Wikipedia talk:WikiProject Elements/Archive 16

Make the group 12 elements poor metals?

 * (@DePiep: my excuse for this new Big Issue™ is that we are now debating on the science. Nothing will be changed yet, if at all.) Double sharp. ;-) for local joke - ok. DePiep.

I still feel that we should rethink our position on group 12 as transition metals. Chemically they are really not very close to the other transition metals and are far closer to the poor metals in their chemical properties. Even IUPAC offers us the choice to exclude group 12 from the transition metals (whereas they say nothing on issues like Sc/Y/Lu/Lr vs. Sc/Y/La/Ac), etc.

Here is how it looks (extended):


 * Technical edits: 1. Use short, simple & fast element cell-1. -DePiep (talk) 22:35, 22 August 2013 (UTC)
 * 2. Demonstrate proposed changes:
 * 3. Voila. Simple & fast for talkpage examples. -DePiep (talk) 23:00, 22 August 2013 (UTC)
 * Wow! Eye candy smorgasbord. Fine piece of work. Sandbh (talk) 02:45, 23 August 2013 (UTC)
 * Thanks. It should be very practical & easy to use. I hope no one comes by to say: "hey, why not this one in article space ...". -DePiep (talk) 16:17, 31 August 2013 (UTC)
 * Updated to reflect agreement to classify group 3 as rare earth metals. For the "lanthanides" category above, read "rare earth metals". Double sharp (talk) 14:20, 7 October 2013 (UTC)
 * Legend /sandbox now shows "rare earth metal" for the same pink lanthanides color. -DePiep (talk) 21:10, 8 October 2013 (UTC)

Issue: For Hg, HgF4 is definitely not a representative compound. Copernicium is described by the experimenters as a typical group 12 element based on physical properties only, right? Should this then be marked as predicted? (Even if we don't implement this, it's still worth considering.) On element 164, it's predicted to show the +4 and +6 oxidation states as well as +2 readily, which would make it irrefutably a transition metal.

As always, feel free to argue. Jensen's paper is here. Double sharp (talk) 14:05, 21 August 2013 (UTC)
 * Support, based on physical, spectroscopic, chemical and theoretical grounds, as comprehensively and eloquently argued by Jensen and as reemphasized or further illustrated by the following:


 * 1. '…the triad Zn, Cd and Hg have more in common with their immediate neighbours in the p block than they do with their neighbours on the other side, in the d block. (Smith 1990, p. 113)


 * 2. '…one of the most conspicuous features in passing from Group IB elements (Cu, Ag and Au) to Group IIB elements (Zn, Cd, and Hg) is the abrupt and considerable reduction in metallic nature from Group IB to Group IIB. Group IIB elements have lower melting points, heats of fusion, boiling points, heats of atomization, and electrical and thermal conductivity than Group IB elements. The availability of d orbital electrons of IB, but not IIB, elements accounts for the difference.' (Sorensen 1991, p. 3)


 * 3. 'The chemistry of zinc, cadmium and mercury is conveniently treated with that of the main group elements because their filled d shells are retained in all stable derivatives.' (King 1995, p. 273)


 * 4. 'In view of the stability of the filled d shell, these elements [Zn, Cd Hg] show few of the characteristic properties of transition metals despite their position in the d block of the periodic table.' (Greenwood & Earnshaw 1998, p. 1206)


 * 5. '…while Cu, Ag and Au are classified as transition elements, Zn, Cd and Hg are not.' (Cotton & Wilkinson et al. 1999, p. 598)


 * 6. Zinc, cadmium and mercury are included in Massey's Main group chemistry (2000, pp. 159–176).


 * 7. 'The zinc group does not fit the general picture of the transition Groups as developed in the last two chapters…In this case, the three elements of this Group resemble the three heavy elements of the boron Group. (MacKay, Mackay & Henderson 2002, p. 385)


 * 8. 'The Group 12 elements have filled d orbitals in all their compounds, so they are better considered as main group metals. Consistent with this assignment, most of the compounds of the Group 12 metals are white, expect when the anion is colored…The only real similarity between the Group 12 elements and the transition metals is complex formation, particularly with ligands such as ammonia, cyanide ions, and halide ions. All of the metals, but especially mercury, tend to form covalent rather than ionic compounds. (Rayner-Canham G & Overton T 2006, p. 567)


 * 9. '…[the] practical consequences [of the presumed preparation of HgF4] for how we view and use the periodic table are essentially negligible.' (Jensen 2008, p. 1183)


 * Cotton FA, Wilkinson G, Murillo CA & Bochmann 1999, Advanced inorganic chemistry, 6th ed., John Wiley & Sons, New York
 * Greenwood NN & Earnshaw A 1998, Chemistry of the elements, 2nd ed., Butterworth Heinemann, Oxford
 * Jensen WB 2008, 'Is mercury now a transition element?', Journal of Chemical Education, vol. 85, no. 9, pp. 1182–83
 * King RB 1995, Inorganic chemistry of the main group elements, VCH, New York
 * Massey AG 2000, Main group chemistry, John Wiley & Sons, Chichester
 * MacKay KM, MacKay RA & Henderson W 2002, Introduction to main group chemistry, 6th ed., Nelson Thornes, Cheltenham
 * Rayner-Canham G & Overton T 2006, Descriptive inorganic chemistry, 4th ed., WH Freeman and Company, New York
 * Smith DW 1990, Inorganic substances: A prelude to the study of descriptive inorganic chemistry, Cambridge University, Cambridge
 * Sorensen EMB 1991, Metal poisoning in fish, CRC Press, Boca Raton, Florida


 * What to do about copernicium. Physically, probably a poor metal; chemically probably a transition metal. Seems to point to the limitations of our current practice of placing each element in only one category, rather than showing a few elements as belonging to two categories. Sandbh (talk) 00:22, 22 August 2013 (UTC)


 * Hg is a transition metal. 03:48, 22 August 2013 (UTC)
 * Hmm. Nergaal: is Au a nonmetal because it can form an anion? Could you please also let us know what you think about reference 9? Sandbh (talk) 04:07, 22 August 2013 (UTC)


 * Entertaining read, but by the guys rationale, the transactinide elements are not actual elements since they are observed under atypical, non-equilibrium conditions. If we are not going to respect the rules set by the appropriate body, then this is nothing else than a forum where people throw around their personal choices. Nergaal (talk) 05:31, 22 August 2013 (UTC)
 * Interesting argument. However, I submit that the comparison isn't valid. Transactindes are elements (and vice-versa) based on what evidence there is. This is vastly different from suggesting that because Hg can behave like a transition element for a 1000th of its chemistry (more like a 1,000,000th or less) it is therefore best regarded as a transition metal, discounting the fact that for the other 999,999 millionths of its chemistry it behaves as main group (poor) metal. IUPAC 'rules' have their place but not when their application produces non-sensical (IMO) outcomes. Sandbh (talk) 06:10, 22 August 2013 (UTC)
 * Seriously, HgF4 is irrefutably atypical chemistry. We have HArF; is anyone going to suggest that Ar is not a noble gas because it can form compounds? (Off topic: were there any proposals to redefine the scope of the noble gas category, or even rename it, after the Bartlett "XePtF6" experiment? Sandbh??) Only in 118 where cationic behaviour becomes the norm do you actually have people questioning its noble gas status. On IUPAC: (sorry for bringing this up, R8R) there was some chaos at a while back based on the phrasing of a IUPAC definition (A has valence x if there's a compound AFx or AHx, was it not?) And also, I would say we should not follow them dogmatically, but actually look at common usage. You will find some authors putting the whole of group 12 into the transition metals and some other authors leaving all of them as poor metals (post-transition metals). Have you ever seen a serious book that covers group 12 chemistry that treats Zn and Cd as poor metals and Hg as a transition metal?


 * Yes, Cn is a problem! My defense is that these colours are for chemical properties of the elements (as evidenced by the "Unknown" label reading "Unknown chemical properties"), and so we should not take physical properties into account when we do this. Indeed, a periodic table organized solely by physical properties would be an absolute nightmare in the p-block! :-) (164 is expected to have similar physical properties to Hg as well.) Double sharp (talk) 10:55, 22 August 2013 (UTC)


 * K is a transition metal, in accordance with IUPAC rules.


 * Seriously, this underscores the folly of applying IUPAC rules on the basis of extraordinarily atypical situations. Sandbh (talk) 02:45, 23 August 2013 (UTC)


 * About Cn, I think the categorisation of the elements is based on considering both their physical and chemical properties, and trying to work out a best fit categorisation as indicated by the literature. Our chemical properties label I interpreted a bit more generically as a chemical element with such and such physical and chemical properties, rather than confining things purely to chemical properties, although for sure these are important. This is all a bit moot in any event. If Cn is predicted to have a predominance of its chemistry in the +4 state then I think that would be enough to justify transition metal status. Same principle applies for mercury which is a liquid but nobody holds that against counting it as a metal; gold with some of its peculiar properties yet still regarded as a transition metal; manganese with its highly brittle comportment, very low electrical conductivity, bizarrely directional crystalline structure, relatively low packing efficiency and e.g. its capacity to form an acid (permanganic), yet still regarded as a transition metal; ditto gallium with its weird Ga2 molecular structure, melt in your hand behaviour, and amphoteric chemistry, yet still regarded as a metal. So, yes, Cn as a transition metal although it may be somewhat of the runt of the TMs. Sandbh (talk) 04:23, 23 August 2013‎ (UTC)


 * I think noble gases used to be referred as inert gases before their actual chemistry was discovered. Nergaal (talk) 14:13, 23 August 2013 (UTC)
 * Yup, I forgot all about that. Oops. Double sharp (talk) 16:22, 23 August 2013 (UTC)
 * Errata: Element 113 Actually, surely by this definition 113 is also a transition metal, as 6d electrons are predicted to be chemically active in 113(II), 113(III) (both with 6d9), and 113(V) (the last from unstable (113)F5 and more stable ). So I changed its colouring here. Double sharp (talk) 12:20, 26 August 2013 (UTC)
 * For some reason the Inert pair effect article was changed to say it refers to poor metals where it previously said post transition metals. The inert pair effect does not apply to aluminium and only applies to post transition metals. A revert I have made has already been reverted by user DePiep and he suggests that I come here- although what the inert pair effect has to do with the group 12 metals escapes me! I hope someone with a knowledge of chemistry gets involved, I do not have the time or inclination to get involved in a wikipedia edit war but it would be nice if the inert pair article could be corrected. Axiosaurus (talk) 13:54, 26 August 2013 (UTC)
 * Fixed back to your version.
 * @DePiep: Thanks a lot for your edits, but do keep in mind that "post-transition metal" and "poor metal" are not exactly synonymous: the former does not include aluminium, while the latter does. In fact our previous periodic table was WRONG because it labelled Al as a post-transition metal!!!
 * All fine. Anyone here may revert me in these things without spending much time explaining. (Your time better spend, and I can stand such a revert -- or should be able to do so. smiley here). -DePiep (talk) 17:25, 15 September 2013 (UTC)
 * On group 12: indeed, the inert pair effect is only expected to extend to group 12 in the 7th period with copernicium. Ref:, p. 82. Hence the predictions about noble-gas Cn. (Cn is expected now to be a metal, but a rather unreactive one, because the 7s electrons are less involved chemically than the 6s ones due to relativistic effects. (It's expected to extend to all the transition metals in the 8th period, so that the group oxidation state is possible, but the group oxidation state minus 2 is much more common!) Double sharp (talk) 14:13, 26 August 2013 (UTC)
 * From what I can see, the low reactivity of Hg has, at least in part, been attributed to the inert pair effect. And, as I recall, the effect isn't confined to either the poor metals or the post-transition metals: it goes as far as into As, a metalloid, and at least Br, a nonmetal. Cooper (1968) includes a periodic table with an inert pair dotted line starting halfway through Hg and extending all the way up to halfway through Cl. OK, that's an old ref which may be in need of supplementation by more recent thinking but for sure our current article doesn't tell the full story. Sandbh (talk) 20:30, 26 August 2013 (UTC)
 * Cooper HG 1968, The periodic table, 4th ed., Plenum Press, New York
 * To clarify, Cooper (pp. 15–16) said the inert pair effect occured in: Group IIb: Hg ("unreactivity, effective valency 0"); Group IIIb: Ga (partly), In, and Tl; Group IVb: Ge (partly), Sn and Pb; Group Vb: As (partly), Sb and Bi; Group VIb: S (partly), Se, Te and Po; and Group VIIb: Cl (partly), Br, I and At. He says, "Sulphur and chlorine are considered to show inert pair behaviour; in their case the pair of electrons is additional to the complete octet (the rather rare 4-covalent sulphur, 8 shared and 2 unshared; 3-covalent chlorine, as in ClF3, 6 shared and 4 unshared).' Sandbh (talk) 00:54, 9 September 2013 (UTC)
 * Expansion really is needed, it seems, for that article! I suppose I now have to change my comment to read that Cn will be much more affected by the inert pair effect than Hg.
 * So, do we think it's OK to make this change? Double sharp (talk) 05:04, 9 September 2013 (UTC)
 * (spammed all the talk pages for the active project members, except those who have already commented) Double sharp (talk) 05:11, 9 September 2013 (UTC)
 * Intermediate check: current proposed changes are: Zn, Cd, Hg → poor metal; Cn → only predicted transistion metal (not proven); 113 → predicted transition metal. -DePiep (talk) 17:17, 9 September 2013 (UTC)

Is there a good book with the sentence:"Zinc is a poor metal." Not with the sentence :"Zinc is a post-transition metal."? Stone (talk) 20:48, 10 September 2013 (UTC)
 * "The Group 13 elements are often designated as poor metals, except boron that is considered as a nonmetal metalloid". And, well, I didn't look that hard. But we can see from this that "poor metals" is alive and in use as a term, it seems, so I think it's OK to start going around calling Zn, Cd, and Hg poor metals too. Double sharp (talk) 03:06, 11 September 2013 (UTC)
 * Looking around, I found:


 * 1. 'Chalcophile elements (e.g. Ag, As, Bi, Cd, Cu, Pb, Zn). Those metals (sometimes called "poor metals") and heavier nonmetals that have a low affinity for oxygen and prefer to bond with sulfur as highly insoluble sulfides.' (Yousif 2007, p. 11).

The only good reference I see. This is Geochemistry and not a Chemistry text book.--Stone (talk) 09:06, 14 September 2013 (UTC)


 * 2. 'Adding…more of the poor metal zinc will degrade score resistance.' (Booser 1992, p. 2) Note however that Booser refers to lead as a 'good element' for increasing score resistance in a copper alloy.

This might work too, but this is material science with good and poor metals.--Stone (talk) 09:06, 14 September 2013 (UTC)


 * 3. 'Among the other metals, while some have excellent technical qualities, the price is prohibitive. The most advantageous combination is, apparently, steel-zinc. It has often been employed; but zinc is a poor metal, very oxidizable and liable to change of dimensions in the course of years.' (The Jewelers Circular 1922, p. 141)

A Jewlers journal for basic chemistry, vey questionable.--Stone (talk) 09:06, 14 September 2013 (UTC)


 * 4. 'Pure zinc is a soft metal, but contrary to the general rule, is a poor metal to work…' (Industrial Press 1912, p. 33)
 * What's the rest of the quote? It seems like 'poor' is being used in its normal, non-technical use.  YBG (talk) 03:19, 12 September 2013 (UTC)
 * '...in this [extrusion] process.' Sandbh (talk) 04:36, 12 September 2013 (UTC)

A poor metal to work is nort a poor metal.--Stone (talk) 09:06, 14 September 2013 (UTC)


 * And the following related quotes:


 * 5. 'Aluminium is one of the borderline or weak metals. These metals, other examples being zinc and tin, have chemical properties of both metals and nonmetals.' (Rayner-Canham & Overton 2006, p. 193)


 * 6. '…a relatively weak metal such as zinc…' (Zahner 1996)


 * 7. 'Zinc is a…metal with a relatively low melting point…and boiling point…When unalloyed, its strength and hardness is greater than that of tin or lead, but appreciably less than that of aluminium or copper. The pure metal cannot be used in stressed applications due to low creep-resistance.' (Porter 1991, p. 36)


 * 8. 'Zinc is weak, costs over twice as much per pound as pig iron or low carbon steel, and even with good corrosion resistance usually needs plating for good appearance.' (Moore & Kibbey 1982, p. 102).


 * 9. '…metals which one would call weak, such aluminium and zinc, have the physical properties of good metals [e.g. ductility, conductivity], but some of their chemical properties are those of non-metals.' (Stott 1956, p. 100)


 * 10. '… any metal which "can be easily cut with a knife" is not particularly good for structural purposes as a metal…It is very much such a metal as zinc or tin.' (Brown 1891, p. 152)

None of those is worth mentioning. --Stone (talk) 09:06, 14 September 2013 (UTC)


 * Booser ER 1992, 'Bearing naterials,' Kirk-Othmer Encyclopedia of Chemical Technology, vol. 4
 * Brown AB 1891, 'Electricity in the production of aluminum,' Transactions of the American Institite of Electrical Engineers, vol. VIII
 * Industrial Press 1912, Machinery's Reference Series, vol. 101-110
 * Moore HD & Kibbey DR 1982, Manufacturing, materials and processes, 3rd ed., Grid Publications, Columbus, Ohio
 * Porter F L 1991, Zinc handbook: Properties, processing, and use in design, 2nd ed., Marcel Dekker, New York
 * Rayner-Canham G & Overton T 2006, Descriptive inorganic chemistry, 4th ed., W. H. Freeman and Company, New York
 * Sott RW 1956, A companion to physical and inorganic chemistry, Longmans, Green and Co., London
 * The Jewelers Circular 1922, vol. 85, no. 1
 * Yousif N 2007, Geochemistry of stream sediment from the state of Colorado using NURE data, ETD Collection for the University of Texas, El Paso, paper AAI3273991
 * Zahner LW 1996, 'The selection, specification, and performance of metals in architecture,' JOM, vol. 48, no. 3, pp. 14–15
 * Sandbh (talk) 06:30, 11 September 2013 (UTC)


 * Just as Sandbh prepared the new poor metals article, would it be helpful to prepare the new Cn and element 113 text with references, somewhere? -DePiep (talk) 13:16, 13 September 2013 (UTC)
 * So, no changes in group 9 then (167?). -DePiep (talk) 13:37, 13 September 2013 (UTC)
 * (I think you mean period 9. Period = horizontal, group = vertical.)
 * Period 9 is expected to be like periods 2 and 3. "Reading between the lines" of Fricke's paper, the 9s electrons should be chemically active in the 9th period and hence 167 should not be a transition metal, having +3 as its main oxidation state. Same goes for 168–170, if you're wondering.
 * Errata: Elements 165 and 166 165 and 166 are trickier. Fricke's more detailed paper says that "From this side E165 and E166 will be members of the groups Ia and IIa. From a more chemical point of view, they will be likely more members of the Ib and IIb groups because of the 7d shell which is more comparable to the elements Au and Hg (but also to the elements E119 and E120) as can be seen from Fig. 14. Therefore, higher oxidation states than 1 and 2 might readily occur.)" Whereas his more overview-like paper softens this to just being a possibility that "might" happen. But I'm more inclined to trust this one. So I have changed them to predicted transition metals here. Double sharp (talk) 13:56, 13 September 2013 (UTC)
 * Sandbh, what do you think of this choice? I think it's defensible and reasonable (isn't Tl(I) similar to K(I) due to similar ionic radii, whereas Tl(III) behaves more like Al(III)? This is just a more extreme version, and involving alkali metal—transition metal, not alkali metal—poor metal.)
 * It's a pity that I can't reflect the fact that 119 and 120 should also form +3 and +4 oxidation states respectively, because those use 7p3/2 electrons and hence don't fit our "transition metal" definition! Double sharp (talk) 14:39, 13 September 2013 (UTC)
 * Yet another problem is that while by this definition 165 and 166 become transition metals, it appears that the 7d electron involvement makes them not have transition metal character, but rather group 13 and group 14 metal character. In light of this I am thoroughly confused on what to call these elements. They seem to fit about as well above 167 and 168 as below 119 and 120. Sandbh, please save me. Double sharp (talk) 07:43, 14 September 2013 (UTC)

Here's a borderline case: Haire (I know very well that he didn't write that chapter of the book, but it seems to be the most readily understood way of referring to that book here) states that the early 7p elements (may be referring to 113–115? He does not make this very clear) may have transition metal character due to 6d destabilization. I have not reflected this because it is a maybe, and is not further treated in any of the following separate sections on the 7p transactinides, save for 113 (which I have duly marked as a predicted transition metal above). Double sharp (talk) 14:01, 13 September 2013 (UTC)
 * Just to be clear: I want the 165, 166 situation to be clear and reasoned and sourced before we conclude. It needs to be fleshed out. No squeezing in. -DePiep (talk) 21:16, 13 September 2013 (UTC)
 * Same for 112Cn and 113. Other than the sandbh sandbox proposal for measured elements (Zn Cd Hg), we do not have a coherent discussion & outcome for these. There is no encyclopedic text. For these predicted properties, I do not see consensus. -DePiep (talk) 23:37, 13 September 2013 (UTC)


 * some new topic
 * There is one guy, a geochemist (Yousif N 2007, Geochemistry of stream sediment from the state of Colorado using NURE data, ETD Collection for the University of Texas, El Paso,) calling zinc a poor metal. No good chemistry text book does so. No discussion in any chemical education journal, nowhere I look. For me this is pure construction without reference. If we change this I seriously doubt that this will stand a closer look for original research. To change a few colours in a periodic table which is used to guide people in wikipedia is our business, but to state that some metal is a poor metal without the overwhelming number of text books on our side is assumption of authority. --Stone (talk) 09:06, 14 September 2013 (UTC)
 * So? By this IUPAC definition, Zn, Cd and Hg are post-transition. None will take issue with that. Much of the categories you see here for the superheavies are "reading between the lines" too. Yet they are all we have, and no one complains about that, because obviously no colouring is less informative than some colouring.
 * On "poor metals", these are often used to mean "main-group metals that aren't in the s-block" (citation needed, but readily available). We already have a source that Zn, Cd, and Hg are sometimes treated as main group metals. So they can be poor metals. We cannot use "post-transition metals" anywhere in this categorization scheme as it leaves Al colourless. Neither should we use such terms as "B group metals" (that's dependent on whether you use the ACS or old IUPAC scheme). I do not see any viable alternative that does not cause even more problems. ("Other metals" is terrible and information-poor.)
 * "Assumption of authority"? This is really the sort of thing IAR was made for. We may not have an explicit mention that Zn is a poor metal. We have definitions of poor metal from sources that, coupled with our new transition metal definition, have to include Zn. (The same applies to Cd and Hg.) Thus we must really put them as poor metals. Double sharp (talk) 12:20, 14 September 2013 (UTC)
 * Poor idea I admit to not following the discussion closely, but I teach inorganic for decades and publish in the area.  I never hear the term "poor metal."  Possibly it is used in some metallurgical context.  Editors need to be cautious about giving undue weight to obscure jargon, although I am sure this is all well intentioned.   I just call them the zinc triad. --Smokefoot (talk) 13:48, 14 September 2013 (UTC)
 * You've got a nice pun there. :-) But is there any other term that would work here? "Post-transition metal" doesn't work because of Al, which isn't after any transition metals. "B group metals" is based on outdated and ambiguous terminology. "p-block metals" not only isn't right here (the Zn group are in the d-block) and is not informative. It tells you what you already can see (if you know what blocks are) or gives you no clue as to their more significant common properties. Same for the related "s-block metals", "d-block metals", "f-block metals". Double sharp (talk) 14:07, 14 September 2013 (UTC)
 * No useful insights from me into the naming issue aside from "zinc triad" Colleagues on this page argue through these things pretty carefully, so if you end up with poor metal, so be it.  I might be out of it, only a few years ago did I hear the term "tetrel" (which now redirects to carbon group]]). --Smokefoot (talk) 15:33, 14 September 2013 (UTC)
 * The literature is tangled. There is no widely accepted term for the well-established concept of a bunch of second-string metals between the transition metals and the metalloids. The closest is possibly 'B-subgroup metals'. But that would compound the situation given, as Double sharp has noted: (a) American/European differences in the use of the A/B nomenclature; (b) the fact that it's a redundant term within IUPAC's Group 1‒18 labelling scheme; and (c) it clashes with the type a/borderline/type b classification scheme for the behaviour of metal ions.


 * The next best label, given the problems with the alternatives (chemically weak metals; metametals; post-transition metals; semimetals) would probably be poor metals. Here's an example from the chemistry literature (Rosca et al. 2009):
 * 'De Vooys et al. studied ammonia oxidation on a series of transition and coinage metals. The coinage metals copper, silver, and gold showed no activity for ammonia oxidation, due to their low dehydrogenation capacity (p. 2235).


