Wikipedia talk:WikiProject Elements/Mixed category elements revisited

This page is a discussion (talk) as part of basic talkpage Elements#Mixed category elements revisited (to subpage), started Nov 16, 2013. It is moved into this separate page to reduce page size and load speed. This page is open for discussion as a regular talkpage.

Ping list (suggestion: add to watchlist): Double sharp, Sandbh, TCO, Nergaal, YBG, R8R Gtrs, John, Petergans, , Headbomb, Legoktm, Axiosaurus, Smokefoot, Bcharles, Sbharris, 194.29.134.246, 79.191.180.224, 95.49.94.5, 95.49.56.163, 208.44.87.91, 70.24.244.158, 71.127.137.171 -DePiep (talk) 13:09, 17 November 2013 (UTC)

Mixed category elements revisited

Categories as content

Content. About: What to mix. The need to describe mixed categories, scientific arguments, encyclopedic considerations. Thoughts per element, or a set of elements. (not: how to color mixes - see below).

-DePiep (talk) 13:24, 17 November 2013 (UTC)

A few of the IP editors have raised the question of mixed category elements. There is an example in the German Wikipedia, here. Se is shown as metalle/halbmetalle; At is shown as metalle/halogene. We (Double Sharp, R8R, DePiep and I) discussed the idea of mixed category elements a while ago. I like them as way of solving thorny categorization questions and avoiding the need for categorisation contortions in order to get every one of the elements into just one our current ten categories. R8R didn't like them for for their lack of clarity; Double sharp didn't like them either; DePiep suggested they could imply a diagonal relationship that may not be there. What's got me going about this again has been the discussions about (1) poor metals & Al: poor metals and where or if Al fits into this category; (2) group 3 and REM: what to about group 3 and Sc, Y, La, Ac, Lu, Lr, and the proposed rare earth category; and (3) Metallicity: the recent posts by the IP editors re the metallicity of the elements.

Before I do much more work re-looking at the possibility of mixed category elements, I'd like to hear what people currently think about this idea. Sandbh (talk) 11:56, 16 November 2013 (UTC)

Stripes: If we're going to stripe, then I prefer the German-style striping, with lots of diagonal stripes: at least this avoids the "diagonal relationship" interpretation. Though I feel that if we need to stripe to be correct, we may need to relook at our category scheme. I just think it should be possible to come up with a classification scheme like that.

How to is continued in: #How to color mixed elements -DePiep (talk) 13:37, 17 November 2013 (UTC)

Al: has some physical-chemical contradiction. Physically it is quite a good metal, chemically it is kind of schizophrenic: its compounds tend to have covalent character, and its oxide is amphoteric, but it is simultaneously strongly electropositive with a high negative electrode potential. I guess I would put it as a pre-transition metal, along with Be, as a sort of outlier in that category, but in that category nonetheless.

Near metalloids: The IP is kinda convincing me to use a "near metalloid" category. I am still thinking about sulfur, iodine, and radon, where their proposed categorization would differ from our current one. I like the way he thinks about diagonals in the periodic table. It is kind of refreshing as I can recite the rows and columns almost without thinking, but diagonals force me to think again like I used to have to some years ago! ;-)

Group 3: personally I feel the physical reasons and some chemical reasons are enough to keep them out of the transition metals. It is not unheard of to do this. And after all, Al has physical reasons and some chemical reasons to be a pre-transition metal: we have a similar situation here. And if you don't consider them transition metals, then one of Jensen's main arguments for including Lu in group 3, periodic trends compared with those among the transition metals, doesn't really hold water. (His main other one is electron configurations, which can be quite easily turned against him, just as I can quite easily split that infinitive for no good reason: do the Aufbau principle's predicted electron configurations really mean anything in the real world?) Now, of course, this is based on my premise that the group 3 elements are not transition metals. You are free to dispute this, as R8R Gtrs does. If you don't accept it, you will probably end up with Jensen's conclusion.

