Talk:Periodic table: Difference between revisions
→Rare earths: r Double sharp: you want a general PT without LnAn, and with REM? |
→Mind the gap (It lives again): 14LaAc shows its stuff |
||
Line 457: | Line 457: | ||
:::::The surpising new A and B forms keep Ln and Ac together by putting exactly all of them in the footnote. Were those form an improvement, it could be worth introducing them to the world through our readers, and having scientists frown & smile. But I see two flaws. Flaw 1: it breaks group 3. The core periodic ''table'' structure is columns & rows first, with reasons. A and B forms break this base: group 3(!) is split over two places: two elements are displaced into the footnote. Only after reconstruction (into 32-col form) are they visibly in the place -- while exactly that is what we ''want'' to show in this detail. IOW, we want to state "group 3 = ./././.", and then we cut that group in half. It is this extra mental step added that imo does not help the reader. Flaw 2: introducing cuts by category. The Ln and Ac together ''looks'' very nice at 1st sight because of their categorycolors being together. But categories are not the structuring base of the PT. Of the 12 categories we show, only one falls together with column(s). Categories are great to show the trend, but they should not be used to make a cut (for practical reason only) in the PT. -[[User:DePiep|DePiep]] ([[User talk:DePiep|talk]]) 06:32, 12 November 2015 (UTC) |
:::::The surpising new A and B forms keep Ln and Ac together by putting exactly all of them in the footnote. Were those form an improvement, it could be worth introducing them to the world through our readers, and having scientists frown & smile. But I see two flaws. Flaw 1: it breaks group 3. The core periodic ''table'' structure is columns & rows first, with reasons. A and B forms break this base: group 3(!) is split over two places: two elements are displaced into the footnote. Only after reconstruction (into 32-col form) are they visibly in the place -- while exactly that is what we ''want'' to show in this detail. IOW, we want to state "group 3 = ./././.", and then we cut that group in half. It is this extra mental step added that imo does not help the reader. Flaw 2: introducing cuts by category. The Ln and Ac together ''looks'' very nice at 1st sight because of their categorycolors being together. But categories are not the structuring base of the PT. Of the 12 categories we show, only one falls together with column(s). Categories are great to show the trend, but they should not be used to make a cut (for practical reason only) in the PT. -[[User:DePiep|DePiep]] ([[User talk:DePiep|talk]]) 06:32, 12 November 2015 (UTC) |
||
::::::Exactly. If we want to show Sc/Y/La/Ac, then we had better see La and Ac in the positions right under Y, and not force the reader to mentally fit them in. Furthermore, I don't see anything wrong with splitting La and Ac from the other lanthanides and actinides if they are coloured similarly. Nobody complains that H is very far away from the other diatomic nonmetals. [[User:Double sharp|Double sharp]] ([[User talk:Double sharp|talk]]) 06:46, 12 November 2015 (UTC) |
::::::Exactly. If we want to show Sc/Y/La/Ac, then we had better see La and Ac in the positions right under Y, and not force the reader to mentally fit them in. Furthermore, I don't see anything wrong with splitting La and Ac from the other lanthanides and actinides if they are coloured similarly. Nobody complains that H is very far away from the other diatomic nonmetals. [[User:Double sharp|Double sharp]] ([[User talk:Double sharp|talk]]) 06:46, 12 November 2015 (UTC) |
||
[[File:14LaAc periodic table sketch.jpg|500px|right]] '''Let us proceed with Sc|Y|La|Ac.''' The lanthanides are reasonably proximate to one another, spatially and chromatically. Membership of group 3 is clear. How the Ln and An fit into the main body of the table is straightforward, and this yields a 32-column form with no IKEA tears (the crying type). Sc|Y|La|Ac is the most common form. The gap between groups 3 and 4 is concordant with the group 3 metals behaving chemically more like the alkaline earths, and the s-block metals generally. Sc and Y remain coloured as transition metals in homage to their d<sup>1</sup>s<sup>2</sup> electron configurations. The footnoted Ln and An line up nicely under the main body group numbers: Ce under group 4; the transition-metal-like earlier actinides---Th, Pa, U, Np, Pu---under groups 4, 5, 6, 7 and 8. Lanthanum and actinium are coloured as, respectively, a lanthanide and an actinide, seemingly out of whack with their d<sup>1</sup>s<sup>2</sup> elecron configuations but consistent with their stature as progenitors of each of their own series of elements. An elegant package all round. And the asterisks line up nicely, too. [[User:Sandbh|Sandbh]] ([[User talk:Sandbh|talk]]) 00:28, 15 November 2015 (UTC) |
|||
====La|Ac and Lu|Lr==== |
====La|Ac and Lu|Lr==== |
Revision as of 00:33, 15 November 2015
This is the talk page for discussing improvements to the Periodic table article. This is not a forum for general discussion of the article's subject. |
Article policies
|
Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL |
Archives: Index, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17Auto-archiving period: 90 days ![]() |
![]() | Periodic table is a featured article; it (or a previous version of it) has been identified as one of the best articles produced by the Wikipedia community. Even so, if you can update or improve it, please do so. | |||||||||||||||||||||||||||
![]() | This article appeared on Wikipedia's Main Page as Today's featured article on February 28, 2004. | |||||||||||||||||||||||||||
|
![]() | The contents of the Placement of lanthanides and actinides in the periodic table page were merged into Periodic table. For the contribution history and old versions of the redirected page, please see its history; for the discussion at that location, see its talk page. |
![]() | This article is of interest to the following WikiProjects: | |||||||||||||||||||||||||||||||||||||||||||||||||||||||
Please add the quality rating to the {{WikiProject banner shell}} template instead of this project banner. See WP:PIQA for details.
Please add the quality rating to the {{WikiProject banner shell}} template instead of this project banner. See WP:PIQA for details.
Please add the quality rating to the {{WikiProject banner shell}} template instead of this project banner. See WP:PIQA for details.
Please add the quality rating to the {{WikiProject banner shell}} template instead of this project banner. See WP:PIQA for details.
|
This is the talk page for discussing improvements to the Periodic table article. This is not a forum for general discussion of the article's subject. |
Article policies
|
Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL |
Archives: Index, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17Auto-archiving period: 90 days ![]() |
Change location of the f-block
Opening statement
As the article currently appears, the f-block (57-71 and 89-103) is placed underneath the table, marked by two asterisks to the left of Lu and Lr, creating a column in-between 2 and 3. Lu and Lr are both members of the f-block, which is why I propose that they be moved into the disjoined rows, and the asterisks take their place. This would shorten the table and make the grouping of the f-block more easily recognizable.
G31r0d (talk) 00:01, 10 September 2015 (UTC)
Initial responses
Talk from 32-column PT. 18-column PT discussions are astrologic to me.Please talk from the 32-column form. 18-column PT with that rectangle below introduces description difficulties, unnecessary (It's like an IKEA cupboard unboxed, deerly needing a manual ;-) ). Now, you mention "f-block", but there also exists the "lanthanides", which does not equals f-block. It also involves the description of "group 3". -DePiep (talk) 00:13, 10 September 2015 (UTC)- (Don't mind DePiep's strong words on the 18-column PT: sure the 32-column one is more accurate, but unfortunately it isn't actually used seriously by anyone to a first approximation yet.) Now, to answer your question: Lu and Lr are not f-block elements - they are d-block elements. (Also how can the f-block have 15 elements per period when an f-subshell can only hold 14 electrons?) The term you are looking for is "inner transition metal". Double sharp (talk) 17:20, 10 September 2015 (UTC)
- Your're right, I rephrased. -DePiep (talk) 17:44, 10 September 2015 (UTC)
- Thank you! ^_^ Yes, 18-column needs a manual, especially for the version the OP advocates (Sc/Y/*/**) - what is under Y, and how does the f-block fit in? Sc/Y/La/Ac or Sc/Y/Lu/Lr 18-columns are at least unambiguous. Double sharp (talk) 17:58, 10 September 2015 (UTC)
- Your're right, I rephrased. -DePiep (talk) 17:44, 10 September 2015 (UTC)
- (Don't mind DePiep's strong words on the 18-column PT: sure the 32-column one is more accurate, but unfortunately it isn't actually used seriously by anyone to a first approximation yet.) Now, to answer your question: Lu and Lr are not f-block elements - they are d-block elements. (Also how can the f-block have 15 elements per period when an f-subshell can only hold 14 electrons?) The term you are looking for is "inner transition metal". Double sharp (talk) 17:20, 10 September 2015 (UTC)
G31r0d, I agree with your suggestion even though the main reason you gave isn't valid, and therefore, the title of this talk page section also isn't valid. I'll be creating a new talk page section below with a valid title to discuss the same proposal for different reasons. As pointed out above, the f-block must have 14 elements per period (and the d-block must have 10). That's a consequence of solutions to the Schrödinger equation when the azimuthal quantum number is three for f and two for d. This means that, if emphasizing the f-block is the goal, then elements (not asterisks) must be placed below Sc and Y. However, if blocks are going to be emphasized then the reader is best served if they are fully emphasized for all blocks, including the s-block. A periodic table layout that does this is already present at Wikipedia at Block_(periodic_table). Different periodic tables in different articles ought to serve the reader in different ways as discussed at Talk:Block_(periodic_table)#Pedagogical_goals.2C_empiricism_and_theory. Flying Jazz (talk) 00:54, 11 September 2015 (UTC)
- On second thought, after reminding myself of other talk page discussions on similar periodic table topics in recent years, I changed my mind about creating a new talk page section about this matter. If a larger community of knowledgeable, focused editors with a reader-focused, pedagogical mindset were present then a focused discussion about this topic might be doable here. Flying Jazz (talk) 02:22, 11 September 2015 (UTC)
- My main gripe with Sc/Y/*/** is that there is no obvious way to translate it to 32-column, as shown in the header and footer PTs on each element article. IMHO it ought to be either Sc/Y/La/Ac (following Greenwood and many esteemed textbooks) or Sc/Y/Lu/Lr (according to the arguments by Scerri and Jensen, among others), if you want to keep those tables 32-column. I do admit that Sc/Y/La/Ac would seem preferred according to an RS count today. It is not my favourite, but my personal preference should not come into this for WP. Double sharp (talk) 12:11, 11 September 2015 (UTC)
- P.S. Yes I know IUPAC uses Sc/Y/*/** in their PT, but they don't think the question is their business IIRC. Double sharp (talk) 12:14, 11 September 2015 (UTC)
A focused discussion would involve goals that editors are attempting to achieve for the sake of a general reader in the broad context of an article about the table and in the even broader context of many other articles where some "generic Wikipedia table" is used. What layout reinforces the most important chemical knowledge at an appropriate level for that hypothetical general reader? Your view seems to be that one particular correct, true, and best identity of group 3 in the 32-column table is so important that group 3 must be presented in the 18-column table in such a way as to maintain that identity. This might be why you frame the discussion as a decision among "Sc/Y/*/**," "Sc/Y/La/Ac," and "Sc/Y/Lu/Lr." Someone with my preference would frame the discussion as a decision among "unify and label the lanthanides/actinides" "separate La/Ac from the other lanthanides/actinides," and "separate Lu/Lr from the other lanthanides/actinides." My view is that the group 3 issue, the 18-versus-32 column issue, and the relationship of both issues to each other are all relatively unimportant matters in introductory chemistry pedagogy when compared to the more central role of the table as an arrangement to spatially unite elements with particular chemical identities. In addition to having united background colors that match united spatial locations, the general reader is also served in my view by having correct and fully meaningful labels (i.e., * = lanthanide and ** = actinide). The discussion among the authors that we've both read is interesting, and I'm familiar with it. But a correspondence between location, chemical identity, and meaningful labels are more important to the general reader than a layout that depends upon our editorial judgement about the outcome of that discussion. Flying Jazz (talk) 20:21, 11 September 2015 (UTC)
- I don't really care which is the best anymore: I just think that all those PTs on WP are meant to be generic ones, not really illustrating any particular point: so they should probably all be consistent with each other, as otherwise the difference raises questions that really are unimportant in context. It's just that if you use "keep Ln and An together", there's no way to make a consistent 32-column table as we already use. Which is perhaps an argument to get rid of the 32-column table in favour of the 18-column ones in our element articles. A much stronger argument for that is that 18-column is what you will find just about everywhere when nobody's trying to prove a point.
