table?". Chances are, it can go on forever and ever under some conditions. (Remember the "stars are elements, too" hypothesis?) Maybe so, maybe not. We don't know for sure yet to imply there is (most certainly) an end.--R8R (talk) 14:11, 20 September 2015 (UTC)

By the way I think we should not confuse present and future tenses here. The answer to the section title (which I did not write) Where DOES the periodic table end? is element 118 as of Sept. 2015, since that is now the highest known element. What is unknown is the answer to the question Where WILL the periodic table end?. Dirac66 (talk) 23:30, 21 September 2015 (UTC)

Heh. Fair enough: today it ends at a "chemically reassuring" element 118, per Emsley. Double sharp (talk) 01:27, 22 September 2015 (UTC)

Placement of hydrogen and helium

(FWIW, though, I think only two placements of H are common IMHO: either in group 1, or floating disconnected from the table. I don't have as strong an opinion on this one, mainly because I have not thought about it as much: but that is a different question.) Double sharp (talk) 12:23, 18 September 2015 (UTC)

Consider the "placing H and He" part. If we follow the current group 3 model, this should also be a bulleted list with four points: a) H goes into the group 1, He goes into the group 18; b) H 1, He 2; c) H (1+)17, He 18; d) H and He just float above the other periods. ANd we would have to give arguments for each version. Yet we're not doing that; and we shouldn't. That said, all of this is true and correct, yet should be discussed into the main articles (period 1 ~~and group 3~~). Here, we should just outline the problem and give a proper Main article link.--R8R (talk) 19:52, 18 September 2015 (UTC)

(just to make sure: does anyone want to extend the effort to represent the group 3 problem at its best to the analogous problem of period 1? or is it staying the way it is because none cares? Myself, I am ready to take part in writing a text a few paras long for it, if someone thinks it is needed.--R8R (talk) 14:18, 20 September 2015 (UTC))

I think that would be a useful addition, though I wonder if maybe a single paragraph might do. Also, it would be good to include partial PT illustrations. YBG (talk) 14:27, 20 September 2015 (UTC)

I've established a draft, here. Give in your comments, opinions, suggestions, or (which would be even better) help fill in the gaps. Is it okay to start with?--R8R (talk) 15:52, 20 September 2015 (UTC)

Of course it's a good idea! I mean, does anyone complain about the position of any element on the PT, other than H, He, La, Lu, Ac, and Lr?

I like your text. With regard to H alone, I think I wrote something at Alkali metal#Hydrogen. Too detailed for this context, but we can extract main points.

I think Seaborg's table main point would probably be the actinide concept, but would dearly love a citation.

Regarding Greenwood and Earnshaw splitting He from the rest of the group: I don't know. I leafed through and didn't find an explanation (though they give one for H). For H they say it is because of the unique electron configuration, resulting in different properties from both the alkali metals and the halogens: maybe this is the case too for He, the only case when s² is a full shell (the other noble gases are ns²np⁶). And indeed it calls 1s² the helium configuration and ns²np⁶ an inert-gas configuration. (Disclaimer: I am looking at the first edition. But IIRC the second one says the same thing.)

So...does He behave significantly differently from the other noble gases, in the way H behaves so differently from the alkali metals and halogens? (For example, the proton is so small compared to other ions that it cannot exist "naked" in condense systems. And of course there is also hydrogen bonding, though Li can kind of do this too.) Double sharp (talk) 16:19, 20 September 2015 (UTC)

I think there aren't any other placing problems.

How could a citation help? I mean, the whole point is the table, not a particular phrase we could extract.

Good one on He. I've given it a look and found kinda the same thing in what we could call an RS.--R8R (talk) 17:16, 20 September 2015 (UTC)

My best guess for what makes He different would be something to do with the first-row anomaly (aka first-element rule, first-member rule). I see that Scerri confirms this. (But then why does that argument work for He, but not Li–F?)

On the first-row anomaly (Li–Ne). Double sharp (talk) 07:30, 21 September 2015 (UTC)

I think a citation would help to state precisely what Seaborg's table was meant to illustrate. My assumption is it's for the actinide concept, but I could very well be wrong.

Thanks for this comment. I checked for it and while I'm sure I saw it before, I can't find it now; other Seaborg's tables don't follow suit (before he introduced the actinide concept, Th and U were thought to be 5d elements, yeah, that's what was so special about that particular table, not the H/He placing), so I'll look for another supporting table. A Seaborg table I saw gives a double 1+17 listing of H; it should be mentioned.--R8R (talk) 12:20, 21 September 2015 (UTC)

(@DePiep: Is there a way to make H and maybe He float above the rest of the table, to illustrate these tables? It doesn't really matter which group they are floating over.) Double sharp (talk) 02:51, 21 September 2015 (UTC)

See also this thread on new evidence supporting H's placement in group 1. Sandbh (talk) 03:56, 21 September 2015 (UTC)

Yes, I remember that one. (In fact I was looking for it, so: thanks!)