BTW, here's my responses to Jensen's list of attributes of La vs. Lu (Table 2 of his paper). I'll start by saying that I don't really buy the first five items in that table. The first (highest common oxidation state) is neutral; (second) Y group vs Ce group (what happened to the Tb group?) is very much an atomic size issue, and so will be caused simply by periodic trends (La is larger than Sc and Y, while the Ln contraction causes Lu to be much smaller); (third) the crystal structures aren't always the same going down s- or d-block groups; (fourth and fifth) different structures of homologous ionic compounds is not prohibitive to group assignment, viz. NaCl vs. CsCl. I think many of these can be explained by group trends.

A stronger argument of his is sixth, that La has low-lying non-hydrogenlike f-orbitals, which Sc, Y, and Lu do not. Please accept a pause while I contemplate this and whether this is an issue. (First thought is that it is an issue, and that if I'm going to make the case for Sc/Y/La I had better think on this and see whether it is a prohibitive issue.)

Will have to do some more reading regarding the last two items. But from what I can see, superconducting pressure lowers down group 2 from Ca to Ba, therefore something similar may be going on from Sc to La, in which Lu breaks the pressure trend again (despite Jensen's article showing La as superconducting and Sc, Y, and Lu as not). I may be completely wrong on this. Have to go look up info on conduction band structure. Double sharp (talk) 14:40, 16 November 2013 (UTC)

P.S. I guess I didn't need to rationalize through all that mess! It appears that La and Lu are actually both superconducting at atmospheric pressure (naturally, at low temperatures)! This contradicts Jensen's paper: I guess Lu's superconductivity must be a more recent discovery than La's. Well, that's one less argument I have to deal with. (Sc and Y can only do it at high pressure.) Double sharp (talk) 14:44, 16 November 2013 (UTC)

P.P.S. My rationalization appears to be somewhat correct: pressure does indeed play a role in superconductivity. (It even cites my examples of group 2 vs. group 3! Sweet!) Double sharp (talk) 14:47, 16 November 2013 (UTC)

I'm with R8R. Group 3 are TM's physically but chemically only marginally. Mackay, Mackay and Henderson (2002, p. 246) write, 'The scandium group of elements has the outer electronic configuration d¹s² and is formally part of the d block of the Periodic Table. However, as the chemistry is dominated by the III oxidation state, which involves the loss of the d electron, this Group is best regarded as forming a transition region between the s elements and the main d-block.' Similary, Phillips and Williams (1966, p. 4) wrote: 'The metals may be divided into four main classes... (1) The pre-transition metals. These occur in Group IA and IIA of the Periodic Table, and for a number of purposes it also convenient to include with these the Group III metals, Al, Sc and Y… (2) The transition metals. These occur from Group IIIA to Group Ib, although transitional character is not very marked in Sc, Y, and La, while the metals of Group IB also show B-metal character.' I've been thinking about treating Group 3 as mixed category elements, i.e. pre-transition metal|transition metal, so as to do away with much categorisation angst. Sandbh (talk) 02:38, 17 November 2013 (UTC)

• MacKay KM, MacKay RA & Henderson W 2002, Introduction to modern inorganic chemistry, 6th ed., Nelson Thornes, Cheltenham, ISBN 0748764208
• Phillips CSG & Williams RJP 1965, Inorganic chemistry, II: Metals, Clarendon Press, Oxford

I think that mixed category is really good solution for too metalloidal nonmetals such as C or Se and it will be good if it will be implied. It could be also implied for poor (chemically) transition metals (such as Au and Pt), which have higher electronegativity in Pauling scale than phosphorus and hydrogen and form relatively stable monoanions. Al and Be are in the poorer chemically group of metals. Group 12 metals are also physically and chemically rather poor.

Lu really looks not like a lanthanoid according to these links. Its popular position is even misleading, maybe erroneous. Lu should be placed below Y. It is quite clear from the papers. Discussion about Lu: http://en.wikipedia.org/wiki/Talk:Lutetium

And link from the discussion does not work... Interesting...