- Once you stop proudly displaying 32-column everywhere, there are no further obstacles to keeping Ln and An together – which is also the form supported by the most RSs, save perhaps only separating La and Ac. And I agree that for the beginning reader, the f-block elements are not a big deal, and the most salient point is showing how similar they all are - which Sc/Y/*/** does best. (Even Greenwood, for example, which discusses La and Ac along with Sc and Y and restricts the Ln and An chapters to Ce-Lu and Th-Lr respectively, admits in those chapters that it is very useful to include La and Ac IIRC. I don't have it with me right now so if I'm wrong, please correct me.)
- tl;dr: use 18-column throughout whenevera standard PT is needed, and use Sc/Y/*/**, as used by many RSs and fits the chemistry the best. Sc/Y/La/Ac is really the only possible alternative by RSs and it gives the false impression that La and Ac act noticeably differently from the rest of the lanthanides and actinides. Double sharp (talk) 08:18, 12 September 2015 (UTC)
- It's been stated before that any of the three layouts under discussion for the 18-column table used in this article can be found in excellent reliable sources. Therefore, separating La/Ac from the other lanthanides/actinides is not the only possible alternative by RSs. At some point in the past, someone may have argued that huge literature surveys will result in a judgement that a particular table is the one supported by sources. In an actual community of science editors at an encyclopedia, an inappropriate literature survey that doesn't serve the reader would be treated with derision. Also, at some point in the past, someone may have argued that our judgement should be based on how to "make a consistent 32-column table." Again, in an actual community of science editors, the idea that the reader is served when editors make decisions based on how one table can make another would be treated with derision. As for displaying the 32-column table everywhere, that is a separate issue for a different discussion. If you've reached a decision about what serves the general reader of an encyclopedia for the template table in the Overview section of this particular article, then you've made a focused editorial judgement about a particular content issue, and that's what we're here to do. Flying Jazz (talk) 13:55, 12 September 2015 (UTC)
- As I said above: for this table, Sc/Y/*/** is my personal editorial judgement, based on what I think would serve the general reader. I do wonder what is so wrong with wanting PTs that serve the same purpose to look the same, but that is a different issue: now you have an answer for what you state we are here to do. Double sharp (talk) 22:06, 12 September 2015 (UTC)
- I'll refactor the talk page in a day or two, wait a few weeks to see if anyone else chimes in, and then make the change. Flying Jazz (talk) 23:24, 14 September 2015 (UTC)
- Although a note: while I think Sc/Y/*/** would be OK in isolation, I still think the consistency argument (if we have both an 18-column table and a 32-column table, and both are meant as simple illustrations, they should show the same layout, having the same purpose) would trump it as long as we have 32-column tables displayed proudly in WP for general purposes. Double sharp (talk) 09:54, 21 September 2015 (UTC)
- I'll refactor the talk page in a day or two, wait a few weeks to see if anyone else chimes in, and then make the change. Flying Jazz (talk) 23:24, 14 September 2015 (UTC)
- As I said above: for this table, Sc/Y/*/** is my personal editorial judgement, based on what I think would serve the general reader. I do wonder what is so wrong with wanting PTs that serve the same purpose to look the same, but that is a different issue: now you have an answer for what you state we are here to do. Double sharp (talk) 22:06, 12 September 2015 (UTC)
- It's been stated before that any of the three layouts under discussion for the 18-column table used in this article can be found in excellent reliable sources. Therefore, separating La/Ac from the other lanthanides/actinides is not the only possible alternative by RSs. At some point in the past, someone may have argued that huge literature surveys will result in a judgement that a particular table is the one supported by sources. In an actual community of science editors at an encyclopedia, an inappropriate literature survey that doesn't serve the reader would be treated with derision. Also, at some point in the past, someone may have argued that our judgement should be based on how to "make a consistent 32-column table." Again, in an actual community of science editors, the idea that the reader is served when editors make decisions based on how one table can make another would be treated with derision. As for displaying the 32-column table everywhere, that is a separate issue for a different discussion. If you've reached a decision about what serves the general reader of an encyclopedia for the template table in the Overview section of this particular article, then you've made a focused editorial judgement about a particular content issue, and that's what we're here to do. Flying Jazz (talk) 13:55, 12 September 2015 (UTC)
- re Double sharp's "I don't really care which ...." full post. Sorry I'm this late, must have been distracted in between. Now your general line, to me, seems to be that an 18-column PT everywhere (as our 'general presentation form', except for special issues of course) is better for the reader. First I note that an 18-column PT does not solve the group-3 question. 18-col form still has to make a point in this (just as it does with the position of He). We'd still need to decide on how to present group 3 even if we abolish and delete every 32-col form! Namely: group 3 = Sc/Y/Lu/Lr OR group 3 = Sc/Y/La/Ac. The suggestion that "group 3 = Sc/Y/*/**" solves it is wrong: it just buries the issue in ambivalence (See my #Sc/Y/*/** in 32-column form below). Then, have we solved & made this unhidable group 3 point, there is the question on why we would prefer to put those 14 or 15 elements (times 2) below. I only hear this argued by people educated before 1990 or so. [warning: I'm charging, to make this clear] That is: those who have internalised the 18-col, and saying "I understand it all right this way, so young students don't need a new form". (Disclosure: I was educated back then too). -DePiep (talk) 22:00, 8 October 2015 (UTC)
Sc/Y*/** once more
Sc/Y*/** should appear in the overview section for the reasons given by Flying Jazz, including general reader interests, and pedagogy. This would be consistent with the 32-column table appearing in the lede and footer, since both tables (18-column and 32-column) are premised on spatially uniting, to the greatest extent practical, chemically similar elements. Unresolved arguments on which elements go with Sc and Y in group 3 are appropriately and well enough summarized in the open questions and controversies section of the article and don't need to spill over into the lede, the overview section, or the foooter. I have argued Sc/Y/Lu/Lr for WP in the past and have now changed my mind in light of discussions on WP and in other forums. Sandbh (talk) 04:23, 5 October 2015 (UTC)
- But again: what or which "Sc/Y/*/**" presentation do you mean? Given the topic, I request that the group number "3" is present (something even Scerri does not do consistantly). I can think of multiple graphical variants (seriously) that would comply. You can link to a 18-, 32-, or x-column variant as you like, as long as it is not ambiguous and not ambivalent wrt this issue. (Is why I discard the IUPAC drawing). btw, I read Sc/Y/*/**. -DePiep (talk) 21:46, 6 October 2015 (UTC)
- I missed what FJ said about pedagogy. But re Sandbh in this, I disagree. There is no reason why the 18-col and 32-col form should be structurally different. Any pupil or student should be able to zoom in on a more detailed PT and discover, but this requires that these are the same. One does not start telling that the earth is flat, and then leave it to the pupil to discover something different. 18-col PT and 32-col PT must represent the same. (See #Sc/Y/*/** in 32-column form). And btw, in PT overview the f-block (two rows and then a step) is more convincing that the Ln, An categories. Categories are loosing their PT-behaviour in the p-block anyway. -DePiep (talk) 22:54, 8 October 2015 (UTC)
- The version appearing in the lede of the periodic table article, top right. Like the template was originally, here. But I would like to hear your view and the views of other editors about this. Sandbh (talk) 09:13, 7 October 2015 (UTC)
- That's the undecided one (bad). Gap column missing. Does not show whether Sc/Y is glued to group 2 or to group 4. Not a new PT. -DePiep (talk) 22:24, 7 October 2015 (UTC)
- Sandbh: I am not a specialist in elements. I am a technician. Now it occurs to me time and time again that scientist in PT (physicists, chemicists) can not reflect. Can not convey. It is a "we all know" attitude --as in "we in the lab"--. Each and every time I ask for a graphical representation here, I get these variants/not-a-variant/whatever/youknowwhatimean things, but *not* an answer. From Sandbh, from IUPAC, that sort of people ;-). -DePiep (talk) 22:41, 7 October 2015 (UTC)
- Yes, the gap column is missing but I think it is more imporant that an encyclopedia aimed at the general reader emphasizes the most important chemical similarities, and reinforces this with joined up background colours. Hence I support showing 15 lanthanides and actinides at the foot of the table. If the curious reader wonders what is going on with the gap in group 3, and just how the lanthanides and actinides fit in the gap they will be able to work this out from looking at the 32-column table in the article footer. If they get really interested there is also the accompanying text in the article on which elements are the period 6 and 7 members of group 3.