 * 'Besides transition and coinage metals, there have been a few studies of nitrate reduction in acidic media on “poor metals” such as mercury, indium, cadmium, and tin (p. 2236).'


 * Now, usage of the term poor metals in the chemistry literature isn't consistent either. For example (Hill and Holman 2000, p. 40):
 * 'The term poor metals is not widely used, but it is a useful description for several metals including tin, lead and bismuth. These metals fall in a triangular block of the periodic table to the right of the transition metals.'


 * In this case, Hill and Holman treat the transition metals as being Groups 3–12, hence the Group 12 metals are treated as TMs rather than PMs.


 * But at least the term is used in chemistry, it is usefully descriptive, and there are good materials science and chemistry grounds, as pretty consistently recorded in the literature, for collectively distinguishing between the Group 11 metals as e.g. transition metals, or coinage metals, or noble metals, and the rest of the metals in question as poor metals.


 * Hill G & Holman J 2000, Chemistry in context, 5th ed., Nelson Thornes, Cheltenham, ISBN 0174483074
 * Rosca V, Duca M, de Groot MT, Koper MT 2009, 'Nitrogen cycle electrocatalysis, Chemical Reviews, vol. 109, pp. 2209‒2244,
 * Sandbh (talk) 00:59, 15 September 2013 (UTC)
 * Supplementary example, for future reference: 'On poor metals such as Zn, Cd, and In, glucose enhances DHMF [2,5-dihydroxymethylfuran] formation; however, its contribution in the presence of Bi, Pb, Sn, and Sb is limited.'
 * Kwon Y, de Jong E, Raoufmoghaddam S & Koper MT 2013, 'Electrocatalytic hydrogenation of 5-hydroxymethylfurfural in the absence and presence of glucose', ChemSusChem, vol. 6, no. 9,
 * Sandbh (talk) 23:41, 24 September 2013 (UTC)

Poor metal article 2.0
I've drafted a proposed version 2.0 of the poor metal article, including Zn, Cd and Hg, in my sandbox. Things to note: Sandbh (talk) 05:46, 12 September 2013 (UTC)
 * commonality of properties
 * aluminium is the most complete entry and includes its 'good' qualities (an approach I intend to take with the other element sketches, where possible)
 * many citations still to follow, including tidying the ones in the notes
 * no pics yet. Sandbh (talk) 07:18, 13 September 2013 (UTC)
 * I note: User:Ds (Ds means Doublesharpium) has made the essential PT template Periodic table (poor metal)/sandbox into a shining clear thing. E.g., it is great in clarifying the "(predicted)" lighter colors (period 7), even without legend. And it is great in showing the environmental elements. -DePiep (talk) 17:44, 15 September 2013 (UTC)
 * Unfortunately, that symbol is taken. :-( Double sharp (talk) 14:45, 23 September 2013 (UTC)
 * I'm pleased to advise that the draft of poor metal 2.0 looks to be launch-ready. Sandbh (talk) 11:58, 26 September 2013 (UTC)
 * That would be great. So the some new topic is resolved, and my 165, 166 quest will be answered & sourced in the articles. -DePiep (talk) 21:30, 26 September 2013 (UTC)
 * 165 and 166 seems to be being dealt with by me currently, since Sandbh has not yet answered me. :-( But given that they seem to behave as Au and Hg's "big brothers" (well, not so much Hg as what Hg would be if it were hypothetically a true transition metal; I would say Cn except that we don't actually know its chemistry), based on Fricke, transition metal is OK. I inserted explanations into alkali metal and period 9 element. I did not put it in alkaline earth metal at this moment because it would result in the unknown elements in the group being described in greater detail than the known ones! Double sharp (talk) 14:23, 27 September 2013 (UTC)
 * Sandbh, are the statements (promises?) by Double sharp OK to go ahead? They are about predicted elements. I must be tough with respected elementary user:Ds. -DePiep (talk) 00:28, 28 September 2013 (UTC)
 * I'm thinking about the classifications of the predicted elements now. Hope to post something shorty. Sandbh (talk) 06:51, 28 September 2013 (UTC)
 * Option 1: I had a brief look through Türler and Pershina (2013). Towards the end of their article they say, 'It was shown that the heaviest elements are basically [italics added] homologs of their lighter congeners in the chemical groups, though their properties may be rather different due to very large relativistic effects on their electron shells.'


 * Their reference to these elements being 'basically homlogs of their ligher congenors' makes me think we should classify them the same as their lighter congenors, unless we are satisfied that the preponderance of their predicted properties falls into a different category. Their reference to 'though their properties may be rather different' made me think about thallium which, despite its similarities to silver and to the alkali metals, we still show as a poor metal.


 * E113, for example, would probably have the physical and chemical properties of a poor metal, plus some transition metal chemistry. The preponderance of properties would lie in the poor metals category. I know this is at variance with the IUPAC definition of transition metal but that definition was never framed with the complications of the super heavy elements in mind.


 * The above represents my early thoughts in this matter. Please free to present more arguments. Sandbh (talk) 14:11, 28 September 2013 (UTC)


 * Türler A & Pershina V 2013, 'Advances in the production and chemistry of the heaviest elements', Chemical Reviews, vol. 113, pp. 1237−1312,
 * re Double sharp, this is my point: Now that Sandbh has made article 2.0 stable (about existing group 12 elements), these predicted elements still creep in as a topic -- but without a stable conclusion. For example, even you write "seems to be ..." above, yesterday. I myself cannot conclude in this, but I just do not want a discussion afterwards. Because that could imply reversals, or partial reversals, or continued disputes, in dozens of pages & graphics. That would be a disaster. So this is why I keep asking for article text (=the sourced base) for the changes w.r.t. predicted elements. -DePiep (talk) 23:41, 28 September 2013 (UTC)
 * Option 2: We've discussed this option before. It was to allow for two-category elements. Nobody who commented on this proposal, apart from me, liked it. It may be worth revisting. A lot of the drama associated with classifying the atypical elements might go away. E113, for example, could be shown as poor metal/transition metal. That would seem to be a concise way of summarising the (predicted-to-be) atypical nature of these elements.


 * Option 3: Hybrid of option 1 and 2. Colour the elements as per option 1. Show more than one category, as per option 2, only in the element's article category box. Sandbh (talk) 01:03, 29 September 2013 (UTC)
 * Vote for option 3. (@DePiep: Yes, I share your desire to have all this sorted out first.)
 * Now here's my detailed Sandbh-style analysis on E113. Fricke's first paper directly compares E113 with Tl and implies that it follows the trend down the boron group: "In the chemical group III the oxidation state is 3, only the valency of T1 is mostly 1 so that Z = 113 is expected to have only the oxidation state 1." His second paper goes into much greater detail: "The chemistry of El13 is expected to be similar to that of the thallous ion although E113+ being a soft Lewis acid should form complexes more easily. The predicted radius of 113+ is 1.48Å, the same as Rb+, and is not much larger than Tl+ itself (1.40Å). The large polarizability and moderately large negative oxidation potential of −0.6V will increase the binding of anions to E113+, but the large radius and the associated electronic repulsion may counteract these effects. Since H2O is a hard solvent, salts of E113 should have low solubilities whereas the nitrate andfluoride should be quite soluble. The E113+ ion may, however, form only a slightly soluble oxide whose solution will be alkaline;(considerable polymerization could be anticipated) and this solution will readily absorb carbon dioxide from the air. Like argentous and aurous oxides, the oxide of E113+ may be soluble in ammonia." But then his first paper goes on to say that Fl(IV) may use d electrons but still act like main group. Based on all this (similarities Tl and alkali metals and transition metals), I think a poor metal classification is warranted, though with an explanatory note in the infobox. (TL;DR: Retracts statement that E113 is transition, except as IUPAC demands, because by IUPAC definition Fl is also a transition metal despite chemical properties not tallying.)
 * E165 and E166: Fricke says "From this side [periodic trends] E165 and E166 will be members of the groups Ia and IIa. From a more chemical point of view, they will be likely more members of the Ib and IIb groups because of the 7d shell which is more comparable to the elements Au and Hg (but also to the elements El19 and E120) as can be seen from Fig. 14. Therefore, higher oxidation states than 1 and 2 might readily occur." But surely periodic trends of ionization potentials have an impact on chemistry! And E119 and E120 can also form higher oxidation states. Changing the classification smacks of over-applying of IUPAC's definition. The similarities to Au and Hg I don't see from his paper; but if we assume he means that E165+ acts like K+ and E1653+ acts like Au3+ or Tl3+, well, I think they could safely be called alkali and alkaline earth metals based on physical properties. Chemically we have a dispute; physical properties are firmly on the side of alkali and alkaline earth metals. (TL;DR: Retracts statement that they are transition.)
 * OK, from looking over this again, I now think the main categories for E113, E165, and E166 should be reverted back to what periodic trends would expect. We can mention other things in the infoboxes or their articles. (Now I think that some of the superheavies should be re-articleized.) For the record, Cn and E164, I think, should stay as transition metals. Double sharp (talk) 06:17, 29 September 2013 (UTC)


 * Option 4. This is the pragmatic IUPAC option. Consistent with the IUPAC definition of a transition metal, silver is a TM as it has (significant) TM chemistry. It is classified as a TM, even though the chemistry of silver in its most common oxidation state of +1 is that of a main-group element. Copper and gold, transition metals both, also have significant MG chemistries. Yet, unless you go with option 2, copper, silver and gold are routinely shown as TMs, and references to their MG chemistry are only given in any accompanying text. So, under this option E113 is a TM, E119 is an alkali metal and E120 is an alkaline earth metal. (Yes, E119 and E120 may have chemistries in higher oxidation states than their group numbers but there is no IUPAC definition that obligates them being classified as anything other than (atypical, admitedly) MG metals. E165 and E 166 = TM, as long as we are satisfed that they are predicted to have significant TM chemistry, and would meet the IUPAC definition: 'An element whose atom has an incomplete d sub-shell, or which can give rise to cations [italics added] with an incomplete d sub-shell.'


 * The only element whose proposed classification I'm not comfortable with is that of E171. I'm not sure how an element with such a high atomic number would be anything other than a metal, at least in terms of its crystalline structure, consistent with the Goldhammer-Herzfield metallicity criterion. The GHMC is a ratio that compares the force holding an individual atom's valence electrons in place with the forces, acting on the same electrons, arising from interactions between the atoms in the solid or liquid element. When the interatomic forces are greater than or equal to the atomic force, valence electron itinerancy is indicated. Metallic behaviour is then predicted. Otherwise nonmetallic behaviour is anticipated. Goldhammer-Herzfield is non-relativistic but I'm not sure what kind of relativistic effect would prevent the onset of metallization in an element with such a high atomic number.


 * E171 (predicted IE 10.2 eV; predicted EA 3.0 eV) may be something like gold (IE 9.22 eV; EA 2.3 eV) which, despite its high IE and EA, has a common oxdiation state of +3 and forms, despite its high EA, very few compounds with an oxidation state of −1. Haire says, 'The common oxidation states of elements 167 to 179 will be 3+ to 6+. Element 171 is expected to have many states from 1− to 7+, as do halogens.' E171 might be a bit like gold, having a common oxidation state somewhere in the range of +3 to +6, and forming some compounds with an oxidation state of −1. I think, given what I would expect would be its metallic structure, this would qualify E171 as a poor metal. If we were proposing to classify E171 as a predicted nonmetal then I think, perhaps, we would need more supporting material. Sandbh (talk) 05:43, 29 September 2013 (UTC)
 * Ah, this is what I was originally thinking. Now E113 would have significant transition metal character in at least the +5 state, though I suppose that the +3 state may well have main group character (cf. Fl having sd hybridization possibly for the +4 state, but no one calls it a transition metal, strengthening my speculation that the d electrons are behaving like the s electrons of the lighter main group elements). Yet the only predicted stable compound where it is in the +5 state is the polyfluoride ion, so I'm not sure if this warrants such a classification. Though it is predicted to have complex (noun) chemistry like Ag so I'm divided on this issue. In addition E165 and E166 are expected to behave as group 13 and 14 metals when they are in the higher +3 and +4 oxidation states, per Haire, and I can reasonably see E119 and E120 acting similarly; Fricke expects a similarity to Au and Hg, though I don't really see it from his diagram of predicted ionization potentials.
 * You make good points regarding E171. My last line of defence is that Fricke predicts H171 to be a hydrogen halide (thus implying that 171 is a halogen), and that 171− would be about as hard a base as Cl− (though, looking at the EA he gives, I have to wonder if he doesn't mean I−). Does H+Au− even form as a stable compound? Is Au− a hard base?
 * As for relativistic effects, I can just about imagine the fact that 8p3/2 and 9p1/2 forming one shell wouldn't act like a split 7p1/2—7p3/2 shell. It is still hard to believe, though, that (as Fricke implies), E167 to E170 are all poor metals but E171 is somehow magically a halogen and hence a diatomic nonmetal! (As you can see, I was unsure of this in the beginning, and not much is required to sway me one way or the other!) Double sharp (talk) 15:39, 29 September 2013 (UTC)
 * P.S. I have to wonder though if the "179" in the Haire quote is not a typo for "170", which would make more sense per the Fricke papers. If that is the case, then the case for a halogenic E171 is strengthened. Double sharp (talk) 13:07, 30 September 2013 (UTC)

Poor idea. The argument for poor metals as a major categorisation of elements that deserves to sit alongside the more widely used, accepted and taught categories of non- metals, metalloids etc. is not made, yet alone the argument to include the grouo 12 metals as well. A much improved version of the poor metal article is in draft and as such will be the most comprehensive literature survey of the previously obscure term "poor metals". The periodic table can be sliced and diced in many ways, often ending up with fuzzy sets of elements where there will be debate and disagreement about the members at the set boundaries. Poor metals is one of these fuzzy groups, and it was hardly known until wikipedia stepped in. In fact reading the poor metals draft has convinced me that "poor metals" actually sits alongside coinage metals and the platinum group etc. fairly comfortably, and that is a good reason to include it in a "metal categories" version of the periodic table. I would prefer to remove the poor metals category completely from the "standard" periodic table, on the basis that the category is ill-defined and inclusion is confusing to the target audience. It is already there in the current wikipedia standard table, and that is bad enough, but taking in group 12? For me that is one step too far. Axiosaurus (talk) 08:09, 30 September 2013 (UTC)
 * But what do we replace it with? Post-transition doesn't make sense because of Al, B-subgroup metals doesn't make sense because of group 11, chemically weak metals doesn't make sense because it leaves out everything that isn't right next to the metalloid line, p-block metals doesn't make sense because group 12 isn't in the p-block, etc. Poor metal is not the best term, but the fact is that there really isn't a good name for these metals that is reasonably descriptive and used at all in chemistry. The notion that there is distinct category of elements between the transition metals and the metalloids isn't disputed; the dispute is what to call it. Given the problems with the alternatives, "poor metals" seems to me to be the best solution, though it is of course not ideal.
 * (Also, we do call it a loose category, and note that the terminology used varies greatly. But at least "poor metal" is descriptive in that these are actually physically and chemically poor metals, and have reasonably well-defined distinguishing properties of having high electronegativities and low melting and boiling points.)
 * As for group 12, that is only because they can hardly be called transition metals, and this is the only category that makes sense in our current scheme. Double sharp (talk) 11:02, 30 September 2013 (UTC)
 * Firstly once you color in the boxes the definition ceases to be loose, typical readers won't go and look at the definition- only sad old gits like me do that. Poor metals is a wikipedia construct, its verging on OR. Secondly the group 12 are transition metals on the edge of their set, they are just more atypical than a lot of the others. My advice, although I expect no-one to take it is leave the group 12 in transition metals, and call the others p block, making the wikipedia "standard" look more like the versions in text books and published by IUPAC. Make a new version if you want of the periodic table to highlight the different categories of metals- showing coinage, platinum - that would be useful.Axiosaurus (talk) 13:16, 30 September 2013 (UTC)
 * Well, if they don't know what "poor metals" are, they will certainly check the definition! (And if they do know what we are using it for, then they don't need to check it.)
 * Group 12 is physically and chemically overwhelmingly not akin to the transition metals. They have far more in common with the metals of the p-block than the metals of the d-block. The availability of d electrons in group 11 but not group 12 for bonding makes a great deal of difference, such as heavily reducing metallic character (e.g. all of them tend to form covalent rather than ionic compounds, especially Hg) – this transition metal–poor metal divide is the main basis for the use of the adjective "poor" in the category "poor metals". The only real similarity is complex formation. These points have all been noted in reliable sources, per Sandbh's first comment in this thread. It is also quite telling, in my opinion, that the supposed "edge, but still transition metal" area group 12 is in doesn't seem to exist on the left side of the transition metals if we take the groups 3–12 definition. On the whole, I think putting group 12 with the transition metals, while giving rise to an aesthetically pleasing and symmetrical periodic table, is undesirable because it gives an inaccurate impression of their behaviour. Textbooks are split approximately 50–50 on the issue anyway. (Does the official IUPAC table actually colour any groups other than the lanthanides and actinides?) Double sharp (talk) 12:54, 1 October 2013 (UTC)

Group 12 elements into poor metals: consensus roundup

 * This is the situation as of 7 Oct 2013, 14:00 UCT. Following discussion above, statements may change. Individual edits are not signed (see page history)

This is a description of the proposed statements, as of this moment. Before consensus is concluded above, this section is not to be used. It is provided here to check and to prepare necessary edits. This section does not discuss or reason the scientific background. it just describes outcomes. -DePiep (talk) 13:12, 13 September 2013 (UTC) It is just a title. Not a factual description. It does not describe exactly all conclusions (for example, elements in period 7, 8 change otherwise).
 * Group 12 elements into poor metals
 * Desription and references
 * The new situation is described in the concept new poor metal article: see User:Sandbh/sandbox.
 * Copernicium is described in its article.


 * Elements changing into poor metal
 * Zinc, cadmium and mercury are poor metals (not post-transition metals).
 * Period 7 changes
 * Category of 112Cn (group 12!) will be changed into "transition metal (predicted)".
 * Period 8 changes
 * No changes in period 8. Element 164, in group 12, stays "transition metal (predicted)", pale pink.
 * Period 9 changes
 * No changes in period 9.

The two categories names and colors (changed to) are: " " (the grey) and " " (the light pink).
 * Not changed
 * Element 164 is not changed. Stays "transition metal (predicted)".
 * Post-transition metals
 * Post-transition metals will include Zn, Cd, and Hg. Still will not include aluminium.
 * Elements that change
 * Edits todo
 * See Wikipedia talk:WikiProject Elements/Group 12 elements into poor metals: edits todo

If not poor, what then?
Writing in the Journal of Chemical Education about how periodic table groups and subgroups should be designated, Fernelius, Loening & Adams said, 'The situation for Zn, Cd, and Hg is different. They form no compounds having incomplete d orbitals and hence have no transition metal character. Still one wishes to distinguish these from Ca, Sr, Ba in some manner. No one has suggested a good term for this. ' (p. 595) [underline added]

They wrote this in 1971 and the situation hasn't changed much (I'm not sure if they knew that Zn, Cd and Hg were sometimes referred to as the volatile metals.)

It seems to me that we either use an existing term such as poor metals (perhaps the front runner in a relatively mediocre field of 15) or, since the concept is well established but there's no widely used short-hand term for it, we use a descriptive phrase, in accordance with WP:NAD:


 * 'In a few cases, there will be notable topics which are well-documented in reliable sources, but for which no accepted short-hand term exists. It can be tempting to employ a neologism in such a case. Instead, it is preferable to use a title that is a descriptive phrase in plain English if possible, even if this makes for a somewhat long or awkward title.'

The shortest descriptive phrase I can think of is low-melting electronegative metals or, in abbreviated form, LME metals. The low melting part is a no-brainer and has been used previously to refer to to metals in this part of the periodic table. The phrase 'electronegative metals' is reasonably common in the literature, for example:


 * 'On the other hand, polar intermetallics, e.g., BaAl4, involve combinations of active, electropositive metals (Ba) with electronegative metals (Al), which form complex two- or three-dimensional networks that indicate some degree of covalent bonding.' (Miller, Lee & Choe 2002, p. 39)


 * '…the deposition of electronegative metals such as Cr and Zn is hindered by poor current efficiencies…' (Silvester at al. 2008, p. 310)

I like 'low-melting electronegative metals' as a reasonably objective descriptive phrase. The lede for the LME metal article would say:


 * Low-melting electronegative metals is a descriptive phrase for the metallic elements in Groups 12 to 16 of the periodic table. These metals are physically and chemically weak, consistent with their location between the 'true metals'[n1] (to their left) and the metalloids (to their right). Among the metals they are distinguished by having a combination of relatively low melting points (all less than 950 K) and relatively high electronegativity values (all more than 1.6, revised Pauling).'

I could see if could work in the quote by Fernelius, Loening and Adams, and I could add poor metals to the 'Related groups' section, and a footnote reference re 'volatile metals'.

Sandbh (talk) 03:07, 7 October 2013 (UTC)
 * Miller GJ, Lee C & Choe W 2002, 'Structure and bonding around the Zintl border', in G Meyer, D Naumann & L Wesemann (eds), Inorganic chemistry highlights, vol. 1, Wiley-VCH, Weinheim, pp. 21‒54 (39)
 * Silvester at al. 2008, 'Technical aspects', in F Endres, ‎D MacFarlane & A Abbott (eds), Electrodeposition from ionic liquids, Wiley-VCH, Weinheim, pp. 287‒352


 * The only problem with this (as noted below) is that it falls flat for Cn, classifying it as an LMEM, despite conflicting chemical properties... Double sharp (talk) 06:23, 13 October 2013 (UTC)

The placement of hydrogen in the periodic table
Are we still going to keep it above a group of elements it shares very little with? :-P

Given that we probably don't want to show elements twice because it's confusing and breaks Scerri's "periodic law" that elements appear in increasing atomic number (otherwise you get into all sorts of issues as showing the main groups twice above the transition metals with C/Si/Ti and C/Si/Ge etc.), I can see four choices:


 * 1) Status quo: keep H above group 1 (alkali metals). Pro: Has one electron in its valence shell, just like the alkali metals. Can behave like the alkali metals in metallic hydrogen. Quite commonly encountered. Con: At STP, H is very different from the alkali metals. Its first ionization energy is very much higher than those of the metals. Hydrogen always bonds covalently and unlike the alkali metals, is never found as a free H+ ion (hydron) except in a vacuum. The first "pro" argument listed makes as much sense as putting helium with the alkaline earth metals due to its having two electrons in its valence shell.
 * 2) Move H to above group 17 (halogens). Pro: Has one electron missing from its valence shell, just like the halogens. Hydrides behave similarly to halides. Con: H does not fit the trends down from F very well. It also does not share the behaviour of the heavier halogens past F (forming oxoanions), never being in an oxidation state that isn't −1, 0, or +1. In contrast to F its +1 state is stable.
 * 3) Move H to above group 14 (crystallogens). Pro: Has half-filled valence shell, just like the crystallogens. H–H, C–H, and Si–H bonds show clear familial relationships. Hydrides show strong links to the methides. H and C form oxonium ions. Makes smooth trends in electronegativity, electron affinity, and first ionization energies. Con: H has many other distinctive properties that this placement cannot explain (hence the arguments for group 1 and group 17 placement).
 * 4) Leave H floating in the middle. Pro: H's properties are very anomalous and it does not fit very well in any group. Also commonly encountered. Con: The periodic law should apply to all elements. This placement seems to exempt H from it.