(A bit of unfinished business from that thread, though: how do Li₂, Na₂, etc. behave compared with H₂?) Double sharp (talk) 04:23, 21 September 2015 (UTC)

(I'm ignoring the noble-gas cores for the next comment. So when I write 2s² for Be, I know that it's really [He]2s², but want to only consider the valence electrons.)

I think the whole point that is at work here with H and He is that the electron configuration argument gets often used with PTs. Now this works swimmingly from the second to seventh periods, where the outer shell cannot go beyond s²p⁶. So s¹ indicates "reactive metal which wants to get rid of that s electron", and s²p⁵ indicates "reactive nonmetal which wants to complete its octet".

But the problem is that, for the first period alone, there is no p orbital. It's not an octet anymore: it's a duet. So for He, 1s² looks analogous to 2s² (Be), but He's electron configuration is actually a full shell unlike Be's. So, the argument goes, place it above Ne (2s²2p⁶), which also has a full outer shell. Maybe there will be differences stemming from He's lack of a p orbital, but it's nowhere near as incongruous chemically as He over Be. And most people are happy with this placement. (Although I wonder if it is a contributing factor as to why He is more reactive than Ne? ... OK, "more reactive" is a silly phrase for these elements. "Less inert", let's say.)

Oh hey! It is! ＼(^o^)／ See this article by Felice Grandinetti, a chemist who seems to support moving He to group 2! And the reasons? For one, Ne is more inert than He. Additionally, "[s]upporting arguments are the isoelectronic analogy (it has two electrons in its outside shell), and the anchoring of otherwise concealed periodic regularities" (oh I would love to know what these regularities are: anything to do with He/Be?). Here's the exact quote from his article: "Neon is bigger than helium, and possesses occupied p orbitals. This is thought to produce less effective electrostatic interactions and higher orbital repulsions, which typically make the neon compounds either unstable or only marginally stable, although the contributions of these factors are still to be further investigated." Double sharp (talk) 04:27, 21 September 2015 (UTC)

But H has a real problem still. At least He's 1s² can readily be compared with the full outer shell of Ne. With H's 1s¹, you have a half-filled shell, that needs only one electron added or removed to be stable (no electrons, or the duet of He). There is nothing exactly like this in any later row. Li's 2s¹ strongly prefers removing the extra electron; F's 2s²2p⁵ strongly prefers adding the extra electron; and C's 2s²2p² is so far from a noble-gas configuration that it resorts to other methods than forming ions like H can. Faced with these facts, a common thing to do is to just throw up one's hands and leave H outside any of the groups, and treat it as a case sui generis. (And if any element deserves to be thrown outside the periodic law, surely it is H, the first of them all!) Double sharp (talk) 04:23, 21 September 2015 (UTC)

Graphic workouts

(re Double sharp on graphics: sure we can show H floating & disconnected from groups &tc. I suggest we keep this thread scientific arguments only, and create a subthread for the graphical fleshing out. Just let me know when to start. Graphics best follow scientific requirements). -DePiep (talk) 08:35, 21 September 2015 (UTC)

We're not arguing to change anything about how we show our PT, AFAIK. I think H on top of group 1 and He on top of group 18 are fine as they are now. What I would like though are some illustrations for the different styles that have been used:

(a) Current WP table: H in group 1, He in group 18.

(b) H in group 1, He in group 2

(c) H in group 17, He in group 18

(d) H shown in both groups 1 and 17, He in group 18

(e) H in group 14, He in group 18

(f) H floating, He in group 18

Can you give me a source that actually used his one?--R8R (talk) 12:20, 21 September 2015 (UTC)

Here's Kaesz and Atkins. Double sharp (talk) 07:58, 24 September 2015 (UTC)

(g) H and He both floating (as in Greenwood). Greenwood has H and He both float, but they are floating adjacent to each other

I recognize that these are a lot of tables, so maybe it is better to just show the first few periods of each. I'm not proposing we change anything in the WP table. The status quo (a) is fine. (I could support (f) or (g), but none of the others.) Double sharp (talk) 09:59, 21 September 2015 (UTC)

Got it, thanks. It must be possible to make these seven illustrations, not too big to go with the prose. -DePiep (talk) 10:43, 21 September 2015 (UTC)

Might wanna cut at period 4 – that way we save some vertical space, and don't drag the group 3 thing into the mess. Double sharp (talk) 10:45, 21 September 2015 (UTC)

(Oh, missed a small point above for (g); Greenwood has H and He both float, but they are floating adjacent to each other.) Double sharp (talk) 10:52, 21 September 2015 (UTC)

Low on inspiration, receiving only one bar out of five. -DePiep (talk) 11:03, 1 November 2015 (UTC)

Group 12: Post-transition metals

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

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

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

See this article. "At high pressure the alkali metals potassium, rubidium, and cesium transform to metals that have a d¹ electron configuration, becoming transition metal-like." Actually, I think you linked it first! This is why I'm not so keen on the d-electron argument: there isn't an obvious place to draw the line (except saying "any d-electron involvement at all counts", which makes Hg a transition metal but Cd not, which is rather silly). If you say Hg doesn't pass it, how do we know if E113 and Fl are going to?

