Wikipedia talk:WikiProject Elements/Archive 32

Archive 25

←

Archive 30

Archive 31

Archive 32

Archive 33

Archive 34

Archive 35

Professor Poliakoff on group 3

Prof Poliakoff has a rather good YouTube video discussing the group 3 question. Early on, the video shows two long two periodic tables in which the d block is split into group 3, and then groups 4 to 10.

The prof discusses the implication of the recent experimental measurement of the ionisation energy of lawrencium, and speculation this would support the placement of Lu and Lr in group 3 [thereby eliminating the possibility of a split d-block in long form depictions of the periodic table].

Regrettably, as per Jensen (2015), the ionisation energy of Lr has no particular bearing on its positioning in the periodic table.

At the end of the video the professor entertains the amusing possibility that the "break" in periods 6 and 7 could occur in different positions i.e. that La might go under Y and Lr under La! He says this would be a real mess for the people that draw periodic tables but that,

"What we are interested in is what nature is like not how easy it is to draw it."

To which I say, "Bravo Prof Poliakoff!" -- Sandbh (talk) 06:34, 7 December 2017 (UTC)

• Jensen WB 2015, "Some comments on the position of lawrencium in the periodic table," Department of Chemistry, University of Cincinnati, Cincinnati
• Prof Poliakoff in the video is needlessly complicating the topic by mixing the scientific statement (different composition of group 3) with the graphic depiction of that statement (the placeholder * stuff, nicely but ambiguously animated in the video).

Simple: if there is a "3" in the top row, that says that all elements in that column are "3". Even those that are valet-parked elsewhere by astrerisks. Because: the asterisks are in that column "3". That is the scientific statement these older PTs make. With the two PTs earlier in the video (animated one and brown-background one), the video sneakily removes elements away from under the "3", suggesting that they are not a "3". This way the prof creates a different structure (scientific statement), that was not in the original drawing; this is done before (and irrespective of) processing the new Lr data.

As I said before, draw the periodic table in 32-column format and you are confronted with the issue right away. And I don't think using "long"-terminology covers this f-block issues well — if at all. "Long" was introduced 100 years ago to differentiate from Mendelev's "short" one, which had groups I–VIII (in 10 columns: VIII had three subcolumns=todays 8/9/10), and two rows (two Reihe) per modern period. (Later these two Reihen were aligned into one modern period (row), creating the A/B groups; This was called the "long" form. No f-block in sight back then).

Interestingly, the Wikipedia periodic table visible in the end of the video is the new one Poliakoff proposes, but we abandoned this one (it had group 3 = Sc/Y/Lu/Lr). -DePiep (talk) 08:45, 7 December 2017 (UTC)

Survey of periodic tables, n = 132

I recently had a look through a random collection of 132 chemistry books dating from 1970.

Results:

Sc-Y-La = 80.5 or 61%
Sc-Y-(La—Lu) = 31 or 23.5%
Sc-Y-Lu = 20.5 or 15.5%

The 0.5 is from one book that showed an La table and an Lu table. The oldest Sc-Y-Lu table was from 1970.

If you exclude the "it’s too hard tables" the breakdown is close enough to 80% Sc-Y-La and 20% Sc-Y-Lu.

I also happened upon a 32-column split d-block table published in 1970. Eric Scerri calls this option a Fly in the ointment, as per here. As long ago as 1965, Hamilton published a periodic table extract with a split d block (the gap is between groups 3 and 4) and said that—without any fuss—this is "the periodic table as it is usually presented". I prefer to think of it as the asymmetric reality of the inconvenient facts of chemistry.

Dates of Sc-Y-Lu tables:

1970 (1)
1974 (1)
1978 (1)

1982 (1)
1986 (1)
1987 (3)
1989 (0.5)

1999 (1)
2000 (1)
2001 (2)
2002 (1)
2003 (1)
2006 (2)
2007 (1)
2009 (1)
2011 (2)

Jensen published his article recommending Sc-Y-Lu in August 1982. Maybe there was a weak flurry of recognition in the late 80s. There is a big gap from 1990 to 1998, followed by a modest resumption of hits in the naughties.

I see websites of the ACS, RSC, and GDCh (German) show Sc-Y-La, as do the big three chemistry compendia Cotton & Wilkinson etc; Wiberg; and Greenwood & Earnshaw. My own chemical society, the RAIC shows La—Lu. None of this means much since, for example, everyone dismissed stomach ulcers as having a dominant bacteriological cause until this was shown to be right in the early 1980s, forcing a rewrite of the textbooks. -- Sandbh (talk) 05:50, 9 December 2017 (UTC)

• Hamilton DC 1965, "Position of lanthanum in the periodic table", American Journal of Physics, vol. 33, no. 8, pp. 637–640, doi: 10.1119/1.1972042

I think you meant to say "n = 132".