What do you think? (I like #3, personally.) Double sharp (talk) 04:13, 16 October 2013 (UTC)


 * Cronyn's paper Double sharp (talk) 04:15, 16 October 2013 (UTC)


 * I vote (1) and you've left out a benefit. Readers are used to it that way.208.44.87.91 (talk) 04:53, 16 October 2013 (UTC)
 * Added. (It does apply to (4) also, though to a lesser extent.) Double sharp (talk) 05:37, 16 October 2013 (UTC)
 * Hmm. There are many arguments in the literature for either group 1 or 17. I marginally favour group 1 over group 17. For a while, having done a fair a bit of research on this question, I favoured group 17 however (as I recall) I went back to group 1 when I read that hydrogen can form metallic alloys with some metals, hence behaving as a metal. Does it have some properties that are different from the alkali metals? For sure, but that is a "super" first-row-element-anomaly, hydrogen being not only the first element in its row but the first element in the whole PT. Hydrogen over lithium is a bit like boron over aluminium except more pronounced. I can understand helium over beryllium from the pov of esoteric electron configuration arguments (which I tend not to support the validity of) however, in any event, I think this is a step too far---pragmatically speaking---and that helium over neon is a better fit. Sandbh (talk) 06:24, 16 October 2013 (UTC)
 * Irritatingly the hydrogen article doesn't mention the history of its placing in the periodic table, though alkali metal does. Double sharp (talk) 09:18, 16 October 2013 (UTC)
 * I remember reading that early chemists thought hydrogen would become a metal if it could ever be frozen. That didn't happen, of course, but it shows how much hydrogen was regarded as having metallic character, back in those days, and explains why it ended up aligned with Group IA/IB. Sandbh (talk) 01:45, 18 October 2013 (UTC)
 * I dunno, but to me speculation of early chemists that turned out to be wrong is not really a good argument. If this was the sole reason why H ended up in group 1, I'd advocate changing it. I'm torn about the placement of H in the periodic table precisely because this is not the sole reason why H ended up in group 1, and there are many good arguments on each side of the debate.
 * As for those metallic alloys with H: are these stable at STP? Because for example at 200 TPa everything is a metal, even the noble gases. And is H the primary component? Otherwise it's not very impressive given alloys of B and C with metals, most famously Fe.
 * I wouldn't call H over Li anything like B over Al. The latter is just a case of how period 2 elements (most significantly from Be to O) behave significantly differently from the corresponding period 3 elements (most significantly from Mg to S, though Li/Na, F/Cl and Ne/Ar have less striking effects too) – also the reason why "pnictogen" and "chalcogen" are commonly used to exclude N and O from their groups, though "halogen" isn't used in this restricted sense. But H over Li is very much more drastic as the trend is really not smooth and the anomalous properties are not easily explainable from trends; unfortunately H over F and H over C have this problem too. Also note that the restricted use of "pnictogen" and "chalcogen" to exclude N and O respectively is quite common, but you will almost never see "alkali metal", "halogen" or "group 14 element" used to include H.
 * (Disclaimer: I have not given this as much thought as I have done to the other important periodic table issues.) Double sharp (talk) 04:17, 24 October 2013 (UTC)

Even though they were wrong, the early chemists thought hydrogen might be a metallic gas based on the properties of hydrogen that they were able to discern at that time. I think this is still a pertinent consideration. Nothing they saw in those days has changed.

H has stayed in Group 1 because nobody has presented a sufficiently strong argument to warrant moving it, since it was first put there. I'm travelling at the moment and don't have access to much. Ferro and Saccone (2008, p. 326) note that hydrogen forms a series of metallic phases/solid solutions with f- and d-block elements, especially where the ratio of metal to hydrogen is around 1:2. This page also has a reference (reliability uncertain) to the alloy forming ability of hydrogen (mind you, I'm no expert when it comes to intermetallic chemistry). My comment about H as an example of a super-first row anomaly stands. In the same way that B is such a contrast to Al, H is even more of a contrast to Li. Excuse my ignorance but, on balance, doesn't H in its reactions seem to act more like a weak metal, rather than as a nonmetal (+1 oxidation state being its most stable oxidation state)?

Agree your comments about names of groups hence the elements in column 1 are sometimes called 'Hydrogen and the alkali metals'.

What did Mendeleev say about H? Sandbh (talk) 22:47, 24 October 2013 (UTC)


 * Ferro R & Saccone A 2008, Intermetallic chemistry, Elsevier, Oxford
 * Mendeleev just put H in its own period, in group 1. Are there any quotable quotes from him about H? I don't know of any.
 * What I'd be really interested to know is how the alkali metals behave in the gaseous state (e.g. Li2), and a comparison of that to H2. Double sharp (talk) 06:01, 25 October 2013 (UTC)


 * I'm for option 1), given that's where people, myself included, would expect it. If you argue that H can't be placed in group 1 because it's properties are too different from e.g. Li, or Cs, one could easily argue that N is too different from Bi to be in the same group. Headbomb {talk / contribs / physics / books} 01:37, 25 October 2013 (UTC)
 * Hmm. Having most common oxidation states +5, +3, and −3, N must really go in group 15 (same valence). H's −1 state is however not too uncommon, though it's not exactly stable; in contrast, Li's is unknown, and that for Na–Cs is very unstable and has to be aggressively stabilized by shielding the anions with bulky ligands IIRC.
 * But you have a good point, and this is why I'm very hesitant about moving H, despite being the one who opened this for discussion here. Double sharp (talk) 06:01, 25 October 2013 (UTC)
 * As far as I'm concerned, this really is not that big an issue. In our tables, H is colored differently than the rest of the alkaline earth metals, so that's pretty much all the warning someone needs to see to known that something's different about this one. But the real nail in the coffin is that Wikipedia should follow what sources do, and sources overwhelmingly favours placing H in group 1 (just do a search for "periodic table" on google images). The place to discuss the classification of hydrogen in the periodic table is, of course, in our article on the periodic table should mention that classifying H in group 1 is somewhat subjective, and that other people sometimes put it in group 17 (halogens). And then mention that other people consider it too different from both group 1 or group 17 elements, and put it in a standalone group. Our articles on group 1 and group 17 could also mention it, but I personally would leave this discussion out of the hydrogen article.
 * Count me in for number one, please. My reasoning has already been mentioned by others entirely.--R8R Gtrs (talk) 20:25, 25 October 2013 (UTC)

In his self-published book, New ideas in chemistry from fresh energy for the periodic law, Henry Bent (2006, p. 173) wrote that the PT first and foremost, at the outset, was a table of maximum states of oxidation, regardless whether or not elements with the same maximum oxidation states were metals or nonmetals. Sandbh (talk) 00:58, 26 October 2013 (UTC)

Signpost Report
The WikiProject Report would like to focus on WikiProject Elements for a Signpost article. This would be the second time. June 2011 was the first Signpost report.--Stone (talk) 12:28, 4 October 2013 (UTC)
 * I haven't posted here for a while, but I've been following some of the discussion, and would be willing to participate. StringTheory11 (t • c) 18:43, 6 October 2013 (UTC)

Fluorine peer review
Please review and fix "Fluorine". If it's too long, just hit a section.-TCO

Kohonen periodic table map


This is a Kohonen neural network (KN) map of 59 of the elements, as published by Chen (2010), but with colours added by me; and thicker lines added around the ferromagnetic metals (at 3,2) and the coinage metals.. Chen's highly condensed 2-page article did his map a great disservice. I remember reading the article when it came out, and neither understanding nor learning anything from it. However, the complete manuscript (22 pages, double spaced) is available as supplementary information from the Journal of Chemical Education page for the article in question.

Chen (or rather the 'trained' Kohonen network) produced the map using 10 physiochemical properties (atomic mass; minimum and maximum oxidation states; atomic radius; EN; state of matter; MP; BP; heat of atomization; and IE). Chen says the map has three major clusters: metals (top half); nonmetals (lower part; with gases lower-right); and semimetals in-between. He notes the network correctly picked the diagonal relationships between Li, Mg and Be, Al; the first row anomaly in Be, B, N, O and F; and the special nature of H. He also talks about some sub-clusters: highly reactive metals; less reactive metals plus Ge; alkali metals; transition metals; the iron group metals (Fe, Co, Ni); the noble gases; and the other nonmetals and arsenic.

Reading all this prompted me to immediately colour his map using our colours. The result speaks for itself.

Chen says the advantages of a KN are that (unlike consensus trees and dendograms) the results are very easy to map and interpret and that (if I understand correctly) the results are not sensitive to properties having a high redundancy with respect to other properties.

He concludes by saying that the KN approach can help students gain a more sophisticated understanding of the relationships among and between the elements compared to what can be gleaned from building a traditional periodic table.

Anyway, I thought his table was pretty cool. I was particularly interested to see where Ge ended up, noting it is sometimes regarded as a metal, due to falling on the wrong side of the metal-nonmetal dividing line. I was also interested to see the coinage metal cluster; and the poor metals cluster (and a few other things). Starting with the alkali metals, and going anticlockwise, see how the alkaline earth metals go next, then the poor metals; and then the transition metals before the metalloids; polyatomic metals; diatomics and noble gases. It's like the metals progress from main group metals (inc. the poor metals) to TMs, before running into the metalloids, rather than what happens on the conventional periodic table (s-block metals -> TMs -> poor metals, then metalloids etc).

Sandbh (talk) 11:27, 7 October 2013 (UTC)
 * Chen DZ 2010, 'A new method for studying the periodic system based on a Kohonen neural network', Journal of Chemical Education, vol. 87, no. 4,
 * It seems as though we are doing something right! :-) Though I would have loved to see where the lanthanides (and maybe actinides) would end up. That would give us a good idea on what to do about group 3. Double sharp (talk) 13:55, 7 October 2013 (UTC)
 * Think it should read "polyatomic nonmetals" and "diatomic nonmetals". YBG (talk) 22:07, 14 October 2013 (UTC)
 * Indeed it should, and I'm now wondering how we all missed that! Double sharp (talk) 10:47, 20 October 2013 (UTC)

Is there a pdf or djvu or whatever so I could at least this table? Also, why just 59 elements? Where would thallium and bismuth go? Can this theory be extended to superheavies like element 118?

If this can be worked out, the table could be interesting.--R8R Gtrs (talk) 19:48, 25 October 2013 (UTC)
 * R8R, in the previous paragraph, what word did you mean to have between 'at least' and 'this table'? Sandbh (talk) 11:25, 29 October 2013 (UTC)


 * I think it could be "read," but I don't think it sounds good that way.


 * In either case, you said that 2-page paper was a bad deal for this table, and the 22-page one explained it a lot better. I don't see a point of this table even after your intro text, just groups of elements in a checkerboard-like table distantly resembling the way they are in real life because they were meant to. I think I could change my opinion (it's not really an opinion now, more like of an impression) after reading a whole version of the 22-page paper, as I would understand it better.--R8R Gtrs (talk) 18:04, 29 October 2013 (UTC)

Has anyone ever seen a picture of promethium?
Only promethium is missing. Has anyone ever found one or made one? Then that sector of our project can move towards the stage of getting better pictures for every element instead of completing the periodic table.

(P.S. I'd be interested in some more Pm compounds or Pm in aqueous solution also.) Double sharp (talk) 06:20, 13 October 2013 (UTC)
 * Is there a pic of 61 in the 60's Time-Life bookie on the elements? Sandbh (talk) 08:31, 16 October 2013 (UTC)
 * Yes, but it shows Pm2O3, not Pm metal. Double sharp (talk) 09:06, 16 October 2013 (UTC)

Sourcing data in the infobox
A reader was looking for a data source (boiling point of boron) and could not find it. As we know, data is linked via the ref link that points to the data overviews page, so there you have to look for "boiling point (data)" (not boiling point), then scroll to "boron" and understand that WEL is code for the source used (not the other codes CRC or LNG), which is specified -- without a ref link -- below the table. Maybe the user is right is asking for improvement.

To improve, I cannot escape the consequence that there will be more linked text in the infobox, one way or another. First proposal (note that the link is anchored to the boron row in the data page): Example 2: -DePiep (talk) 10:47, 6 October 2013 (UTC) add example 2 -DePiep (talk) 10:52, 6 October 2013 (UTC)
 * Added: square brackets around " source " -DePiep (talk) 18:32, 6 October 2013 (UTC)
 * What do you think of the system we use at superheavy elements (e.g. Template:Infobox ununtrium), where none of the info is in these data pages? Because then we can have a link to the direct source there, and if you really want a source comparison, the "ref" link still exists if you want one (at the data pages). Double sharp (talk) 12:15, 6 October 2013 (UTC)
 * re About your 2nd half: I do not like the four-plus-step ref link we now provide (read my intro story above. Comical and true). It was a good setup in 2005, but we can do better sourcing & linking today. Below I describe a 2-ref solution that solves the general linking and the e113 incidental ref link. I think is does what you ask for. -DePiep (talk) 19:20, 6 October 2013 (UTC)
 * Right. So there is also the need for a local, on-page reference (that will look like ). These two refs are independent! They do not interfere, they don't know about one another. That makes it easier for us.
 * A. The [source] link will be added automatically by the template infobox element, and we'll have to add in code: " only do not show when atomic number >96". Can be done. This is a non-regular reference (it links to another page).
 * Then each individual infobox (like infobox ununtrium) should be able to show an individual reference on-page (it should look like ). This is a regular reference, as we know it.
 * B. About the individual, regular refs we talked here some months ago Wikipedia_talk:WikiProject_Elements/Archive_14. I researched it back then, to position the reference from current " " to " ". Well, it is not possible to "take out the reference link code from boiling point K input, and paste it at the end (after the K)". Not in a stable way. So we should introduce an infobox parameter boiling point ref. Then infobox ununtrium can have  boiling point ref. The parameter puts the ref nicely after the K unit, not before as is today.
 * Questions left:
 * 1. Is the linktext [source] good enough? (shorter? there will be dozens in an infobox)
 * 2. Can the local reference be once at the end (after all K, C, F values), or should it be after K? (so with extra options 2. after C, 3. after F). I propose one at the end.
 * 3.While we are at it, what to do with the addition like "(predicted)"? Introduce boiling point comment? Or type that into the proposed new "ref" parameter?
 * Examples 3A - 3E:


 * New parameters, proposed:
 * boiling point ref
 * boiling point comment


 * -DePiep (talk) 19:11, 6 October 2013 (UTC) refined -DePiep (talk) 20:03, 6 October 2013 (UTC)
 * Support 3B and 3D. Double sharp (talk) 06:49, 7 October 2013 (UTC)
 * OK. You think we should use the second parameter for the comment, or add the text to the new ref parameter? Of course this change will be added to all infobox entries. -DePiep (talk) 08:18, 7 October 2013 (UTC)
 * Up for improvement, but I plan to use [ref] not [source] for the wikilink (to the point, shorter, and it catches the plural too). Also, the on-page links like [1] will be placed after the general ref link. -DePiep (talk) 19:14, 12 October 2013 (UTC)

Weird that none has come up with this: --R8R Gtrs (talk) 20:57, 12 October 2013 (UTC)
 * Glad to see you proposal, but really why a basic reference like [1] in a separate column? -DePiep (talk)


 * Because infoboxes aren't text. They are made so people can easily scan information (figures) from it. [1] is a visual distraction, especially when not at the end of the line, but even when it is, it remains a distraction. Giving it a separate column allows faster info scanning. If you don't believe me, you draw a table of, say, 10 rows, in both version and see which one is easier to scan. Even the looks are still OK (maybe a bit unusual, but not too unusual). Most people don't need refs, they're okay with the fact they exist; and when someone needs a ref, it is still easy to find, in the same row (and way better than now).--R8R Gtrs (talk) 22:04, 12 October 2013 (UTC)
 * My list of responses. Mostly these are independent issues (that is, topic A can be treated independently of topic B; that makes it easier for us)
 * infobox, not text: indeed. Though, in text the regular wiki ref[1] is as distracting as in the infobox (in printed books footnotes do not distract me &mdash;o holy print world&mdash; but this is internet).
 * Scan a table. Tables scan better your way indeed, sure. My first thought was: better columnize the K/C/F values then! Just that is what readers scan (look for, are familiar with).
 * Wider. It will make the infobox wider. Maybe a wider box is justified, but for this reason? Deviate from standard ref rule ("ref right after the punctuation")? My preference is that regular wiki format should prevail. Exactly that is what helps readability. Ref link in an extra column is irregular. I say: no exception for elements of for an infobox.
 * X[ref]: the original question was: how to ref to the true source page. This ref link is needed in most element infoboxes. It will wide the box always. I example:


 * You read a text. You scan a table. That is why [1] is okay in a text... just for starters, it occurs where a text is supposed to break, near punctuation, which invented for that very reason, in 99.99% events. I guess we just have to handle those 0.01%. A table is not supposed to break, it is supposed to be as clear as possible.
 * You can't columnize temperatures. There are just two in a usual infobox (they also look columnized in most cases, if it matters).
 * Not having these refs at all is better for readability still, less distraction still. Again, most readers don't need sources, just the figures, so figures should have the no 1 priority, not refs. Note: I have seen this formatting in actual handbooks.
 * I would solve the problem with [ref] in first place by replacing the ref wikilink at the bottom of an infobox with a complete phrase. "Complete references list," for example, or whatever. I don't see why this shouldn't be done. --R8R Gtrs (talk) 23:28, 13 October 2013 (UTC)

Separate column gets my vote. It encourages having sources, makes checking the sources easier, etc. It's still relatively mischief safe since we use the transcluded templates, but it shows better sourcing in article for the verifiability sticklers. It also reads better with all the numeric text. I'm sure there will be some very rare occasion when there are a huge number of refs, but that can be handled by bundling refs into a note. (Anticipating another kvetch), and yes you may have situations when there are two facts and ref A is not right next to fact A, but who cares...it's still verifiable and easily used. In the real world, I look up citations and then figure out what they apply to.71.127.137.171 (talk) 00:12, 13 October 2013 (UTC)
 * How would the sourcing be improved by using an extra column? Remember, today the source is not linked at all (see my introduction above: five clicks away, when knowingly thinking). Please explain why examples 3B versus 3I improves sourcing. The second half of your post I do not understand (kveth?). -DePiep (talk) 23:00, 13 October 2013 (UTC)

TCO response

 * If we don't go with a column, then I think the first proposal is better as less obtrusive. I just think columns look cleaner, but not a huge deal.  See State reptile table for an example of ref column in a table.  Yes, refs are slightly annoying in running text, but really only slightly.  In a table, especially a chemical infobox, info density and number density much higher, so a separating column is more ergonomic.  But the first proposal is an improvement over second in viewability.


 * I think having the table will make it more likely to get refs in ("don't leave it empty"). As it is now, some articles (F, Cf) have refs in and others don't and even F and Cf are not perfect.  A table would drive better practice.  Like a kanban square.


 * I'm not 100% clear your scheme, but would love it if the references show in article (how fluorine does now). With yours are they click through to another page or at bottom of element or both?  I prefer if they are a bit more prominent in article as they will be more likely to be used (even for reasons other than verification, just usage).  Also, they will get checked more if at least duplicated into article.  And thus improved.  We've had some mistakes in the 2005 stuff and more eyes drives better practice.  I still think our scheme of the transcluded templates is enough to retard mischief.


 * In any case, versus the status quo (off in Mat Sci's tables from 2005), I prefer either of yours.


 * I'm not too bothered by a little wider infobox. If it makes the dagger go a little shallower (strip on the right not as long) that is good.  In some cases, the squeezing down of text is beneficial.  In a few others (images on left) it is not.  But I usually do centered images if there is an infobox clash anyway.  See Painted turtle for a little wider infobox.

In summary: columns>option1>option2>status quo. ;-)

71.127.137.171 (talk) 23:34, 13 October 2013 (UTC) (TCO)
 * note: First examples now are numbered too for easiness. I understand "option 1" = 1A, and "option 2" = 1B. -DePiep (talk) 07:51, 27 October 2013 (UTC)
 * re in general: I am pondering this third-column idea, and its consequences.
 * re If we don't go with a column, [use A1] (reflink next to "b.p." text, so in lefthand column). It keeps the infobox clean indeed, not so much mixup of texts. Still, it would defy the purpose of a ref: next to the statement (as we do in running text, so as expected). But this is just an "if", let's try out that 3rd column first. So if 3rd column does not work out for some reason, we'll revisit this option.
 * re having the table will make it more likely to get refs [added]. Agree... but. 1. the intention is to add the referenced automated (build in in the ibox), for all available data. So there will be a ref often. We have ~20 data pages, so there could be ~20 automated links (this b.p. one is our trial of course). Now for example fluorine (big ibox, and also a popular page: as most old elements) has ~30 sourceable entries, plus some rows that won't need a source. So a number of ref cells will stay empty (idle), while still inviting as you say. I find this a bit of a drawback in the visual appearance.
 * re your scheme (where will the ref actually appear: on page or on datapage? Any ref grouping maybe?). Good point. My proposals now link to the datapage while going to the right row (like a section link). Try the boron example [source] above. That is now.
 * Indeed we should explore the idea to add the source to the page (as a regular ref), still automated. That would be Good Encyclopediing! First thoughts:
 * Nice: Fl has 12 refs (re)used 25 times! Also there is repetition over the elements (CRC, Webelements.com). So such repetition is easy & usefull to automate in the iboxes.
 * Now b.p. is an easy example: one size fits all elements &lt;E97. I expect less simple refs in other ibox data. That is: element-dependent differences. When the ref is element-dependent (say, a different source page number) it is not easy or even useless to automate that in the infobox. Automation does not like exceptions.
 * I'll work on this for b.p. examples. Do you have a suggestion for a second dataref to try? Something more complicated?
 * Of course, individual ref adding (in one ibox) will always stay possible.
 * And, when all refs are in, the "ref" link at the bottom should go.
 * re versus the status quo [2005] - me too. If we can make it work.
 * re a little wider infobox [is OK] - me too. And reducing fontsize is also in my mind. I will propose something in this shortly, but first I'll see what the content-discussions (4 images) will give. Mixing those two change topics destroys process overview (both in talking and in editing). This topic could appear elsewhere too, outside of this datraref thread, but the third column is related to ibox width.
 * re In summary. As said, we'll explore the third-column here in trial.
 * Not addressed: visual effects. If we add a 3rd column, it will add a cell separator line (as the other table cells have). This may be nice for ref links, but visually ugly for empty cells. The eye perceives this as an omission first thing always, triggering a mental wondering; all subconscious and difficult to reason out because we are not there to explain it. We'll have to keep those effects in mind (you mentioned others). I still want to compare the effects with regular ref adding: right after the fact as we have, inline.
 * Concluding, todo: 1. Build 3rd column examples; 2. pursue automated on-page refs. 3. keep ibox (font) resizing in mind.

-DePiep (talk) 09:45, 27 October 2013 (UTC)
 * Somehow I suffer a lack of inspiration to do something for someone else. -DePiep (talk) 10:00, 12 November 2013 (UTC)

Quantum graphics
The subpage discusses how to represent electron shells bin a better way (not just circles). Latest suggestions are:




 * Moved to subpage. -DePiep (talk) 21:12, 11 October 2013 (UTC)
 * I should spend more time with this baby. -DePiep (talk) 19:03, 13 November 2013 (UTC)

Option 17
This option covers some old ground, re groups 1 and 2, which I know at least R8R Gtrs, Double sharp and Nergaal may be familiar with. I'm posting this time partially in light of the comments by Axiosaurus and mainly because option 17 seems like a natural evolution from what has been covered so far. Feel free to use my own arguments against me.

(a) Let groups 1, 2 and Al be pre-transition metals. [Al will be in good company with Be, as will Li with Mg.] (b) Let Sc, Y and the lanthanides be rare earth metals [Nice progression from pre-transition metals to rare earths] (c) Let Zn, Cd, Hg; Ga, In, Tl; Sn, Pb; Bi and Po be post-transition metals [Nice progression from transition metals to post-TMs] All these category names are found in the literature:
 * 'Aluminium and the elements of groups 1 and 2 are classed as pre-transition metals…' (Cox 2012, p. 188). Cox also discusses the properties of the pre-transition metals as a whole, as did Deming (1940, pp. 650‒672), the guy who popularized the medium-long form of the periodic table (except he called them light metals). Deming wrestled with Al too. Originally (1923, p. 163), he grouped Al with the heavy metals (= transition metals + post-transition metals).


 * 'Pre-transition-metal oxides (e.g. MgO, Al2O3, etc.) usually are good insulators and inert to redox gas molecules.' (Wang & Gouma 2012, p. 169)


 * 'In this scheme, the alkaline-earth metals, Ca, Sr, and Ba, may be regarded as "pre-transition metals", and the noble metals, Cu, Ag, and Au, as "post-transition metals".' (Collings 1984, p. 46)


 * 'Hopefully, these four volumes, and any which may follow will make a major contribution to our progress in understanding these exotic and fascinating elements. In writing these chapters the authors have been asked to use the term "rare earhs" to include Sc, Y and the elements La through Lu, and the tern "lanthanides" when referring to only the elements La through Lu.' (Gschneidner & Eyring 1979, p. vi)


 * 'This concept will be considered in Chapter 2, but suffice it to say here that complexes of the pretransition metals (Na+, K+, Ca2+, Mg2+, Ba2+, Al3+) are held together by electrostatic forces…' (Eichorn 1973, p. 4)


 * 'It is convenient to deal with the subject by considering the derivatives in turn of the three main types of metal: (a) Light and Pre-transition Metals (Li, Mg, Al etc.); (b) The Post-transition Metals (Zn, Sn, Pb etc.); (c) The Transition Metals (Ni, Pd etc... ' (Nyholm 1970, p. 35)

This option addresses a number of discussion points and would also reduce the number of categories by one. Of course, we would still have articles about the alkali metals, and the alkaline earth metals, and the group names would still be shown on DePiep's showcase periodic table. We would need a new article for the pre-transition metals, however.