I wouldn't say Zn, Cd, and Hg have no TM properties. Zn forms stable complexes with O-, N-, and S-donor ligands as well as halides and CN⁻, like a transition metal, and Cd is rather similar. Additionally, many compounds of Hg^II (and Cd^II to a lesser extent) are highly coloured, a characteristic of transition metals. It seems like the easiest way to go is to call everything from groups 3 to 12 TMs, but call Sc/Y/La/Ac and Zn/Cd/Hg borderline TMs.

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

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

Thank you for stating where the line is drawn: does it happen at standard conditions, or not? Now we have a completely sound basis to exclude group 12, so now I agree with you. (Really: apart from not being sure where to draw the line, I don't really have any objections to taking out group 12).

Now how do we colour copernicium? Minor element, I know, but it's going to show up on all our periodic tables. Physically it is known to behave like a typical group 12 metal, except that it is predicted to have Cn(IV) as a significant part of its chemistry. The trouble with flerovium is that Fricke's old predictions mention sd hybridisation as a possibility, but the recent sources all assume Fl(IV) would go for sp³; this seems to imply that new predictions have the 6d energy levels too low. E113 OTOH is still predicted to have 113(V) AFAIK. But these are all still predictions, so maybe we ought to just colour them like their groups (post-transition metals) until we get evidence otherwise. After all, the only thing we know about them so far is that physically, they behave like their congeners in their groups. As such we would colour Cn and Fl as post-transition metals (and E113 as a predicted PTM) until evidence comes in for Cn(IV), 113(V), and Fl(VI). This is what Japanese Wikipedia does (or at least it will until Dlgkstmf ar1103 insists on restoring his/her favourite classification from his/her favourite science magazine, which apparently is not aware of the existence of chemical studies on At, Fr, and Fl but spontaneously invented studies on Mt, Ds, and Rg out of thin air, and has some penchant for keeping Ge and Sb out of the metalloid club while putting Se in it, and before I revert him/her again). In fact, they're very similar to your current proposal for the right side of the periodic table (but calls those just "other metals", and puts Be and Mg in that club; is that traditional there?). (talk) 15:20, 13 November 2015 (UTC)

So it got protected at The Wrong Version. Oh well. Time to write more screeds on the talk page (alas in English). (I do find the classification of Be and Mg outside the alkaline earth metals interesting. I will ask about that.) Double sharp (talk) 06:04, 16 November 2015 (UTC)

Data for group 3 trends

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

Properties of selected elements
Z	Symbol	Atomic weight	Electronegativity (Pauling)	1st Ionization Energy (eV)	Atomic radii(pm)	Covalent radii(pm)

19	K	39.10	0.82	4.34066	220	196
20	Ca	40.08	1	6.11316	180	174
21	Sc	44.96	1.36	6.5615	160	144
22	Ti	47.87	1.54	6.8281	140	136
23	V	50.94	1.63	6.7462	135	125
37	Rb	85.47	0.82	4.17713	235	211
38	Sr	87.62	0.95	5.6949	200	192
39	Y	88.91	1.22	6.2171	180	162
40	Zr	91.22	1.33	6.6339	155	148
41	Nb	92.91	1.6	6.75885	145	137
55	Cs	132.9	0.79	3.8939	260	225
56	Ba	137.3	0.89	5.2117	215	198
57	La	138.9	1.1	5.5769	195	169
71	Lu	175.0	1.27	5.4259	175	160
72	Hf	178.5	1.3	6.82507	155	150
73	Ta	181.0	1.5	7.5496	145	138

Atomic number & symbol
1	2	3		4	5
19 K	20 Ca	21 Sc		22 Ti	23 V
37 Rb	38 Sr	39 Y		40 Zr	41 Nb
55 Cs	56 Ba	57 La	71 Lu	72 Hf	73 Ta

Atomic Weight
1	2	3		4	5
39.10	40.08	44.96		47.87	50.94
85.47	87.62	88.91		91.22	92.91
132.9	137.3	138.9	175.0	178.5	181.0

Electronegativity (Pauling)
1	2	3		4	5
0.82	1	1.36		1.54	1.63
0.82	0.95	1.22		1.33	1.6
0.79	0.89	1.1	1.27	1.3	1.5

1st ionization energy
1	2	3		4	5
4.34066	6.11316	6.5615		6.8281	6.7462
4.17713	5.6949	6.2171		6.6339	6.75885
3.8939	5.2117	5.5769	5.4259	6.82507	7.5496

Atomic radii
1	2	3		4	5
220	180	160		140	135
235	200	180		155	145
260	215	195	175	155	145

Covalent radii
1	2	3		4	5
196	174	144		136	125
211	192	162		148	137
225	198	169	160	150	138

Ionic radii
1	2	3		4	5
152	114	88.5		74.5	68
166	132	104		86	78
181	149	117.2	100.1	85	78

Melting and boiling points, degrees Celsius
1	2	3		4	5
63.5 759	842 1484	1451 2836		1668 3287	1910 3407
39.3 688	777 1382	1526 2930		1855 4377	2477 4744
28.5 671	727 1637	920 3464	1652 3402	2233 4603	3017 5458
27 677	700 1140	1051 3197	1627	2100 5500