Did you really see "[group 3 =] Sc-Y-La" in 61% of those? Is that really the asterisks being not under the "3" columnheader? (Or did you read those different from how I do?) That surpises me, because most (internet) PTs I see have Sc-Y-(La—Lu) squeezed into the group 3 column. As has IUPAC. -DePiep (talk) 11:29, 9 December 2017 (UTC)

I saw Sc-Y-La, with Ce to Lu footnoted in 61% of the tables. The La would have an asterisk, or an arrow, or some such to indicate Sc to Lu fit between Ba and Hf, although whether this means all 15 elements are squeezed into the one box under Y, or whether it means the 15 elements occupy one box each in the 32 column form of table, between Ba and Hf, is rarely explained. Sandbh (talk) 06:49, 11 December 2017 (UTC)

Wrong. Which elements are in group 3 in any specific periodic table (drawing), is defined by the elements in the column with header "3"'. Exactly that is the scientific statement being made. Replacing elements with an asterisk, and moving them to a "footnote" (to a separate set drawn in the bottom), is only a graphical (editorial) effect, and does not change the column "3" content. As you describe it, the counted numbers are wrong.

Must say, we can assume that many of those 132 PTs you checked were not discussing group 3, they just copied this detail unwittingly an earlier form. This implies that their PT was not a statement on group 3 at all, and better not be used as a !vote. It was Seaborg that started using those asterisks in 1945, to show other issues.

Periodic tables stating that group 3 is Sc-Y-Lu respectively Sc-Y-La explicitly can do so by using a non-numbered column (header = "") for the asterisks (as does Periodic table). Or, of course, since and when group 3 is the topic, draw it in 32-column form. -DePiep (talk) 11:05, 11 December 2017 (UTC)

I just saw this post and will recheck a sample of the 132, to see if there any interpretation issues. Sandbh (talk) 09:36, 16 December 2017 (UTC)

Group 3 "mess" cleanup

re the quote from the professor as mentioned by Sandbh in the OP:

At the end of the video the professor entertains the amusing possibility that the "break" in periods 6 and 7 could occur in different positions i.e. that La might go under Y and Lr under La! He says this would be a real mess for the people that draw periodic tables but that,

"What we are interested in is what nature is like not how easy it is to draw it."

(end of quote).

Of course, I can not let this go without response. The short reply would be:

"Prof Poliakoff, the mess for those who draw will never be greater than the mess the scientists ask them to draw". -DePiep (talk) 21:27, 9 December 2017 (UTC)

Now for the longer reply. I use subsections for ease of editing. I use the 32-column as base, not the 18-column periodic table. That is because the 18-column is just a crippled and distracting version always — at best. At worst it hides crucial information into ambiguity. Also, I think the 32-column table (ratio) fits a classroom better than the old 18-column PT. I use a cutout of the relevant periods & groups.

Current group 3 alternatives

Today, there are three group-3 compositions in view. One is the "Group 3 = Sc/Y/*/**" (including all Ln and An), the others are "Group 3 = Sc/Y/Lu/Lr" and "Sc/Y/La/Ac".

• The last two are depicted below (of course, they can not exist together in one table: one must choose one).
• The large one is not considered relevant an more by IUPAC, and is not shown below.
• Note that nicely, both alternatives do not break the PT because of the whitespace in periods 4 and 5. Any shell filling order indication is maintained.

Periodic table
	2	3	n/a [14 col]													3-alt	4
4	20 Ca	21 Sc														21 Sc-alt	22 Ti
5	38 Sr	39 Y														39 Y-alt	40 Zr
6	56 Ba	57 La	58 Ce	59 Pr	60 Nd	61 Pm	62 Sm	63 Eu	64 Gd	65 Tb	66 Dy	67 Ho	68 Er	69 Tm	70 Yb	71 Lu	72 Hf
7	88 Ra	89 Ac	90 Th	91 Pa	92 U	93 Np	94 Pu	95 Am	96 Cm	97 Bk	98 Cf	99 Es	100 Fm	101 Md	102 No	103 Lr	104 Rf

Table Basic group 3 options

Group 3 = Sc/Y/La/Lr (1)

As Prof Poliakoff says in the video (see quote), there is a sound argument (shell filling) that group 3 should consist of Sc/Y (unchanged), and ₅₇La and ₁₀₃Lr. This would be alternative number 4 for this group. Given the traditional positions of elements 57 and 103 (quite not in the same column, with elements in between), this is a disturbance.