- I further submit that Lu shares more in common with the rest of the lanthanides than it does with Sc and Y, or the other period 6 transition metals. If we accept that the purpose of our 18-column table is to (as best we can) group like with like then this requires showing 15 lanthanides at the foot of the table. Sandbh (talk) 11:52, 8 October 2015 (UTC)
- ...I think I'm convinced. Sc/Y/*/** it is as default. 32-column will stay as it is showing Sc/Y/Lu/Lr. Double sharp (talk) 15:12, 8 October 2015 (UTC)
- re Sandbh above: "If the curious reader wonders what is going on with the gap in group 3, and just how the lanthanides and actinides fit in the gap" -- uh, there is no gap in the Sc/Y/*/** form (eg the IUPAC graph). The reader is lead into thinking that the PT, asterisks expanded, looks like #Sc/Y/*/** in 32-column form
. Right above the asterisks, it says "3". -DePiep (talk) 22:11, 8 October 2015 (UTC)
- re Sandbh above: "If the curious reader wonders what is going on with the gap in group 3, and just how the lanthanides and actinides fit in the gap" -- uh, there is no gap in the Sc/Y/*/** form (eg the IUPAC graph). The reader is lead into thinking that the PT, asterisks expanded, looks like #Sc/Y/*/** in 32-column form
Done: changed
{{Periodic table}}
and{{Periodic table (18 columns, large cells)}}
to show Sc/Y/*/**. Double sharp (talk) 15:16, 8 October 2015 (UTC)- I reverted. This "Sc/Y/*/**" is not the result of this discussion. And I noted elsewhere and often, it introduces an ambiguous form for group 3 and the f-block. As explained (e.g. in #Graphic_presentations), the 18-column form should follow from a 32-column form, and so represent it (the gap column does that). If the general 32-column shows group 3 being Sc/Y/La/Ac: fine, and we'll move the gap column to the right (btw, this is how dewiki has their general PT). Now this is what happens if you read that "Sc/Y/*/**"
presentation made into a 32-column: -DePiep (talk) 20:31, 8 October 2015 (UTC)
- Source: Sc/Y/*/** (IUPAC)
- I reverted. This "Sc/Y/*/**" is not the result of this discussion. And I noted elsewhere and often, it introduces an ambiguous form for group 3 and the f-block. As explained (e.g. in #Graphic_presentations), the 18-column form should follow from a 32-column form, and so represent it (the gap column does that). If the general 32-column shows group 3 being Sc/Y/La/Ac: fine, and we'll move the gap column to the right (btw, this is how dewiki has their general PT). Now this is what happens if you read that "Sc/Y/*/**"
- ...I think I'm convinced. Sc/Y/*/** it is as default. 32-column will stay as it is showing Sc/Y/Lu/Lr. Double sharp (talk) 15:12, 8 October 2015 (UTC)
Template:Periodic table (32 columns, Sc-Y-*-** from)
- -DePiep (talk) 20:58, 8 October 2015 (UTC)
- We know the gap is made ambiguous by Sc/Y/*/**. We know it results in an inconsistency with the Sc/Y/Lu/Lr 32-column periodic table, but the main argument for Sc/Y/*/** is that it emphasizes the chemical similarities of the lanthanides and actinides, as well as Sc and Y. Then both the 18- and 32-column periodic tables unite similar elements as much as possible (e.g. Yb is next to Lu in Sc/Y/Lu/Lr 32-column, but not in Sc/Y/Lu/Lr 18-column, whereas it is in Sc/Y/*/** 18-column), given the universally accepted constraint that one element takes up exactly one cell. Double sharp (talk) 05:36, 9 October 2015 (UTC)
- P.S. It's not that I don't think your argument regarding 18/32-column consistency has merit. I guess Sandbh+FJ+my position of yesterday is that we should have 18- and 32-column tables constructed on the same principles, whereas yours is that the end result should look the same (i.e. both should have Sc/Y/Lu/Lr, or Sc/Y/La/Ac). I guess it depends on which you think is more likely to confuse a reader: two slightly differently-looking periodic tables, or two periodic tables that appear to imply different chemical similarities. It may perhaps be arguable that the latter is a subjective interpretation, while the former is a fact about how the tables look. But I'm not going to take a side on this any longer. Whatever a majority consensus thinks is fine, I will accept. Double sharp (talk) 05:50, 9 October 2015 (UTC)
- Double sharp: with all due respect, times two: no. That is: no. I don't have the time, nor the plan, nor the patience, to explain the same thing again to you. Must say I am quite astonished that you keep opposing (even by edits) without discussing. -DePiep (talk) 23:05, 10 October 2015 (UTC)
- What is this if not a discussion? What is this if not a 3:1 consensus? Double sharp (talk) 04:55, 11 October 2015 (UTC)
- Double sharp: with all due respect, times two: no. That is: no. I don't have the time, nor the plan, nor the patience, to explain the same thing again to you. Must say I am quite astonished that you keep opposing (even by edits) without discussing. -DePiep (talk) 23:05, 10 October 2015 (UTC)
- P.S. It's not that I don't think your argument regarding 18/32-column consistency has merit. I guess Sandbh+FJ+my position of yesterday is that we should have 18- and 32-column tables constructed on the same principles, whereas yours is that the end result should look the same (i.e. both should have Sc/Y/Lu/Lr, or Sc/Y/La/Ac). I guess it depends on which you think is more likely to confuse a reader: two slightly differently-looking periodic tables, or two periodic tables that appear to imply different chemical similarities. It may perhaps be arguable that the latter is a subjective interpretation, while the former is a fact about how the tables look. But I'm not going to take a side on this any longer. Whatever a majority consensus thinks is fine, I will accept. Double sharp (talk) 05:50, 9 October 2015 (UTC)
- We know the gap is made ambiguous by Sc/Y/*/**. We know it results in an inconsistency with the Sc/Y/Lu/Lr 32-column periodic table, but the main argument for Sc/Y/*/** is that it emphasizes the chemical similarities of the lanthanides and actinides, as well as Sc and Y. Then both the 18- and 32-column periodic tables unite similar elements as much as possible (e.g. Yb is next to Lu in Sc/Y/Lu/Lr 32-column, but not in Sc/Y/Lu/Lr 18-column, whereas it is in Sc/Y/*/** 18-column), given the universally accepted constraint that one element takes up exactly one cell. Double sharp (talk) 05:36, 9 October 2015 (UTC)
- -DePiep (talk) 20:58, 8 October 2015 (UTC)
- C'mon. You claim 'discussion' and you do 'vote-counting'? -DePiep (talk) 22:16, 11 October 2015 (UTC)
- I see three people (me, Sandbh, and FJ) who have discussed it with arguments and came to a conclusion, and one (you) who opposes that conclusion. Looks like a discussion and consensus to me. The counting is just a rough estimate to get a sense of how clear the consensus is, but it seems pretty clear that there is one. Double sharp (talk) 04:27, 12 October 2015 (UTC)
DePiep, there is no need to stretch the Sc and Y cells. This is not done for the 32 column table in the footer. A simple note added to the Sc|Y|*|** table will point the reader to an explanation (with pictures) on how the lanthanides and actindes fit under Y (see below). I agree Sc|Y|*|**, at face value, is ambiguous. However, I think this is outweighed for reasons others have noted previously. And Sc|Y|La|Ac or Sc|Y|Lu|Lr have their own problem. Sandbh (talk) 02:09, 11 October 2015 (UTC)
Group | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Alkali metals | Alkaline earth metals | Pnictogens | Chalcogens | Halogens | Noble gases | |||||||||||||
Period |
||||||||||||||||||
2 | ||||||||||||||||||
3 | ||||||||||||||||||
4 | ||||||||||||||||||
5 | ||||||||||||||||||
6 | ![]() |
|||||||||||||||||
7 | ![]() |
|||||||||||||||||
![]() |
||||||||||||||||||
![]() |
||||||||||||||||||
N.B. The occupancy of the group 3 positions below yttrium is discussed in the article. | ||||||||||||||||||
P.S. If you look at some historical periodic tables, you will find that the "*" under Y is sometimes explained as La and the following elements: so you can even argue that the form suggested is in fact a Sc/Y/La/Ac 32-column table, but with La and Ac put together with the (other) lanthanides and actinides, which are similar. No need to stretch cells. Similarly it can also be explained as Lu and the preceding elements. But since there is a disagreement it is best to stay neutral on the matter unless we are trying to illustrate a point – and here we are not. Double sharp (talk) 05:02, 11 October 2015 (UTC)
- re Double sharp there is no need to stretch the Sc and Y cells - Yes there is, because that is what the graph says. The IUPAC Sc/Y/*/** graph has "3" (i.e. group 3) over the Sc an Y, and over the asterisks. So your IUPAC form says it. It is not even a "need", it is a fact. -DePiep (talk) 21:32, 11 October 2015 (UTC)
- Double sharp. To cut things short: how does the 32-column PT look like for your Sc/Y/*/** 18-column version? Graphs only. -DePiep (talk) 21:38, 11 October 2015 (UTC)
- Like this:
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | |||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | H | He | ||||||||||||||||||||||||||||||
2 | Li | Be | B | C | N | O | F | Ne | ||||||||||||||||||||||||
3 | Na | Mg | Al | Si | P | S | Cl | Ar | ||||||||||||||||||||||||
4 | K | Ca | Sc | Ti | V | Cr | Mn | Fe | Co | Ni | Cu | Zn | Ga | Ge | As | Se | Br | Kr | ||||||||||||||
5 | Rb | Sr | Y | Zr | Nb | Mo | Tc | Ru | Rh | Pd | Ag | Cd | In | Sn | Sb | Te | I | Xe | ||||||||||||||
6 | Cs | Ba | La | Ce | Pr | Nd | Pm | Sm | Eu | Gd | Tb | Dy | Ho | Er | Tm | Yb | Lu | Hf | Ta | W | Re | Os | Ir | Pt | Au | Hg | Tl | Pb | Bi | Po | At | Rn |
7 | Fr | Ra | Ac | Th | Pa | U | Np | Pu | Am | Cm | Bk | Cf | Es | Fm | Md | No | Lr | Rf | Db | Sg | Bh | Hs | Mt | Ds | Rg | Cn | Nh | Fl | Mc | Lv | Ts | Og |
|
- DePiep, Unless we find a WP:RS that shows a 32-column PT with stretched cells, attempting to show it on WP would cross the line into WP:OR. Even if we find such a RS, we should be careful that we are not violating WP:UNDUE. YBG (talk) 03:32, 12 October 2015 (UTC)
- It is the other way around, YBG. The 18-col version Double sharp is promoting itself has "3" above the asterisks, i.e. all lanthanides and actinides are in group 3, by drawing. Not me. Also, it has Sc and Y glued to both group 2 and group 4 (no space --gap-- between). So when the reader expands it to 32-col form, these properties are to be maintained with the result. It is not me who drew them. In general, the two forms should be representing the same PT. As for shortcut arguments: the 18-col form is OR or FRINGE for putting those elements in group 3. -DePiep (talk) 08:18, 14 October 2015 (UTC)
- What? If IUPAC is fringe, then what is Jensen?
- And to be clear, I am assuredly not the only one here promoting this. And how do you know the first asterisk means "all the lanthanides" and not "La and the following elements" (creating Sc/Y/La/Ac) or "Lu and the preceding elements" (creating Sc/Y/Lu/Lr)? Greenwood and Earnshaw doesn't treat La as a lanthanide, incidentally. Double sharp (talk) 12:16, 14 October 2015 (UTC)
- re "What? If IUPAC is fringe, ...": Ask IUPAC why they draw group 3 this way. Not me. -DePiep (talk) 19:48, 17 October 2015 (UTC)
- So re Double sharp: your 32-col differs by taking f-block (14 columns of Ln andAN). out of group 3. Also it states that SAC/Y are to the right hand, while that does not follow from the 18-col form. It is not the same PT, you have added statements that are not in the 18-col form. -DePiep (talk) 08:21, 14 October 2015 (UTC)
- I prefer to think of it this way: Sc/Y/*/** cannot expand to your form, as an element cannot take up more than one cell: the only PTs that have done this are pretty fringe/specific-in-intention (with the exception of duplicating H in group 1 and 17) and not meant for a beginner's overview (unlike Sc/Y/*/**). The answer is that Sc/Y/*/** is a form to use for beginning, when the f-block elements don't really matter. This keeps similar elements together. When you put them in the main body of the table, though, you have to stop equivocating over La and Lu, so the best way to preserve the table's intentions (keeping similar elements together) would be Sc/Y/Lu/Lr if you want to keep the blocks together. But doing this in the 18-column table tears Lu away from its fellow lanthanides. Double sharp (talk) 12:16, 14 October 2015 (UTC)
- If IUPAC is fringe - WP putting all Ln and An in group 3 is OR or FRINGE. IUPAC doing that is wrong - they have drawn the PT wrong. -DePiep (talk) 19:32, 15 October 2015 (UTC)
- If we're going to throw around acronyms: you are moving into WP:IDIDNTHEARTHAT mode. Double sharp (talk) 02:15, 16 October 2015 (UTC)
- If IUPAC is fringe - WP putting all Ln and An in group 3 is OR or FRINGE. IUPAC doing that is wrong - they have drawn the PT wrong. -DePiep (talk) 19:32, 15 October 2015 (UTC)
- I prefer to think of it this way: Sc/Y/*/** cannot expand to your form, as an element cannot take up more than one cell: the only PTs that have done this are pretty fringe/specific-in-intention (with the exception of duplicating H in group 1 and 17) and not meant for a beginner's overview (unlike Sc/Y/*/**). The answer is that Sc/Y/*/** is a form to use for beginning, when the f-block elements don't really matter. This keeps similar elements together. When you put them in the main body of the table, though, you have to stop equivocating over La and Lu, so the best way to preserve the table's intentions (keeping similar elements together) would be Sc/Y/Lu/Lr if you want to keep the blocks together. But doing this in the 18-column table tears Lu away from its fellow lanthanides. Double sharp (talk) 12:16, 14 October 2015 (UTC)
- It is the other way around, YBG. The 18-col version Double sharp is promoting itself has "3" above the asterisks, i.e. all lanthanides and actinides are in group 3, by drawing. Not me. Also, it has Sc and Y glued to both group 2 and group 4 (no space --gap-- between). So when the reader expands it to 32-col form, these properties are to be maintained with the result. It is not me who drew them. In general, the two forms should be representing the same PT. As for shortcut arguments: the 18-col form is OR or FRINGE for putting those elements in group 3. -DePiep (talk) 08:18, 14 October 2015 (UTC)
- Sandbh, I add a new argument here, so as not to disrupt your summary.