Sandbh (talk)
 * Collings EW 1984, The physical metallurgy of titanium alloys, American Society for Metals, Metals Park, Ohio
 * Cox PA 2004, Inorganic chemistry, 2nd ed., Instant notes series, Bios Scientific, London
 * Deming HG 1923, General chemistry: An elementary survey, John Wiley & Sons, New York
 * —— 1940, Fundamental chemistry, John Wiley & Sons, New York
 * Eichorn GL 1973, Inorganic biochemistry, vol. 1, Elsevier, Amsterdam
 * Gschneidner KA & Eyring L (eds) 1979, ''Handbook on the Physics and Chemistry of Rare Earths: Nonmetallic compounds II", North-Holland Publishing Company, Amsterdam
 * Nyholm R 1970, ‘Some advances in organo-metallic chemistry and their industrial application, Journal of the Royal Society of Arts, vol. 119, no. 5173, pp. 28–43
 * Wang L & Gouma P 2012, 'Selective crystal structure syntheseis and sensing dependencies', in MA Carpenter, S Mathur & A Kolmakov (eds) 2012, Metal oxide nanomaterials for chemical sensors, Springer Science+Business Media, New York, pp. 167‒188


 * I have re-thought about this in my process of reconsidering your older ideas. (Why the jump from 10 to 17, by the way? Were there previous versions of this?)
 * On pre-transition metals: characteristic properties please? If we have reasonable ones like we already have for the poor metals I may be convinced. I trust an article is possible without always breaking them into groups.
 * Your graphic in your sandbox with EN and m.p. values shows considerable kinship of Al with the post-transition metals (now I can't use the acronym PTM anymore...); are we considering the wrong properties here? (Be and Mg don't show up with Ca, Sr, Ba, and Ra on this graph either.)
 * The diagonal relationships are more major only for the first few periods, aren't they? But are there significant differences between group 1 and group 2 aside from typical oxidation state? Correcting for that, don't they behave quite similarly, with some exceptions like Li, Be, Mg, Al (diagonal relationship territory, but they don't fit anywhere else and fit reasonably where they are?)
 * Support, though asking for further info and a definition by physical and chemical characteristics of the pre-transition metals. Double sharp (talk) 12:29, 8 October 2013 (UTC)
 * Am still working on this. Trying to compare Be and Al to see which falls better into which camp (pre- or post-). Tricky. Option 17 came after the fifteen titles set out in my sandbox, plus option 16 above (LME metals). Oh, LME metals won't work, since copernicium, which we agree would be transition metal, would also likely meet the LME criteria (unless its predicted electronegativity is less than 1.6). Sandbh (talk) 13:10, 11 October 2013 (UTC)
 * Argh, Cn, why do you have to be so difficult and throw alphas in the works? :-( Hg's high EN is (at least partially) a relativistic effect, isn't it? If so then yes, since relativistic effects in the 7th period are predicted to be at a maximum at Cn (at least on 7s, which are half the valence electrons anyway), Cn would probably have an even higher electronegativity (though I've never actually seen a value). Rg(−1) is predicted to be possible like Au(−1), which in my mind strengthens the case for an electronegative Cn.
 * Cn: OK that kills LME metals. Sandbh (talk) 04:29, 13 October 2013 (UTC)
 * Ok, so call them low-melting main-group electronegative metals then = LME metals. Sandbh (talk) 11:50, 14 November 2013 (UTC)
 * Yes, Be and Al are tricky. Japanese Wikipedia classifies Be, Mg, and our poor metals (current + zinc group) without Po as "base metals" or "other metals" (they're inconsistent). What is interesting there is their treatment of Be and Mg with the poor metals. I can understand this for Be, given its brittleness, formation of an amphoteric oxide and hydroxide, a strong tendency to bond covalently (because the Be2+ ion would be extremely polarizing), its chloride reacting with water instead of just dissolving (typical of nonmetal oxides; MgCl2 exhibits this very weakly too) and an anomalously high electronegativity (just below the 1.6 cutoff). OTOH it has a high melting point that is quite close to those Mn and the yttrium and terbium group metals actually, and it has a pretty ideal hcp crystal structure (among the poor metals, only Tl has that ideal too). But I sympathize with calling Be a post-transition metal. Mg has far fewer credentials; save for its low melting point, I don't see any strong reasons to follow the Japanese Wikipedia here. (There is a gap between it and Ca–Ra, but in general it's quite an OK simplification to consider Mg–Ra one typical happy group family. Be, on the other hand, is a problem child.)
 * (The Japanese Wikipedia also has Cn as a poor metal; a very generous interpretation is that they are basing it on the expected physical properties, but their article seems to just cover Cn's initial discovery and little else...and to my mind, +4 and 0 being the most stable states and not +2 is the main dealbreaker here.)
 * Japanese Wikipedia: Base metals won't work; other metals is as bad as other nonmetals; agree Be has a bit of "poor" in it but also has reasonable metallic qualities, as you note (also very high strength-to-density ratio, and high specific modulus = aerospace applications). Mg is no poor, although it melts at around the same temperature as Al, and chemically shows a wiff of poor, as you note. Mg is otherwise, however, a chemically strong metal. Sandbh (talk) 04:29, 13 October 2013 (UTC)
 * Indeed, "base metals" is a bad choice given its historical meanings. "Other metals" and "other nonmetals", as we've discussed last year, are not even worthy of consideration. Double sharp (talk) 06:05, 13 October 2013 (UTC)
 * You've explained very well in your sandbox all of Al's poor-metallish properties, so I'll refrain from doing a quick roundup of the main ones here. :-) I'm not going to do pre-transition properties roundups of Be, Mg, and Al until you write up a quick lede paragraph. :-) (Mg got dragged in because of the whole group II bifurcation issue, splitting as it does into groups IIA and IIB. I'm using the old IUPAC nomenclature here because it eliminates the homophony of "II" and "2".)
 * The main issue for me here is that it seems to me that you cannot logically call Be or Al post-transition metals, because they don't succeed any transition metals. Yet if we consider them to pattern with these respective groups, then I think we should classify them as such. (You've already searched for so much alternative nomenclature, and I don't think any of these will work well, because of annoying superheavies like Cn...) Double sharp (talk) 15:53, 11 October 2013 (UTC)
 * Pre-transition metals: Summary of Cox and Deming follows. Pre-transition metals are 'typical metals', very electropositive in character, invariably found in oxidation states expected for ions in noble-gas configuration (Na+, Mg2+, Al3+ etc). Occur in nature widely as silicate minerals, although weathering gives rise to deposits of other compounds: halides, carbonates and hydroxides e.g. AlO(OH). The cations of the pre-transition metals are resistant to oxidation and reduction hence remained undiscovered for many centuries as ancient civilizations made use of Au, Ag, Cu and Fe, and later peoples became familiar with other heavy metals. Post-transition metals are less electropositive; typically found as sulfides. Form compounds with oxidation state = d10 ions (e.g. Cd2+, In3+, Sn4+) but these are less ionic in character than pre-TMs. In solution, post-transition metals form stronger complexes than pre-transition metals; lower oxidation states also common (e.g. Tl+, Sn2+). Post-transition metals often have lower coordination numbers than expected by comparing to pre-TM ions of similar size, and appreciable covalency.


 * Inside the pre-transition metals, three sub-categories can be discerned: (1) Alkali metals; (2) Be, Mg, Al; (3) Rest of Group 2 metals:
 * (1) Alkali metals are completely basic (no amphoterism). They have no structural uses. Their densities, melting and boiling points, and enthalpies of atomization are all lower than the alkaline earths. They all have BCC structures (I assume Fr too), unlike the alkaline earths with HCP (Be, Mg), FCC (Ca, Sr) and BCC (Ba, Ra).
 * (2) Be and Mg are distinct from other alkaline earths as seen in, per your comments, the bifurification phenomenon. The chemistry of Al is very close to the chemistry of Be; one author (House 2013, Inorganic chemistry, p. 351) after discussing the chemistry of Group I and 2 elements, then discusses Be and Al together, due to 'numerous similarities in their properties and reactions.' Be, Mg and Al would be the only pre-transition metals with structural applications.
 * (3) Rest of Group 2 metals (Ca, Sr, Ba; and I presume Ra) are closer to the Group I metals (Greenwood & Earnshaw 1998, p. 113), than Be and Mg. Historically, Be and Mg were not classed as alkaline earth metals (Steele 1977, The metallic elements, p. 34).Sandbh (talk) 04:29, 13 October 2013 (UTC)
 * Fr and Ra are kind of weird, not following the trend down to the 6th period. Their electronegativities are slightly greater than those of the elements above them, and they have some covalent character (e.g. in FrO2 and RaF2). Sadly RaO is uncharacterized for some reason.
 * Francium: Commonly painted by explosion-loving students as a stereotypical temperamental alkali metal, a close investigation shows this not to be the whole story of francium. The 7s electron is relativistically destabilized, and the 6p electrons actually are involved in the bonding in francium superoxide. While it is still a group 1 metal at heart, this bucking of the trend sets the stage for the outright anomalous behaviour expected of E119 and E165, behaving like their 5th and 4th period congeners respectively; E119 should have EA around 0.66 eV, higher than any known alkali metal. Fr seems to behave as if it were intermediate between Cs and Rb, rather than below Cs. (And yes, Fr is predicted to crystallize in bcc.)
 * Radium: Ra's chemistry is much better developed than Fr's. Some properties transfer down from the lighter alkaline earth metals, such as an incredibly insoluble sulfate (RaSO4, even less water-soluble than the stereotypical insoluble salt BaSO4). The behaviour of the Ra2+ and Ba2+ ions is very similar. Like the group 2 elements (and Li, thanks to its diagonal relationship with Mg), it reacts with the nitrogen in the air to form a black nitride Ra3N2. The ion is highly basic and does not form complexes readily. The reaction with water is violent, as expected. Yet it is more volatile than Ba, and 6s and 6p electron involvement gives covalent character to RaF2 and RaAt2. In conclusion, these two elements, while not going too far off the stereotypical group 1 and group 2 moulds respectively, set the stage for the anomalous properties one is expected to encounter in the eighth and ninth periods. Double sharp (talk) 06:05, 13 October 2013 (UTC)


 * Be, Al: Cutting to the chase. Place both in pre-transition metals. Call poor metals 'post-transition metals'. Post-transition metals then have no elements with structural applications, appropriately enough for the second string metals located between the TMs and the metalloids. Al is a bit borderline being so close to the metal-nonmetal dividing line. Yes, it does have some poor metal/post-transition character but, upon reflection, has enough structural uses and is sufficiently active chemically, to warrant falling over the line into the pre-transition metals, noting many similarities with Be. The latter is close to the metal-nonmetal dividing line too, so also has some 'poor' in it but is strong enough as metal to not be categorised with the poor metals/post-transition metals.
 * The graphic in the poor metal article cld be adjusted, by moving the MP line down and the EN line to the right, so that Al shows as a moderately fusible electropositive metal, but still close to the poor-metals/post-transition metals. Mg and Be are already distinct from the cluster of the remaining alkaline earth metals, in any event (Be more so then Mg). Poor metals/post-transition metals than all have EN 1.65 or more (or > 1.5 on the Alfred-Rochow scale). PS. I am also thinking about adding a dotted boundary line to delineate the alkaline earth metals (Ca, Sr, Ba, Ra) from the rest of the moderately fusible electropositve metals, and to add an explanatory note about the historical distinction between these elements as the original alkaline earth metals, and the latecomers Mg and Be. Sandbh (talk) 04:29, 13 October 2013 (UTC)
 * Be as post-transition metal/poor metal is largely motivated by its generally covalent character. If we consider physical properties too, as we certainly ought to, we should definitely consider it along with the pre-transition metals. I suspect it could technically be used structurally, and the only reason why we don't is because of toxicity concerns – witness how much structural usage its bigger brother Al has! Double sharp (talk) 06:05, 13 October 2013 (UTC)


 * P.S. on IUPAC nomenclature: My main issue with the typical beginner's-chemistry numbering (old IUPAC minus A's and B's) is that it fails to show the group II bifurcation, and the proper place of Zn, Cd, and Hg as main group elements, IMO. (Cotton & Wilkinson's opinion also, but not Holleman & Wiberg's.) So we have 2A and 2B because of this, making it nicer and simpler. (Of course the simplest solution is to just pretend that the group 12 elements are transition metals. But I find this overly distorting the facts.) But then we also have similar analogies between groups IIIA and IIIB (and also IVA and IVB, and so on). So each group really bifurcates, but only in the case of group II are both the A and B group main group. (I'm not totally convinced that IB is right anyhow. These elements do not have +1 as their maximum valence and it may be more prudent, if less symmetrical, to classify them under group VIII, as Mendeleev did.) But of course this is confusing again for beginners...still not sure what to do here. Double sharp (talk) 15:53, 11 October 2013 (UTC)

Graphic below. Read "pre-transition metals" for "alkaline earth metals" (and remove "alkali metals", and have "poor metals" become "post-transition metals" again). Also read, as now usual, "rare earth metals" for "lanthanides". Double sharp (talk) 12:40, 8 October 2013 (UTC)
 * Tech question: Shall I create a color + legend for pre-transition metal (s)? Can do. There better be a page then. -DePiep (talk) 20:44, 8 October 2013 (UTC)
 * Tech note: please fellow editors, do not feel restrained by any color issue. Above the "alkaline earth metal" color, soft orange, is (ab)used to illustrate a proposal. Great. A definite color we can choose later on, if and when this matures. We can make a new legend color (for talk purposes) easily. -DePiep (talk) 22:06, 8 October 2013 (UTC)


 * IMO, this has nothing to do with the group 12 topic. We should keep it separate. Why not create another thread (instead of this subthreead)? -DePiep (talk) 20:33, 9 October 2013 (UTC)
 * Indeed, and I have now responded to your call. Double sharp (talk) 11:19, 10 October 2013 (UTC)
 * Point is solved. -DePiep (talk) 18:48, 12 October 2013 (UTC)

size of non-element images in our infoboxes
I hope this does not bring the gnomes down on me. I don't want any Wiki wars or to make people sad or have them take it out on me. But here goes:

We get a lot of criticism on the size of our infoboxes. Now, having an upfront list of properties is great and I will go to war to defend an upfront tabular chemical almanac (as serving the reader, who say...just wants a melting point, not an essay). But the frills are a bad idea. They take up space and don't add much. For the scientist or searcher of a number, they're not helpful. And for the general reader, they're cryptic. Plus they might be better used in the context of discussion.

What I'm talking about:

1. The shell diagrams. They're HUGE for one thing, for the small atoms like F or N. For the big ones like U, who is going to really count all those electrons. Is it really that helpful to put them in the INFOBOX (very valuable real estate) as duplicative of the electron configuration itself. Why not within some discussion of the electron configuration in text? Or at least make them smaller (make them scale as needed) for stuff like F and N. But really, gone would be better.

2. The little icons for hexagonal and the like. (Wire diagrams.) All the same applies, duplicative, cryptic for the general person and not needed for the scientist. And the terms (hexagonal) are blue linked. In many cases, it's not even that relevant (solid form of a gas) or there are elements like oxygen that have a multitude of forms based on temperature. Makes the strip on the right, go that much further down the page.

3. Spectral lines: pretty, but not usable at the scale shown and really...who is going to use that? If important, can be shown and DISCUSSED in article body discussion of spectroscopy aspects.

4. (keep this, but make it better) Even the periodic table. On Internet Explorer on a big laptop, I can barely see which atom is the one in the article. I really want to show where the element is in the table, but there has to be a better way to display. I end up seeing 2 thin black lines (not even all the way around the square). Why not make it something that works better at the scale involved here (which is tiny). Make the whole table beige and the element (the whole square, not the outline) grey. I don't need/want all the other stuff about different element groupings in THIS table. Since this table is so small, all I want is to know where my element is and then the shape of the table is good enough. Oh...and the clickable feature is kind of pointless at this scale. Who has the muscle control/memory/eyes to pick a particular transition metal? Can indulge with the table down below for those features. But at this scale, let's think about ergonomics and simplify to communicate better. [More] And why the extended width table for a 180px image? Surely the reader is more familiar with the lanthanides and actinides "below" on his HS chemistry periodic table...and you have a tiny space but use a wide image??

71.127.137.171 (talk) 00:03, 13 October 2013 (UTC)


 * 4. "Who has the muscle control/memory/eyes to pick a particular transition metal?" Well, I seem to be able to do this for some reason! But can appreciate that it's very difficult (I was only able to do this from last month). Then again it is kinda convenient so that you do not need to bother scrolling down to the bottom.
 * 3. Yes, kill that. It is better discussed separately if at all. Most articles do not even have it, despite the fact that spectrums are known for every element up to Es except At.
 * 2. Agree. (BTW double hexagonal close packed seen in some lanthanides should be marked better; now it's merely shown by a change of image.)
 * 1. I do want to be consistent about this. You are proposing we remove them all, right? From F and U alike? In that case, I could get behind this proposal. The pedagogical value is quite important too (it's been discussed before); nevertheless it is still misleading, sort of. Feel free to discuss; you're convincing me (as long as it applies to all the elements). Double sharp (talk) 02:23, 13 October 2013 (UTC)
 * 1. Yeah, get rid of them all (sorry).  It's very valuable real estate and we should not duplicate information (written electron configuration) within that table and make it grow further down the page.  Within text, a discussion of the electronic arrangement, might use a diagram.  I find myself in agreement with that John fellow.  (And nothing to do with who is first mover or BRD or AGF [I don't care], just on the content.)98.117.75.177 (talk) 16:48, 13 October 2013 (UTC)


 * On the little Bohr diagrams, I dislike them because I think they overemphasize an oversimplified and outdated pictorial way to think of an atom. Beyond about sodium you can't use them to explain the properties of materials or their chemistry, so they should not be in the infoboxes. They would be fine in the body of the articles for the lighter elements, where they are useful. Carbon's should be tetrahedral, of course. --John (talk) 20:21, 13 October 2013 (UTC)


 * Support except for the "remove links from the mini PT" part.


 * I (an editor) actually do use the mini PT in the infobox to get to other element articles. Keep the links. They certainly do no harm. Other ideas seems to be rational and fine, I support them, just keep wikilinks in the mini PT if you gray out other elements and if you don't.--R8R Gtrs (talk) 22:34, 13 October 2013 (UTC)


 * Leaving the click links is fine. They do no harm.71.127.137.171 (talk) 23:14, 13 October 2013 (UTC)


 * Question: IP 71.127.137.171 are you the same contributor as IP 98.117.75.177? -DePiep (talk) 23:25, 13 October 2013 (UTC)
 * OP (original post): We get a lot of criticism on the size of our infoboxes you say. Link and quote, or retract. -DePiep (talk) 23:30, 13 October 2013 (UTC)
 * re OP IP 71.127.137.171.
 * Welcome to the internet. Over here at internet, we have pages that you can read like a book. Above that you can click somewhere on a word or picture on that page, and that action will open for you another page -- exactly the detail of what you clicked on. For example, on page Germany you can click on Berlin (the capital) and then you arrive at the Berlin page. Bingo! Very important is: you don't have to click. Only click when and where you want to! -DePiep (talk) 23:50, 13 October 2013 (UTC)

I just found one link that talks about chemboxes too long at chembox template talk. I've seen the crit other places in discussions. It's not a huge thing and I'm happy to fight for having a prominent table with melting points and the like (serves the reader). But when you load it with icons, that's not good use of the imposition the box makes on the page.71.127.137.171 (talk) 00:08, 14 October 2013 (UTC)
 * I suggest you stop fighting, even if it makes you happy. And my hexagonal is not an 'icon', it is a scientific graph/scheme/depiction/youchoose. -DePiep (talk) 00:23, 14 October 2013 (UTC)
 * Oh please, share the link and stop using code like "crit". -DePiep (talk) 00:37, 14 October 2013 (UTC)


 * re 1-2-3-4: I find not a single argument for any change. I read you only have issues with image size, basically (and you do not get the link-to-detail click). Apart from the bohr circles (already addressed as a WT:ELEM topic; please contribute here). For example, you even want to remove the spectral lines picture from the infobox. You want to remove an identification, really? And what is the problem with showing a hexagonal crystal? (disclosure: I handcrafted that hexagonal pic myself, geometrically correct. It is on my userpage ever since). -DePiep (talk) 00:17, 14 October 2013 (UTC)

I'm not fighting man. If anything I'm really holding back from going nit by nit through the points. Cause I really don't want to hurt your feelings. I appreciate your technical skills. It's great if they are harnessed in the interests of good. How can you say there are "no arguments" when I've clearly said the issue is how the table gets longer and longer (which interferes with layouts) and how the information is duplicative (a Bohr icon next to an electron config, but used in a place of very high prominence. I'm not trying to get rid of the Elements infoboxes.  I just want them tighter.  I've really made the point and will leave it as is.  (I get this want of you to engage point by point on the details, but then I don't want to upset your sensitivity with a view of criticizing something you do. Let's let a few others engage, k?)
 * Sez ho? -DePiep (talk) 02:16, 14 October 2013 (UTC)


 * Yes, please get rid of the shell diagrams. Too big, hard to read, unphysical, ... --99of9 (talk) 01:05, 14 October 2013 (UTC)
 * Q IPs:, , , did you ever edit wikipedia under another name or id? Are you related to user:John? -DePiep (talk) 01:26, 14 October 2013 (UTC)
 * I'm about 99.99999% sure he's the IPs are TCO. Not a problem with me. Double sharp (talk) 03:31, 14 October 2013 (UTC) (edited for clarity by Double sharp (talk) at 08:30, 15 October 2013 (UTC))
 * Yes, as an admin and long-term editor since 2006 I would definitely engage in sockpuppetry to influence this trivial, though irritating, formatting issue. The Bohr diagrams are unnecessary and unsightly. No proper consensus was ever achieved to add them. They should be removed, or returned to their previous small size where they were less intrusive. --John (talk) 13:02, 14 October 2013 (UTC)
 * I am perfectly in place to ask whether three commenting IPs (one not signing) are three different persons. Also, for tone an attitude similarities, I can ask for John's connections. The fact that John points to his adminship is dubious and does not add to trustability. As he demonstrates in writing here, he cannot separate the tasks & responsabilities. My questions stands: are these four independent editors or not. -DePiep (talk) 22:35, 14 October 2013 (UTC)
 * The IPs are all the same. (I don't know how to engage.  Ignoring enrages you, but so do responses.)  I've gone ahead and teed up a vote.  Sorry, if you don't like my style.  I like your demeanor actually.208.44.87.91 (talk) 00:56, 15 October 2013 (UTC)
 * re IPs; The IPs are all the same -- thanks. So my question into this was to the point. Different IPs make it difficult to follow your line of reasoning. It is sort of giving me the homework job to sort that out. That combined with the understructured setup, jargon/abbreviations used, a figthing attitude in a discussion: makes it even more difficult. On top of that, experienced user John started throwing personal attacks . -DePiep (talk) 08:08, 16 October 2013 (UTC)
 * Shell diagrams: never noticed them. Lighter ones seem quite big. With, for example, U would it be more economical to just show the Rn core in the centre and the residual electrons around that? Icons for crystal structures: like them. Adds information. Spectral lines: don't like these so much. Detract from the beautiful pictures. They don't add any information. Periodic table: like. Looks good on my ipad. Resonates with PT in PT article. Keep the links. I suppose the lanthanides/actinides cld go below the main body of the table: would leave more room for the element boxes. Sandbh (talk) 02:28, 14 October 2013 (UTC)

Vote on options to improve the Elements infobox
(I will advertise at Chemicals and Chemistry also...they are our brothers.)
 * I object to this setup. The topics are only introduced half heartedly above. So far there is no substantial discussion that leads to these black-and-white statements; also the setup mixes votes and proposals -- which will make it difficult to conclude anything sensible. And of course there is WP:NOVOTE. -DePiep (talk) 19:33, 15 October 2013 (UTC)
 * I guess things are being done the other way round? Having the vote first and then the discussion. So some of the votes could be expected to move around. Sandbh (talk) 10:36, 17 October 2013 (UTC)

Anybody seen Theodore Gray's (2008) 'The photographic card deck of the elements'? On the back of each card is included a colour category medium-long PT; a crystal structure wire diagram (as well as the name of the crystal structure); and a shell diagram similar to ours except that the size of the shell diagram is proportional to the atomic radius. Oh, and on the PT, the element box in question is blacked out---stands out a mile. And I just noticed that the horizontal colour bars in each of our element info boxes nicely correspond to the colour category of each element. Sandbh (talk) 10:36, 17 October 2013 (UTC)
 * I have seen a sample. It looks pretty cool. We could try this here too. Double sharp (talk) 11:10, 17 October 2013 (UTC)

Shell diagrams in infobox

 * Closed as delete all from infobox element. No editors opposing since the discussion was opened over a month ago. The earlier suggestion to keep the lower numbered ones was not brought forward here again. I was involved in this talk, but this closure is contradicting my earlier opposition, and that opposition I dropped weeks ago. -DePiep (talk) 10:52, 21 November 2013 (UTC)

F sized: atomic radius = 50 pm U sized: atomic radius = 156 pm Sandbh (talk) 21:05, 17 October 2013 (UTC)

(pick one choice please)

Eliminate

 * 1) TCO (IP)208.44.87.91 (talk) 00:56, 15 October 2013 (UTC)
 * 2) Weak support. Once we get the quantum graphics ready, we have a good replacement for these. Double sharp (talk) 01:42, 15 October 2013 (UTC)
 * 3) Delete. Make articles look like they were designed by a primary school student, not the impression we want to give. --John (talk) 09:44, 15 October 2013 (UTC)
 * 4) Ditto per Double sharp Sandbh (talk) 10:30, 15 October 2013 (UTC)
 * Question for and : do you mean "weak support" for deletion or for the image (to stay)? -DePiep (talk) 19:38, 15 October 2013 (UTC)
 * Support for deletion given the proposed quantum graphics replacement. I like the fluorine picture but I also had a look at the E118 picture and that was too much, ditto U. So, I like graphic accompaniments in general that convey some useful information. Sandbh (talk) 21:15, 15 October 2013 (UTC)
 * Delete, but only because our quantum graphics are coming. Double sharp (talk) 02:43, 16 October 2013 (UTC)
 * re "given the proposed quantum graphics replacement" -- these are not fit for articles. Will take months not days to get there. And they are indeed intended to eplace these circle graphics, maybe somewhere else on the page. Do you two want to delete before there is a replacement? -DePiep (talk) 14:26, 18 October 2013 (UTC)
 * I would rather we wait on this till we have a suitable replacement, though I could probably accept deletion before the replacement enters. Double sharp (talk) 10:38, 20 October 2013 (UTC)

note: as written below, I withdraw my !vote. You all can conclude here as you think right. (I understand that the "keep the ones below sodium, and make smaller" idea is not supported either - fine with me). -DePiep (talk) 12:24, 30 October 2013 (UTC)
 * 1) +1--R8R Gtrs (talk) 14:20, 18 October 2013 (UTC)

Modify (make smaller, etc.)