• A crude first try would be to boldly put ₁₀₃Lr below ₅₇La (see graph).

Periodic table N
	2	3	n/a [14 col]														4
4	20 Ca	21 Sc															22 Ti
5	38 Sr	39 Y															40 Zr
6	56 Ba	57 La	58 Ce	59 Pr	60 Nd	61 Pm	62 Sm	63 Eu	64 Gd	65 Tb	66 Dy	67 Ho	68 Er	69 Tm	70 Yb	71 Lu	72 Hf
7	88 Ra	103 Lr	89 Ac	90 Th	91 Pa	92 U	93 Np	94 Pu	95 Am	96 Cm	97 Bk	98 Cf	99 Es	100 Fm	101 Md	102 No	104 Rf

Table ₁₀₃Lr repositioned

• Then the other Ln's ad An's could be rearranged and spaced to restore columns (Ce above Th: add some white cells).
• But even before we do this, there is this law breaking: the prime order of the PT, by atomic number, is broken by E103. End of story.
• Sure one is free to rearrange elements like this, or for example by atomic weight, or alphabetically, but then it is not a periodic table any more.
• If I got this right, more elements have a irregular shell filling (or atomic weight). - DePiep (talk) 21:27, 9 December 2017 (UTC)

Sc/Y/La/Lr (2, solution)

Still, for the single Sc/Y/La/Lr question, the PT could be redrawn like this:

Periodic table (46-column, accomodating group 3 = Sc/Y/La/Lr)
	2	n/a [14 col]														3	n/a [14 col]														4
4	20 Ca															21 Sc															22 Ti
5	38 Sr															39 Y															40 Zr
6	56 Ba															57 La	58 Ce	59 Pr	60 Nd	61 Pm	62 Sm	63 Eu	64 Gd	65 Tb	66 Dy	67 Ho	68 Er	69 Tm	70 Yb	71 Lu	72 Hf
7	88 Ra	89 Ac	90 Th	91 Pa	92 U	93 Np	94 Pu	95 Am	96 Cm	97 Bk	98 Cf	99 Es	100 Fm	101 Md	102 No	103 Lr															104 Rf

Table PT Sc/Y/La/Lr

• This would be a 46-column PT.
• No Z-order is broken, group 3 is as intended by Prof P.
• This being Ln's and An's, the question could be whether the old columns (like: Ce, Th) should be reintroduced, e.g. by adding relevant column headers. Still this is major: the PT is not build by shell filling sequence.
• If I got this right, there are more irregular shell fillings often in the f-block.

Conclusions

• So it is possible to draw the Sc/Y/La/Lr group 3 in the PT, without breaking anything. However, this may be incidental (incidentally ending good, this time). Because, in general, the PT is not ordered on actual shell filling, and so that may not be enforced without loosing the "periodic". OR loosing the Z-ordering. - DePiep (talk) 21:27, 9 December 2017 (UTC)

Group 3 as Sc,Y, La, Lr?

The 46-column Sc/Y/La/Lr PT is astonishing to behold, and contemplate.

Jensen (2015) discusses, as reported in the 9 April 2015 edition of Nature, the experimental confirmation of the ionization energy 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. Double sharp and I, in our submission to IUPAC (with R8R), agreed with Jensen that this finding is inconclusive.

The group 3 metals, whether Sc-Y-La, or Sc-Y-Lu, behave like main group metals. Basicity increases going down all main groups. This is the case for Sc-Y-La, but would not be the case for Sc-Y-La-Lr, or Sc-Y-Lu. The case for either of the latter two options has not been made.

Sc-Y-La-Ac:

• maintains the principle that a block starts upon the first appearance of the relevant electron thus s-block = H; p = B; d = Sc; and f = Ce.
• results in a pleasing double periodicity as seen in the analogous +2 ions of Eu and Yb, and the +4 ions of Ce and Tb (Shchukarev 1974, p. 118; Ternstrom 1976). A similar, but weaker, periodicity is seen in the actinides (Wiberg 2001, pp. 1643–1645), with a half full 5f shell at Am and Cm, and a full shell at No and Lr. Horovitz and Sârbu (2005) observe a similar double periodicity with an Sc-Y-Lu composition for group 3 but to achieve this they need to counterintuitively arrange Yb to Gd in descending rather than ascending order of atomic number, which is naff as the Brits would say.
• results in the number of f electrons in Ln ions matching their position, with no exceptions, which is quite pleasing e.g. Ce, as the first Ln, has a +3 configuration of [Xe]4f¹; Lu, as the last Ln, has a +3 configuration of [Xe]4f¹⁴;