- I note that, say, elegance of the result is mentioned (i.e., for cutting or not cutting same colored groups). Of course, if factual things are the same, we are free to choose an elegant result from options. However, elegance should not be an argument to form a drawing some way, into making it wrong/distorted/ambiguous.
- I add that another point of elegance is to be mentioned: showing the steps clearly (two-period steps). Some weeks ago it looked like the User:Sandbh/sandbox seemed to end up preferring Sc/Y/Lu/La, which would nicely show the steps, and keep colors & blocks together (as opposed to the alternative there Sc/Y/La/Ac). Again, this does not add to the conclusion, but is says "nice if our preference has this too". -DePiep (talk) 15:12, 16 October 2015 (UTC)
Summary of arguments
Sc|Y|*|**
Pros | Cons |
Spatially unites elements with similar chemical identities | |
Unites background colors that match united spatial locations | Not clear how the Ln and An fit in |
Sc|Y|Lu|Lr
Pros | Cons |
Clear how the Ln and An fit in | Spatially separates elements with similar chemical identities |
Spatially separates elements with similar background colours |
Is this a fair summary? Sandbh (talk) 10:49, 13 October 2015 (UTC)
- Thanks for the summary of a very long set of discussions. I think that an encyclopedia article on the periodic table should show the various presentations which are common in books and on websites, whether or not they seem logical to the editors of the article. That includes at least three 18-column formats: Sc|Y|La|Ac, Sc|Y|Lu|Lr and Sc|Y|*|**. The first (Sc|Y|La|Ac) seems to have disappeared from the article, although it is the most common in textbooks and even on classroom walls. I do not agree that Sc|Y|*|** preserves neutrality - it implies that * and ** are the correct notation rather than La and Ac or Lu and Lr. Dirac66 (talk) 14:03, 13 October 2015 (UTC)
- So here's a summary for Sc/Y/La/Ac (which I would like, if Sc/Y/*/** continues to be unacceptable for whatever reason, because this way we have Greenwood & Earnshaw and Holleman & Wiberg etc. behind us):
- I added "Contradicts other authors who assign Lu and Lr to group 3" and moved the other two con entries down one row to be more consistent with the second table. Sandbh (talk) 10:54, 14 October 2015 (UTC)
- I amended the neutrality "pro" to make its meaning clearer. Sandbh (talk) 10:59, 14 October 2015 (UTC)
- I added "Contradicts other authors who assign Lu and Lr to group 3" and moved the other two con entries down one row to be more consistent with the second table. Sandbh (talk) 10:54, 14 October 2015 (UTC)
- So here's a summary for Sc/Y/La/Ac (which I would like, if Sc/Y/*/** continues to be unacceptable for whatever reason, because this way we have Greenwood & Earnshaw and Holleman & Wiberg etc. behind us):
Sc|Y|La|Ac
Pros | Cons |
Clear how the Ln and An fit in | Spatially separates elements with similar chemical identities |
Spatially separates elements with similar background colours |
- (I still like it myself, as Sc and Y are rather peripheral transition metals, almost main-group-ish, and La would fit a main-group-style trend better than Lu. But the key point is that this way we don't equivocate on the composition of group 3, if you don't like that, and we would have more support from authorities using Sc/Y/La/Ac than if we chose to use Sc/Y/Lu/Lr.) Double sharp (talk) 14:25, 13 October 2015 (UTC)
- The current PT in the lead shouldn't be considered to be the standard to comply with; it can be altered as well. I agree we should show the three in this article, but since we will use just one in most circumstances, I would advocate using the /Lu/Lr one. If the group 3 topic is not all that important for most readers (it is not), it shouldn't be a problem to have the /Lu/Lr version. Except can we (DePiep (talk · contribs)) make the cell between groups 2 and 3 thinner (use thinner cells?), so it's not too intense at attracting attention?--R8R (talk) 09:00, 14 October 2015 (UTC)
- Thinner would be great. I tried that a few times, but could not get it done in the fixed-colwidth setting. (btw, it is that column that I call "gap", for disconnected cells). -DePiep (talk) 19:36, 15 October 2015 (UTC)
- I removed the "consistent with/contradicts table in the lede" arguments as I agree these arguments are neither here nor there. Sandbh (talk) 10:48, 14 October 2015 (UTC)
Since Sc/Y/*/** continues to be unacceptable for DePiep, my second choice is Sc/Y/La/Ac along with those famous textbooks that I mentioned. With /Lu/Lr, I think we're going to have questions. I learnt the table with Sc/Y/La/Ac (which I still see a lot) and have seen Sc/Y/*/** many times too, but I never once saw Sc/Y/Lu/Lr before encountering WebElements (and then got interested in the issue). Double sharp (talk) 11:56, 14 October 2015 (UTC)
Analysis of arguments
- I agree with User:Dirac66, who said, "I think that an encyclopedia article on the periodic table should show the various presentations which are common in books and on websites, whether or not they seem logical to the editors of the article." This is a separate issue that can be addressed in the Layout variants section of the article.
- I agree with Flying Jazz (retired), who said, "At some point in the past, someone may have argued that huge literature surveys will result in a judgement that a particular table is the one supported by sources. In an actual community of science editors at an encyclopedia, an inappropriate literature survey [i.e. one] that doesn't serve the reader would be treated with derision." For that reason I have struck out the pro and con arguments about consistency with the literature.
- I agree with Flying Jazz, who said, "My view is that the group 3 issue, the 18-versus-32 column issue, and the relationship of both issues to each other are all relatively unimportant matters in introductory chemistry pedagogy when compared to the more central role of the table as an arrangement to spatially unite elements with particular chemical identities…a correspondence between location [and] chemical identity are more important to the general reader than a layout that depends upon our editorial judgement about the outcome of that discussion." For this reason I have struck out the pros or cons to do with the group 3 issue.
- I agree with User:Double sharp's observations that Sc|Y|*|** tables (imagined or real) that widen the Sc and Y cells to span the 15 Ln and Ac elements, "are pretty fringe/specific-in-intention (with the exception of duplicating H in group 1 and 17) and not meant for a beginner's overview (unlike Sc/Y/*/**)" and "The answer is that Sc/Y/*/** is a form to use for beginning, when the f-block elements don't really matter."
- Looking back on the tables 1–3 as they appear now it seems to me that the general reader's interests would be best served by featuring Sc|Y|*|** in the article.
- And I think the Layout variants section would be more clear if it made mention of Sc|Y|La|Ac and Sc|Y|Lu|Lr.
- User:DePiep, I respect your right to disagree and ask you to place the interests of the general reader ahead of your concern about graphical integrity in the region of group 3. My preference is to instead make mention of this concern in the Layout variants section. Sandbh (talk) 10:57, 15 October 2015 (UTC)
RBR Gtrs comment: For me, one point is not clear. So if the f block does not matter, why does it help anyone advocate the /*/** position? I think if it is not, we would be fine with any of the three. And if I am correct and it actually is not all that important for a newcomer, we could also aid a more educated person by not choosing the ambiguous variant? We're an encyclopedia, after all, we may be user-friendly (personally, I am doing my best to ensure user-friendliness in my writing), but not on expense of anyone else.--R8R (talk) 11:20, 15 October 2015 (UTC)
- I'll try and address your comments as best as I can. All the relevant considerations for a general reader are set out in tables 1–3. Yes, you are right about the f block. Catering for a more educated reader is not really on the agenda of this particular discussion. That is why I struck out the reference in table 1 to the membership of group 3 being unclear, as this is not an issue for the general reader. I was almost going to say the same thing about User:Dirac66's mention of having the article show the various presentations which are common in books and on websites etc as this was somewhat of another distraction from the core discussion but I capitulated because I thought it could be quickly addressed. Now a couple of things about the more educated reader. I shouldn't be saying anything about this because it's another distraction from the central theme of the discussion. Anyway, 1. A more educated reader would likely be comfortable with a degree of ambiguity. And if this causes some wonderment they can read more about it in the article. 2. If we choose a less ambiguous variant then this results in spatially separating elements with similar chemical identities and spatially separating elements with similar background colours which, for the general reader, is bad. I don't want to go on about these side issues because---in good faith, and trying to be civil---we end up going down energy-sucking and argument-distracting rabbit holes. Sandbh (talk) 10:41, 16 October 2015 (UTC)
Dirac66 comment: I agree with Sandbh that the Layout variants section would be a good place to present the different versions used today: Sc/Y/La/Ac, Sc/Y/*/**, Sc/Y/Lu/Lr and perhaps 32-column and Janet too. Dirac66 (talk) 11:49, 15 October 2015 (UTC)
YBG comment: I think ambiguity is one of the strongest arguments for */**. Experts seem to disagree as to the correct membership of group 3; the /*/** form, by its very ambiguity, neatly avoids this issue. YBG (talk) 02:18, 16 October 2015 (UTC)
- With respect, YBG, ambiguity is a side issue that is not relevant at the general reader level. Sandbh (talk) 10:41, 16 October 2015 (UTC)
Double sharp comment: I think the idea is that with Sc/Y/*/**, all the lanthanides (which are similar) are together, while Sc/Y/La/Ac tears La away from Ce and Sc/Y/Lu/Lr tears Lu away from Yb in their 18-column forms. In a 32-column form, there is no obstacle to Sc/Y/Lu/Lr, as Yb and Lu remain together. Double sharp (talk) 02:22, 16 October 2015 (UTC)
DePiep comment. There is a flaw in this setup. It suggests there is an equal 3-way choice at hand. However, this is not the case. There is the choice between group 3 = Sc/Y/Lu/Lr and group 3 = Sc/Y/La/Ac. Both have a scientific base, and the choice would only be for all of our general PT's being presented in the same way (as a preference). Of course, our general reader is best-served if our PT's are consistently congruent in this. The non-chose option would be explained in article group 3 (most likely). The text is being prepared by Sandbh in User:Sandbh/sandbox. I have no preference between these two (except that we should pick just one for that purpose). And I thought it was converging to one of these.
Important to note is that these two variants are based on scientific (chemical, physical) arguments, both options being well-based. Once a preference is chosen, we can draw whichever one (see here). They are two, say, structural variants. This has a big big analogy with the positioning of He in the PT: multiple options are well-based in science, but we use only one in our general PT's. For this, the group 3 = Sc/Y/Lu/Lr or Sc/Y/La/Ac variants do not belong in the Periodic_table#Layout_variants section.
However. The third option under discussion, named Sc/Y/*/** here, is *not* about a scientific third variant. It is a presentation variant. Now this could be an improvement, given that we know what structural PT we want to present. And to cut this short: up until now I have not been able to get the structural variant it represents (seeing the IUPAC 18-col drawing). I repeat: it does not show which group 3 variant it uses. Even worse: it has two graphic features that are undeniably wrong or ambivalent: "group 3" is said to contain 32 elements, and it says that Sc and Y are spanning 15 columns. This is unacceptable. Another sign of bad drawing is that no one has been able to show how its 32-col drawing looks like. This goes against the second argument mentioned: the Good of our Reader. Well, inconsistent PT's are helping no one. (At this point, I see no need to evaluate 18-col vs 32-col. Useless if they are inconsistent).