79.191.180.224 (talk) 18:23, 16 November 2013 (UTC)

IP, next time keep the cynicism with you and write something helpful. Thanks. -DePiep (talk) 12:38, 17 November 2013 (UTC)

Huh? What cynicism? The post seems all right to me. Double sharp (talk) 14:11, 17 November 2013 (UTC)

Again about German Wikipedia. I think that there are some errors:

The addition "... Interesting..." is. -DePiep (talk) 15:24, 17 November 2013 (UTC)

• In it only Se is marked as even a "metalloid" (at least C and P have to be also), in fact they (C, P, Se) are the best examples of "near-metalloids" :- elements which are intermediates between nonmetals and metalloids,

• poor metals are named just as "Metalle" ("metals"; other groups of metals, such as alkali metals, lanthanoids and transition metals are also metals)),

• below yttrium there is lanthanum (not lutetium) in German Wikipedia,

• astatine is just a halogen (even not a metalloid)

Example: http://de.wikipedia.org/wiki/Stickstoff (no double-categorised elements now (?))

79.191.180.224 (talk) 19:24, 16 November 2013 (UTC)

Hmm, what happened to the double-categorized elements in de.wp? I could have sworn they were there just a few days ago...

I think that yes, most of these are errors. The only one of these I can see a case being made for is La under Y. I've attempted to make a case for it. Try to convince me that it's wrong! We can only benefit from discussion. Double sharp (talk) 03:58, 17 November 2013 (UTC)

@DePiep: Do you have any thoughts as to a better way to depict mixed category elements, should such a thing eventuate? I'm sure there are other ways to do this besides stripes (which I find to be quite garish), and diagonals. Sandbh (talk) 01:59, 17 November 2013 (UTC)

Re Sandbh: see how to color them below. -DePiep (talk) 13:37, 17 November 2013 (UTC)

• Since categorization is a scientific issue, we should accommodate any outcome that says: "elements can be in multiple categories". Presentation problems can not prohibit a scientific statement. And since I have seem many elements in the PT being discussed on this, it is time to say: we should make possible to present such mixing. Some questions that arise:

What kind of facts to expect?. How do such mixings appear in facts? Is it always 50/50 over just two categories? Could there be three? Would it need addition of another category to the dozen we have (example: suppose we have a regular cat "PTM", but Al must be in "mixed Poor metal and Metalloid". That would add another cat "PM", and only in a mix at that. May I prohibit this for reasons of being too-detailed? IOW, can you agree there must be found an other solution?). Should we prepare for "expect every variation"?

Always 50/50? If an element is not simply mixed 50/50 over two, do we want that represented? The metalloid border already shows four grades of mixing. I also note that apart from categorizing mixes as facts, we will need some standard article phrasing for all instances. And maybe a section Metal#Categorisation in the periodic table.

It's a detail. The mixed categories of an element are just a detail. In a larger picture, this detail might be omitted. For example, it will be mentioned in the infobox and in a large PT (as we do with, say, atomic mass), but in the smaller PT's (navigation box at end of page) we could forgo this detail. There the element should have a single color (of a 51% category). In intermediate size PT: tbd. -DePiep (talk) 14:16, 17 November 2013 (UTC)

How to color mixed categories

About: how to follow and implement the content decisions. Coloring is just a representation of scientific content. Expect examples in color-to-category, mixing colors, layout options.

-DePiep (talk) 13:24, 17 November 2013 (UTC)

List of examples

Please add your example links to this central bulleted list. Number them distinctly "[demo 1], [demo 2], ..." for easy reference, and maybe add a small description. Discussion is below the list, not in the list. -DePiep (talk) 14:42, 17 November 2013 (UTC)

• Earlier discussion: Wikipedia_talk:WikiProject_Elements/Archive_15.
• [demo 1]: File:Periodic_table_(German)_EN.svg. Mentioned by Sandbh in original post. -DePiep (talk) 14:42, 17 November 2013 (UTC)
• [demo 2] de:Vorlage:Navigationsleiste Periodensystem dewiki, today: no mixing
• [demo 3] archive 15, metalloid discussion: German striping, as Double sharp saw it in July 2012.