Restrepo (2017, p. 103), who is on the IUPAC project looking at the group 3 question, and on the basis of an analysis of 4,700 binary compounds of 94 chemical elements recently wrote that

Scerri has discussed about the element at the beginning of the third row of the transition elements, which in some tables is La and in others Lu…Schwarz and Rich have stated that Lu cannot be considered a lanthanoid, for it does not fill f orbitals as they are already filled; and have suggested that Lu should be regarded as a transition metal…According to our results, La appears in between two clusters, one of 11 lanthanoids and another of transition metals, namely {Y,Sc}. Lu is part of the clusters of 11 lanthanoids and the smallest cluster containing it is {Ho,Er,Lu}, which shows that Lu is more similar to lanthanoids than to transition metals, while La share similarities with lanthanoids and with transition metals. Therefore La must be the element located at the beginning of the third row of transition metals [bold added] if chemical resemblances is what it is to be emphasized.

I see websites of the ACS, RSC, and GDCh (German) show Sc-Y-La, as do the big three chemistry compendia Cotton & Wilkinson etc; Wiberg; and Greenwood & Earnshaw. My own chemical society, the RAIC shows La—Lu.

Nobody has yet made a conclusive case for overturning the dominance of Sc-Y-La.

--- Sandbh (talk) 07:42, 11 December 2017 (UTC)

• Horovitz O & Sârbu C 2005, "Characterisation and classification of lanthanides by multivariate-analysis methods", Journal of Chemical Education, vol. 82 no. 3, pp. 473–483
• Jensen WB 2015, Some comments on the position of lawrencium in the periodic table, Department of Chemistry, University of Cincinnati, Cincinnati
• Restrepo G 2017, "Building classes of similar chemical elements from binary compounds and their stoichiometries" in MA Benvenuto and TC Williamson (eds), Elements old and new: Discoveries, developments, challenges, and environmental implications, ACS Symposium Series, American Chemical Society, Washington, DC, pp. 95–110
• Shchukarev SA 1974, Neorganicheskaya khimiya, vol. 2, Vysshaya Shkola, Moscow (in Russian)
• Ternström T 1976, "Subclassification of lanthanides and actinides", Journal of Chemical Education, vol. 53, no. 10, pp. 629–631
• Wiberg N 2001, Inorganic chemistry, Academic Press, San Diego

Also related to the Restrepo quote. Double sharp (talk) 04:59, 26 February 2018 (UTC)

Yet another Group 3 option: Sc/Y only

I don want to disrupt the current IUPAC (Scerri) research wrt group 3, nor anyones holidays. Consider this a Christmas Seasonal Holidays puzzle.

Coming from the graphical approach, we can draw a group 3 as follows:

Periodic table group 3 = Sc/Y only
	2	n/a [15 col]															3	4
4	20 Ca																21 Sc	22 Ti
5	38 Sr																39 Y	40 Zr
6	56 Ba	57 La	58 Ce	59 Pr	60 Nd	61 Pm	62 Sm	63 Eu	64 Gd	65 Tb	66 Dy	67 Ho	68 Er	69 Tm	70 Yb	71 Lu		72 Hf
7	88 Ra	89 Ac	90 Th	91 Pa	92 U	93 Np	94 Pu	95 Am	96 Cm	97 Bk	98 Cf	99 Es	100 Fm	101 Md	102 No	103 Lr		104 Rf

Table Group 3 being Sc/Y only

This says:

• Group 3 only consists of scandium and yttrium.
• All lanthanides and actinides have no group number ("n/a"). That is: 15+15, full f-block
• This would be a 33-column periodic table.
• Periodic table structural integrity is maintained: Z-ordered, groups, periods, and most of block & step presentation.

This opens a new route: there is no need to sqeeze any Ln or An element into group 3 behaviour while downplaying other properties. *All* An's and Ln's, with their irregular, non-cooperative properties and even law-breaking (seen from periodic table point of view; so periodic law that is) are away from group and block inclusion considerations. So this is coming from graphical development department, not from scientific claims. The puzzle is, for those luckily away for three weeks in a log cabin near Rovaniemi doing sauna talk: is there any scientific thruth & usefulness in this? -DePiep (talk) 12:28, 26 December 2017 (UTC)

Good question.

A scientist might say to the graphical development department, "This appears to suggest that there are some missing elements such as 71½ and 103½, or that we have decided it's too hard to work out which elements go under Y, or that we have decided that the periodic law breaks down in group 3. Nice try, but we are concerned with collecting facts about the elements, classifying those facts, and seeing what regularities and patterns can be observed."