I think the order of choices is 1. which structural variant do we want to use in general; 2. By what layout (graphic features) do we draw that. -DePiep (talk) 15:45, 16 October 2015 (UTC)
- DePiep, as I see it you are re-presenting an argument that has been previously discussed and set aside i.e. re the composition of group 3. Your attempted distinction between "scientific" and "presentation" variants is a dressed-up version of the same theme. All three variants are based in science; they just differ in which scientific aspects are emphasized. Concerns about the composition of group 3 are relatively unimportant in introductory chemistry pedagogy. A correspondence between location and chemical identity is more important for the general reader. Sandbh (talk) 00:01, 18 October 2015 (UTC)
Sandh comment: I'll see if I can draft a Sc|Y|*|** version of the periodic table article in my sandbox. This would include Dirac66's suggestion to say some more about the other major variants. I'll try and do justice to DePiep's comments by seeing if I can refer to the graphical ambiguity issue, without blowing the general reader's mind. I'm not sure yet but this may mean moving the Layout alternatives section further down the article, nearer to the Open questions and controversies section, possibly amalgamating it with the Alternative structures section. Sandbh (talk) 10:30, 21 October 2015 (UTC)
I'll await this result before making any analysis or argument wrt these replies.-DePiep (talk) 10:14, 23 October 2015 (UTC)- I'm sorry. I can't hold my breath any longer. So Sandbh is stating that "Sc/Y/*/**" represents a third scientifically based PT structure? How and when did that #3 creep into the User:Sandbh/sandbox? I must say, today I do not understand the Sandbh sandbox any more. Up to you, Sandbh. -DePiep (talk) 22:37, 23 October 2015 (UTC)
- You did well holding your breathe for so long. The sandbox version is still evolving. Things may come and go. I don't know yet what will happen to the draft Sc/Y/*/** content. I'm trying to say what needs to be said in the "Layout variants subsection" without duplicating what's in the "Period 6 and 7 elements in group 3 subsection." This is hard, so bear with me. Sandbh (talk) 06:09, 24 October 2015 (UTC)
- OK, and it feels good to be breathing again. You know, if you want me to sketch weird graphical things I'll loyally make views I might not yet agree with. -DePiep (talk) 19:56, 25 October 2015 (UTC)
- Thank you DePiep. Are you able to make 15LaAC, 14CeTh and 14LaAc forms, to go into the periodic table article draft Layout subsection? I had in mind the same style as the two 32-column tables in the Period 6 and 7 elements in group 3 subsection, except that these need to be separate 18-column tables, like the existing svg image in the Layout subsection. Sandbh (talk) 10:58, 27 October 2015 (UTC)
- OK, and it feels good to be breathing again. You know, if you want me to sketch weird graphical things I'll loyally make views I might not yet agree with. -DePiep (talk) 19:56, 25 October 2015 (UTC)
- You did well holding your breathe for so long. The sandbox version is still evolving. Things may come and go. I don't know yet what will happen to the draft Sc/Y/*/** content. I'm trying to say what needs to be said in the "Layout variants subsection" without duplicating what's in the "Period 6 and 7 elements in group 3 subsection." This is hard, so bear with me. Sandbh (talk) 06:09, 24 October 2015 (UTC)
Three forms
- In svg overview form, they are:
- You want them in micro-form?
- Does anyone else think it would look better if the "*/**" in the main table were aligned vertically with the "*/**" in bottom? Of course, what is actually used in the published literature is infinitely more important that what I think would be visually appealing. YBG (talk) 04:04, 28 October 2015 (UTC)
- You want them in micro-form?
- re YBG.
Of courseIn the processs I considered that, but this is my outcome so far. The job of the 'asterisks' (big dots here) is to explain the displacement. Nothing more, nothing less. Now to make the re-gathering of the 'footnote' (as the bottom set is sometimes called) the simplest mental step possible, I prefer to hint only a vertical displacement of the leftmost footnote elements (is what you see). Note that we tend to read left-to-right. Then, it is easily to grasp that all the elements aright of the asterisks need a vertical shift to allow the footnote in. Aligning the 'asterisks' vertically would imply that the footnote is shifted one column to the right too (a horizontal displacement). Needless complication in the mental step. - Even better, consider this. Since the one and only connection between the footnote and the main table are the asterisks (and specifically not the downward extended column/group suggestion), we should stress the disconnection even further.
OrWe must prevent that the reader, in their mental re-gathering step, shifts the whole right-of-asterisks bunch of elements one row up to accommodate the footnote in (period 4 Sc or Ti ending up in period 2!). This bad suggestion/possibility we must prevent. I even plan to draw the gap-column (i.e., the column with only the asterisks) smaller, intentionally breaking the suggestion that the footnote-groups (columns) are a continuation of the true group columns above. Also R8R asked about this. - Finally, this is not about "visually appealing", I am not lead by aesthetics (iirc, above such arguments were mentioned like keeping colors together). I only use aesthetics after the What of the PT is in, and after I have a How-to-graph-it. Next, to use "what is actually used in the published literature" as you suggest, we have problems in there. 1. As Jensen (2008) notes, he sees that quite often textbook text does not represent that textbook's PT graph. (I claimed that the IUPAC 18-column has this error). 2. I note too that any google search for an 18-col image shows dozens of graphical variants of the same PT (times the two or three structural variants we are talking about here). In any PT you see out there (books, internet), just look at exactly how the footnote is graphically linked to its place the main table. Quite often graphically ambivalent or wrong. These are unfit for a vote count on 'usage in literature'. 3. We are talking about a graphical structure, not a Van Gogh painting. The graphics must reflect the underlying structure. We can not describe a certain PT structure in the texts, and then pick a 'I like this face' PT from internet. -DePiep (talk) 11:06, 28 October 2015 (UTC)
- Points well taken. Thanks for bearing with me. YBG (talk) 17:11, 28 October 2015 (UTC)
- Yes, agree. Could you please go ahead DePiep. Sandbh (talk) 20:57, 28 October 2015 (UTC)
- Edited and corrected my text for clarity. -DePiep (talk) 16:47, 29 October 2015 (UTC)
- DePiep, could you also do a micro-form version of the 32-column table as per Jensen i.e. with the cells for Sc and Y stretched across La-Lu? Thank you. Sandbh (talk) 23:27, 29 October 2015 (UTC)
- You mean to show the PT I wrote here at "20:58" [search term]? -DePiep (talk) 01:30, 31 October 2015 (UTC)
- Does not Jensen (2008) say that this form was abandoned after 1946 (admitting that Seaborg wrote it that way back then?). -DePiep (talk) 02:03, 31 October 2015 (UTC)
- DePiep, could you also do a micro-form version of the 32-column table as per Jensen i.e. with the cells for Sc and Y stretched across La-Lu? Thank you. Sandbh (talk) 23:27, 29 October 2015 (UTC)
- Edited and corrected my text for clarity. -DePiep (talk) 16:47, 29 October 2015 (UTC)
- Yes, agree. Could you please go ahead DePiep. Sandbh (talk) 20:57, 28 October 2015 (UTC)
- Points well taken. Thanks for bearing with me. YBG (talk) 17:11, 28 October 2015 (UTC)
- re YBG.
- Yes, 20:58 is the one. As I read Jensen, he refers to it as antiquated, "chemical nonsense", rather than saying it was abandoned. It was still used by, for example, Housecroft and Sharpe (2008), Inorganic Chemistry, 3rd ed. They show "La-Lu" and "Ac-Lr" under Y. As well, the IUPAC table, which shows "57-71 lanthanoids [sic]" and "89-103 actinoids" in the two positions under Y is essentially the same. I've included a place for a micro-form 20:58 table in my sandbox, to illustrate what Jensen was describing, and to do justice to your concerns about the graphical ambiguity issue. PS: Could we use a dagger † rather than two asterisks to refer to the actinides? Sandbh (talk) 04:11, 31 October 2015 (UTC)
- re Sandbh This reply may be telegraph-style. May sound like making claims - so be it. No obstruction intended in there, but these days I don't have the time or patience to reply carefully & with due care, as the topics & you fellow editors deserve.
- - dagger - no, not now. It's complicated enough. January.
- - this edit by Double sharp, read the es. Pending this discussion, I'll leave it alone, and not discuss it here, again for reason of unneeded complication. However, once we have an outcome, it might be reversed. (I'd happily do that. Double sharp says that cerium is in group 4).
- - Sandbh asks for micro-PT's to have 18-col variants. Today, I don't think that shows well in a webpage. For example, take a look at the extended PT in micro form, which uses asterisks for for footnote elements (predicted elements 118+, this instance). It is way too tiny and scribbled to explain something to the reader. Also, the template does not scale at all (one cannot set the size). Let's consider this: the three svg-forms above are fine and can show anything we want. Now they don't have hyperlinks to articles (cannot click on Hg to get to mercury). But that can be added to the svg (I'm experimenting). So we can have scalable, clickable PT images. Great IMO, even and especially in this whole topic.
- - Sandbh also asks for PT's that are "chemical nonsense" (dixit Jensen), and not to be used on the PT history page (fun section)! I don't get the proposal (by Sandbh/sandbox). This is a tough
notnut to crack. I need time, and a clear & fresh mind to understand what this is leading to. - - Concluding. Technically (as in: graphs & webpage-smart) things are limited and can be handled, when isolated. But there are content issues too. I have not enough energy to solve the questions, nor to respond fresh and sound. Is why I can not promise to deliver all tries, for now. Nor can I make serious replies to content issues. OTOH, most of my arguments are on this wiki already. -DePiep (talk) 19:19, 1 November 2015 (UTC)
- Best way forward is when Sandbh drops the graph issue, and firstly has the group 3 scientific aspects into articles. (IOW, do-not-touch the 18-col issue). -DePiep (talk) 22:34, 3 November 2015 (UTC)
Draft Sc|Y|*|** version ready
I think the draft in my sandbox has got to a more or less stable state (aside from persnickety formatting and linking fussiness). The main changes appear in section 5 "Alternative structures", which is now called "Different periodic tables", and has two subsections, 5.1 "Layout variations", which summaries the three main forms of 18 column table; and 5.2 "Alternative structures" (no change). I've added a paragraph to section 6.6 "Period 6 and 7 elements in group 3" about the attempted use of the 32 column table to address this question. I've tried to do justice to DePiep's graphical concerns by including content about the apparent implications of some 15LaAc tables. Comments welcome. Sandbh (talk) 11:23, 5 November 2015 (UTC)
- I'll go and take a look. Must say, after what I met last weeks and you throwing together, as I read here, in #5 "Layout variations" and "Alternative structures" (in that order!!!), I'll prepare for a disappointment. -DePiep (talk) 21:51, 5 November 2015 (UTC)
- Nature gave us elements in a 32-column PT structure. Nature did not mind classroom size, book page format or as-I-learned-it habits. That PT was gradually and gently unfolded (discovered) by Mendeleev's "8-col" etcetera. Group 0 (=18) could be added without being disruptive. Ln's: more unfolding. Seaborg: more of the same (though he didn't mind the graphic back in '46). Some stuff is still scientifically ambivalent: position of He, categorisation of metallishness, and group 3 composition. Fine. We should aim to present this PT, in its forms & variants as based, and by detail when & where required, to our readers. And, by the way, as far as I know a classroom wall has the right format for a 32-column PT (longitude:height), much better than for an 18-column.