 * 1) Keep - into neutral now, see below. Not wrong. Resize might be an improvement, size by radius (number of circles) could look nice.
 * Confusing is that the argument flips from "size" to "relevance". Anyway, it is not a "non-element graph" the OP suggests. A bit confusing that in the lower numbers it would be OK and instructive, then inexplicably the same systematich approach is said to fail in the higher numbers. And yes, we need a better graphic, as is being worked on (instructive and more correct wrt quantum mechanics). Till then these could stay. -DePiep (talk) 15:26, 22 October 2013 (UTC)
 * I change to neutral (i.e. withdraw my !vote). Since I am the only keeper now, against people in the know, I better not block the change. Hey, these crowded ones are my babies! Kill them gently please ;-) -12:24, 30 October 2013 (UTC)

Crystal structure type wire diagrams in infobox

 * Closed as: ''no consensus. No changes. This also after reading the arguments in the top paragraphs of this thread. Note: I contributed in this discussion. -DePiep (talk) 11:07, 21 November 2013 (UTC)
 * Note: less relevant crystal structures, including images, were removed . -DePiep (talk) 11:21, 21 November 2013 (UTC)



(please pick one)

Eliminate

 * 1) TCO (IP)208.44.87.91 (talk) 00:56, 15 October 2013 (UTC) (in particular, annoying in articles on gases)
 * Box size (you opened the thread) is not an argument. You have not elaborated any other (content) criteria for including/excluding data. -DePiep (talk) 09:48, 12 November 2013 (UTC)

Leave as is

 * 1) Kinda useful in helping you understand the structure if you don't already know what the words mean. This way we don't have to lose our readers to another article. Double sharp (talk) 01:42, 15 October 2013 (UTC)
 * 2) Ditto Sandbh (talk) 10:22, 15 October 2013 (UTC) The crystal structures of gases cld be for the element at its freezing point, with commentary in the text, if there are additional structures at lower temperatures. Sandbh (talk) 09:33, 18 October 2013 (UTC)
 * 3) A wire scheme for a crystal structure, what could be more illustrating? They are iconic. And think of our allotropes (any reader who has not been told about the C ones?). Yes, use word and inage is sort of "duplicative", as in "picture and caption" is a duplication. We could link the image to the article, not to the image page. (Then we could even leave out the bluelinked word (e.g. hexagonal), to illustrate my point).
 * OP (first IP post) notes that these icons have the same issues as the electron config images. That would be, by the same post: huge, who counts electrons, make smaller, can be in article with the text. Well, first three clearly don't apply. Why mentioned at all? The fourth issue can be said of every item in the infobox, it even can be part of quality writing. So I don't see "same" issues at all. Then, if there are issues with individual elements, solve them per element. One is invited to propose a nicer presentation of the temperature-variants mentioned. Just saying it is "not relevant" is not quite an argument, except that I could write "it is relevant" as well. On what scale is relevance seen wrt this? I add this idea: We could remove the algebra from the icons this size, they were not made specifically for this usage.
 * Concluding, I think using the icon-like image is to the point as used here. Presentational improvements can be made (layout &tc.). -DePiep (talk) 07:40, 16 October 2013 (UTC)


 * 1) I actually think this would be better. You have to imagine what, say, cubic fcc is. having a pic allows a faster representation.--R8R Gtrs (talk) 14:20, 18 October 2013 (UTC)

Spectral lines in infobox

 * Closed as no consensus. No changes from here. No extra arguments found in top paragraphs. There is a small support to keep or add them for gaseous elements, which is not denied by this outcome. Note: I contributed to this discussion. -DePiep (talk) 11:35, 21 November 2013 (UTC)



(please pick one)

Eliminate all

 * 1) TCO 208.44.87.91 (talk) 00:56, 15 October 2013 (UTC)
 * Box size (you opened the thread) is not an argument. You have not elaborated any other (content) criteria for including/excluding data. -DePiep (talk) 09:56, 12 November 2013 (UTC)


 * 1) I could make them all. But what's the point? What do they show? I mean, outside H, it's not like they are discussed in detail in the article – nor do I foresee this being done. Double sharp (talk) 01:42, 15 October 2013 (UTC)
 * Whether it is discussed in the article text decides if it is in the infobox? Will we drop 80% of the data in that box then, including the main picture? They are just facts, and if there is nothing more to say about (say) the boiling pint, then we do not. Bad argument to me. -DePiep (talk) 16:26, 24 October 2013 (UTC)


 * 1) I agree with Double sharp's reasoning Sandbh (talk) 10:19, 15 October 2013 (UTC)
 * 2) Nergaal (below) raises an important question, about gases (is a reasonable question). I would rather say this could be discussed (briefly) in the article text, and the pic could go there easily.--R8R Gtrs (talk) 14:20, 18 October 2013 (UTC)

Leave as is (infrequent usage)

 * 1) I would keep them for the gaseous elements (I mean we have pictures of light bulbs not of the actual element for these). Nergaal (talk) 10:39, 15 October 2013 (UTC)
 * 2) As Nergaal says. btw, this one too is quite not a "non-element" image. And re Double sharp, "What do they show?" -- they are an identification for the element, right? could be in the block "General propoerties" -DePiep (talk) 14:31, 18 October 2013 (UTC)
 * Change: I suggest we add all of them. see below. -DePiep (talk) 13:12, 26 October 2013 (UTC)

Feature in all

 * 1) Show all known. They are an identification of the element, right? Could be in the block "General propoerties". Also I have the impression that deletion is proposed to save space. That is an irrational route of reasoning because it uses a different argument. Again, if we gonna remove data from the infobox we should do consistently so over all boxes and all data. -DePiep (talk) 13:12, 26 October 2013 (UTC)
 * 2) *My main issue is that it is very hard to get useful info out of the picture. It would be better to have a numerical list of the wavelengths of the spectral lines along with the picture. Double sharp (talk) 14:15, 31 October 2013 (UTC)
 * The picture is the info. Adding numerical data - wondering what gut arguing TCO has for or against it. He knows the mysterical arguments for what/not to include. -DePiep (talk) 10:04, 12 November 2013 (UTC)

Periodic table in infobox


(sorry, this is not exactly what we display, closest I could find)

(pick amongst various binary options)
 * Explained: there are three independent topics, each with options. There is: 1. f-block inline or below?, 2. Mark current element by outline or by shaded cell?, 3. Keep or remove category colors? -DePiep (talk) 09:24, 16 October 2013 (UTC)
 * Do I understand you want to show the this picture but without the four black items? -DePiep (talk) 08:29, 16 October 2013 (UTC)
 * PT structures (not size) discussed are this (right?):
 * Since this change does not reduce infobox heigth, I guess you are asking this for other reasons. How would you use the extra width space (reduction of number of columns)? Improve clickability of element cells? -DePiep (talk) 08:29, 16 October 2013 (UTC)

Now that I use IE today (unknown version), I understand what TCO and Double sharp noted: bad visibilityy of the mini PT. In this IE it has no cell borders, and the element marker only shows two cell border lines. (while on my regular Firefox cells have nice 1px white borders a bit like the top picture here, and the marker is a full square black line). This should not be part of the discussion here, it should be resolved as a bug. Must say I have little online time these days to look into this. -DePiep (talk) 11:44, 17 October 2013 (UTC) Added: screenprint of how it should look like for As. -DePiep (talk) 10:28, 26 October 2013 (UTC)
 * IE bug? [[File:WP-ELEM_talk,_mini_periodic_table_-_intended_effect_(EI_bug_showing_PT_wrong).png|thumb|right|200px|This is what the infobox should look like ([[arsenic]], reduced size here): The marker is full square and all cells have white borders.]]

1. A. Change to narrower version (actinides and lanthanides below)

 * 1) TCO208.44.87.91 (talk) 00:56, 15 October 2013 (UTC)
 * Please specify what to achieve. See also images I added, above. It would make the infobox longer, while you aim at shorter iic. -DePiep (talk) 09:12, 16 October 2013 (UTC)
 * I like the picture. Yes, it makes the box longer (bad) but in a good cause...this is an image which everyone agrees should exist and is the most prominent part of the article, needs to work.  Your side by side, doesn't show the proper comparison as the aspect ratio is different (need to show the size each has in infobox (same width)  [This is also why I want it more iconic, less detailed...but any changes to help the reader are ones I take]  Reasons (which have been given): (1) Better viewing (this is a tiny space with a lot of detail in the image, we can barely make out which atom is the one of the article), (2) better clicking, (3) more common presentation for the reader.  208.44.87.91 (talk) 12:00, 16 October 2013 (UTC)
 * re: which picture of the three, now numbered, you like most as proposal (for structure, not about size)? Sure there are reasons given somewhere, but also contradicting ones so I asked. (1): if this issue is in IE, there might be a bug (see my note above); the bug should go but not the mini PT show. (2) clicking: cells are 6&times;8px, doesn't look too small to me for access. e.g., reference notes like [1] are about equal or smaller in size. (3) Inline f-bock is not that uncommon. That less widestructure was standard when 3a f-block was not discovered and 3b print pages were more rectangular than HTML pages need to be. (add 4) Please take note that the extended mini PT (see element 121) already folds out with external rows (superactinides, ...) that would make four sub-rows more confusing. -DePiep (talk) 11:44, 17 October 2013 (UTC)
 * I like the narrow table best because for the reasons I gave: it is a very small space and there is a huge amount of detail and I want the reader to see where THE ARTICLE's atom is.  Also clickability is difficult because of the lack of differentiation as well as just the sheer number of little tiny pixes.  I mean imagine a rectangular array of 120 footnotes!  and it's not about if you can click it or see it...but about READERS in general.  How to best ergonomically serve them.  Not like techie people who don't understand users...but like editors of content who think about the reader and care for him.
 * Using the narrow gives you more room (I already acknowledge this takes up more space...so what...I am a sophisticated thinker...I am not up for arbitrarily cutting but for having the best use. So cut the crappy parts and keep the good parts.  Similarly, I want to keep the pictures of the elements (but some of the bloat in properties could go bye bye.  It's a risk reward type thing.)
 * Your table's are not drawn to scale. You need to show them at the width of in the infobox (whatever that is, I don't know...but it's equal):
 * Periodic Table overview (standard).svg
 * Periodic Table overview (wide).svg
 * If you can fix the IE bug (in Wiki, MSFT is not recoding for us), great. Serious, it's something.  But bottom line is we are serving up something really bad looking for a vast slice of the readership...and have been for months now.  A design fix by being more iconic would be even better and certainly the way to go if you can't fix the bug.
 * I actually tried knocking something up (as a special graphic) in article, because the graphic in the table was so lousy that I'm not even really doing a good job of helping the reader to understand one of the simplest things to orient all the chemistry (periodic table location, relation to column especially). It didnt' quite work and I'm not using it...but I commissioned the work in the Graphics Lab, because the Infobox is not getting it done.
 * Fluorine with significant periodic table neighbors.svg
 * "The inline f block is not that uncommon": Well it sure as heck is LESS common.  Look at the file names (the narrow table is actually labeled standard).  That's what people saw on the high school chemistry lab.  Why start the article with stuff that is less familiar?
 * As far as DS saying that the tables should be the same bottom and top, that makes ZERO sense. At one scale, you are incredibly cramped.  The other you have all the space in the world.  Bottom is fine to have wide, with symbols, numbers, all that!
 * 208.44.87.91 (talk) 00:02, 18 October 2013 (UTC)
 * It's quite a bit taller than the wide table. We do not need such a big PT. The focus is on the element and its region in the PT. Even that is quite ameliorated by discussion of period and group and trends. See curium for an example.
 * (Did I say that? I'm not sure where, if I did at all.) Then why do you feel the need to blow up the neighbours to show symbols, numbers, etc.? Because they're the most important in relation to the element you're writing about? Then it looks weird and is even more of a spacehog in the infobox than the electron shell images.
 * Why not start it with the wide version? It stays very neutral to the differing conventions for group 3. Already the 18-column table is not completely standard in the form we present it (many textbooks still show the oudated Sc/Y/La/Ac grouping, and some still float H in the middle instead of putting it over the alkali metals, and let's not even talk about us wanting to move group 12 out of the transition metals, which only about 50% of textbooks do). The 32-column table helps and better reflects the quantum mechanics. We use the wide version at the bottom too: is that a problem there? Then why is it a problem in the infobox? (Also giving navigation and a sense of position in the table.)
 * I don't think cutting would be best. This takes up more space that I feel is not needed, the wide table giving you the SAME info for less space. If other properties are helpful in understanding the element I feel it is good. Books like the CRC handbook exist for a reason. Why not show the data? It may not always be interesting – for example, it's really not surprising that the only known oxidation state of barium is +2 – but it should be there for consistency, and when it is especially interesting, we can talk about it.
 * Similarly for crystal structure: it may be annoying to you for articles on liquids and gases, but these are not exceptional elements. They are just elements that happen to have melting and boiling points below 25 °C at 1 atm. By removing it you are giving readers the impression that the crystal structure is either not known (which is not true) or that these elements are special and don't have standard crystal structures in the solid state (which is also not true). Whereas it's not very reasonable to conclude that they have crystal structures in the gaseous state from the infobox – the fact that gas particles can move around freely is elementary school stuff. Double sharp (talk) 07:17, 18 October 2013 (UTC)
 * "We do not need such a big PT." We do. That may be just an impression, but yeah, DePiep and TCO are right when they say there's a lotta times when people oppose changes because they're used to what they have, and this seems to be the case. A larger table is more useful, since it's easier to look at and more recognizable. And the cells are easier to click. Group 3 is a minor question. Even for lutetium, I'd use a more readable table than a less controversial. There is no consensus anyway. And you can do Ba * Lu Hf; Ra ** Lr Rf. But whatever, this is a minor question overall and a VERY minor here. Readability is far more important. Especially in an infobox (which is essentially a bunch of data standing out in a colored box so they can be found quickly. For something like that, readability is essential.)
 * Same thing with 18 vs. 32 from the point of view of science. This may be an important question, but it's none of the infobox's business. It is there so a reader can orientate easily and for no other reason.
 * P.S. It's Internet. You have as much space as you want. We can afford a bigger table...
 * Please sign your edit. Why make me do for you? -DePiep (talk) 08:22, 26 October 2013 (UTC)
 * P.S. It's Internet. You have as much space as you want. We can afford a bigger table...
 * Please sign your edit. Why make me do for you? -DePiep (talk) 08:22, 26 October 2013 (UTC)


 * 1) TCO is completely right as far as I'm concerned.--R8R Gtrs (talk) 14:20, 18 October 2013 (UTC)

OK, you're starting to convince me. :-)

Last line of defense I have: in the extended periodic table, we have to rip out the g-block. It would look confusing if we ripped out both the f-block and g-block, going with the 18-column version, and I think the 52-column version is awful. :-) And to be consistent I think we should show it as 32-column both ways, because (1) we should be consistent and not disorient the reader when they click on the link at E118 to E119 and (2) we should pick the solution that works best for both. 32-column seems to be the only one that works reasonably in both scenarios. Double sharp (talk) 10:30, 20 October 2013 (UTC)


 * Yes, I see where that is coming from. That's far from unreasonable, sure. And of course, fifty is waaaaaay too many :)


 * But I don't think we should be consistent. Think of this in this way:


 * 98% of element articles views come from the first 118 (that's an estimate; feel free to correct the figure, I'm sure the unstables have even less than 2%). I think we are fine to think about these 118 more than about those remaining <2%. Almost all readers who read oxygen or silicon will never ever ever read ununennium (let's change those stupid generic names already). I think we're okay to give that majority a better table without making it worse for the minority. It would be much different, if the ratio were 70/30 or like that. But really, a lot of people don't even care or even know about the fact elements past 118 exist. Even transuraniums are out of interest for a most of people. These elements are kind of a hidden level within a more general one (common elements), a lot of people don't know, many don't look for it, some just don't care.


 * Because of that, I believe we can think of these 98% article views first. We can give them a beter article. As for those <2%, they, at least, have a differing table already, with all those other elements and new lines of g block. We can leave a 32-column table there. It's okay... no one would ever want a 18-column one when g block is present :0


 * If we change that for just first 118, everyone wins. 118 get a better table, 119+ stay with a better table. (Better for their purposes, which are different for oxygen and 120, I think it's obvious.)--R8R Gtrs (talk) 19:39, 25 October 2013 (UTC)
 * And you have officially convinced me to support your proposal! 118 and below can use 18-column, and 119 and above can use 32-column. (DePiep, is this usage OK?) Double sharp (talk) 09:03, 26 October 2013 (UTC)
 * No, as I explained below: 32-col is the PT structure, and 18-col is a compromise and a horror to teach. -DePiep (talk) 09:52, 12 November 2013 (UTC)

(arbitrary break for editors using their cell phones)
Coming back in after a few days out. Is the black/white proposal still in topic (the one here with a looking glass to illustrate F)? Oh and please don't refer to it as "your" (my) table. That is incorrect and it spoils the atmosphere. -DePiep (talk) 20:23, 25 October 2013 (UTC)
 * My replies re the mini PT.
 * re "looking glass" example. a. Width as in the infobox is too small to use any PT text in the element cells. A most small clickable PT (wikitable, like here), even only 18 column, needs half page width to be readible (where infobox is ~33% page width). In image form (unclickable then), it could be a bit less wide. (see this 18 column one we use in PT). Still, as we use it it is also 50% page width. I conclude: we cannot squeeze a text-filled PT into infobox width. b. The looking glass is a visual disruption. c. I disagree that the PT in the infobox is about neighbouring elements only. It is also about its place in the PT structure. This includes its group, period, metallishness (category color), and more. It is not up to us to claim that the structural overview is not needed. On top of this, if it were true that only the neighbours are relevant as the looking glass says, we could drop this whole PT from the infobox. d. The caption of the mini PT says it nicely: "fluor in the periodic table" (there is also this example of the same). Also note that in most browsers (don't know about IE) mouse hovering helps the reader by showing the element name plus some properties (possible because a clickable table). So I do not want to go that route.
 * re f-block inline or out of line. I found two arguments.
 * On argument is: f-block below is a very common style, and is seen very often. That is true within limits. But it is also old-fashioned, we should not keep it just because it was so earlier (and we are used to it). This layout has two backgrounds: 1. the narrow form had stabilized before f-block elements were understood or discovered as a block. 2. Printing (books) were based on a more rectangular page size. Printing f-block below requires a ~(7+2)&times;18 space (page ratio 0.5). Printing it inline requires ~7&times;32 (page ratio 0.22). A regular page like A4 has ratio 0.7 -- close to 0.5 then. So, a printing page was screaming for the more fitting table, e.i. f-block below. But since we publish on internet, we are not limited by print page ratio.
 * Also note, the usage of old-style/new style PTs may differ between print and internet. Within this wiki we use the inline layout whenever whereever possible. Only when we are forced to do otherwise we have moved f-block below. For smilar reasons, I wonder if other internet-based PTs still use that old style layout in the same rate.
 * The natural structure is with f-block, now fully known, inline. That is what the PT tells us: the structure. There is no need for the mental leap to reorganize a PT, except for those very involved in the PT (they/we have learned that leap into intuition). A mental step it stays.
 * Second argument is: bigger cells are better clickable. That is true in itself. And reducing width from 32 to 18 cells would give a cell, in a same fixed width, ~75% more (32/18). That would be 6&times;8px &rarr; ~10&times;14px (everything else asame). But I do not see why that would be needed. Not needed so much that we should change the PT layout for that (note that until some years ago we had an picture in top, unclickable. The picture now atop this thread. It was f-block inline).
 * Now I have not found a WP:ACCESS guideline that says what clickable area is acceptable or not. If there were a guideline (minimum size), we'd apply that of course. But now the tradeoff (between element cell size and PT layout) is just a personal preference. I am not convinced current size it is too small.
 * There are some other interesting options wrt this mini PT size. I will address them separately. Same for using category colors.
 * Based on all this, I don't want the mini PT layout changed. -DePiep (talk) 09:51, 26 October 2013 (UTC)
 * Please see below in to discuss my sandbox3 PT proposal (cells bigger & still f-block inline). -DePiep (talk) 12:37, 26 October 2013 (UTC)


 * Hmm.
 * What you write makes sense. But I would try to take a different view.
 * There are people familiar with the 18-column table and unfamiliar with the 32-column table. They are students or weren't good at school. On the other hand, virtually all of people familiar with the 32-column table are also familiar with the 18-column table. There are reasons why this happened (namely what you wrote about book formats), but if we want to be nice to as many people as possible, we should use the better-known table. Even though it is older. Even though it is not as good scientifically (it's a very minor difference anyway). When usage changes, we, of course, should match that, but not before. Re cells, why would you need to be convinced cells are too small to improve them? Bigger = better (until it's ugly, but that's not our case). And they won't become just easier to click (although I miss sometimes the one I'm aiming for, I really do), they will be easier to see. You may have a good vision, but not everyone does.
 * As for sandbox, I'm fine with it as with a temporary solution (until we make a 18-column table).
 * (When it comes to Wiki, I want readers to be more important than scientific bagatelles.)--R8R Gtrs (talk) 10:31, 2 November 2013 (UTC)
 * re R8R: There are people familiar with the 18-column table and unfamiliar with the 32-column table. They are students or weren't good at school. Exactly! 18-Column people are bad at school. And this is why: because the 18-column periodic table is a distorted periodic table. They were shown the wrong table! It is way more complicated than necessary. 32-Column people recognize and fill in mentally the satellite f-block into the right place yes. Not by natural intuition, but because they have learned to do so (after they learned 18-columns, a time wasting and bad teaching sequence). Still, uch a fill-in takes an extra mental step or two (exception: some people have their brains wired in a PT pattern. They don't need that step. These people are invited to Oslo). And conversely: no person seeing a 32-column PT would think of cutting out the f-block for more overview or insight. Ever.
 * All other PT quirks (such as: positioning H, Janet left step thinking, use atomic number not weight for ordering) are physics/chemical facts. They can be discovered and taught.
 * But setting apart the lanthanides and actinides deliberately is breaking the system, vandalising the PT essence, is taking both the "periodic" and the "table" out of the "periodic table". What remains is a "chaotic bag". And then there are the newly introduced dirty details: where to put Li and Lr?, don't forget to label that island of mental instability with "period 7, 8", explain "group 3" issues on two locations, explain that these 14 (or 15) elements are not to be crammed into that one cell, and of course remember when inserting that block back in we need to cut the PT and reform it to make space. That's just details. One will learn how to address these details some five years after they were introduced to you when they showed you that bad 18-column table at day one (reversing the original waste of time - doubles it).
 * This is a circle (1) - circle with gap.png


 * How to teach the obvious. Please imagine. In a classroom of ten year olds, the picture right is the very first drawing of a circle a young pupil sees. Sure factually correct, after an additional explanation by the teacher. How instructive is that? Is there really no better way to introduce and learn about a circle?