Used in de:Alternative Periodensysteme. Uses File:Metal-Metaloid.svg and File:Nonmetal-Metaloid.svg. -DePiep (talk) 15:35, 17 November 2013 (UTC)

• [demo 4]: File:Eight category periodic table (Mk2).png: from archive 15, metalloid discussion, Sandbh upload July 2012. -DePiep (talk) 14:42, 17 November 2013 (UTC)
• [demo 5], multi-color gradient: rude example of fade-over of colors (using gradual transparency; a stable technique). -DePiep (talk) 15:26, 17 November 2013 (UTC)
• [demo 6]: mixed categories in the main periodic table page, in other languages?(iw:periodic table):

-sq:Sistemi_periodik_i_elementeve (sq=Shqip=Albanish): the German striped version. -DePiep (talk) 06:32, 18 November 2013 (UTC)

• [demo 7] all PT images: commons:Category:Periodic_table. -DePiep (talk) 07:12, 18 November 2013 (UTC)
• [demo 8] commons:File:PSE.png uses a narrow colored bar in the left of cells. It is fading, and not overlapping the text. It shows the columns (groups) like "pnictogens" (not mixed cats), which we do in the column header. -DePiep (talk) 07:12, 18 November 2013 (UTC)

Options

Before choosing a striping pattern, let's find as many options as possible.

• Split by diagonal, NW-SE. See demo 4 by Sandbh. Could be NE-SW too.

Creates the grouping in a glance visually, more or less (esp. the Ln's and when in 32-column layout). At the same time this introduces a visual cut across the PT right through elements. This is the "diagonal" that possibly misleads the eye and the grasping of the PT. -DePiep (talk) 15:26, 17 November 2013 (UTC)

• German striping, archive 15, metalloid discussion|demo 3 (in 2012, second PT).

This surely puts the mixing on the map and into your eye. Even into a distracting level I'd say. It is grabbing all your attention, to just this minor aspect. Even with (our) lighter cat colors, this effect is present. That is because we want those cat colors to be different. A gentle pair of colors that would make a softer striping, would defy the purpose of coloring categories recognizably different.

Also, any striping background makes reading or recognizing text very difficult. It surely is a setback in accessability. It is a technique used in captcha forcing us to spend extra concentration, and we'd multiply that for dozens of elements!

• Fade-over. demo 5: a color gradient. The direction (horizontal in the demo) can be any degree. Also the gradient can be radial (like waves from a stone thrown in the water), though I don't see an advantage in that for us. -DePiep (talk) 15:26, 17 November 2013 (UTC)
• Other wikis for inspiration. I have checked every other language wiki's Periodic table page (the iw: links). In these PT pages, only the Albanian PT page (demo 6, sq:wiki=Shqip) uses any sort of color mixing; it uses the German striping. Todo: the striping could be on other pages (like in de:wiki, in "Alternative PT"). -DePiep (talk) 06:32, 18 November 2013 (UTC)
• demo 8 shows a small, fading colored bar, left side of the element cell. Nice option, since it does not interfere with the text that much. The bar could be at any side (or even top+right). A smaller bar like this is difficult to express a 50/50 mix, that would interfere with the text & symbol. Could use some examples into out topic. -DePiep (talk) 07:12, 18 November 2013 (UTC)
  • In the 32 column PT . Mixing categories/legend/colors is adding one more complication to the PT. To keep oversight, the mixing in graphic PTs should be demo'ed & applied in a 32+ column PT only, not in an 18 column. (and possibly in the large cells, that is: a PT wider than page). -DePiep (talk) 08:18, 19 November 2013 (UTC)
• Please continue reading below: the tetris option. -DePiep (talk) 07:51, 21 November 2013 (UTC)

Tetris markings

• Let's use tetris markings! Not exactly coloring, but who does not understand this:

Tetris markings

Hydrogen Helium Lithium Beryllium Boron Carbon Nitrogen Oxygen Fluorine Neon Sodium Magnesium Aluminium Silicon Phosphorus Sulfur Chlorine Argon Potassium Calcium Scandium Titanium Vanadium Chromium Manganese Iron Cobalt Nickel Copper Zinc Gallium Germanium Arsenic Selenium Bromine Krypton Rubidium Strontium Yttrium Zirconium Niobium Molybdenum Technetium Ruthenium Rhodium Palladium Silver Cadmium Indium Tin Antimony Tellurium Iodine Xenon Caesium Barium Lanthanum Cerium Praseodymium Neodymium Promethium Samarium Europium Gadolinium Terbium Dysprosium Holmium Erbium Thulium Ytterbium Lutetium Hafnium Tantalum Tungsten Rhenium Osmium Iridium Platinum Gold Mercury (element) Thallium Lead Bismuth Polonium Astatine Radon Francium Radium Actinium Thorium Protactinium Uranium Neptunium Plutonium Americium Curium Berkelium Californium Einsteinium Fermium Mendelevium Nobelium Lawrencium Rutherfordium Dubnium Seaborgium Bohrium Hassium Meitnerium Darmstadtium Roentgenium Copernicium Nihonium Flerovium Moscovium Livermorium Tennessine Oganesson