- However. Proposal User:Sandbh/sandbox#Different_periodic_tables has this: It describes graphical variants and structural variants of the PT as one sort of. It says: "5. Different_periodic_tables" (and, sic, in the reversed order of relevance at that). We must recognise: structural variants like 14CeTh and 14LaAc are scientifically based, and throw a nice new light on the PT. As do Janet's Left Step and ADOMAH. A graphical variant on the other hand, is just an illustrators choice: pic font and background colors, choose decomposition into footnote elements maybe. By definition, graphical variants do not alter the structure being presented.
- In short: that is what the "Sc|Y|*|**" (or "15LaAc") graph version does/so badly. It does not state which structure it represents. Or, even worse: it claims/suggests that group 3 consists of all 32 rare earth elements. And now Sandbh proposes this as some "3rd variant". But I see no source for such a structural variant (no Jensen, no Scerri). The check questions are: A. So what is in group 3 then?, and B. how do Sc and Y position in your 32-col form? (and C one can check the accompanying textr for discrepancies. e.g., the published IUPAC graph says group 3 = REM, but the text never says so).
- On top of this: OR. The linked paragraph opens like this: "The 18-column periodic table featured in this article [and elsewhere in this wiki, I add - DePiep] has been referred to as the 15LaAc form.[79] It shows 15 lanthanides and 15 actinides at the foot of the table". Sure it does, but why was this form chosen to represent it at all? Nowhere its 32-column equivalent is shown or referenced (of course not: it does not exist, in science). The only reference, #79, is jolly titled "Flyleaf [i.e. of textbooks] periodic table" (btw, I hope I understand the frivolity right. How is this a scientific ref?). So the source still has book formatting for base. The proposal also writes "other common forms": 'common' is not a source. Really, not.
- I propose to remove the Sc/Y/*/** form from the proposal altogether, and from all our enwiki pages. It can stay in the curiosities section of History of the periodic table. For our kids to laugh at their parents: why did you have to learn the IKEA version of the PT?
- -DePiep (talk) 23:04, 5 November 2015 (UTC)
Mind the gap (It lives again)
![](http://upload.wikimedia.org/wikipedia/commons/thumb/c/c6/15LaAc_with_LaAc_in_group_3.jpg/220px-15LaAc_with_LaAc_in_group_3.jpg)
@DePiep: I suspect your concerns would be addressed by either option A (left) or option B (right)? If I recall correctly you have no particular preference for either Lu|Lr or La|Ac lining up under Y in group 3. (talk) 03:18, 7 November 2015 (UTC)
- DePiep response: re Sandbh, "De Piep, I suspect your concerns would be addressed by ..." - indeed. So please throw the 15LaAc out. My fly over: my response above ("Nature gave us elements in a 32-column PT structure", [23:04]) was against current sandbox proposal [1] that treats three forms alike (the 18-cols: 15LaAc, 14CeTh and 14LaAc). I repeated my note (from "There is a flaw in this setup", [15:45]) that 14CeTh and 14LaAc are sound scientific variants, but 15LaAc is not a new variant, e.g. per my "The check questions are", above; it is a historic curiosity. So I concluded that 15LaAc should be removed from the proposal.
- R8R Gtrs comment: I am observing this discussion, rather than actively participating in it; but one thing I want to be clear is that we can't throw away the 15LaAc option. We may say it is not as natural in the sense of how it places 15 elements under one, and describe it as an unnatural construct (except not using these terms) used for particular reasons, but we can't ignore if we think it's wrong. People would use a description of PT variants in an encyclopedia, and we should provide one, since we happen to be editors of an encyclopedia.--R8R (talk) 15:41, 7 November 2015 (UTC)
- Double sharp comment: We cannot throw out 15LaAc; it has been used by enough reliable sources that it has to be mentioned. We can of course point out its problems, but it has to stay as a prominent mention. The most we can do is not use it as our main format.(talk) 16:21, 7 November 2015 (UTC)
DePiep, 15LaAc should not be removed for the reasons mentioned above but we can use its unambiguous form (option A or B above) instead of the confusing version currently used in the lede. Sandbh (talk) 22:12, 7 November 2015 (UTC)
- Then, which RS states that group 3 contains 32 elements? (don't forget what Jensen wrote: text may not correspond with the accompanying graph [2]). We also do so in this file (I disapprove). -DePiep (talk) 22:47, 7 November 2015 (UTC)
- Let me be more specific, re Double sharp: "We cannot throw out 15LaAc; it has been used by enough reliable sources that it has to be mentioned" and Sandbh: "should not be removed for the reasons mentioned above". I propose to throw it out of the topic/sandbox, and move it to the history department because sure it is/was used widely. The question is: which RS uses this graph variant while claiming this describes group 3 correctly? Of course, we should be alerted that the group 3 issue often was not considered at all (e.g., a sort of carelessness or sloppyness in drawing), and that the text might be inconsistent with the graph (IUPAC has, I claim). I also know that the number "3" is missing often, which reduces the preciseness of a PT. The notion "commonly used" may be true, but does not qualify as an RS. -DePiep (talk) 03:11, 8 November 2015 (UTC)
- re Sandbh we can use its unambiguous form (option A or B above) instead of the confusing version currently used in the lede (this one). - Yess, this is progress! Glad to have your signature on this. Because the lede PT is an 18-column PT saying group 3=Sc/Y/*/**. It is this 'ambiguity' (or worse) why I want that version out, out of the whole sandbox. (Now after that removal, we should talk about best replacement: one of A/B or those. That's a n independent topic. Really). -DePiep (talk) 03:29, 8 November 2015 (UTC)
- YES!!! You have it! High five time! Dip me in molasses, cover me in feathers and call me a lucky duck! Happy days are here again. I'll see if I can refactor the A or B discussion into a new section or subsection, unless someone else beats me to it. Sandbh (talk) 04:06, 8 November 2015 (UTC)
- re Sandbh we can use its unambiguous form (option A or B above) instead of the confusing version currently used in the lede (this one). - Yess, this is progress! Glad to have your signature on this. Because the lede PT is an 18-column PT saying group 3=Sc/Y/*/**. It is this 'ambiguity' (or worse) why I want that version out, out of the whole sandbox. (Now after that removal, we should talk about best replacement: one of A/B or those. That's a n independent topic. Really). -DePiep (talk) 03:29, 8 November 2015 (UTC)
- Let me be more specific, re Double sharp: "We cannot throw out 15LaAc; it has been used by enough reliable sources that it has to be mentioned" and Sandbh: "should not be removed for the reasons mentioned above". I propose to throw it out of the topic/sandbox, and move it to the history department because sure it is/was used widely. The question is: which RS uses this graph variant while claiming this describes group 3 correctly? Of course, we should be alerted that the group 3 issue often was not considered at all (e.g., a sort of carelessness or sloppyness in drawing), and that the text might be inconsistent with the graph (IUPAC has, I claim). I also know that the number "3" is missing often, which reduces the preciseness of a PT. The notion "commonly used" may be true, but does not qualify as an RS. -DePiep (talk) 03:11, 8 November 2015 (UTC)
- Comment on the new graphs A and B (top of this section). Sandbh is mixing up graphic presentation and scientific base. 1. Quite disturbingly, in here the notion "15LaAc" is used in a new definition, thereby introducing mistakes and mixing up two opposite statements. As Jensen wrote about 15LaAc (as opposed to 14LaAc & 14CeTh): In the case of the 15LaAc form, however, an entirely different interpretation is placed on these elements. The 30 elements are treated not as a separate independent electronic block but rather as degenerate members of group 3 of the d-block. 2. A and B still do not explain any scientific base for the "Sc/Y/*/**" variant. 3. There is an inconsistency wanting to group all lanthanides and actanides together in the footnote. Because this form in general of the PT has blocks not categories together. Categories are not the structure basics in here. It is an addition of complexity to start using categories as drawing criteria. 3 (minor at the moment but still): the re-introduction of "minor long form" terminology does not help, it is ambiguous.
- More worrying is that lot of scientists around here seem unable to keep independent issues independent. Instead, even more variants by different issues are created by mixing up. -DePiep (talk) 23:11, 7 November 2015 (UTC)
- I confess I don't understand Sandbh's PTs here. If you want to show Sc/Y/La/Ac, and yet use an 18-column table, then doesn't the gap have to be between groups 3 and 4, and La and Ac put under Sc and Y? That's how Greenwood and Earnshaw does it. Double sharp (talk) 02:39, 8 November 2015 (UTC)
- (I understand, after some studying, that they are both graphic variants of our familiar scientific group-3 variants: Sc/Y/La/Ac=14CeTh and Sc/Y/Lu/Lr=14LaAc. In essence, the variants have the Ln and An together. Earlier graphs of the same, you might recongnise more easily, are shown in top of #Graphic presentations below). -DePiep (talk) 02:59, 8 November 2015 (UTC)
- @Double sharp: DePiep's explanation is correct. In option B the lanthanides and actinides can only be moved into the main body of the table if La and Ac end up under Y in group 3 (and the end result would be a 32 column table). Does this help? Sandbh (talk) 03:33, 8 November 2015 (UTC)
- It does, but I'm still concerned about it, because I've never seen this form of Sc/Y/La/Ac 18-column before. Can you point me to a reliable source that uses it? In my experience, the most common 18-column table that seeks to put La and Ac under Y simply does so and has a an asterisk inserted in the cells of La and Ac, obviously intending to signify a gap between La/Ac and Hf/Rf. Double sharp (talk) 04:19, 8 November 2015 (UTC)
- No, I can't Double sharp, as I wasn't specifically looking for such an example when I was working on the problem. My focus was instead on seeing if I could eliminate ambiguity. The best I can offer is an example of its 32-column equivalent in: MacKay, MacKay & Henderson (2002, p. 196), Introduction to Modern Inorganic Chemistry, 6th ed., Nelson Thornes, Cheltenham, p. 196. Could you elaborate your concern? Not that I've ever looked closely, but it seems to me that there are effectively innumerable ways of representing a Sc/Y/La/Ac 18-column table, involving (either singly or in combination) asterisks in various positions, daggers, arrows, └> { , vertical pill-box slits, dotted lines, bars, connectors, and sundry other graphical artifacts and doodads, or sometimes even nothing. The tables involved are the equivalent of written forms of communication (are they not?) that all attempt to effectively say the same thing, but in different ways, presumably in an effort to be as clear as possible in their graphical representations. Hence option B. Sandbh (talk) 11:02, 8 November 2015 (UTC)
- re Double sharp "It does, but I'm still". My concern is this. At hand at this point are two pairs of graphs:
- one pair says "group 3= Sc/Y/Lu/Lr" (graphs 14CeTh and Sandbh's 'A')
- other pair says "group 3= Sc/Y/La/Ac" (graphs 14LaAc and Sandbh's 'B')
- not at hand now is: any "Sc/Y/*/**" form.