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

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

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

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

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

Ionization energies (1st for group 1, 1st+2nd for group 2, etc.)
Ionic radii
Redox
mp/bp

Any effect for our group 3 presentation from this? -DePiep (talk) 02:09, 2 December 2015 (UTC)

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

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

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

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

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

Young's modulus, GPa
1	2	3		4	5
3.1	2.0	74		116	128
2.4	15.7	64		68	105
1.7	13	37	69	78	186
	13.2	25

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

Group 3 in detail

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

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

I think this material should be included because it concerns the classification of an important part of the periodic table, and it is an open question in the literature. If it is one-sided, perhaps it can be improved and made more neutral, but that is not a reason to delete it entirely. Dirac66 (talk) 11:13, 18 September 2015 (UTC)

I like that brief overview of group 3, giving a few reasons for each, but I've got two objections: firstly, it's biased towards Sc/Y/Lu/Lr (while it may be the best, it is not obviously better than Sc/Y/La/Ac and Sc/Y/*/**), and secondly, the referencing is terrible for an FA. So I would keep it, but address these problems. It's not obviously worse than the H/He section, which also states all alternatives and gives reasons for each. Double sharp (talk) 12:23, 18 September 2015 (UTC)

~~That said,~~ all of this is true and correct, yet should be discussed into the main articles (~~period 1 and~~ group 3). Here, we should just outline the problem and give a proper Main article link.--R8R (talk) 19:52, 18 September 2015 (UTC)

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

@Double sharp: Sandbh is cleaning up the prose in userspace (see above), so I have no objections anymore to the material being removed. Karl Dickman ^talk 05:49, 19 September 2015 (UTC)

OK.

Done. Double sharp (talk) 09:25, 19 September 2015 (UTC)

I love the development now visible in User:Sandbh/sandbox. The group 3 issues described well. I want to add some notes:

The topic & the prose is quite complicated. I had to make notes to get it. Can it be lighter?

Also, I think we could add two images: a PT with group 3 being Sc/Y/La/Ac and Sc/Y/Lu/Lr respectively. This could be parial only, say groups 1-4 and periods 4-5-6-7 only, f-block included. Shall I make some sketches, Sandbh?

The IUPAC PT is a horror [1]. Not about its status (being "not the only official one"), but because it is ambiguous graphically. It ~~suggests~~ states that, by substituting (inserting) the two rows from below, the whole f-block is group 3 and Sc, Y elements somehow span 15 columns (R8R wrote: Sc/Y/*/**). I am not convinced this is what it purports to say. We do this in our first periodic table image too. A second ambiguous hiccup are the two dashed lines that vertically make a connection between group 3 and La, Ac (it is screaming: but what about the other 14+14 elements?).

-DePiep (talk) 11:22, 20 September 2015 (UTC)

Thank you DePiep. I will try and make the prose clearer. Pls go ahead with the partial images and let's see how they look. We don't necessarily have to link to the IUPAC table. Sandbh (talk) 12:03, 20 September 2015 (UTC)

Maybe they were thinking of it meaning Sc/Y/La*/Ac**, i.e. Ln = La and the following elements? How very obfuscated indeed. It seems like an attempt to offend no one, with it possibly implying both Sc/Y/La/Ac (where the f-block is Ce–Lu) and Sc/Y/*/** (where somehow the f-block elements have become degenerate d-block elements).

Of course Sc/Y/*/** is pretty good for an introduction, where the f-block is unimportant. Unfortunately it also implies that since the lanthanides and actinides are in group 3, they won't exhibit common higher oxidation states, which is not true (e.g. UF₆). Oh, the urge to make everything show Sc/Y/La/Ac like Greenwood and Holleman-Wiberg. Might annoy some people, but given how common this layout is no one can say it does not serve the beginner well (if not, why is it in very many introductory textbooks?) And it would score one over Sc/Y/*/** for the consistency argument (18- and 32-column tables that are both for general purposes should look the same), given that I'm not sure we can easily get rid of 32-column here thanks to the width issue. Double sharp (talk) 13:36, 20 September 2015 (UTC)

...sigh. No, as much as that would work, I don't like Sc/Y/La/Ac anymore (I don't feel like repeating all the arguments), and IUPAC gives me a perfect excuse not to use it. Sc/Y/*/** it will have to remain, then. Double sharp (talk) 13:52, 20 September 2015 (UTC)

I think the idea of partial PT's to illustrate this section is a great one. Originally, I was thinking of suggesting a side-by-side comparison akin to Periodic table § Layout variants, but having seen the sandbox, I now think having a separate PT for each paragraph would be best. YBG (talk) 14:12, 20 September 2015 (UTC)

About the IUPAC PT. as DS writes: Maybe they were thinking of it meaning .... Says it all: a useless ambivalent presentation by IUPAC. Since Sandbh suggests it can be left out, I'm convinced it does not offer a new or different variant (next to the two discussed). So we can leave it out without loss of information (instead, it's cleaning up this neighborhood!). I support it being removed from the sandbhox.