 * This is a circle (2) - complicated, distorted circle.png to teach the circle with many more aspects present (like center dot, radial colors, a rotation of the segment below and squeezing the gap for some reason -- one can see the PT parallels, right). Would any sane teacher show first day this example? The question is: "Why take out that section in first place?". And still this is what happens when we keep the 18-column PT the basic source of PT teaching.


 * Yes there are situations that require a split table. That is when we are forced to trade off contradicting requirements (such as print page size). Still then, the 32-column overview (in some mini size) better be present nearby, or I will not give permission to print that.


 * This is a circle (3) - simple circle.pngng the f-block back in, we can teach the periodic table like this image. Teach a child the 32 column PT, and he or she forever will understand the degraded 18 column PT.
 * (I will turn this into an essay :-) ) -DePiep (talk) 13:14, 4 November 2013 (UTC)


 * re other points, I will be back later. -DePiep (talk) 13:14, 4 November 2013 (UTC)
 * re why would you need to be convinced cells are too small to improve them? - Didn't say that. I said 'too small by accessability standards? - no'. From there it is an option, not an obligation, to enlarge. Now we can and have to trade it off against its drawbacks. One drawback is, that enlarging results in an obligation to move the f-block into outer space. That is a negative consequence (as I wrote above), and so an argument against enlarging. So, enlarging is not free. Having said this, I find it weird that you do not support the cell enlargement I propose in /sandbox3 (not reproduced here for page speed reasons; just click). Here the cells grow to 48/70=146% clickable area. Why is this not "Bigger = better"? 7px&times;10px is  - compare x.
 * re we want to be nice to as many people as possible - not exactly nice. I'd rather change that in to something like "... inform usefully ...". I don't think we are helping people by repeating the ill-formed, hard-to-understand PT while we can use the original version at hand. Of course the split version is not scientifically wrong. It is a worse form of presentation. Let me state it this way: "everyone who understands the 18-column PT outline, understands the 32-column PT". And: "Given the 32-column PT, no one needs the 18-column PT to understand the PT better". I think I'll have these engraved in my buckle. -DePiep (talk) 17:49, 4 November 2013 (UTC)


 * They were shown the wrong table It is not wrong. Maybe confusing (for you), but not worse than that.


 * 18-Column people are bad at school. That is a very bald thing to say. I'd argue against that. a) Many 18-column people have grown to be very acknowledged. Glenn Seaborg is an example. Few, if any, kids in the Western world learn not the 18-column table first. We still have chemists today, though; b) The difference is not very important. s and p electrons are a subject of a focus in school, d electrons are as well, but to a lesser extent (partially because the d electrons filling progression is pretty much like the p one, but the geometry is more complicated.) f elements don't receive much attention in either case (pretty analogous to d, except for being even more complicated. Also, there are no very important f elements (but there are d elements like iron, copper, or gold); c) I had a 8-column table in school, I still managed to keep up with that they wanted me to know. That said, this is a very minor issue when you think of teaching.


 * vandalising the PT essence chat with Mendeleev about that. He had an 8-column form.


 * where to put Li and Lr? Li is a no brainer. Lr is a problem no matter what form you have (check German infoboxes).


 * that island of mental instability ha ha. That one actually made me smile :) (no serious talk here)


 * How to teach the obvious. Yaah, sure. It does work with a circle. Not with a waaay too wide table when it changes so little (see above). Maybe I'm even wrong, but I think this is why everyone uses the 18-column form. Which is a reason for us to use it. I actually believe this is the reason.--R8R Gtrs (talk) 17:48, 13 November 2013 (UTC)
 * Re Bad at school: I took that from you: "There are people familiar with the 18-column table and unfamiliar with the 32-column table. They are students or weren't good at school." I use your point for my main thesis: better teach the 32 column straight away, cutting it to 18 col can be done easily -- instead of the opposite direction. And no, Mendeleev always kept all in one table, he never split off parts (and of course this is not about column counting, but about conveying the structure). No it is not about me not understanding (When I take enough time and mentally prepare, I can reconstruct the true PT from the broken 18-col two-table aftermath); what I do not understand is how the 18-col is a natural, primary way to show it.
 * re where to put Li and Lr? is about where to put them (graphically) in an 18-col PT: in the big one or in the blocks below? In a 32-col PT there is no discussion about their position (funny: your German example Lawrencium is not shifted in the table, Sc and Y are -- proves my point that re-entering the block, from below, causes confusing details left to explain, and that you, "18-columnist", misread their PT. ;-) ).
 * Circle, good example but not here. So my circle demo works for you as you say (to explain something). Well, I wanted to illustrate that the second "circle" picture is a parallel for the broken PT: logically connected & systematic parts are moved away. (here I can not follow your reasoning "when it changes so little ... is the reason" What changes little?). I say the periodicity & the table is broken & distorted for no reason.
 * And yes, Many 18-column people have grown to be very acknowledged. Glenn Seaborg .... That is why I said 'Oslo'! His mind was wired for the 32-col table! You say: see, 18-col people can get a Nobel prize. I say, a 32-column bred person can win the Prize one year earlier, not having to mentally translate the broken 18-col-PT in the true table always. Seaborg put the f-block in the PT himself. Yes, in there.
 * For me, I have encountered only two reasons why "everybody" uses the 18-col: forced by circumstances (e.g., book page ratio) or by peer's habit (teachers, scientists who have learned it that way). -DePiep (talk) 18:57, 13 November 2013 (UTC)

I actually meant there are poor students, and they are poor no matter what table they have. Seriously, looking back at my school education, there is no difference. I wouldn't be any better if I had 32 columns. Just to make it clear: people are bad at chem NOT because their PT is poor (read my previos comment, f-block is a very minor issue). They're either not listening to the teacher (which is common is school if you have a bad teacher) or, well, chemistry is not their thing (I suspect, it's teacher's fault anyway).

Lawrencium thing: if we have 32 columns, we have end up having it under Y (as we do today) or under a void (as in German tables). Two options. That was all I wanted to say re your lack of ambiguity in 32-column table.

Circle: okay. Won't argue about this.

The argument about Seaborg was bad. But look at chemistry before him, it was growing and expanding. With that poor table. But whatever. My core argument is: 18-column table is used today by everyone, and therefore we should use it as well. That's the only reason for me (other things are nice chatting). Just to make clear: I am not against 32 columns. If we were a part of a group which makes the decision how to teach kids, I would not be arguing against 32 columns. But it's Wiki in here. We should reflect changes, not make them (which is a core policy). When 32 columns become the education standard (which, I believe, is not so unreal in a long-term perspective), I'll advocate switching. But only after that.--R8R Gtrs (talk) 19:21, 17 November 2013 (UTC)

Oh yes, re your sandbox. (I though I had said something before, sorry) I am not 100% sure about not having the Sc<-Y->Lu (vertical) column, but I think it may be a deal of a habit, so I am okay with it (at very least, I will not oppose). But only as with a temporary solution before we get the 18-column table.--R8R Gtrs (talk) 18:09, 13 November 2013 (UTC)
 * As I said elsewhere, there will be no "temporal" version "before we ...". Funny that cell size was an argument, but now that I have enlarged them (px 48/70=146% clickable space) size doesn't matter any more. -DePiep (talk) 19:11, 13 November 2013 (UTC)


 * Re size. I don't follow. Just to make sure you get me right: I am in favor of larger cells, that is not to be questioned. That thinking about the left navigation column was on my head, so I mentioned it. I could've (should've) make that more explicit, yeah. Well, I hope I make my point clear now.
 * Tl;dr Size does matter. If it didn't, I wouldn't support the removal of the column in first place.--R8R Gtrs (talk) 19:21, 13 November 2013 (UTC)


 * Whoa, this is getting somewhat hot on 18- vs. 32-column!
 * Actually I like Mendeleevian 8-column short periodic tables, as they show indirect relationships like C/Si/Ti. But I don't think they will catch on in the West. :-( I guess you could call me a weak 8-columnist, and a more strong 32-columnist.
 * 32-column, I think, is better than 18-column, but 8 I think is somewhat superior to both. I would not use it in the infobox since not many will recognize it and 32-column is clearer.
 * I submit that the electron configuration of Lr doesn't actually create problems. It behaves like an actinide chemically. And the German WP, with its consistent Sc/Y/La/Ac choice, doesn't seem to have problems with it (though neither do we with our Sc/Y/Lu/Lr choice).
 * The wide version's main fault was always its bad aspect ratio for print work. Here we are not constrained by it. Double sharp (talk) 03:59, 15 November 2013 (UTC)

1. B. Stay with wide version

 * 1) I just think it's less confusing to the reader. Without labels, how does he tell at a glance if you're using the Sc/Y/La/Ac or Sc/Y/Lu/Lr convention? And it fits in the space well, filling the width but not having inappropriate length. An 18-column periodic table would either not fill the entire width or extend inappropriately long. Double sharp (talk) 01:42, 15 October 2013 (UTC)
 * 2) Mild support for keeping wide because there is enough room (and it is good to be reminded of what the long version of the PT looks like). Sandbh (talk) 23:57, 15 October 2013 (UTC)
 * 3) Wide version preferred. See my note on possible IE bug above (bad borders showing). Note that period 8-9 elements show these periods below now, which would then lead to 4 of these shifted rows -- not very clear.
 * Possible improvements: a. group and period "numbers" could be added as clickable dots-in-a-cell above/beside this mini PT. b. neighbouring elements could go from the infobox when clickable? -DePiep (talk) 11:44, 17 October 2013 (UTC)

=Sandbox 3: cells a bit bigger
=
 * Suggestion. I have prepared a new version of the mini PT: see sandbox3. The element cells are from 6×8px; into 7×10px;. (not transcluded here because of page load speed.)
 * Of course the PT is 32px wider: from 192 to 224px. To keep the width of the infobox reasonably small, we can remove the vertical neighbors symbols (eg in chlorine the vertical Fl ← Cl → Br), and consistently the horizontals (below) too. The resulting widening of the infobox will be a ~10px. -DePiep (talk) 12:32, 26 October 2013 (UTC)

2. A. Make article's element more prominent (e.g. shade entire square)

 * 1) TCO208.44.87.91 (talk) 00:56, 15 October 2013 (UTC) (hard to find the element of concern)
 * 2) Support. I believe this was done some years ago, and I'm not sure what happened. We should bring the full shading back. Double sharp (talk) 01:42, 15 October 2013 (UTC)
 * Originally changed to 2B at 10:34, 15 October 2013 (UTC), then changed back. It should be OK given the infobox header showing the colour category. Double sharp (talk) 11:10, 17 October 2013 (UTC)
 * 1) Nergaal (talk) 11:03, 15 October 2013 (UTC)
 * 2) Support, given colour bars in each info box match element colour category. Sandbh (talk) 10:36, 17 October 2013 (UTC)
 * But Sandbh, shading the whole square (element cell) makes the category color unrecognisable ?! -DePiep (talk) 13:30, 26 October 2013 (UTC)


 * 1) Readability.--R8R Gtrs (talk) 14:20, 18 October 2013 (UTC)
 * Note: the element marker should look like pictured above. EI might be buggy. -DePiep (talk) 13:30, 26 October 2013 (UTC)

2. B. Leave article's element shown as is (outline)

 * 1) Sandbh (talk) 10:27, 15 October 2013 (UTC) Changed to 2A. Sandbh (talk) 10:36, 17 October 2013 (UTC)
 * 2) Oppose – it covers the colour category for the element if you shade the entire square. We should make that important information clear. Double sharp (talk) 10:34, 15 October 2013 (UTC) Changed back to 2A. Double sharp (talk) 11:10, 17 October 2013 (UTC)
 * 3) Use outline! (keep as is). Of course the IE bug should go (see top), it should go whatever we choose here. I'll put up a good example for IE editors.
 * Why do I want to keep it? The element color nicely links that color's meaning to the category color we use in the infobox title bars. It is a direct (and the only) association to that category (e.g. As in metalloid group with other metalloids). Making the cell black would remove that info from the PT, info about the topic element. -DePiep (talk) 10:03, 26 October 2013 (UTC)

3. A. Get rid of the different colors for elements other than the one of interest

 * 1) TCO208.44.87.91 (talk) 00:56, 15 October 2013 (UTC) (a more iconic presentation works better at this size scale)
 * 2) First of all, a color of one element (the vote straight above) will mix in with other colors (I really don't understand why this and the previous vote weren't done as a single one. It simply won't work (will work for you guys, and won't for newcomers and non-professionals). As for this vote, decoloring other elements makes more sense... the element in question is easier to find then. Others serve as a colored distraction. Moreover, they're not important for the one I'm here for (maybe a few homologs are, but not the whole table).
 * Please sign. And the colors are mixing? Probably the IE bug (see above). There should be white borders between all element cells. -DePiep (talk) 10:09, 26 October 2013 (UTC)

3. B. Leave colors as is (detailed colors)

 * 1) I think it's more understandable to the reader if the small table is an exact mini-version of the large table. This way they can get oriented if they want to click somewhere. (There should be a way to keep the cells clearly distinct, so that you can easily pick out one specific transition metal, instead of the cells running into each other like they do now.) Double sharp (talk) 01:42, 15 October 2013 (UTC)
 * Nope. It's nothing for unexperienced reader (i.e. one who doesn't actually know cells are clickable or doesn't care for it), while distracting him from the important one. And still... if you have to look for samarium, you're in trouble anyway, but looking for lead (p-block is the only one to benefit) is much easier. But whatever, it is a minor feature. Orientating the reader goes first.--R8R Gtrs (talk) 14:20, 18 October 2013 (UTC)
 * re R8R: "distracting from the important one"? Which one? Category (metallishnesh) is the third structural fact that is in this PT (after group and period). It is part of the PT core. Even when a reader is not interested in or familiar with the categories, it does no harm. Because: most of our PT's have this same color scheme (and almost all PT's in the world use some color scheme; nothing unexpected). It actually helps orientation, because after associating color with fact (where else?), it shows similar elements wrt category (eg As in metalloids). It is the only link between the expressive background color we use (in title and subtitle bars) and a property. -DePiep (talk) 10:41, 26 October 2013 (UTC)
 * The most important thing this mini PT does is show where our element is. All the rest is secondary. I mean, reading the infobox apart from the PT, you can always identify the category element belongs to if you know our color scheme, and even if you don't, this is still easy, because one of our top lines is "element category". I would be fine if you also wanted to color a new neighboring elements that are chemically related to the one in question with our current colors. Just don't go over the whole table. That table could do better.--R8R Gtrs (talk) 09:52, 2 November 2013 (UTC)


 * 1) Sandbh (talk) 10:24, 15 October 2013 (UTC)
 * 2) Since we use the color scheme over almost all PTs, it is consistent and so good for orientation. -DePiep (talk) 20:30, 25 October 2013 (UTC)
 * Also, after associating color with fact, it shows similar elements wrt category (eg As in the metalloids). It is the only link between the expressive background color we use and a property. It replaces a legend. We do not want the background color to become a frivolous illustration. -DePiep (talk) 10:41, 26 October 2013 (UTC)

Mixed category elements revisited (to subpage)
-DePiep (talk) 12:47, 17 November 2013 (UTC)

Metallicity of nonmetals
I am irritated when I see a periodic table in which selenium is counted as (semi)metalloid, but carbon at phosphorus only as nonmetals. These three elements have the same level of general metallic properties (physical and chemical) - they are most typical "half-metalloids". I propose other marking of some nonmetallic elements: C, P and Se as half-polyatomic nonmetls and half-metalloids, I nd H (due to chemical properties) as half-diatomic nonmetal and half-metalloid, Rn as half-noble gas and half-metalloid (due to chemical properties).

In German Wikipedia is an example of mixed classification:

http://de.wikipedia.org/wiki/Arsen

Or in Latvian:

http://lv.wikipedia.org/wiki/Ars%C4%93ns

Both classification are erroneous due to marking only selenium's metallicity. Metallicity of carbon and phosphorus have to be marked also.

95.49.56.163 (talk) 15:14, 11 November 2013 (UTC)


 * WP ate the previous iteration of this post (which was better). Something really needs to be done about the loading issues on this page. We're adding more discussion faster than we can archive it away!
 * Yes, I am annoyed by that. We don't do that. Though if you want to revamp our classification, I've read through your comments on Talk:Metalloid, and I think you have a good case (though I think you should use "near metalloid" instead of "semimetalloid", as the former is attested while the latter is not).
 * In general it seems as though metallic character has a strong correlation to the difference between the period an element is in and the number of valence electrons it has. Negative differences generally indicate metals, poor or not so poor; differences close to zero usually indicate metalloids; and positive differences generally indicate nonmetals. Metametals and near metalloids seem to occupy the space around −2 to −1 and +1 to +2 respectively. Relativistic effects also reduce the valence core for Po to four electrons most of the time; therefore the values for period 6 and 7 elements are lowered by 2, which corresponds to increased metallic character just as you note for At and Rn. As a result I think E117 may be safely classified as a poor metal, and E118 as perhaps a metalloid. Double sharp (talk) 14:56, 13 November 2013 (UTC)
 * If C, P, Se and I are regarded as near metalloids on the basis that they look like metals (P as black P, the most stable form) then I'd also be inclined to count H as a near metalloid, on the basis that it appears to have a metallic chemistry. I'd still be inclined to treat Rn as a noble gas, using the Au, Pt argument that these are still regarded as metals even though they can form stable mono anions. Would that then leave the fabled category of 'typical nonmetals', being N, O, F, S, Cl and Br? [Not sure about any of this; thinking out loud] Sandbh (talk) 01:50, 14 November 2013 (UTC)

Yes, hydrogen also presents metalloidal behavior chemically, even compound with a metalloid (H2Te) is an acid. Due to chemical behavior Au and Pt should be in the worse group of metals. Elements generally tend to be more metallic physically than chemically (exceptions are H, Rn). There should be a clasification based on summarized level of metallic properties (Be, Pt, Au should be then marked as "worse" due to their chemical behavior, not to the position in the periodic table). Iodine is more near-metalloid than sulfur due to physical properties (such as appearance and conductivity), but sulfur is also an anomaly in comparison to N, Br and Xe (other elements from nitrogen diagonal).

There is a screen with the text "Wikimedia error" after clicking "Save page" button.

194.29.134.246 (talk) 11:28, 14 November 2013 (UTC)

I think that term "near metalloid" is even better than famous "nonmetal" to describe elements such as C, P, Se, eventually I and H. They have some properties very atypical for a nonmetal, for example: very high melting, boiling and sublimation points for C (and thermal conductivity for diamond and graphite, first have also metalloidal hardness and density, second has metalloidal electrical conductivity and apperance (but usually is seen as black and dull, not greyish and shiny), metallic appearance and photoconductivity for Se, strong cationic tendences for H. These elements could be even more properly marked as half-nonmetals and half-metalloids. At can be half-poor metal and half-metalloid. Rn (chemically) and S (physically) are also quite weak nonmetals.

79.191.180.224 (talk) 18:09, 16 November 2013 (UTC)
 * I'm not really a fan of double categorization. Which category is more relevant? I'd rather use the category that gives the most insight on the actual properties of the element in question. Double sharp (talk) 04:05, 18 November 2013 (UTC)

Astatine and element 117
See Talk:Ununoctium, Template talk:Infobox astatine, etc. Basically, the IP is proposing changing our colouring of At as a poor metal, and I argue that on that basis E117 should also be a poor metal. The At poor metal is based on the recent studies showing metallic band structure. I'm quite, but not totally, convinced (Bi is a poor metal, but has a semimetallic band structure). I suspect this is a case of physical and chemical properties conflicting, just like Bi and Cn. By periodic trends then E117 would also have a metallic band structure, and with −1 being an uncommon and not very stable state like in Au, we would have it as a poor metal. Double sharp (talk) 04:56, 11 November 2013 (UTC)

Picture (At, E117 poor metals):

I'm pretty sure about E117, but wavering on At. It will not take much more to convince me one way or the other... Double sharp (talk) 04:56, 11 November 2013 (UTC)
 * There is enough literature to justify At as a metalloid. There is some speculation, and one paper (based on relativistic modelling) predicting At will have the band structure of an FCC metal—a true metal band structure, not that of a semimetal like Bi. I'm not sure speculation plus one paper is enough to warrant showing At as a metal. It may be a metal, in which case it would join the other metals capable of forming mono anions (Pt, Au, Po), but is a 'might be' enough to show it as such, especially for a period 6 element? I would be inclined to leave it as a metalloid but in the article say that there is even some research predicting it will be a metal. Having said that, what do others think? Sandnbh (talk) 07:34, 11 November 2013‎ (UTC)


 * I completely agree.--R8R Gtrs (talk) 20:01, 13 November 2013 (UTC)

Gold and platinum are still counted as transition (good, not weak or poor) metals despite forming monoatomic anions. In the Pauling scale Pt (2,28) and Au (2,54) have also higher electronegativity than At (2,2). Astatine will also have some cationic chemistry. If it is a semimetal (such is Sb) it is more appropiate to count At as metalloid, but formation of true metal FCC structure at ambient conditions for me (definately) will rule astatine out of metalloids.

95.49.56.163 (talk) 14:39, 11 November 2013 (UTC)
 * Double sharp, what conflicts do you see in Bi and Cn? Sandbh (talk) 21:23, 11 November 2013 (UTC)
 * Bi has semimetallic band structure, but chemically is more of a poor metal. Cn is probably a poor metal physically, but a transition metal chemically. Double sharp (talk) 05:37, 12 November 2013 (UTC)

What about melting and boiling points of astatine in the infobox? I do not think that it has so low boiling point if it is a metal (even lower than Hg) and so narrow liquid range (lower than Br and Cl). Thermal conductivity probably is also much higher than about 1,7 - 2 W / m / K. — Preceding unsigned comment added by 194.29.130.244 (talk) 08:29, 12 November 2013 (UTC)
 * In that you are wrong. Experiments show that these values of At's melting and boiling points are actually overestimates, not underestimates. Double sharp (talk) 13:57, 12 November 2013 (UTC)

If it has so low liquid range and boiling point, it probably is diatomic. But metallic allotrope should be stable at normal pressure accoriding to some calculations. Maybe they had a contact with diatomic allotrope? If metallic allotrope is stable at normal pressures, astatine must be named as poor metal. It is too big for a metalloid - having metallic band structure at ordinary pressures.

95.49.94.5 (talk) 00:39, 13 November 2013 (UTC)

Maybe it is better to mark astatine as "unknown", not as "metalloid"?

Some interesting questions are:


 * appearance of astatides - are compounds such as potassium astatide or calcium astatide colorless and soluble in water, such as iodides? What are properties of other astatides? Metal-like or they are salts?


 * properties of compound with hydrogen - is it acidic such as hydrogen iodide, or not? Is it volatile?


 * is it possible that metallic astatine can not exist at ambient pressure?

194.29.134.246 (talk) 11:40, 13 November 2013 (UTC)


 * You're back to questions I can answer, being radiochemical. ;-)
 * AgAt is unstable as the At tends to oxidize. I'm guessing that the only stable astatides would be with the group 1 and group 2 metals: these would, I think, be truly ionic astatides. I think most astatides would not behave this way.
 * AtH is also unstable, and when it dissociates quite often the positive charge ends up on the At atom. But it is a strong acid, even stronger than HI.
 * I don't think that's possible. Double sharp (talk) 12:04, 13 November 2013 (UTC)

AtH or HAt can dissociate at astatide anion? It is very strong acid? Even if, I think that it is still a poor metal (band structure over all). It could be also marked as poor metal and metalloid, such as C, P and Se could be marked as polyatomic nomentals and metalloids. In both cases elements looks more metallic physically than chemically.

If true metallic astatine can exist as an element with metal band structure, then At is unambigously a (very poor) metal, but a metal (which is rather breaking of the trend - true metal structures should be stable at normal conditions only for elements at least from beryllium diagonal (or, obviously, these which lie more to the left)). Metalloidal character is only chemical.