We know that Sc and Y behave chemically like trivalent group 1 or 2 metals. Therefore we would expect the elements below Y to continue the same trend going down Sc and Y as we see going down groups 1 and 2. Therefore La goes below Y and Ac goes under La. This means that group 3 is chemically distinct from group 4, hence there is a split d-block in the 32-column form of the table.

Consider also that there is likely to be a split f-block too, should elements 120+ ever be synthesised, as g electrons might not appear until 123 or higher.

I'll say a bit more about the tyranny and limitations of blocks elsewhere on this page. Sandbh (talk) 07:59, 5 January 2018 (UTC)

As a minor nitpick, I'd say that the difference between group 3 and 4 is a lot more solid with Ti than with Zr and Hf, because the latter two in aqueous solution are pretty much always tetravalent; the early actinides can take such large charges and come off no worse than Fe³⁺. It is a lot easier to justify Ti as a transition metal than Zr and Hf, just as it is a lot easier to justify Be as not being an alkaline earth metal than Mg or Ca (though the arguments can still be made for Mg). I think I wrote something about this somewhere and will give a more substantial treatment, but it is essentially once again the thing about basicity being a very "s-block" characteristic and getting larger with decreasing ionic charge and increasing atomic radius (hence Cs as the ultimate base); I will soon write about how that manifests itself across "bloc" boundaries. Double sharp (talk) 09:55, 5 January 2018 (UTC)

re Sandbh's ... hence there is a split d-block in the 32-column form of the table: Time and time again there is this repetitive issue on how to draw the periodic table, especially related to group 3 and lanthanides/actinides/elconfig/shellfilling. This issue seems to up every once in a while, with intermediate periods shortening. I want to kill this periodical trend. From now on, I claim by DePiep's Law of Non-periodicity, that each and every paper & discussion touching group 3 must be described in a 32-column formlike periodic table, not an 18-column one. (If you cannot draw in in 32-column form, it probably has a fault). OR else: propose a new periodic table by structure. - DePiep (talk) 21:55, 2 February 2018 (UTC)

32-column Sc/Y/La/Ac table with split d-block

I must confess that I'm not sure what you mean here. With a Sc/Y/La/Ac table, group 3 is a d-block group, since Sc, Y, La, and Ac all have three valence electrons and the occupied valence orbital with the highest angular momentum is a d-orbital. Group 4 is likewise a d-block group, since Ti, Zr, Hf, and Rf all have four valence electrons and the occupied valence orbital with the highest angular momentum is a d-orbital. But between these we have the columns headed by Ce, Pr, Nd, and so on, where the occupied valence orbital with the highest angular momentum is an f-orbital. Hence we have one column of the d-block (the column with La in it), then fourteen f-block columns (the columns headed by Ce through Lu), and then the other nine columns of the d-block (the columns with Hf through Hg in them). It seems to me that this is a split d-block regardless of how you draw it; in particular, it is also a split d-block in the 32-column form, as shown to the right. Double sharp (talk) 07:03, 3 February 2018 (UTC)

The sections now under Group 3 observations used to follow here, with the first header "Split f-block?" originally being a level 4 rather than a level 3 heading; I have moved them at the suggestion of DePiep, as they start a new topic. Double sharp (talk) 14:59, 25 February 2018 (UTC)

Is crystallogen a case of citogenesis?

I am afraid it could be (and that I may have played a part in it through assuming a bit too much good faith). T_T This having been nearly six years ago, I think I've learned my lesson to shoot first and ask questions later when it comes to uncited and not obviously citable material. Double sharp (talk) 14:38, 2 March 2018 (UTC)

International Year of the Periodic Table of Elements

FYI, the UN has announced that 2019 will be the International Year of the Periodic Table of Elements. The organisation has global-scale plans to raise awareness of chemistry’s pivotal role in solving the world’s biggest challenges.

The celebrations coincide with the 150th anniversary of Dmitry Mendeleev’s discovery of the periodic system, as well as the 100th anniversary of the International Union of Pure and Applied Chemistry. It will also mark a number of other milestones in the history of chemistry, including the discovery of phosphorus 350 years ago, Antoine Lavoisier’s categorisation of 33 elements in 1789 and the formulation of the law of the triads by Johann Wolfgang Döbereiner 190 years ago.