Currently, only pre-known groups, categories, and the element-self page can be marked this way. And only in the micro PT (!). After expanding the options

• We can point each and every element for marking. Just list them like |mark=1,15,21,33.
• We are not limited to categories or groups. Any range of elements is accepted. This can be used for any overlapping area. As the example shows: showing group 13 & cat colors (this is a big feature in the concept: we are not limited to a single dimension like "category").
• Expect that the inner borders will not show (the outline remains).
• Applied to every PT, in every size (while respecting current occurrence-borders).
• Disjointed, scattered areas? Yes. (I'm a Tetris level 5+ gamer).
• Option for a second marker (dotted border) for whenever needed.
• Also: mark the s-block in a category-colored PT? Yes. Mark the Ln's, with the discussed ones apart (dotted), while a category is colored REM: yes.

Saves us from having to pre-define any mixing, any striping. Discussion on what to "mix", and on which page, in which detail -- all becomes more of a content discussion (not a question of how to squeeze it wikitechnically into PT).

Will not be available in 2013. -DePiep (talk) 07:51, 21 November 2013 (UTC)

Near-metalloids again

• I wonder why people so eagerly mark carbon as "just a (typical) nonmetal" when many of its properties are in clear conflict with typical nonmetal traits: it remains solid at extremely high temperatures have thermal conductivity on the range of metals (both diamond and graphite), is vey hard and dense (relatively, to its atomic number) as diamond and well electrical conductive and metal-like looking as graphite. These properties are too alien for a pure nonmetal. There are things such as graphite foils (metal-looking elastic material), expanded (flexible) graphite or thin nanotube threads. Even very light, volatile oxides (gases at STP) are not so soluble and acidic.

• Phosphorus is commonly known as a white, low-melitng and low-conducting soft substance, not so dense and very toxic, red (more metalloidal polymeric allotrope) is also well-known. Black, metalloidal, most stable allotrope is not, but its properties are more interesting - rather metalloidal than nonmetallic. Low (or just the lowest) electronegativity, in comparison to nonmetals, in almost all scales is also worth to note.

• Selenium is not so metalloidal as it is often thought. It is well-conducting only in the light or (such as many other (rather) nonmetallic elements) under high pressure. Grey allotrope has good (even silverish(?)) appearance, its thermal conductivity is not so special (lower than its of black (metallic) phosphorus), it has relatively low melting and boiling point (in that it looks like heavier analogue of sulfur). Oxides are well soluble in water and signifiantly acidic, even its hydride is relatively strong acid.

• Iodine looks really metalloidal (quite bright grey color when pure, metallic luster). It is diatomic and volatile, quite heavy element (Z=53) It has better conductivities than sulfur. Iodine definately has some properties, which are atypical for a nonmetal. It is highly acidic (HI solution in water is very strong acid). Many iodides are soluble in water and do not decompose so easily (unlike carbides, phosphides, selenides, even sulfides).

• Radon is (like hydrogen) strongly nonmetallic physically. But these two elements have the same electronegativity in the Pauling scale (2,2) and show marked cationic behaviour (noble gases could be also depreciated to other categories of nonmetals).

From these elements, C, Se and P are most metallic, next are: I, S, then H, Rn.