Maybe it is better to place helium (element from block s) above beryllium, not above neon (element from block p)? Hydrogen, helium and (probably) astatine have so small or so big degree of metallicity (according to they position in the periodic table). It will present the breaking of the diagonal trends by H and He, which should be metalloids, but are (especially physically) nonmetals.

What about diatomic astatine? Can it at all exist at normal temperature and pressure? Maybe At has two different allotropes - metallic and metalloidal (diatomic). Maybe metallic At has higher boiling point than diatomic? Maybe only diatomic astatine has been produced?

95.49.248.238 (talk) 20:25, 13 November 2013 (UTC)


 * The evidence for (or against) At2 is sparse and inconclusive. Some sources state that it has not been prepared or even cannot exist, while some assert (or imply) that it has. We are really going into the backwaters of chemistry here, what with At's extreme radioactivity. It is possible that both allotropes are stable at STP, and that the melting and boiling point values were for diatomic astatine.
 * You're looking at it from a more physical perspective. I tend to look at things from a more chemical perspective, which is my main basis for calling At a metalloid. HAt can dissociate as either H+At− or At+H−; while it is a strong acid, this is perhaps because when it's prepared it starts out as the former.
 * Now I'm wondering what you think of copernicium, which is predicted to behave as a semiconductor (band gap ≥0.2 eV), though would chemically be a transition metal.
 * I'm a bit hesitant to support placing He over Be. Physically and chemically it is very hard to justify, apart from electronic configuration. H over Li is easier to justify. I would keep putting He over Ne: the trends at least fit reasonably among the noble gases. Double sharp (talk) 01:19, 14 November 2013 (UTC)

And copernicium may also be a gas in STP... If it is a semiconductor and as gas, it could be a metalloid (if there is no stable metallic allotropes in STP). Cn can be next trendbreaker in PT. What is electronegativity of Cn in Pauling scale? First ionisation energy of Cn (1154.9 kJ·mol−1) is predicted to be higher than of carbon (1st: 1086.5 kJ·mol−1), which is even nonmetallic property. At probably has very inequal properties (chemical and physical). Physically it can be a real metal but chemically it can be poorer than recognised metalloids ("full" metalloids, not such elements like P or Se).

194.29.134.246 (talk) 11:44, 14 November 2013 (UTC)


 * What we know about Cn is that it behaves like a typical group 12 element in that it is very volatile and its adsorption (or is it absorption? I think it's the former) properties are similar to Hg. However, beyond this, it displays very many differences chemically. Now I am thinking if we should explicitly colour it in with a known chemical property (likely metalloid). Because the thing is, we know some of its properties, but not all; with an incomplete picture of what is known alone, we must misleadingly classify it as a transition metal, while with a more complete picture (but relying on unknown but predicted properties), we arrive at a wholly different conclusion.
 * I dunno about the EN of Cn. In fact, I'm not sure if predictions have been published for anything beyond Lr. But Cn is expected to be relatively inert and hard to oxidize or reduce from the neutral state. Additionally the electron affinity should be very low due to the quasi-noble gas closed shell electron configuration of [Rn] 5f14 6d10 7s2. Hopefully this info will help. Double sharp (talk) 15:50, 17 November 2013 (UTC)

It is interesting if astatine can form diatomic molecule at STP. If diatomic molecule and metallic allotrope are stable at STP, it will be very curious situation.

194.29.134.246 (talk) 12:43, 18 November 2013 (UTC)


 * Personally I would expect this table to reflect observed physical properties, with studies to rationalise these. If the physical properties of an element are not known because insufficient material has been made and all that is available to justify any categorisation is just a study(!!) then leave it blank. Studies without evidence are just speculation.Axiosaurus (talk) 18:56, 18 November 2013 (UTC)
 * My logic was always that since the "unknown" category said "unknown chemical properties", those were the primary. In which case At is rather more clearly a metalloid. Physically the properties are unknown, which is why I am so hesitant to give it a poor metal (post-transition metal) classification. Double sharp (talk) 07:26, 19 November 2013 (UTC)

Infobox fontsize and more
I have changed the fontsize in the infobox. It now uses the fontsize as a regular infobox. In effect, the font is smaller. Just to note: the Fl box is 6 regular textlines smaller (6/114=-5%), mostly for reduced needs for wrapping (e.g., in the isotope box). I can note that this improvement was not induced by anyone. Nor are the next two.
 * In this department (general layout), two other options are available:
 * We can remove the bolding from the data names, left hand column. (demo on request; few lines will unwrap; box 2 lines smaller). See German Fluor.
 * We can remove the borders: This is standard infobox layout.
 * See Infobox element/sandbox3/testcases.
 * The frwiki has this: Fluor.

The options can be chosen independently (any one or both). Ideas? -DePiep (talk) 11:34, 12 November 2013 (UTC)

I'd love to see the remove the bolding version. Also, as an idea, could we move the degree values of structure pics from above the structure itself to the left/right from it? That would save a little space and look better IMO.--R8R Gtrs (talk) 19:35, 17 November 2013 (UTC)
 * Eh, "degree values of structure pics"? -DePiep (talk) 19:53, 17 November 2013 (UTC)


 * That should be "angle values" (e.g. α = 90°). I'll go get me some sleep.--R8R Gtrs (talk) 20:45, 17 November 2013 (UTC)
 * Agree the text on these crystal graphs could be better. That is: because it is just footnote information, not image title. But not for space reasons, they better be larger to show their essence. -DePiep (talk) 07:44, 19 November 2013 (UTC)

Nickel electron configuration
Yea, now we shall verily have that hair-splitting section that I promised unto you some sections ago. Turn thither (the end of the section "Why do we use weird names like unbiseptium? [sic] (part two)") if thou willst know more about how this idea of mine started.

Seriously though, should we label the electron configuration of Ni as [Ar] 4s2 3d8 or [Ar] 4s1 3d9? See Nickel. And I'll quote that section below, as it's short:

Double sharp (talk) 15:14, 8 November 2013 (UTC)
 * Bump. Double sharp (talk) 02:31, 25 November 2013 (UTC)

Caesium
There is a new addition to the article that nodeling shows under 30 Gigaopascal CsF5 could exist. Do we really need this? I was reverted one time my the user but I still think it should go.--Stone (talk) 21:58, 23 November 2013 (UTC)
 * As a footnote would do, after the currently non-existent content (in the article in question) about caesium fluoride. Sandbh (talk) 00:52, 24 November 2013 (UTC)
 * This content is now in note 4, qualifying the statement that Cs's only common oxidation state is +1. If these predictions do get realized, then Cs will be known in all oxidation states from −2 to +6 (really, no CsF7 despite XeF6 and Cs here behaving as a supposed seventh 5p element?)
 * I suspect the reason why the caesium halides aren't covered by the article (with the exception of CsCl) is that they're quite conventional alkali halides. Though CsF certainly deserves a mention, being useful as a fluorinating agent. Double sharp (talk) 07:14, 25 November 2013 (UTC)
 * I've cobbled together a short paragraph on CsF. Not finding very much to say about CsBr and CsI that isn't either (1) general to every alkali halide or (2) too specific to the individual compounds such that they fit better there than at the main Cs article... Double sharp (talk) 07:22, 25 November 2013 (UTC)

Make the group 12 elements poor metals? (continued)


Latest version of proposed periodic table, can be edited. Here is how it looks (extended):


 * Edit: I have added another period 7, above "7e". It should display the colors we use when not extended. (in most PTs we color the "predicted" cells grey, not a "x (predicted)" lighter shade).
 * We can use lightgrey to illustrate. -DePiep (talk) 08:50, 24 November 2013 (UTC)

Do we go ahead and make this change now, or do we wait for the rest of the discussion to avoid a "poor metal" backlash? I do want a more correct placement for Zn, Cd, and Hg. Double sharp (talk) 13:46, 11 November 2013 (UTC)
 * Don't know about "poor metal" fallout (i.e., reopening content discussion afterwards). For me, it is about having to revisit all PT templates and images twice for the edits. So I am waiting for the other discussion(s) below to finalize. I note that each decision must have proposed Article text(s) with sources as a base, say in a sandbox. Did I move one to the archive ;-)? Maybe writing a formal proposal "Option X" could make the fat lady sing. -DePiep (talk) 15:45, 11 November 2013 (UTC)


 * Situation so far (Nov 14): made a draft PM article text in his User:Sandbh/sandbox (last major edit: 6 Oct 2013). Acceptance would make this article 2.0 a base for group 12/PM changes everywhere (PT above is just an reflection of that). Its discussion at archive #Poor_metal_article_2.0 showed opposition by  ("poor idea"). Last edit: Oct 1, 2013. So the proposal has not reached a consensus, and talks are dead. Unless I missed something in the earlier subsections in archive, this is the situation wrt group 12/PM. -DePiep (talk) 09:14, 14 November 2013 (UTC)
 * As I see it, the talk about poor metals is...having a kip/pushing up zzz's. I tried to follow up with Axiosaurus on his talk page but rec'd no response. I believe he was expecting the change to poor metals anyhow. I've let the talks go into snooze mode, out of respect for his opinions while I was working through the options. RL has been a factor too. I expect to have time to revisit this over the end of year break if not before. I presume the delay won't be an issue? I'm also keen to have a more correct placement for Zn, Cd and Hg, and to avoid having to revisit all PT templates and images twice. Sandbh (talk) 09:33, 14 November 2013 (UTC)
 * Picking this up later should be no problem. As it is no cut-and-dried thing, we canot force a decision. -DePiep (talk) 15:37, 19 November 2013 (UTC)
 * About the only thing we seem to have a consensus on is that Zn, Cd, and Hg aren't transition metals and should not be counted so. What exactly should they be put into? Therein lies the problem. I suspect we're not getting a "poor metal" backlash yet simply because it's not yet in mainspace. Which leads me to think that maybe we should simultaneously roll out "pre-transition metal" and "rare earth metal" (the latter if only for quelling the dispute on whether group 3 is best classified with the main group elements or the transition metals without a need for a mixed categorization). Double sharp (talk) 04:15, 20 November 2013 (UTC)
 * I'm been admiring this periodic table, thinking how elegant it looks, not just aesthetically but also in terms of the tidiness of its taxonomical categories. Upon further reflection I think that the way the rare earths are shown above is a (very) clever solution to the group 3 mixed category dilemma. Sandbh (talk) 21:32, 20 November 2013 (UTC)
 * Not because it was originally your idea, I hope, before I appropriated it? ;-) Seriously I also think it's a good idea. It sidesteps the whole are-group-3-elements-TMs-or-not debate by simply not taking sides! The only danger is fallout from the term "poor metal", but I think it's useful enough and reasonably defined in your sandbox. So, consider this a go-ahead from me. Double sharp (talk) 02:19, 21 November 2013 (UTC)
 * I'll see if I can do a summary of the arguments that have resulted in this conclusion. It's helpful that the exclusion of Group 12 from the TMs is IUPAC compliant. Also that 'rare earth metals' (Sc, Y and the lanthanides) is an IUPAC approved name. Sandbh (talk) 06:12, 22 November 2013 (UTC)
 * Please do so, Sandbh. I'd prefer a proposal, not a recap of arguments (to get a conclusion).
 * Tech: in the PT above, I have added period 7 once more, to have it colord for not-extemnded (se enote there) -DePiep (talk) 08:50, 24 November 2013 (UTC)
 * I'm glad to have you two split share five-pointed barnstar for the idea, as long as you remember that you used the 32-column periodic table to explain & convey. :-) -DePiep (talk) 09:10, 24 November 2013 (UTC)
 * I understand the core referenced text proposals are in (1) User:Sandbh/sandbox for the group 12 outcome, and (2) in Rare earth metal (?) for the Ln/REM change. Am I right, and does this covers all?
 * We might expect fallout too about this: "What? No lanthanides in the PT? While actinides are in?" Such a fallout is not about a wrong, but more about expectations so less problematic than PM/PTM issues. To pull REMs more apart from Lns, I think it better to give REM a different color (not the Ln purple). That would be a much bigger change (more & bigger edits to be done in one go after this concludes; more like the option-10 change last August). You think it worth? -DePiep (talk) 09:10, 24 November 2013 (UTC)
 * I've added color option "rare earth metal" = to the available categories. It is almost the same color as Lanthanide pink: . Almost: the last difit changed by one, so that we can my mousepointer can detect (show) on a page which cells have been transformed correctly. (IOW, my mousepointer can show the color number when hovering; with this minorly different color number I can do checks without a visual change for the reader). -DePiep (talk) 11:07, 27 November 2013 (UTC)

On the question of group 3
@Sandbh: after thinking about the edges of the d-block a little, I now have a question for you. :-) In what ways would you say group 3 is transition metal–like, and in what ways are they not? They don't form ions with d electrons in them, so transition metal character could, to my mind, be weakened. (If it is weakened enough to warrant a classification of group 3 as not being transition metals, we could again consider your proposal to group Sc, Y, and Ln as the rare earth metals and An as the actinides!) Double sharp (talk) 12:54, 1 October 2013 (UTC)
 * (P.S. To my mind they are still transition metals, still having a fair number of the characteristic properties.) Double sharp (talk) 13:54, 1 October 2013 (UTC)
 * (P.P.S. Have not completely made up my mind on this issue, and would like to hear your comments. Here is an extended PT example – read "rare earth metal" for "lanthanide".) Double sharp (talk) 15:03, 2 October 2013 (UTC)

Sc, Y and Lu are physically transition metals. Their chemistry is largely (but not exclusively) not that of transition metals. Physical and chemical properties together would make them (marginal) transition metals. So, I agree with your initial assessment. Lr depends on its electron config. 'Should' be [Rn]7s25f146d1; but quantum mechanical research suggests [Rn]7s25f147p1. That would make Lr a poor metal. Sandbh (talk) 02:59, 3 October 2013 (UTC)
 * Proposed periodic table here -- To speed up page load, the example template is moved to subpage Wikipedia talk:WikiProject Elements/On the question of group 3. You can edit there too, as part of this discussion. When the talkpage is shrunk (by archiving), it could be put back here, alive. -DePiep (talk) 08:19, 26 October 2013 (UTC)
 * (After some thinking about this...) What I am not completely at ease with is that the colouring suggests that Sc and Y are transition metals, but Lu is not, despite considerable homology of Sc and Y to the lanthanides (including Lu!). That is what makes me think it might be better to include them into our current pink-coloured region and change the "lanthanides" category into "REMs". Furthermore, due to the lanthanide contraction, Y tends to act in terms of physical properties and chemical reactivity as though it were actually a lanthanide between Tb and Dy! These are very close similarities – Y is closer to the lanthanides than to its immediate left and right neighbours on the periodic table, even correcting for different valences – that this colouring obscures. As for Sc, its being slightly larger than the other transition metals results in it behaving significantly differently from them: its complexes have higher coordination numbers (e.g. 7 in [Sc(H2O)7]3+) and it tends to act like a smaller version of Lu. Even the coordinate chemistry of Sc and Y – the very property that the Zn group share with the transition metals – is often discussed in relation to that of the lanthanides rather than to that of the transition metals. I'd submit that labelling group 3 as rare earth metals with the lanthanides reflects their chemical and physical properties far better than labelling them transition metals: they may qualify as "marginal" transition metals, but surely that's not their main category. (Here's my source for this.)
 * I'm not sure that just having p valence electrons makes an element a poor metal. Element 113 should have valence electron configuration 6d10 7p$1 1/2$, and is expected to act like a group 13 element except when it's in the +5 state (which is really just in one fluorine compound and its analogous polyfluoride ion), not a transition metal. So Lr could conceivably be an actinide. Also, the 7p1 configuration is just a suggestion. Lr (s2p) should be less volatile than Lr (s2d), and have comparable volatility to Pb, and indeed experiments suggest that Lr is non-volatile; on the other hand, its enthalpy of absorption is much higher than what theoretical calculations predict for Lr (s2p). The Chemistry of the Actinide and Transactinide Elements (3rd ed.) gives Lr's electron configuration as 5f14 6d1 (7p1) 7s1, so I would submit that there is still a fair amount of doubt surrounding the actual ground-state electron configuration of Lr. (I do also note that the s2p configuration is supported by Eliav et al's 1995 CCSD calculations, which gave s2d2 for Rf after some doubt, so while the jury is still out there, I personally bet on s2p.) Chemically Lr behaves like your average trivalent actinide, forming a trivalent ion in aqueous solution and extracting into the organic phase over the pH range of 3+ ions with Cf and Fm. I don't think the (admittedly limited) evidence is strong enough to justify labelling it as a poor metal. Double sharp (talk) 08:56, 3 October 2013 (UTC)
 * Still thinking about all of this. Difficult question. Rare earth metals is attractive in some ways however I need to think through all of the options/do more reading etc. Sandbh (talk) 01:10, 4 October 2013 (UTC)
 * Yes, I support your suggestion to categorise Sc, Y and lanthanides as rare earths. First two are currently shown as TM but chemistry is largely atypical of TM. Sc is said to be intermediate in properties between Al and Y. Yttrium plus the lanthanides are regularly called the rare earths. Sc, due to its intemediate position, sometimes is or isn't included with the rare earths. It's smaller than the other rare earths so it's the least basic, and atypical in some other respects. What to to? Ignoring electron configurations, I had a look at 25 physio-chemical properties that I could find data for, on Al, Sc, Y, La and Lu. These properties were: density; hardness (Brinell); ductility; electrical conductivity; thermal conductivity; superconductivity at normal pressure; thermal expansion; crystalline structure; heat of atomization; solubility of H2; ionic radius; nature of hydride; melting point; boiling point; ultimate tensile strength; electronegativity (revised Pauling); electronegativity (Allred-Rochow); 1st ionization energy; sum of 2nd and 3rd ionization energies; electrode potential; composition of aqua-cation; oxide melting point; oxide structure; chloride melting point; and chloride structure. I assigned from 1 to 4 points to one of Al, Y, La and Lu, with 4 points going to the metal that scandium was closest to in that property, 3 points to the next closest, and so on. Results were: Al (42.5); Y (80); La (48); Lu (79.5). This supports the notion that Sc is closer to Y than Al, and hence merits the inclusion of Sc with the rare earths. In doing my reading I found it odd that some authors would call attention to the atypical behaviour of Sc as a reason to question its membership of the rare earths whereas none quibbled about e.g. amphoteric and covalent-compound-forming Be being an alkaline earth metal, or the other first-row elements being members of their respective groups, despite their anomalies. Sc seems to have gotten the short-end of the stick, for no good reason that I could discern from the literature apart from a suggestion that its chemistry is the least studied of the 3d-block elements.


 * Next question: Which camp does Lr fall into (actinide or rare earth, if not poor) and why? Sandbh (talk) 06:55, 6 October 2013 (UTC)
 * It seems we have two questions here: whether the actinides are rare earth elements, and how to classify Lr, the last actinide, in particular.
 * On the first question, I would say that the actinides are not rare earth elements. The (at least early) actinides' ability to reach much higher oxidation states than the lanthanides causes them to behave more like the transition metals than your average rare earth element. There also seems to be 5f/6d/7s/7p overlapping, especially in the U–Am region; this is not something the lanthanides have to the same extent. Actinides form covalent complexes more readily than lanthanides, including with π-bonding ligands. And their electronic structures are usually unclear – their atomic spectra are very difficult to interpret, their electron configurations can vary depending on whether you are considering the pure element or a compound, and it is very hard – often impossible – to say precisely what orbitals are being used in bonding. Now I do realize that most of this really applies mostly to the early actinides (until about Am or Cm), but they behave so little like their corresponding lanthanides that I feel a separate classification is warranted and necessitated. And for the late actinides, they are divalent metals from Es onwards, except Lr (whereas I think only Eu and Yb are divalent in the metallic state among the lanthanides).
 * Lr seems to be very closely related to Lu, as expected from group relationships (so are Sc and Y), but it also continues the elution sequence of the trivalent actinides, appearing just after Md (No is divalent). So I think an actinide classification should be given, but much mention should be made of its close kinship with Lu, visible in the metallic state (similar enthalpy of sublimation and atomic volume) and its having +3 as its only stable state in aqueous solution and solids. So I would say actinide. Also, IUPAC has all but defined the actinides as elements 89–103 on its official periodic table (one of the only two categories they choose to explicitly define there, the other naturally being the lanthanides), so I would not be comfortable with calling Lr a non-actinide. Double sharp (talk) 14:20, 7 October 2013 (UTC)

Group 3 in periodic table
I've mentioned this before (in 2011), but there seems to be an inconsistency between the ways we present our 18-column and 32-column periodic tables. For the 18-column periodic table places all the lanthanides and actinides under Sc and Y, whereas the 32-column one (correctly IMHO) places only Lu and Lr in that position.

So I propose we return to the old (pre-2005 or thereabouts?) version of the PT on WP, with Lu and Lr in the main body of the periodic table. As usual, read "rare earth metals" for "lanthanides":
 * See also: periodic table/sandbox, as it would look. -DePiep (talk) 14:00, 8 October 2013 (UTC)

Potential objections:


 * (1) Confusing: "lanthanide" and "actinide" labels before the pulled-out rows (57–70 and 89–102) do not correspond precisely to the real meaning of these terms. Response: True; this is a real issue. (Ideas for solutions?)
 * (2) Original research: IUPAC recommends no particular placement for group 3, so we should not take sides. Response: The depth of the sources – and the chemical properties of the elements in question, as seen in reliable sources – is firmly on the side of Lu and Lr. (See Jensen especially.)