- The surpising new A and B forms keep Ln and Ac together by putting exactly all of them in the footnote. Were those form an improvement, it could be worth introducing them to the world through our readers, and having scientists frown & smile. But I see two flaws. Flaw 1: it breaks group 3. The core periodic table structure is columns & rows first, with reasons. A and B forms break this base: group 3(!) is split over two places: two elements are displaced into the footnote. Only after reconstruction (into 32-col form) are they visibly in the place -- while exactly that is what we want to show in this detail. IOW, we want to state "group 3 = ./././.", and then we cut that group in half. It is this extra mental step added that imo does not help the reader. Flaw 2: introducing cuts by category. The Ln and Ac together looks very nice at 1st sight because of their categorycolors being together. But categories are not the structuring base of the PT. Of the 12 categories we show, only one falls together with column(s). Categories are great to show the trend, but they should not be used to make a cut (for practical reason only) in the PT. -DePiep (talk) 06:32, 12 November 2015 (UTC)
- Exactly. If we want to show Sc/Y/La/Ac, then we had better see La and Ac in the positions right under Y, and not force the reader to mentally fit them in. Furthermore, I don't see anything wrong with splitting La and Ac from the other lanthanides and actinides if they are coloured similarly. Nobody complains that H is very far away from the other diatomic nonmetals. Double sharp (talk) 06:46, 12 November 2015 (UTC)
- re Double sharp "It does, but I'm still". My concern is this. At hand at this point are two pairs of graphs:
- @Double sharp: DePiep's explanation is correct. In option B the lanthanides and actinides can only be moved into the main body of the table if La and Ac end up under Y in group 3 (and the end result would be a 32 column table). Does this help? Sandbh (talk) 03:33, 8 November 2015 (UTC)
- (I understand, after some studying, that they are both graphic variants of our familiar scientific group-3 variants: Sc/Y/La/Ac=14CeTh and Sc/Y/Lu/Lr=14LaAc. In essence, the variants have the Ln and An together. Earlier graphs of the same, you might recongnise more easily, are shown in top of #Graphic presentations below). -DePiep (talk) 02:59, 8 November 2015 (UTC)
- I confess I don't understand Sandbh's PTs here. If you want to show Sc/Y/La/Ac, and yet use an 18-column table, then doesn't the gap have to be between groups 3 and 4, and La and Ac put under Sc and Y? That's how Greenwood and Earnshaw does it. Double sharp (talk) 02:39, 8 November 2015 (UTC)
Let us proceed with Sc|Y|La|Ac. The lanthanides are reasonably proximate to one another, spatially and chromatically. Membership of group 3 is clear. How the Ln and An fit into the main body of the table is straightforward, and this yields a 32-column form with no IKEA tears (the crying type). Sc|Y|La|Ac is the most common form. The gap between groups 3 and 4 is concordant with the group 3 metals behaving chemically more like the alkaline earths, and the s-block metals generally. Sc and Y remain coloured as transition metals in homage to their d1s2 electron configurations. The footnoted Ln and An line up nicely under the main body group numbers: Ce under group 4; the transition-metal-like earlier actinides---Th, Pa, U, Np, Pu---under groups 4, 5, 6, 7 and 8. Lanthanum and actinium are coloured as, respectively, a lanthanide and an actinide, seemingly out of whack with their d1s2 elecron configuations but consistent with their stature as progenitors of each of their own series of elements. An elegant package all round. And the asterisks line up nicely, too. Sandbh (talk) 00:28, 15 November 2015 (UTC)
La|Ac and Lu|Lr
---Extracted by me from the preceding section: "Mind the gap (It lives again)"--- Sandbh (talk) 11:43, 8 November 2015 (UTC)
I was an Lu|Lr in group 3 supporter, having regard to Jensen. Having looked closely into this question again—including doubts raised by Scerri on some of Jensen's arguments—I now (tentatively) support La|Ac in group 3 i.e. option B. As I believe User:Double sharp intimated, the chemistry of Sc, Y, La and Ac has much more in common with the alkaline earth metals in group 2, and the s-block metals more generally, than it does with that of the transition metals proper in groups 4 to 11.
If that is the case, then slightly "gapping" the d-block into portions of one and nine groups would be a more natural grouping of like with like.
Some of Jensen’s arguments for Sc|Y|Lu versus Sc|Y|La are highly questionable e.g. his comparison of periodic trends in atomic radii; sum of the first two ionization potentials (why these two given we are talking about group 3 elements?); melting point; and electronegativity. Here, Sc|Y|La clearly resembles Ca|Sr|Ba rather than the pattern seen in the transition metals.
I further like the fact that in the set {Sc, Y, La and the rest of the lanthanides}, the elements are in atomic number order, whereas this is not the case for the set {Sc, Y, Lu and the rest of the lanthanides}.
I also like an argument I stumbled upon in the 4th edition of Shriver & Atkins, here, re ionic radius and its influence on chemical properties. Here, Sc|Y|La is a better fit than Sc|Y|Lu.
I do agree that Sc, Y, La and Ac are physically more like transition metals (and I’m OK, I think, with Jensen’s arguments in this regard); however it seems to me that---these days---chemical properties trump physical properties when it comes to organising the periodic table.
Hence, to my surprise, I now find myself thinking that the composition of group 3 appears to indeed be better regarded as Sc, Y, La and Ac. I also like the fact that in option B, La and Ac are lined up under group 3. And, most of all, there is nil ambiguity as to the composition of group 3. Sandbh (talk) 03:18, 7 November 2015 (UTC)
- A bit strange that Sandbh now dives into the 14CeTh/14LaAc dispute without addressing
this[the above] 15LaAc issue first. I'd expect arguments on why this third "variant" should be in there -- or not. Once this is solved, we can talk about the remaining choice. Note that in this I avoid the question about "it should be in xx-column form", as this complicates the topic with irrelevant arguments (cluttering). -DePiep (talk) 13:35, 7 November 2015 (UTC)- And yes, Sandbh, I think it was I who said that! Even physically, though, we have Ti, Zr and Hf being similar hard refractory metals with high melting points, while Sc, Y, and La show a clearer trend (decresing melting point). And unlike Ti, Zr, and Hf being similar in reactivity, we have a clear trend going down from Sc to Ac of increasing reactivity. My favourite comparison is to see how Be/Mg/Zn/Cd/Hg looks like a p-block group, while Be/Mg/Ca/Sr/Ba/(Ra) looks like an s-block group, looking at densities and melting points. If you consider Sc and Y to be more like transition metals than main-group elements, then Lu makes for nicer-looking trends; but as you say, Sc, Y, La, and Ac have a chemistry more like their s-block neighbours, so it makes sense to choose the trends for Sc/Y/La resembling those of Ca/Sr/Ba. Double sharp (talk) 16:21, 7 November 2015 (UTC
- I have added these micro PTs to help me follow DS's thought processes. Perhaps others might find them helpful.
- (Side note: the above illustrations might have been improved if {{periodic table (micro)}} not only had
|mark=
but also|mark2=
to provide two contrasting markings.) - YBG (talk) 22:17, 7 November 2015 (UTC)
Rare earths
---Extracted by me from the La|Ac and Lu|Lr section---Sandbh (talk) 07:19, 13 November 2015 (UTC)
Actually this is making me think of using rare-earth metals as a classification again, given how Cotton and Wilkinson group Sc and Y with the lanthanides and say "the properties of Y are extremely similar to, and those of Sc mainly like, those of the lanthanide elements proper, and quite different from those of the regular d-block elements.", corroborating what has been said above. But this may be too hard to explain, so we perhaps would be better off leaving a note that Sc and Y are marginal transition metals, like what we do for group 12 (treating them otherwise leads to logical problems with K, Rb, Cs, predicted E113 and Fl, etc.) [see Group 12 section, below, re logical problems]. Double sharp (talk) 16:21, 7 November 2015 (UTC)
- Using "rare earth metals" (REM) as a category to encompass Sc, Y and the lanthanide elements (La to Lu) is a potentially neat and interesting option for carving up that part of the periodic table. But it means the lanthanides would not show as a separate colour category and I don't know if that would be a good or bad thing, or neither. I recall R8R Gtrs was not in favour of using a REM category since only older Russian professors apparently still use this term. Certainly, the general reader is more likely to have heard of the rare earths due to the amount of popular press coverage this term tends to attract ("China, a Rare Earths Giant, Set to Start Importing the Elements"; "Once scarce the world is suddenly awash in rare earths"; "Rare earth demand to increase 50% by 2020" etc). Sandbh (talk) 23:06, 11 November 2015 (UTC)
- I wouldn't be too sad about losing lanthanides as a separate category, as Sc and Y are so similar to the lanthanides (and are often covered together). The press coverage can't hurt. But getting rid of one borderline group (group 3) and not the other (group 12) does bother me a little. Double sharp (talk) 06:56, 12 November 2015 (UTC)
- re YBG "And yes, Sandbh, I think it was I ..." [in this section, above] About showing REMs as a separate category(-color). I do not doubt its scientific base and encyclopedic relevance. But this is off topic, because recoloring (i.e., using different categorisation-criteria) does not relate to the core question: what constitutes group 3 and how do we present that? Recoloring does not state or clarify anything with this. In general, groups and categories are per se not interchangeable (only once our PT says "group=category"!). Illustrative, we solved a lot by saying "halogens is a group name, but not a category name" (At). And remember, this is independent of the exact memberlist of any category, usually not definitive. Categories do not follow the periodic table structure (cols & rows), except for showing a definitive trend in periods (same colors are always adjacent in a period!). That said, this topic started by YBG may very well be a sound encyclopedic article or section. For sanity if the main topic discussion, I ask us all to keep this subtopic isolated & separate. Better not add another opion-multiplier that comes from the wrong preposition. -DePiep (talk) 05:59, 12 November 2015 (UTC)
- re Double sharp: "using rare-earth metals as a classification again", Sandbh: "Using "rare earth metals" (REM) as a category ... is a potentially neat and interesting option for carving up that part of the periodic table". First of all, this issue is only a consequence for trying to write an 18-column PT. In 32-col format this does not exist. (I maintain that they should be interchangeable, not stating different things). Then, the REM-grouping was done this way by Seaborg 1946 (says Jensen 2008). It is still present in the "Sc/Y/*/**" presentation form used today. So I can be repetitive about this: 1. approach by category introduces an extra criteria, thereby adding complexity to the graph. Except for the tetris-like color effect, we should not try to add more 'explanation' (incidental at that, b/c not done in the main group) to the basic PT. Just as we don't use natural occurrence (cell border) of state (Z font color) to reorganise the PT graph. 2. It has this huge, huge disadvantage that it is stating group 3 wrongly. 3. I see no advantages, except for elegance of grouped colors. That's not enough. -DePiep (talk) 08:15, 13 November 2015 (UTC)
- Huh? In a 32-column Sc/Y/La/Ac PT, it is quite easy to use REM as a category: it would encompass Sc, Y, La, and Ce-Lu, without disturbing the alignment of La below Y. Maybe I am misunderstanding you, but using REM would require no more than taking Template:Periodic table/Sc-Y-La-Ac/sandbox and colouring Sc and Y pink as well (and changing the legend from "Ln" to "REM"). This doesn't imply that we're putting all the lanthanides and actinides into group 3; the fact that we currently colour all the actinides the same way in a Sc/Y/Lu/Lr periodic table (where only Lr is in group 3) does not suddenly make Ac-No group 3 elements too. The only thing Sc/Y/La/Ac with REM would change is that La and Ac would be the group 3 elements of period 6 and 7, not Lu and Lr. There would still be only one element in each position below Y. Now [Sandbh] for group 3: do you think we should take Sc and Y out of the TMs and lump them with the lanthanides into REMs? I'm leaning towards yes, as they are also quite borderline as TMs. Double sharp (talk) 15:20, 13 November 2015 (UTC)
- re "it is quite easy to use REM as a category" -what? WHAT? are we drawing the PT by reason of easyness? -DePiep (talk) 21:45, 13 November 2015 (UTC)
- What we should be doing according to Wikipedia policy is following reliable sources. So how about someone doing a survey of recent (say 0-10 years) textbooks and review articles to see how many use Rare earths as a category? Dirac66 (talk) 22:20, 13 November 2015 (UTC)
- re "it is quite easy to use REM as a category" -what? WHAT? are we drawing the PT by reason of easyness? -DePiep (talk) 21:45, 13 November 2015 (UTC)