I knew about the first one but not about the other anniversaries. Sandbh (talk) 07:41, 18 February 2018 (UTC)

Oh my! I guess I really had better finish period 3 off with Mg, P, and S! ^_^ Double sharp (talk) 11:50, 18 February 2018 (UTC)

On the diving of the 1s subshell into the "negative continuum" and whether this actually spells an end to the periodic table

There is a very interesting discussion taking place between Droog Andrey and myself at Talk:Extended periodic table#Elements 173 and 174 on this matter that some others among us may also be interested in. Double sharp (talk) 15:01, 25 February 2018 (UTC)

Yellow colour of caesium

Just edited Relativistic quantum chemistry, but there are many other attributions of the yellow colour of caesium to relativistic effects on WP. I guess the real reason is the large plasmonic wavelength of about 420 nm, which makes caesium notably transparent at the violet end of visible spectrum, hence more absorption here. Anyway, I haven't found any sources about the relativistic nature of the yellow colour. Droog Andrey (talk) 22:20, 2 March 2018 (UTC)

For when elements 119 and 120 are discovered

An earlier discussion about this topic, in January 2018, is at /Archive_31#RIKEN started attempt on element 119 in December 2017. [1].

An interesting link re RIKEN was recently added by Double sharp here "Japanese element hunters are trying to extend the periodic table".

-DePiep (talk) 09:27, 12 March 2018 (UTC)

Since the previous thread on the matter ended up with a different topic, I'd like to ask these questions again, so that we know what to do when or if these situations come about.

1. Suppose element 119 has just been discovered. Now,
   • a. Do we put it on our periodic tables? (Excluding the extended ones, where it already is there.)
   • b. How do we colour it on the periodic table, if we put it on?
2. Suppose element 120 has been discovered, but not yet element 119:
   • a. Do we put the undiscovered element 119 in the table?
   • b. If so, how do we colour and border it?
3. Suppose elements beyond 121 are discovered, necessitating the opening of a g-block. How do we deal with this?

And here are my choices:

1. I would personally prefer to put element 119 on the table under Fr (similarly for 120 under Ra), and colour it as "unknown chemical properties" just like we currently do for Og. However, this depends on what other reliable sources do. If, as Droog Andrey has suggested (echoing Theodore Gray's comments, incidentally), elements in the 8th period are ignored by most periodic table manufacturers, then we should follow suit and simply write below the table something like "Some short-lived artificial elements beyond oganesson have been reported, but are not usually included in illustrations of the table; they would open an eighth period. The known period 8 elements are listed below."
2. Following the precedent of putting element 117 in the table back when it was undiscovered, fill up the blanks with a transparent background with the names and symbols still there in a borderless cell. (But, again, if most reliable sources do not do this, I will change my mind.) This is of course assuming that 119+ elements are drawn at all (see my answer to question 1).
3. This is going to be a frightful mess no matter how it is done, never mind that there are lots of calculations and they don't quite agree. I have no preference other than that we follow reliable sources.

If I were making a periodic table poster, I think the most sound strategy to deal with the 119+ elements is to simply leave the table at 118 until enough new elements are discovered to justify drawing the eighth row in. (While 119 and 120 are easy for everyone to draw regardless on viewpoints on group 3, the extra padding on the s-block makes things look ugly.) In fact, if I was making it for kids, I'd even use that as a hook, and float a box somewhere on the poster saying "Scientists have discovered more elements beyond the 118 shown here. These elements should start an 8th period in the table and begin filling the 5g orbitals, starting a new block with totally unknown chemistry. We still do not know how these elements should behave and where they should be placed: perhaps one day you will find out! The known 8th period elements are as follows: [list with entries like '119. Ununennium, Uue']." But again, this is Wikipedia, and we'll follow reliable sources.

Comments? Double sharp (talk) 07:14, 6 March 2018 (UTC)

If a g-block element is discovered before 119-121 (which is unlikely), My opinion would to have an isolated cell below the 7th period, with the elements being arranged in accessing order of protons. Should 120 be discovered before 119 (which is possible), the best way is to put it below Radon and have an empty cell in place of 119. On where to put the g-block, it can be put underneath the actinides? 1.02 editor (talk) 09:41, 6 March 2018 (UTC)

I suppose you mean to put 120 under radium (Og is under Rn). I could agree with some of this idea: it's consistent. Certainly the g-block can be put as a separate footnote under the actinides, but we shouldn't align it there. We shouldn't forget that while the eighth row drawn at Extended periodic table looks beautifully simple, we currently know next to nothing about even the calculated chemistries of the elements beyond 122, and it is very possible that when we do manage to discover their chemistries we will shuffle the row around considerably, just like we did for the seventh row back when Seaborg started filling it up beyond uranium. (About that article, I'm sorry for not having updated the section about what happens past element 173; I promise to get to it, but I have a lot of things to reply to, including your Si review, and not quite enough time to do it.) That is why I'm not too keen on drawing in a g-block until we really know how they act chemically, even if the half-lives will not allow us to do it as quickly as we did for the actinides. But for 119, 120, and 121 (assuming we keep -La-Ac, which I still personally support), I think putting them in may be all right. In the end, what I really want to see is what periodic table manufacturers do, and this may be a problem because I suspect that many might take their drawings from Wikipedia. ^_-☆ Double sharp (talk) 10:46, 6 March 2018 (UTC)