DS, Sc/Y/*/** is pretty good for an introduction - I strongly disagree, because it has the IUPAC ambiguity. IMO, we must show a gap. Compare: this (bad) vs. this (good). The variant Sc/Y/La/Ac should have the gap to the right of group 3 (good), but will not be our general presentation. -DePiep (talk) 09:15, 21 September 2015 (UTC)

Yes, if we need to be consistent (and I still think that's a good argument, despite FJ comments), Sc/Y/*/** is ruled out. (I say it's an OK introduction as some have used it that way, and because when you first learn chemistry the f-block is really unimportant. But once consistency with 32-column is needed, it shatters.) I mention Sc/Y/La/Ac as a possibility for general presentation only because (1) it ought to keep the FJ-ish derision towards Sc/Y/Lu/Lr out, as it's been used as a general-purpose PT by such esteemed textbooks as Greenwood & Earnshaw and Holleman & Wiberg, and (2) it allows for a table with a clear gap between groups 3 and 4, as you say. Double sharp (talk) 09:56, 21 September 2015 (UTC)

Yes. Sc/Y/La/Ac will be described in detail in the section Sandbh is making. There will be two equal graphs available. After that, we can choose one variant to be the general one for out PTs. (Can you give article(s) that have the two sources?). -DePiep (talk) 10:40, 21 September 2015 (UTC)

@Sandbh, a text suggestion. Maybe leave out "rare earth metals" completely, even making the quote like: ... "closer in their chemical properties to lutetium [...] than they are to lanthanum." REM is another new term to be looked up when reading, adding to the complications. It should be defined elsewhere in the article sure. -DePiep (talk) 10:40, 21 September 2015 (UTC)

What's wrong with "closer in their chemical properties to lutetium and the other heavy [lanthanides] than they are to lanthanum."? Heavy REM = heavy Ln, as the only REMs that are not Lns are Sc and Y, which are light (though Y acts as a heavy one). Double sharp (talk) 10:51, 21 September 2015 (UTC)

Nicer reading indeed. It is a literal quote from a source, so maybe remove the "'s before rephrasing. -DePiep (talk) 11:16, 21 September 2015 (UTC)

So Double sharp, do you know one or two articles that have the links to Greenwood & Earnshaw and Holleman & Wiberg sources you mentioned? Would be great. -DePiep (talk) 23:16, 21 September 2015 (UTC)

Try fluorine or yttrium. (The latter perhaps more relevant.) I have no doubt that there are many more. Double sharp (talk) 01:32, 22 September 2015 (UTC)

Graphic presentations

The essential two structures are:

Group 3 = Sc/Y/Lu/Lr

Tin

Corresponding 18-column graphs: designated 14LaAc (right, and {{this}})

Group 3 = Sc/Y/La/Ac

{{Periodic table (32 columns, micro)/Sc-Y-La-Ac/sandbox|mark=Sc, Y, La, Ac}}

Corresponding 18-column graphs: designated 14CeTh (right, and {{this}})

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

Yeah, electron configuration is kind of the point here. Sc = [Ar]3d¹4s²; Y = [Kr]4d¹5s²; La = [Xe]5d¹6s²; Ac = [Rn]6d¹7s²; Lu = [Xe]4f¹⁴5d¹6s²; Lr = [Rn]5f¹⁴7s²7p¹. Double sharp (talk) 10:11, 21 September 2015 (UTC)

OK. -DePiep (talk) 10:32, 21 September 2015 (UTC)

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

Side-issue (confirmed)

Centuries from now, our distant descendants will remember 21 September 2015 as the day when all of humanity (as represented by two Wikipedia users) agreed to create a stable, uniform, and consistent table for their use, and those future people will celebrate and worship that date along with the user names of DePiep and Double sharp! No, of course they won't. This is just another example of DePiep creating a "Recap" that he can use in future arguments about whether a consensus was reached. The English and German Wikipedia tables both had IUPAC's editorial choice as the main table of our Periodic Table articles for about a decade before DePiep changed ours in 2014, and trying to discuss it with him is impossible because he doesn't even know what it means to "frame a question." But the reality is that the Sc/Y/Lu/Lr version is pretty good too, and a hodgepodge of inconsistent tables reflects both the scientific community and Wikipedia. I should be grateful that other editors here deigned to consider the woefully ignorant choice made by the moron at IUPAC who created their website. This paragraph contains mockery in the first sentence and the sentence before this one. Flying Jazz (talk) 17:00, 4 October 2015 (UTC)

@Flying Jazz:, you acknowledge that your comment contains mockery, and I appreciate that acknowledgement. I know you are interested in making WP a good encyclopedia, and it seems to me you are especially concerned that WP meet readers' needs. So I'm wondering, is there a connection between these two? Is there some way you were expecting mockery in talk page space might lead to improving quality in article space? Or maybe there was another reason for the mockery. YBG (talk) 23:33, 4 October 2015 (UTC)

This isn't a Periodic Table issue, so I've continued the discussion at User_talk:Flying_Jazz#Mockery.2C_ridicule_and_derision. Flying Jazz (talk) 00:45, 5 October 2015 (UTC)