- Huh? In a 32-column Sc/Y/La/Ac PT, it is quite easy to use REM as a category: it would encompass Sc, Y, La, and Ce-Lu, without disturbing the alignment of La below Y. Maybe I am misunderstanding you, but using REM would require no more than taking Template:Periodic table/Sc-Y-La-Ac/sandbox and colouring Sc and Y pink as well (and changing the legend from "Ln" to "REM"). This doesn't imply that we're putting all the lanthanides and actinides into group 3; the fact that we currently colour all the actinides the same way in a Sc/Y/Lu/Lr periodic table (where only Lr is in group 3) does not suddenly make Ac-No group 3 elements too. The only thing Sc/Y/La/Ac with REM would change is that La and Ac would be the group 3 elements of period 6 and 7, not Lu and Lr. There would still be only one element in each position below Y. Now [Sandbh] for group 3: do you think we should take Sc and Y out of the TMs and lump them with the lanthanides into REMs? I'm leaning towards yes, as they are also quite borderline as TMs. Double sharp (talk) 15:20, 13 November 2015 (UTC)
- re YBG "And yes, Sandbh, I think it was I ..." [in this section, above] About showing REMs as a separate category(-color). I do not doubt its scientific base and encyclopedic relevance. But this is off topic, because recoloring (i.e., using different categorisation-criteria) does not relate to the core question: what constitutes group 3 and how do we present that? Recoloring does not state or clarify anything with this. In general, groups and categories are per se not interchangeable (only once our PT says "group=category"!). Illustrative, we solved a lot by saying "halogens is a group name, but not a category name" (At). And remember, this is independent of the exact memberlist of any category, usually not definitive. Categories do not follow the periodic table structure (cols & rows), except for showing a definitive trend in periods (same colors are always adjacent in a period!). That said, this topic started by YBG may very well be a sound encyclopedic article or section. For sanity if the main topic discussion, I ask us all to keep this subtopic isolated & separate. Better not add another opion-multiplier that comes from the wrong preposition. -DePiep (talk) 05:59, 12 November 2015 (UTC)
- I wouldn't be too sad about losing lanthanides as a separate category, as Sc and Y are so similar to the lanthanides (and are often covered together). The press coverage can't hurt. But getting rid of one borderline group (group 3) and not the other (group 12) does bother me a little. Double sharp (talk) 06:56, 12 November 2015 (UTC)
- REM is an existing category, RS & no dispute. But to
useintroduce it as a graph organiser for the 18-col only, I oppose. (graphs please) -DePiep (talk)- I don't think it should be used in 18-col only. If we end up using it, it ought to be in both 18-col and 32-col. Double sharp (talk) 11:11, 14 November 2015 (UTC)
- REM is an existing category, RS & no dispute. But to
- So, Double sharp, you really maintain that we should drop Ln and An categories completely, and in the process declare Sc and Y non-TM because they are --undisputed-- REM's? That is bad categorization because it mixes up category schemes (as if we'd recolor the PT by occurrence for only some elements). But could you be clear: do you propose & support & push each and every PT to be "Sc/Y/*/**" still? (Would save me a lot of time, not having to repeat my arguments). -DePiep (talk) 23:59, 14 November 2015 (UTC)
I went ahead and made a graph of usage of the phrases "lanthanide," "lanthanoid," and "rare earrh element" (2000–2008, since the service doesn't scan in later books). Lanthanides undisputably won.
- Well, in compounds we get terms like "rare earth mineral" and "rare earth magnet", so it may be fairer to use just "rare earth", in which case "rare earth" actually wins. Double sharp (talk) 11:13, 14 November 2015 (UTC)
- Hmm. Indeed. .0000256% vs. .0000155% is a larger share.
- However, this can't be called an indisputable victory: I have the feeling "lanthanide" is better in the context of chemistry, and "rare earth" is now a geology-related word. And both terms you suggest are not chemistry really: they're geology and electromagnetism, correspondingly (which was how I came to this thinking). The whole term "rare earth" is geology. We are concerned with chemistry for now, so I would pick the good old "lanthanide" for element categorization.--R8R (talk) 12:51, 14 November 2015 (UTC)
Maybe I was too emotional against the REM as a term, but I still stand against it being used here. Not just because it is used so infrequently—we have overlapping categories here, we have to make our editorial choices—it removes scandium and yttrium from the TM category, a rare approach to the problem these days as well. Two elements carved from TMs is sort of the least bad solution we have, in part because it is so unnatural. Four is a certain structural decision, and a really questionable one. (I would say d electron being involved in basically every group 3 compound is more than enough to call them TMs.)--R8R (talk) 07:13, 14 November 2015 (UTC)
- I don't believe nor think this occurrence statitics says anything about how we should present our PT. What's wrong with thinking & arguments? -DePiep (talk) 20:02, 14 November 2015 (UTC)
Group 12: Post-transition metals
---Extracted by me from the Rare earths section--- Sandbh (talk) 23:24, 14 November 2015 (UTC)
The group 12 metals (Zn, Cd and Hg) have effectively zero transition metal (TM) properties. They would be better categorised as post transition metals (PTM), but see below. E113 and Fl would count as TMs given the predicted involvement of their d electrons in compound formation.
In the transition metal article I would probably include the group 3 metals (Sc, Y, La, and possibly even Ac), and the group 12 metals (Zn, Cd, Hg) for comparative purposes but show them coloured as , respectively, rare earth metals and post transition metals.
What logical problems did you foresee with K, Rb and Cs? Sandbh (talk) 23:06, 11 November 2015 (UTC)
- See this article. "At high pressure the alkali metals potassium, rubidium, and cesium transform to metals that have a d1 electron configuration, becoming transition metal-like." Actually, I think you linked it first! This is why I'm not so keen on the d-electron argument: there isn't an obvious place to draw the line (except saying "any d-electron involvement at all counts", which makes Hg a transition metal but Cd not, which is rather silly). If you say Hg doesn't pass it, how do we know if E113 and Fl are going to?
- I wouldn't say Zn, Cd, and Hg have no TM properties. Zn forms stable complexes with O-, N-, and S-donor ligands as well as halides and CN−, like a transition metal, and Cd is rather similar. Additionally, many compounds of HgII (and CdII to a lesser extent) are highly coloured, a characteristic of transition metals. It seems like the easiest way to go is to call everything from groups 3 to 12 TMs, but call Sc/Y/La/Ac and Zn/Cd/Hg borderline TMs.
Double sharp, thank you. Your reference to Hg as a transition metal lacks any rigour. In this case you've quoted me out of context. The full story is that Nergaal regarded mercury as a transition metal on the basis that it could form HgF4 in a cryogenic matrix at a few degrees above absolute zero. I responded by saying that, ergo, K was a transition metal since it becomes one at high pressure i.e. greater than 30 Gpa. These are both ridiculous statements given periodic table categorisations are self-evidently based on the properties of the elements in ambient conditions. Regarding element 113 and flerovium, these may well qualify as transition metals based on their predicted chemistry, as I understand from reading their articles. These two elements are distractions in any event---we could easily leave them as post-transition metals until stronger data comes in.
Your argument about the properties of group 12 metals i.e. complex formation and some highly coloured compounds is lame. As I understand it, all or nearly all metals and metalloids can form stable complexes. And while many transition compounds are strongly coloured, some main group compounds are too e.g. SbI3 (ruby red); cesium oxide and sub-oxides ("brightly coloured"); and lets not forget brightly coloured Zintl phases The best that can be said about the group 12 metals is that they are capable of forming some stable complexes, just like e.g. the alkaline earth metals and group 13 metals can, and they are also capable of forming some brightly coloured compounds, just like some main group metals can. To suggest that these properties warrant calling them transition metals is like calling an orange and apple. Greenwood and Earnshaw, whom you were rightly quoting from, say these elements show few of the characteristic properties of transition metals; the only ones they discuss are complexes and highly coloured compounds and they then go on to note the drawbacks of having d10 configurations on complex formation. Cotton and Wilkinson go further and say that the group 12 elements are not classified as transition metals and that while their capacity to form complexes is "reminiscent" of transition metals, it is significantly qualified. Group 12 as transition metals on par with groups 4–11? Phooey! Sandbh (talk) 10:43, 13 November 2015 (UTC)
- Thank you for stating where the line is drawn: does it happen at standard conditions, or not? Now we have a completely sound basis to exclude group 12, so now I agree with you. (Really: apart from not being sure where to draw the line, I don't really have any objections to taking out group 12).
- Now how do we colour copernicium? Minor element, I know, but it's going to show up on all our periodic tables. Physically it is known to behave like a typical group 12 metal, except that it is predicted to have Cn(IV) as a significant part of its chemistry. The trouble with flerovium is that Fricke's old predictions mention sd hybridisation as a possibility, but the recent sources all assume Fl(IV) would go for sp3; this seems to imply that new predictions have the 6d energy levels too low. E113 OTOH is still predicted to have 113(V) AFAIK. But these are all still predictions, so maybe we ought to just colour them like their groups (post-transition metals) until we get evidence otherwise. After all, the only thing we know about them so far is that physically, they behave like their congeners in their groups. As such we would colour Cn and Fl as post-transition metals (and E113 as a predicted PTM) until evidence comes in for Cn(IV), 113(V), and Fl(VI). This is what Japanese Wikipedia does (or at least it will until Dlgkstmf ar1103 insists on restoring his/her favourite classification from his/her favourite science magazine, which apparently is not aware of the existence of chemical studies on At, Fr, and Fl but spontaneously invented studies on Mt, Ds, and Rg out of thin air, and has some penchant for keeping Ge and Sb out of the metalloid club while putting Se in it, and before I revert him/her again). In fact, they're very similar to your current proposal for the right side of the periodic table (but calls those just "other metals", and puts Be and Mg in that club; is that traditional there?). (talk) 15:20, 13 November 2015 (UTC)
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 [3]. Not about its status (being "not the only official one"), but because it is ambiguous graphically. It
suggestsstates 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.
- 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)
- 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)
- OK.
- 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)
- 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)
- 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)
- 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)
Graphic presentations
The essential two structures are:
- Group 3 = Sc/Y/Lu/Lr
![](http://upload.wikimedia.org/wikipedia/commons/thumb/7/78/Periodic_table_overview_%2818-columns%2C_group_3_%3D_Sc-Y-Lu-Lr%29.svg/140px-Periodic_table_overview_%2818-columns%2C_group_3_%3D_Sc-Y-Lu-Lr%29.svg.png)
- Corresponding 18-column graphs: designated 14LaAc (right, and {{this}})
- Group 3 = Sc/Y/La/Ac
![](http://upload.wikimedia.org/wikipedia/commons/thumb/7/78/Periodic_table_overview_%2818-columns%2C_group_3_%3D_Sc-Y-La-Ac%29.svg/140px-Periodic_table_overview_%2818-columns%2C_group_3_%3D_Sc-Y-La-Ac%29.svg.png)
Template:Periodic table (32 columns, micro)/Sc-Y-La-Ac/sandbox
- 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)
- 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)
|
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)
- 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)
- "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)
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)
- @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)
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)
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)
- 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)
- 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)
- 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)
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 article
s(period 1and 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'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)
- 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?)
- 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)
- See also this thread on new evidence supporting H's placement in group 1. Sandbh (talk) 03:56, 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)
- Can you give me a source that actually used his one?--R8R (talk) 12:20, 21 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)
- 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)
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)