I'm not really an expert on chemistry, but I doubt anything above 122 would be created in the near future and hence we can worry about it later. Periodic table drawers can easily come up with their own designs without the need for Wikipedia. 1.02 editor (talk) 11:09, 6 March 2018 (UTC)

Of course they can; I'm just not sure they will. I suppose this is more nuclear physics than chemistry, but I share your reservations about how quickly we can make these new elements. I'm willing to nevertheless bet that we'll cross the 120 barrier in the 2020s, and maybe get to 124 by 2030. ^_^ Double sharp (talk) 12:01, 6 March 2018 (UTC)

Replies by DePiep.

Having read your original answers, I'll note some principles first:

A. We go by RS.

B. When an element is discovered, it is added to the (basic) periodic table.

C. Visual effects nor "other publishers" should define our presentation. (being "Ugly" is not that useful)

For 119, 120 and 121 each, when discovered, this leads to:

1. Add to their group, right below Fr, Ra or Ac respectively, in period 8 (also added).

2. Per the discovery RS their occurrence is synthetic, so their border shows the dotted line (same as 95–118)

3. As long as chemical properties are not known, the background ("element category") is light gray: "unknown chemical properties" (same as 109–118, only Cn has a known category)

This solves the main questions. Remaining questions:

4. Undiscovered elements with a lower Z (to the left of a discovered element)? I'd say add the cell, border=none (already today meaning: "Undiscovered"), background category "unknown chemical properties", or create new category "not discovered", white(?)

5. In Extended periodic table: wait, we don't have a sign (border) in that distinguishes between "discovered" and "undiscovered". Open question.

6. If definitions on group 3 change, E121 could have to follow or not.

Then there is the g-block issue for 122 and up. Double sharp introduces floating element cells (I understand that is: not tied to the 1–118 PT structure).

My reply: as I understand the reasoning, that is because g-block chemical behavior is highly unknown. But why would g-block position be set by chemical behavior? I'd think its position is determined first by atomic number (no order changes, even if E124 were discovered today), and the PT graph is by block structure (electron configuration, not metalloid behavior). Also, physical properties, especially elements being in g-block, is not unknown right? Unless g-block is proven to start below column Ce-Th, then we'd have to move E124 indeed (with more graphical consequences). i.e., f-block and g-block are groupwise related — similar to current group 3 issue BTW. Also similar to Seaborg's additions, we don't have to get it right first day, pending future research (especially physical properties).

So I don't see a problem adding E124 and up, g-group, as is done in extended periodic table (pushing apart group 3 and the 14+14 lanthanides, actinides). Graphically this could be done with asterisks; I don't think current 18-column graph is suitable for g-block (because two, disjunct blocks are to be transposed: confusing).

DePiep (talk) 20:37, 6 March 2018 (UTC)

The thing about determining block position solely by atomic number is that that sort of thing leads rather to the naïve Aufbau extrapolation where E168 ends up below Og, which all calculations unanimously agree is unlikely to happen; the chemical behaviour of E168 should be more like that of Sn. The periodic law is after all about the periodicity in chemical and physical properties when plotted against atomic number, so starting from the atomic number seems to me rather like having the tail wag the dog. ^_-☆ If we use Nefedov-style E172 below Og as the basis, things are better, but then consider: if the f-block is footnoted, then the g-block needs to be a separate footnote before the f-block. Where could we put E122 on an 18-column table without making it look like it's floating irrelevantly? This problem is at least not as bad as it would be if we didn't show undiscovered elements to fill gaps (then we might have E123 without E122 and an even bigger maze of confusion). Because of the problem of showing a g-block in the 18-column format, and that these elements really would have no known chemical and physical properties save their atomic mass and the half-life of each isotope (and are hence irrelevant for most purposes), I suspect that most periodic table designers would sweep the problem under the rug by just not showing elements past 118 for a while. But again, I would recommend going by RS (IUPAC would be the best, except that it wouldn't recognise the new elements immediately and so its answer would be greatly delayed). Double sharp (talk) 00:32, 7 March 2018 (UTC)