Interesting, even today. -DePiep (talk) 22:52, 26 November 2015 (UTC)

Yep. -DePiep (talk) 21:21, 9 December 2015 (UTC)

The IUPAC immature drawing

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

"We need to mention the confusing IUPAC one as well" -- noooo, please kill that gem from now on. It is not even one of the 700 true PT variants that exist, not even fringe opinion, just a very sloppy one. Where its it mentioned that we would have to mention it at all? -14:48, 21 September 2015 (UTC)

We are here to represent the current science, not to build a new one. That's the main idea of Wiki.--R8R (talk) 14:55, 21 September 2015 (UTC)

We have to mention it, because it is in use by a significant number of people. You don't have to like it – I don't either – but that cannot get in the way of documenting what has been used. Double sharp (talk) 15:06, 21 September 2015 (UTC)

DS: "in use by a significant number of people" -- I object, and will do so by capslock. The IUPAC PT image is not part of the group 3-issue at hand. At all. It should not be in the Sandbh/sandbox prose. Doing the two variants correct, we are fine here at wiki. (And then again and again I ask, Double sharp: actually WHAT FORM or INTERPRETATION of that iupac PT you refer to?). -DePiep (talk) 21:50, 21 September 2015 (UTC)

And yet it is in Sandbh's sandbox, in the last paragraph, as it should be. I refer to the interpretation of it as Sc/Y/*/**, i.e. with every lanthanide and actinide being somehow a group 3 element; there does not seem to be any other common one. This is mentioned by Jensen, who calls it "15LaAc", and of course criticizes it roundly (as 32-columnizing it means that the Sc and Y cells have to stretch 15 columns). (He calls Sc/Y/Lu/Lr "14LaAc" and Sc/Y/La/Ac "14CeTh", referring to the number of columns in the resultant f-block portrayal, and then its first column.) Double sharp (talk) 01:23, 22 September 2015 (UTC)

(outdent)

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

re "These are issues of interpretation and how to frame the issue". What 'framing' are you talking about? We already have the issues "scientific PT" (like: Mendeleev, Mendeleev +, Jane, Adomah) and the graphic forms like 18-column or 32-column. Now you really want to introduce a third dimension of confusion called "frame"? Please explain & convince. Oh, and if you care to reply: stop using words like 'ridiculous'. That's not an argument. -DePiep (talk) 00:36, 4 October 2015 (UTC)

@DePiep:, I think that by "framing", Flying Jazz means that there are different ways of presenting a problem (how the PT should look) that are "framed" or presented by what question you ask. YBG (talk) 02:16, 4 October 2015 (UTC)

If so, he should have said so. -DePiep (talk) 22:28, 7 October 2015 (UTC)

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

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

Suggestion that PT grouping applies to footnote elements

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

Now Sandbh and Double sharp earlier promoted to:

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

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

PT claiming groups for f-block elements

Periodic table
Group	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
5	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	Lanthanum57La138.91	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	Actinium89Ac	Rutherfordium104Rf	Dubnium105Db	Seaborgium106Sg	Bohrium107Bh	Hassium108Hs	Meitnerium109Mt	Darmstadtium110Ds	Roentgenium111Rg	Copernicium112Cn	113Uut	Flerovium114Fl	115Uup	Livermorium116Lv	117Uus	118Uuo

group, stated (N)				4	5	6	7	8	9	10	11	12	13	14	15	16	17
				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
				Thorium90Th232.04	Protactinium91Pa231.04	Uranium92U238.03	Neptunium93Np	Plutonium94Pu	Americium95Am	Curium96Cm	Berkelium97Bk	Californium98Cf	Einsteinium99Es	Fermium100Fm	Mendelevium101Md	Nobelium102No	Lawrencium103Lr

Undisputed in this:

group 3 = Sc/Y/La/Ac

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

By what RS?
Why should enwiki make this a general feature (a common presentation)?
How are these group numbers/statements in any 32-col PT? (RS PT's, enwiki PTs, ..)
~~Does the switchover by Lu/Lr from group 17 to group 3 make sense?~~

Better: If I pick the (scientifically correct) group 3=Sc/Y/Lu/Lr structure, where does that leave Ce, re group number & footnote graphical position? And all the others in the footnote?

-DePiep (talk) 22:52, 24 November 2015 (UTC); (better:) -DePiep (talk) 20:54, 25 November 2015 (UTC)

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

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

Ba   La   *    Hf   Ta
Ra   Ac   **   Rf   Db

          *    Ce   ...
          **   Th   ...

and

Ba   *    Lu   Hf   Ta
Ra   **   Lr   Rf   Db

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

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

re "I'd like to say I don't like ...": well I do. I am academic in sciences, and I know in science &tc we treat independent topics independently. That is how we solve & explain things in science. If you can point out why & where these topics are related: please do so. But until today, the group 3-composition has nothing to do with graphing Ce in a group. They are independent. (And thank nature & our mind: that is the way we can understand those things!). No use or need to throw all topics into one soup. -DePiep (talk) 21:15, 26 November 2015 (UTC)

@R8R Gtrs: That is my (YBG's) understanding, also: it is how I understood what Sandbh (and Double sharp) wrote, which I have copied here from above.