What do you think? Double sharp (talk) 11:08, 8 October 2013 (UTC)
 * Tech note: introduced the adjusted legend Periodic table legend/Category/sandbox. That sandbox can be edited. -DePiep (talk) 11:58, 8 October 2013 (UTC)
 * re (1): We can not let these imprecisions stand. (same with "inner transition metal" in the legend). Solution step one: remove the words Lanthanide and Actinide. Simple and correct. Step 2: Do Lans and Acts need to be pointed out? We have another partially overlapping categories: poor metals and post-transition metals (partially, so not interchangeable names). If we need to mark them indeed, we can add another group marking for this (not using any bg color). e.g. box outline. Or move write the links in the two asterisk-boxes. I'll do this in my next edit (so it's easy to revert). -DePiep (talk) 12:10, 8 October 2013 (UTC)
 * Step 1 was a good call (the original was fraught with too many dangers), and what I thought of myself as an "emergency" first solution to make it correct. Step 2: I personally do not think the lanthanides and actinides (standard abbreviations: Ln and An respectively) need special marking – after all, we don't have special marking for any other groups in the PT, and I suspect they only had them because it was easy to show (not a good rationale).
 * On "inner transition metals", now that we have changed Ln to REMs (rare earth metals), it is technically not correct anymore as REM includes Sc and Y but inner transition metals does not (only including Ln and An). So we can get rid of it now. Hooray. (It was the only non-coloured top-level category.) Double sharp (talk) 12:34, 8 October 2013 (UTC)
 * "inner transition metal" is gone. Consider this: we will remove the words & links Lanthanides and Actinides everywhere (all PT templates and images and category overviews). Only in dedicated locations they can stay. That is: when they are topical in text.
 * About re-using the lanthanide (pink) bg color for REM I have my doubts. We do not want any confusion or mixup between the Ln and REM category identifications. The reader better be informed by a serious color change (into say something green). Also it could send a confusing message to other wikis. It's a hell of a change then. First we'll see how the content proposals here settle down (element category groupings). -DePiep (talk) 14:00, 8 October 2013 (UTC)
 * Yes indeed, but it does show the close relation to violet actinides. Maybe a noticeably different shade of a pinkish or purplish colour, to keep this relationship but remove confusion? Double sharp (talk) 14:35, 8 October 2013 (UTC)
 * We do not use or promise associations by color (with reasons). To tear us away from this suggestion, I'll use green here for REM, next edit. -DePiep (talk) 01:57, 12 October 2013 (UTC)
 * Nice shade of green, also suggestive of the f-block WebElements colour. I like it. And it's also quite distinguishable from the lighter "Polyatomic nonmetals" colour (although as a precaution we might want to make that even lighter.) Double sharp (talk) 04:41, 12 October 2013 (UTC)
 * P.S. on your edit summary: all the lanthanides are rare earth metals. The former is a subset of the latter. Kind of like the poor metals vs its subset post-transition metals. (And shouldn't we really change the colour for that too for option 17?) Double sharp (talk) 05:54, 12 October 2013 (UTC)
 * I understand the overlap, but I think we should not keep referring to that except in specialized topics (like lanthanide). It would be very confusing to the reader to mix them before separating them. If we want to mark the lanthanides too in a PT, we need an extra graphic form (e.g, a box line around them).
 * The green is not the definitive proposed color per se, I just needed a color that brutally breaks us away from lanthanide thinking. (btw it is simply "G" in the talkpage-simplified PT; I think it is a bit too bright). -DePiep (talk) 08:55, 12 October 2013 (UTC)
 * Then again when we redrew the boundaries in the p-block ("other nonmetals" and "halogens" becoming our new polyatomic/diatomic classification) we didn't change the colours, just the shades of the colours. And it could be argued that (1) if you've seen our PT before, you'll notice our change in colouring Sc and Y and will look in the legend and (2) if you haven't seen it before, of course you will read the legend. And our classification would be mentioned in text, even if we just used another shade of pink.
 * Green tends to collide with either polyatomic nonmetals if it is bright or eka-superactinides if it is too dark. We could change the latter far more easily. One problem is that we are almost exhausting the entire space of easily distinguished colour categories (black and white are disallowed: then we have red, two greens(!), yellow, blue, brown, orange, pink, purple, grey, and cyan). One of option 17's virtues is that it frees up one colour category... Double sharp (talk) 09:28, 12 October 2013 (UTC)
 * About colors (not about cat definitions)
 * Main point: a differently defined category must have a different color (please read this again, or --threat-- I'll make it bold).
 * Therefor, for example, the diatomic/polyatomic cats may be a green and a yellow again indeed, but they are different colors (to me at least, and by number). We have a 1:1 relation between cat-id and color-id. That's good (this is why we changed the wiki-commons name for these elements: "polyatomic" green is not "other nonmetal" green).
 * Interestingly, if a cat changes content, the color can stay the same. This happens in the group 12 changes above: elements switch color from TM (pink) to PM (grey). But no cat definition changed --> same colors used! (this is why the group 12 change is relatively simple in edits-to-do).
 * Note that we better not start using "associations" with the colors. That will tie us up even more, reducing out freedom of choice into impossibles (too many contradicting requirements).
 * This about: differently defined categories must have a different colors (here, I made you read it again without bolding).
 * Now up to the next step. That is: our color scheme is not fit for the purpose. The ten colors we use are not the best chosen set. Today we use five reds, a grey and a brown, and the three other distinctive ones once (green, yellow, blue). We better redesign the whole palette from scratch - tacitly. Actually this is what I intend to do. But this cannot be done as long as cat definitions are discussed (as is happening now for REM, Ln, AM, AEM). So once the new cats are stable, I will start that talk ("propose change cat-x=color-3 to cat-x=color-5; times ten").
 * This means any color change is temporally. Change 20 elements to green for a few months?
 * Now this consequence: since we cannot change the color from lanthanide-pink to REM-green easily, I propose to use the old lanthanide-pink for the REM category (as it was some days ago). Once stabilized after the content edits, we can overhaul the whole cat color spread. -DePiep (talk) 19:56, 13 October 2013 (UTC)
 * OK, I like this suggestion. So we resolve all this content, come up with a fixed categorization (still using existing colours), and then completely overhaul the colours we use for those categories. Double sharp (talk) 03:46, 14 October 2013 (UTC)


 * Wait, wait. This has nothing to do with the group 12 topic. It should be a separate section (like == group 3 == top level, or so. Not four ==== deep). We better reorganize this first. -DePiep (talk) 20:27, 9 October 2013 (UTC)
 * Indeed, and I have now responded to your call. Double sharp (talk) 11:19, 10 October 2013 (UTC)
 * ✅ Point solved. -DePiep (talk) 18:33, 12 October 2013 (UTC)

I like the current 18-column version, for the reason that it has no 'annoying' gap between the s- and d-blocks. Sandbh (talk) 11:01, 15 October 2013 (UTC)
 * But it's still inconsistent with the long table. The long table shows Sc/Y/Lu/Lr as one group, whereas the medium one shows Sc/Y/*/**. The only way you can translate that directly to a long table is putting super-stretched Sc and Y cells above all the lanthanides and actinides, and that looks odd. Besides the Sc-Y-Lu choice makes more sense as it follows the trends in the rest of the early and middle transition metal groups (Ti-Zr-Hf, V-Nb-Ta, Cr-Mo-W, Mn-Tc-Re, Fe-Ru-Os, Co-Rh-Ir, Ni-Pd-Pt). (And we can make the gap narrower. DePiep?) Double sharp (talk) 11:07, 15 October 2013 (UTC)
 * Re the gap column: cannot be different from other column widths. Because, we force every column to be equal in width (independent of text/content). That option has no exceptions. -DePiep (talk) 15:46, 26 October 2013 (UTC)
 * Oops, yes we can! Must be set in very first row. -DePiep (talk) 16:21, 26 October 2013 (UTC)
 * ... trouble in this paradise of elements: such a column exception must be set in the very first row. In most of our PTs, that is a full-width title. Will require some further experiments. -DePiep (talk) 06:09, 2 November 2013 (UTC)
 * re Sandbh: an annoying gap? It is there for a reason: the 18-column PT is derived from a 32-column PT always -- not the other way around (is why we are having these talks repeatedly). Once you take a block out, there better be a visual clue where to put it back. (Or do not show them below at all; then group 2 and 3(Sc/Y/Lu/Lr) columns can be glued indeed, without distortion of information). btw, the German wiki positions Sc/Y tightly next to group 2 Scandium.
 * In general: first solve the real issues in a 32-col graph, and then come back to see if it can be spolit into an derived (distorted, complicated, unnecessary) 18-col graphic variant. I pity the persons who have to explain these core issues in an 18-col table. -DePiep (talk) 14:20, 14 November 2013 (UTC)

Intermediate, singular proposal
I want to propose a change that does not interfere with the rest of this topic. If we say Yes, no other issues are affected or preconditioned into something. Here it is:
 * Propose to remove the category subgroup name inner transition metal from the categorisation (-legend).
 * Though the subgrouping may be correct, it is too much a detail, visible in the legend only at that. It is not reflected in the Table itself!
 * As for content, no category names will change through this (no REM introduced/no Ln removed &tc).
 * PTs and legends in this talk thread already show the removal.
 * Technically, since it is not used outside of the legend box intself, a few edits will do and no categories will have to change. The topic Inner transition metal can be in a specialised article (as the Redirect points already). -DePiep (talk) 11:53, 17 November 2013 (UTC)
 * Sure, get rid of it. (We'll have to do it sooner or later, because Sc and Y are not inner transition metals. So, to my mind, we might as well do it now.) Double sharp (talk) 05:33, 21 November 2013 (UTC)
 * Support. Sandbh (talk) 08:31, 21 November 2013 (UTC)
 * ✅ some days ago. DePiep (talk) 04:21, 1 December 2013 (UTC)

Should lanthanum be in group 3?
Copied from User talk:R8R Gtrs

I've been thinking about it yet again. ;-)

It is a really tough decision. Group 3 certainly acts like a main group in that its elements almost exclusively form the +3 state. The choice of Sc/Y/Lu/Lr is closer to the transition metal groups; that of Sc/Y/La/Ac is closer to the main groups (specifically 1 and 2). As you know I was originally a Sc/Y/La/Ac proponent; you and Jensen may have converted me, but that's not going to stop me from reëxamining the matter!

I'm beginning to feel that the group actually bifurcates into 3A (Sc/Y/La/Ac) and 3B (Sc/Y/Lu/Lr), just like how group 2 bifurcates into 2A (Be/Mg/Ca/Sr/Ba/Ra) and 2B (Be/Mg/Zn/Cd/Hg/Cn). Parallels may be drawn with how the d- and f-block insertions in 2B cause kinks in the trend in density there, whereas the absence of them at 2A cause a smooth trend in melting point like you have in group 1; similarly the f-block insertion in 3B causes some trends to go wrong (e.g. density) and some to go right (e.g. melting point). I may make a huge table like I did at group 12 element to illustrate this. T Jensen's main argument for choosing 3B as the real group 3 is that group 3 should be counted as part of the transition metals. This is not a universal definition, and I do not agree with it because of their distinct physical and chemical properties: I consider the transition metals to only span groups 4–11 (4–7 early; 8–10 middle; 11 late). And while 3B fits the trends better in groups 4–10, 11 bucks some of the trends, and 3A shows a kinship with 1 and 2A.

I'm not too sure where B and Al fit into this. Maybe there are two bifurcations, into 3 (B/Al/Sc) and 3' (B/Al/Ga), and then into 3A (B/Al/Sc/Y/La) and 3B (B/Al/Sc/Y/Lu)? Off to do some more reading on the matter. Double sharp (talk) 05:59, 25 October 2013 (UTC)


 * This is a bit off-topic, but your use of the term 'kinks' got me to thinking ... would there be a way of group trends graphically in such a way that these kinks and bifurcations actually showed up visually? Not sure just how to make it work, but if it could be pulled off well it would be a thing of beauty. YBG (talk) 06:33, 25 October 2013 (UTC)


 * Jensen did it on his paper in group 12, I think, showing 2A vs. 2B. We also have an example (group 13 and 14 atomic radii, thanks to the d-block contraction) at File:D-block contraction--EN.png. The kinks are at periods 4 (d-block) and 6 (d-block and lanthanide). Double sharp (talk) 11:51, 25 October 2013 (UTC)
 * Aren't you talking about something like the Zmaczynski & Bayley periodic table? Sandbh (talk) 12:00, 25 October 2013 (UTC)
 * Kind of. Here's what I think of its bifurcations. I'm taking the standard 18-column PT as a default, and hence only address the bifurcations it does not show.
 * IMHO, the primary bifurcations are in groups 2 (s/d-block) and 3 (d/f-block). I think these are the ones where we cannot ignore the bifurcation, as chemical and physical properties do not favour one choice over the other.
 * The others (C/Si/Ti, etc.) are quite distinct chemically and don't fit so well together in a group. These are the secondary bifurcations. I think they're called "isodonors" in Bayley periodic table terminology: they are the same number of columns away from the left of the periodic table and hence donate (or share) the same number of electrons when forming bonds (e.g. Al3+ and Sc3+; HClO4 and HMnO4). This doesn't carry well into the lanthanides past cerium, but it does work for the early actinides too (e.g. tungstic acid H2WO4 and uranic acid H2UO4).
 * Then you have the oddballs like Li/Na/Cu and Ca/Sr/Yb. These are, I think, called "isoacceptors". They are the same number of columns from the right of the periodic table, and hence (supposedly) accept the same number of electrons. But TBH it doesn't work too well given that these are all metals and hence are predisposed to forming cations. The nonmetallic examples are already in the same group as the halogens, chalcogens, and pnictogens. Only the former s/d-block example makes any chemical sense, I think, and it is the only one shown by Bayley.
 * Due to relativistic effects, it would also be prudent to show similarities like Pt vs. the chalcogens, Au vs. the halogens, and Hg vs. the noble gases. These are isoacceptors (cf. Ba2+Pt2− and Cs+Au−), although Bayley won't tell you that! This arises from the relativistic stabilization of the 6s subshell. Double sharp (talk) 14:18, 25 October 2013 (UTC)

This version:

Double sharp (talk) 15:28, 30 October 2013 (UTC)
 * This discussion is pointless. There is no one PT that will satisfy all criteria. La is both a group 3 element and a lanthanide: group 3 because of the electronic structure of the atom, lanthanide because its chemistry fits in with the chemistry of the other lanthanides in the +3 oxidation state . Petergans (talk) 13:54, 31 October 2013 (UTC)
 * Yes, it is a lanthanide. We're not disputing that. (It has been disputed elsewhere, but we all seem to agree here that it is.) Whether La is in group 3 is disputable: electronic structure is equally an argument for Lu (cf. Al over Ga vs. Al over Sc). And that is what we are discussing: whether we should primarily show the Y–La relationship or the Y–Lu relationship as group 3. Both have reasons in their favour, and we want to use the one that will provide the most insight on the properties of the elements in question. I'm still thinking about it, like we have on the pre- and post-transition elements.
 * (Also: I agree with you that there is no ideal PT in the sense that it satisfies all physicochemical and other criteria. I'm trying to find one that will satisfy as many as possible ;-). And yes, I am a Bayley periodic table fan.) Double sharp (talk) 14:40, 31 October 2013 (UTC)

Something fishy is going on. Why rare earths? They are confusing (see actinides below) and confusing (Sc and Y are normally considered TMs outside Wiki, but not in this scheme).--R8R Gtrs (talk) 12:15, 2 November 2013 (UTC)
 * The chemistry and physics of Sc and Y are quite distinct from those of the transition metals and they share much more in common with the lanthanides, especially Y. Both terms are used by chemists simply because Sc and Y are so closely related to their heavier cousins the lanthanides. The actinides are much more variable in terms of physical properties and oxidation states and, to my mind, warrant a separate classification. Additionally, no one calls them rare earths: the closest I've seen is "radioactive rare earths" in a thoroughly non-technical popular science book that I can't remember the title of, and it seems like shoehorning. Indeed we lose the parallel between the lanthanides and actinides, but that is to my mind a small loss.
 * It is not unheard of to classify Sc and Y as not being transition metals, and according to this guy (Jim Clark of chemguide.com), a majority of UK-based chemistry syllabi use the definition that requires transition metals to form stable ions (not atoms, as in the IUPAC definition) with d electrons. By this definition, Sc and Y are not transition metals, always forming the Sc3+ and Y3+ ions with noble gas configurations. Whereas Ti, for example, doesn't behave very much like a transition metal ion when as Ti4+, but has another reasonably stable state in the form of Ti3+ and in this state it exhibits transition metal-like behaviour (cf. the violet colour of TiCl3 solution). Group 3 elements, on the other hand, don't have lower oxidation states that are stable alone in solution. Granted you have Sc(II), but that's only stable in salts like CsScCl3, and Sc–Sc bonding is involved here, so this is more like Ga(II) in GaS than Ti(III) in TiCl3, which has no Ti–Ti bonds (though ZrCl3 does have Zr–Zr bonds, due to the larger size of the Zr atom). Also, with this definition, we don't need to stripe to be correct. :-) Double sharp (talk) 17:07, 2 November 2013 (UTC)


 * Fine, you may be right with physics (it occurs to me it's all just a gradient anyway. when going red to green you pass through something distantly reminding of grey), but okay. Also note Zr(II) and Hf(II) normally, if not always (I didn't do a very hard job to make sure there's no exception, but the rule normally holds), feature M-M bonds. Zr(III) and Hf(III) are too reactive for water... not too stable (also commonly, if not always, feature weak M-M bonds, notably in halides (I've consulted Greenwood)). Think for a sec about depriving them from the TM status :) And of course, actinides shouldn't be "radioactive REMs" :) (re small loss. maybe, but then REMs have to have a non-pink color. Brains associate Sc and Y with TMs per geometry of the categories (how unscientific that must be sounding))


 * I know about this opposition who don't want Sc among TMs. But whatever. Most of our readers don't come from UK. It occurs to me that having Sc and Y among TMs is more common today and not. And my fair belief is, this reason alone is enough to justify things they way they are today. OTOH, if you can prove me wrong, that would be enough to legitimatize the proposed scheme. I believe Wiki should represent human knowledge the way it currently is, not making it up. Before Wiki, I would be really surprised not to see Sc and Y among TMs. If there are more people amazed now seeing them in than people who would be if we had them out, though, sure.--R8R Gtrs (talk) 22:34, 2 November 2013 (UTC)


 * You are right about lower oxidation states of Zr and Hf, but that can be explained by (1) their larger atomic radius, so that they can form M–M bonds where Ti cannot in its analogous compounds (the Ti atoms are too far away from each other), and (2) the general tendency in the transition metals for the higher oxidation state to become more stable on descending the group. In groups 4 and 5, the chemistry of the second- and third-row TMs are absolutely dominated by the higher oxidation state (Ti is dominated by +4, but V is dominated by lower oxidation states); this is no longer true in group 6 (+4 and +5 begin to be important for Mo and W), and by groups 7 and 8 the group oxidation state is easily reduced. In groups 9 and 10 you don't even see the group oxidation state (well, maybe in Ir and Mt, though as yet unknown), and even the +6 state is extraordinarily oxidizing in the hexafluorides. By the time of group 11, the highest oxidation states are +4 and +5, and these are mostly one-compound wonders (K2AgF6, AuF5): the main oxidation state is +1, with some +2 and for Au +3. Though what you say then gives me Jensen's idea to go back to Mendeleev and once again limit "transition metal" to groups 8–11! :-P
 * Sure, we will go back and relook at the colours shortly after all this gets decided. We're still at the "debating-the-science" phase.
 * Well, we're not making it up. Even some who call Sc and Zn transition metals call them "non-typical transition metals" or "not true transition metals". (Same for the pairs of Y and Cd, and La and Hg). And it's mentioned quite a lot that the trends in group 3 are similar to those in group 1 and 2 (this is an artifact of choosing La as the heavier congener of Y, and so you'll only find this statement in books that choose that way). And it's quite often stated even more strongly for group 3 than for group 4 and 5: e.g. this textbook calls the chemistry of group 3 "almost exclusively" M3+, whereas it only calls that of groups 4 and 5 "dominated by" M4+ and M5+. So the divide between group 3 and groups 4–11 is pretty universally acknowledged to exist; what differs is whether it is treated as sufficient basis to remove group 3 from the transition metals altogether (and it seems it's mostly just UK textbooks who dare to go that far). Double sharp (talk) 03:54, 3 November 2013 (UTC)

You are right about lower oxidation states of Zr and Hf, but that can be explained so what? it suits the definition you use :) I'm just showing that definition isn't perfect, either, just admit the downside. I don't consider that to be the main reason not to use it.

Well, we're not making it up. We're not. I exaggerated. That was unprofessional. Sorry.

 it seems it's mostly just UK textbooks who dare to go that far This is the main reason not to use it. It is not the most widely accepted one; let's use the most widely accepted one.--R8R Gtrs (talk) 18:17, 13 November 2013 (UTC)
 * Double sharp: Do you like the layout of this periodic table (see p.4/39)? I've had it as a clipping for a couple of years. because when I saw it I had an eye-candy attack. In light of R8R's comments, it looks OK to me. I like the fact that La and Ac still line up under Group 3.Sandbh (talk) 11:02, 14 November 2013 (UTC)
 * It's not bad, especially if you are pro-Sc/Y/Lu/Lr. I am still somewhat on that side, but am not afraid to attack it as strongly as I can to see if it will be able to prove itself to be superior. ;-) But on La and Ac lining up under group 3, while it is a nice touch (especially as it also means that the pseudohomology of the early actinides to the transition metal groups is showable), it's not the deciding factor for me. Double sharp (talk) 09:30, 19 November 2013 (UTC)
 * The links gives me a 6-page pdf (pp. 35-41 from a 2011 NewScientist), not 139 p. The PTR in there looks off-topic here. Is there an interesting PT feature I missed? -DePiep (talk) 06:09, 1 December 2013 (UTC)
 * Oh, I think what I found interesting was the 57-70 box and the 89-102 box, as discrete boxes between the s-block and the d-block. I don't believe I've seen Sc-Y-Lu-Lr done quite that way before. Sandbh (talk) 10:22, 1 December 2013 (UTC)

Nomination for deletion of Template:Periodic table (Hindi)
Template:Periodic table (Hindi) has been nominated for deletion. You are invited to comment on the discussion at the template's entry on the Templates for discussion page. DePiep (talk) 17:10, 2 December 2013 (UTC)

Infobox element
FYI, there's a request to change elementbox at WPCHEMISTRY. See Wikipedia talk:WikiProject Chemistry for the discussion -- 65.94.78.9 (talk) 20:26, 7 December 2013 (UTC)
 * Seen and responded. Watching that page. Double sharp (talk) 09:07, 8 December 2013 (UTC)

Perspectives on the group 12 metals
I've listed these extracts from the literature (two by giants; two by lesser-known writers) for reference purposes. I agree with the author of extract 3: 'Textbook writers have always found difficulty in dealing with these elements.'

1. Treated as main group metals. 'The elements Zn, Cd, and Hg…[each have] a filled (n−1)d shell plus two ns electrons. While Cu, Ag, and Au all give rise to ions or complexes in which one or even two d electrons are lost, that is to compounds in oxidation states II and III, no such compounds have ever been isolated for the Group 12 metals [see my note, below]. Thus, while Cu, Ag, and Au are classified as transition elements, Zn, Cd, and Hg are not…Although the formation of complexes with ammonia, amines, halide, and psuedohalide ions is reminiscent of transition metal ion behaviours, the ability of Group 12 ions to serve as dπ donors is so low that they form none of the other typical sorts of transition metal complexes, such as carbonyls, nitrosyls, or π-complexes with olefins. The only exceptions to this are [Hg(CO)2][Sb2F11]2 and [Hg2(CO)2][Sb2F11]2, which are relatively stable but show no evidence of Hg–CO π bonding.'

Note : HgF4 has also since been synthesized at conditions close to absolute zero, so in this sense Hg could be regarded as a transition metal however this can give rise to the proposition that, for example, K, Rb and Cs are transition metals on the basis that they adopt d1 configurations, at high pressure. Based on current understanding, it is reasonable to conclude that Hg is not a transition metal at, or near, ambient conditions.


 * Cotton FA, Wilkinson G, Murillo CA & Bochmann 1999, Advanced inorganic chemistry, 6th ed., John Wiley & Sons, New York, p. 598

2. Treated as post-transition metals but included with the transition metals. 'The term transition metal, according to the International Union of Pure and Applied Chemistry (IUPAC), refers to an element whose atom has an incomplete d sub-shell or which can give rise to cations with an incomplete d sub-shell…The above definition…specifically excludes the Group 2B elements zinc (Zn), cadmium (Cd) and mercury (Hg), which are often referred to as post-transition metals…However it is useful to include the post-transition metals in any discussion of transition metals, as they provide interesting similarities and contrasts in chemical behavior…'


 * Bullen TD 2011, 'Stable isotopes of transition and post-transition metals as tracers in environmental studies', in M Baskaran (ed.), Handbook of Environmental isotope geochemistry, vol. 1, Springer, Heidelberg, pp. 177−204 (177)

3. Either transition metals or main group elements, as the occasion demands. '…the triad Zn, Cd and Hg have more in common with their immediate neighbours in the p block then they do with their neighbours on the other side, in the d block. Textbook writers have always found difficulty in dealing with these elements. In this book, they will be placed with either the transition elements or the Main Group elements, as the occasion demands.'

Smith DW 1990, Inorganic substances: A prelude to the study of descriptive inorganic chemistry, Cambridge University, Cambridge, p. 113

4. Atypical transition metals. 'In view of the stability of the filled d shell, these elements show few of the characteristic properties of transition metals (p. 905)[see note, below] despite their position in the d block of the periodic table. Thus zinc shows similarities with the main-group metal magnesium, many of their compounds being isomorphous, and displays the… characteristic of complexing readily with O-donor ligands. On the other hand, zinc has a much greater tendency than magnesium to form covalent compounds, and it resembles the transition elements in forming stable complexes not only with O-donor ligands, but with N- and S-donor ligands and with halides and CN– (see p. 1216) as well. As mentioned…cadmium is rather similar to zinc…mercury…has a much greater tendency to covalency and a preference for N-, P-, and S-donor ligands, with which HgII forms complexes whose stability is rarely exceeded by any other divalent cation.'

Note re characteristic properties of transition metals : '(i) They are metals and as such are lustrous and deformable and have high electrical and thermal conductivities. In addition, their melting and boiling points tend to be high and they are generally hard and strong; (ii) Most of them display numerous oxidation states which vary by steps of 1 rather than 2 as is usually the case with those main-group elements which exhibit more than one oxidation state; (iii) They have an unparalleled propensity for forming coordination compounds with Lewis bases.'

Sandbh (talk) 10:53, 12 December 2013 (UTC)
 * Greenwood NN & Earnshaw A 2002, Chemistry of the elements, 2nd ed., Butterworth-Heinemann, pp. 1206, 905


 * "Numerous" oxidation states?! I was only aware of +1 and +2 (and +4 for Hg), other than the trivial 0 that every element has. Double sharp (talk) 11:39, 12 December 2013 (UTC)
 * Yes, the 'numerous oxidation states' referred to the characteristic properties of the TMs (properties which the group 12 metals show few of). Sandbh (talk) 11:48, 12 December 2013 (UTC)
 * Not reading carefully again, sorry... :-( Double sharp (talk) 15:12, 12 December 2013 (UTC)

Metalloid FAC
OK, so having another go. Sandbh (talk) 04:55, 17 December 2013 (UTC)
 * Good luck! (I see you got published! Nice!) Double sharp (talk) 08:22, 17 December 2013 (UTC)