(Is it OK and helpful to expand on g-block here? Looks like the topic is 119, 120 and even 121.( -DePiep (talk) 10:55, 7 March 2018 (UTC)

My main point in mentioning the g-block in my question 3 is just to affirm that I am declining to make a decision on it. As I wrote, "I have no preference other than that we follow reliable sources." That's why the header only mentions 119 and 120 (and I guess 121 if we continue with Sc-Y-La-Ac), because only those are likely to come soon enough and be easy enough to include that we can make preemptive decisions now. Double sharp (talk) 11:35, 7 March 2018 (UTC)

Clear. I'd say, some other time and place this will return. (I wasn't blaming for introducing it, but wondering if huge follow up texts would help). -DePiep (talk) 13:54, 7 March 2018 (UTC)

Yes, I'm willing to bet that we'll be worriedly discussing about this again sometime in the mid-2020s. ^_^ In the meantime, what do the others think about my proposals for dealing with E119 and E120 (eka-Fr and eka-Ra) when they are discovered? Double sharp (talk) 14:21, 7 March 2018 (UTC)

I have created a demo in {{Periodic table/sandbox119}} using existing options.

On second thought, I think we should not show 119 while undiscovered and 120 is. Because: in a large cell PT, we can use the border all right ("no border=undiscovered"). However, in the smaller non-extended PTs, we do not have that option (no border, less details in general). BTW, we are not talking Extended PTs here. - DePiep (talk) 09:49, 8 March 2018 (UTC)

I think that if an element 119 or 120 is discovered, we should just put it where we normally would, just below Fr/Ra, coloring it "unknown" as we normally would. Hiding them under a "there's more" label implicitly hints that these elements are somehow less worthy than the existing ones. If 120 is discovered before 119, we have to admit that 119 is still undiscovered and not put it in the table.

I don't think, by the way, there'll be enough RS to form what we would call a RS-backed opinion, unlike with group 3. So we must be free to make our decisions on that. If I turn out wrong, that will be obvious.--R8R (talk) 09:17, 8 March 2018 (UTC)

I have created this in {{Periodic table/sandbox119}}. - DePiep (talk) 09:49, 8 March 2018 (UTC)

Atomic weight in periodic tables

At last, after a suggestion by Double sharp, the Standard atomic weights are added to our basic periodic table [2]. Even better: the values used are those as formally published by IUPAC — almost.

• Which value is used? The IUPAC-published Standard atomic weight can be long (fluorine has 18.998403163(6)) or an interval (carbon has [12.0096, 12.0116]). These, and only these, are the 84 Standard atomic weights. I repeat: only these 84 values may be named "Standard atomic weight" (it used to be Nobel stuff). We publish them in the element infoboxes.

IUPAC also formally publishes shorter forms, with 4 to 5 digit numbers. They are in the source's table 2 and 3, named Abridged and Conventional value. When combined to one short value per element, here at enwiki, I call these "formal-short". See Atomic weights used at enwiki. So per element we have multiple, formally published values only varying in precision.

The formal short form can have an uncertainty of (1) which is omitted i.e. implicit, or an other uncertainty that is written: Ca has 40.078(4). To save on text length in the small PT cells, I have removed these uncertainties such as the (4) in Ca (I call this "rounded"). So "we" have changed the formally published values; hence I wrote "almost". But at least we use the correct numbers.

For the (118−84 = ) 34 instable elements without a published standard atomic weight, we use the massnumber of the most stable isotope, written like [209].

• Large cell PT's: We have three large cell PT's that already showed the Standard atomic weight. Recently we have changed these cells to contain both the literal Standard form, and the "formal short" form. See the large cell PTs: {{PT 18 columns}}, {{PT 32 columns}}, {{Extended PT (Nefedov, 54 columns)}}. Incidentally, GKFX did a great job in improving the scrolling of these PTs, and their showing in mobile!

• Legend: I have added the theme "Atomic weight" to the legend boxes. So far, there are two texts used: basic cell and large cell. See here for the two text options used. They may be up for improvement. Discuss here?

• To consider: Add short form to the infoboxes? Today we publish the short "Conventional" form when Standard is an interval (like carbon; 12 such elements). We could also add the short "Abridged" form (5 digits with/out uncertainty) to the other 72 elements (like fluorine). Thoughts?

TL;DR: Our basic Periodic table now shows the atomic weights! It is the formal short number as published by IUPAC, but we have removed the uncertainty ("rounded"). The legend has an entry for the Atomic weight. Atomic weights are also in three large-cell periodic tables.

-DePiep (talk) 17:43, 14 March 2018 (UTC)

References