@Sandbh: Please correct me if I have misuderstood you. You wrote on 00:28, 15 November 2015 (UTC) "The footnoted Ln and An line up nicely under the main body group numbers: Ce under group 4; the transition-metal-like earlier actinides---Th, Pa, U, Np, Pu---under groups 4, 5, 6, 7 and 8." (emphasis added) You later added on 23:58, 21 November 2015 (UTC) "The actinides do not have group numbers but lining up Th, Pa, U etc with Hf, Ta, W etc has a strong historical basis and is consistent with the early actinides showing some similarities to transition metals." In the first quotation, I believe that by under, you did NOT mean under in a membership sense, but rather in a spatial sense. Please correct me if I am wrong.
@Double sharp: Please correct me if I have misunderstood you. You answered Sandbh on 14:29, 22 November 2015 (UTC) "Exactly. And furthermore, if it truly does not matter where the footnote is, then they can very well be aligned this way in the 18-column table and not in the 32-column table." So I believe you also agree that the 'footnote elements' do not belong to any of the groups 1-18.

So, Double sharp & Sandbh, please correct me if I am wrong in my understanding that both of you believe that 'footnote elements' do not belong to any of the groups 1-18, but that it would be beneficial, all else being equal, if those elements could be placed physically below non-footnote elements with which they share some chemical similarities.

Again, please, if I have misunderstood what anyone has stated, please correct me. YBG (talk) 08:34, 26 November 2015 (UTC)

Yes, thank you, both of your understandings are correct. Sandbh (talk) 09:10, 26 November 2015 (UTC)

Yes, you're absolutely right. Double sharp (talk) 13:02, 26 November 2015 (UTC)

Nicely shows the issue at hand. YBG is rephrasing: "is group 4" into "but ... they share some chemical similarities", with Sandbh and Double sharp confirming. This introduces a new, secondary-group-thing. I don't deny those similarities exist, duh. Still, none of my questions are answered by now. Not one of four. To check our thinking process, I add: why would we add this information to the graph, while not actually confirming it (eg by repeating group number?). Only suggesting (I must assume it extends to all 14 footnote columns)? There are more stable facts that a general PT does not or can not show. -DePiep (talk) 20:57, 26 November 2015 (UTC)

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

~~group 3 = Sc/Y/La/Ac~~

~~elements 58(Ce)–71(Lu) and 90(Th)-103(Lr) do not belong to any periodic table groups~~

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

I have struck out my post above as it seems to have caused some misunderstanding. Let me try to restate:

So it seems that we all agree that

elements 58(Ce)–71(Lu) and 90(Th)-103(Lr) do not belong to any periodic table group.

YBG (talk) 21:44, 26 November 2015 (UTC)

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

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

"Why is this undisputed;" - No, bad quote. I wrote: "Undisputed in this: "group 3=Sc/Y/La/Ac". That is: For this topic, we can assume that (sake of reasoning). Also, in my objection #4 I mentioned the alternative (whith possibly more devastating effects). How wold your reasoning be, with this? (actually, I can't follow it because of this). -DePiep (talk) 21:05, 26 November 2015 (UTC)

@DePiep: Thank you for clarifying. I had completely overlooked your phrase "in this", which I now understand better. Thanks! YBG (talk) 21:44, 26 November 2015 (UTC)

Thanks. m. You just won one bonus point: 24h of patience from me ;-). -DePiep (talk) 22:44, 26 November 2015 (UTC)

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

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

Misunderstanding, R8R Gtrs? This section is *not* about group 3 at all. It is about: do we show & state that "all Ln/An are in a group" (eg Ho is in group 13)? -DePiep (talk) 00:29, 27 November 2015 (UTC)

@R8R Gtrs: No, the problem is not that you didn't make yourself clear, but rather that I was too myopic to remember and not thoughtful enough to craft my words carefully. Let me try again to express more carefully what I believe we all agree on.

In the 3=Sc/Y/La/Ac PT arrangement being considered here, the footnoted elements 58(Ce)–71(Lu) and 90(Th)-103(Lr) are not considered to be a part of PT groups.

To this it seems one could add:

(a) nevertheless, all else being equal, it would be nice to place the footnoted elements physically underneath chemically similar ones.
(b) consequently, it is best to place the footnoted elements in such a way as to make it obvious that no group membership is implied.

In this, it seems that Sandbh and Double sharp would agree with (a) and DePiep would agree with (b). Am I right in this? I'm not quite sure which of these R8R Gtrs would agree with. YBG (talk) 02:15, 27 November 2015 (UTC)

Oh. I didn't realize that at first (nor at second). In that case, if I was to decide, I would choose the former option for the reasons already said, as I have little to add to what I already posted on the matter. The fact no membership is implied is best shown by some vertical space between period seven and the lanthanides.--R8R (talk) 10:07, 27 November 2015 (UTC)

Please, can we come home to my original Four Questions? (Fm is in group 14 then?) -DePiep (talk) 22:35, 27 November 2015 (UTC)

@DePiep: your questions are based on the following statement that you posted above:

Now Sandbh and Double sharp earlier promoted to:

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

@Sandbh and Double sharp: does this statement accurately reflect your POV? YBG (talk) 05:34, 28 November 2015 (UTC)

No. Ce is not in group 4, although it shows some similarities to the group 4 elements. Sandbh (talk) 05:45, 28 November 2015 (UTC)

No. Okay...the main reason I support this, with Sandbh, is that while Ce is not in group 4 (nor indeed any group), it is more similar to the elements of group 4 than those of any other group. Same for Th, Pa, U, Np, and Pu in groups 4 to 8 inclusive. This also complements the similarity of the lanthanides and actinides to group 3 in general, as shown by the position of the asterisks (they come right after group 3). You cannot show both similarities in a 32-column table, but I think that if it is possible to show something, we should – especially since displacing these elements by a column or two aligns them under elements they are not closely related to. Fm in group 14 isn't true, but this is the only possible location for the f-block where at least some of these relationships are true. So it is a nice touch for the reader who knows this, and for the reader who doesn't, it doesn't really matter where the footnote is.

But TBH, this is a minor issue. I don't mind if it doesn't go through. (In fact I'm getting rather close to not really caring about any of this, save the question of group 12 as transition metals – which is a separate topic that I will not go into further here.) Double sharp (talk) 12:02, 28 November 2015 (UTC)

re Double sharp saying "in groups 4 to 8 inclusive": so the statement is explicitly not valid for all footnoted elements. The thing that is being illustrated is not to be read for elements Eu to Lu (under group 9 and up+). I'd say that is a misleading presentation then. -DePiep (talk) 18:39, 28 November 2015 (UTC)

Ce "shows some similarities to the group 4" - being true, isn't that a weak statement to be made in a general PT? In parallel, there are other statements like "Xx is considered a metalloid by multiple sources" is correct too, but we don't cram that in either. We'd have to add extra explaining legends or, er, footnotes. -DePiep (talk) 18:49, 28 November 2015 (UTC)

Apart from this all, there is a graphic bad issue with this group 4 (etc) suggestion. That is: this way, once the 18-col 14CeTh from is drawn, the CeTh column is tied to group four! One can not freely remove it (the whole footnote rectangle 30 elements), freely, from its position without breaking the original '(bad) intent'. Not into a 32-col PT! What is it: are the asterisks to be followed, or is the "4, ..." column numbering to be followed? My mind hurts. -DePiep (talk) 02:19, 2 December 2015 (UTC)

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

Hoffman DC 2009, 'The Periodic Table Key to Past "Elemental" Discoveries—A New Role in the Future?', Journal of Chemical Education, vol. 86, no. 10, pp. 1122–1128
Hoffman DC & Lee DM 1999, 'Chemistry of the Heaviest Elements—One Atom at a Time', Journal of Chemical Education, vol. 76, no. 3, pp. 331–347

Great! Thanks for this Hey DePiep, here's a sketch. Still, what you claim is OR and not RS. Could be Fringe. Bye. -DePiep (talk) 22:52, 11 December 2015 (UTC)

This is not OR as it has been sourced from reliable, peer reviewed sources. Positioning the Ln and Ac so that Ce-Th are aligned with, but separate from, Ti-Zr-Hf-Rf is quite common and clearly not fringe. Even the dreaded table appearing on the IUPAC site features such an alignment. Sandbh (talk) 03:42, 12 December 2015 (UTC)

I did not contest or deny (ever) the claims Hoffman (2009) makes wrt these similarities in behaviour. And she even has them graphically supported correctly (text and graph correspond, great!).

I do claim that (1) adding this information to the graph is way too detailed (unimportant) for our general purpose PT. Of course they can be described in some dedicated article or section, maybe we'll see Hoffman there too. (2) If we want to add this info, it must be done graphically correct. i.e., the graph must explicitly show the data correct. The Hoffman graph does so. Hoffman also has the feature that it does not use the asterisks, so it does not point to replacement (footnote elements in the main body ie into 32-col). This way she prevents the bad issue that the graph would have contradicting features: does Ce belong below group 4 by position or somewhere between group 2 and 4 by asterisks? But the proposed PT does have this ambivalence and there for is graphically incorrect (elements are taken out into a footnote, and then the footnote is tied to a different position wrt the main table). Also, the proposed PT does not have the stairs. So it does not differentiate between Ln's, nor between Ln-An in a column (Hoffman does). About the IUPAC PT: why do you conclude that they want to convey this information? How is that put into their PT? What does it say about the gradual change of fact both in period and in group (column) for these? So far, IUPAC only happens to show it this way. And, not unimportant, for other reasons other topics their PT already is dismissed. -DePiep (talk) 12:35, 14 December 2015 (UTC)

TL;DR Given the undisputed scientific facts, the Hoffman graph has everything right. Transplanting any of her graphic features into the our general (18-column only) PT is wrong always. -DePiep (talk) 09:42, 19 December 2015 (UTC)