Periodic table [by IP, Nov 18]

H

He

Li

Be

B

C

N

O

F

Ne

Na

Mg

Al

Si

P

S

Cl

Ar

K

Ca

Sc

Ti

V

Cr

Mn

Fe

Co

Ni

Cu

Zn

Ga

Ge

As

Se

Br

Kr

Rb

Sr

Y

Zr

Nb

Mo

Tc

Ru

Rh

Pd

Ag

Cd

In

Sn

Sb

Te

I

Xe

Cs

Ba

La

Ce

Pr

Nd

Pm

Sm

Eu

Gd

Tb

Dy

Ho

Er

Tm

Yb

Lu

Hf

Ta

W

Re

Os

Ir

Pt

Au

Hg

Tl

Pb

Bi

Po

At

Rn

Fr

Ra

Ac

Th

Pa

U

Np

Pu

Am

Cm

Bk

Cf

Es

Fm

Md

No

Lr

Rf

Db

Sg

Bh

Hs

Mt

Ds

Rg

Cn

Uut

Fl

Uup

Lv

Uus

Uuo

Alkali metals

Alkaline earth metals

Lanthanides

Actinides

Transition metals

Post-transition metals

Metalloids

Other nonmetals

Halogens

Noble gases

 

194.29.130.244 (talk) 08:02, 18 November 2013 (UTC)

Hmm. The stripes are beginning to grow on me. I can see an advantage in that the underlying category is preserved, which covers off on one of DePiep's points. In this scheme I would probably unstripe S and just note its metallic attributes in its article. Also change post transition metals to poor metals since Al is not a post-transition metal, it's a pre-transition metal. The name of the category of other nonmetals is very bad. I think you could change that to reactive nonmetals since they are in comparison to the noble gases. As well, no need to keep halogens as a category, although we still keep it as a group name. Astatine I would leave as a metalloid, for now, until we have more evidence that it's a metal. Oh, keep the noble gases category. Sandbh (talk) 20:27, 18 November 2013 (UTC)

• It is a pity that there is no stripping "transition metal + poor metal", which would be useful to describe "anionic" electronegative metals (Pt and Au), even Au and Ag might be classified as "half-poor". For Be there also should be stripping.

• Maybe we should leave controversial astatine and element 118 as just elements with unknown properties?

• Sulfur lies not so near to the typical metalloids, it has phosphorus, not a typical metalloid, on the left side (hydrogen has nothing), but H practically borders to the metal-nonmetal dividing line. Sulfur is tricky. It has lower electronegativity than iodine, higher boiling point as is polyatomic, can be even a polymer. Iodine is a volatile diatomic element with metalloidal appearance and much better conductivity than S. In fact, there are two (or even three) classed of poor nonmetals. First consists of C, P and Se and they are typical, polymeric near-metalloids. Second class (weaker general metallic character) consists of I, S and two "electropositive" gases: H and Rn. Difference between I and S is not so great! I, Rn, (H) are "positionally" near-metalloids, but S is then "near-near-metalloid" which has two near-metalloids at neighbours (P on the left and Se below). S is almost anomalously metallic! It should be diatomic such as its diatomic neighbors from the diagonal (N and Br), but can be very polyatomic (not so polyatomic as (near-)metalloids). Electronegativity of sulfur is also closer to values of near-metalloids (2,58 in Pauling scale). S has not markedly metallic allotrope. It is in agreement with its position in the periodic table.

83.6.21.120 (talk) 22:22, 18 November 2013 (UTC)

Of course, the mixing would be applied to our newer PT with diatomic and polyatomic categories, right? This looks like an old one. -DePiep (talk) 01:19, 19 November 2013 (UTC)

IP (TCO?), please add how far you meant this to be an older version. Is that part of the proposal, or do we need a category transformation first? -DePiep (talk) 07:55, 19 November 2013 (UTC)

No, I don't think the IP is TCO. For one, TCO would never start a long talk about categorization. ;-) Double sharp (talk) 04:09, 20 November 2013 (UTC)

I think this is the old German PT template. So yes. But what about this:

No need for mixing!

I'm pretty sure that E118 is already at the metalloid point; I'm not sure if Rn is there yet.

My main issue is that there's no distinguishing from near metalloids coming from the metallic side (e.g. Bi, Cn) and those coming from the nonmetallic side (e.g. I, Rn). Double sharp (talk) 07:17, 19 November 2013 (UTC)

I'd be inclined to leave 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. S as a near metalloid seems odd---the metallic properties are not obvious unlike C, P (black), Se and I. Not having near metalloids on the metal side is OK; we call them post-transition metals. Cn remains a potential issue. Sandbh (talk) 10:06, 19 November 2013 (UTC)

Updated per your feedback. Cn is indeed an issue, but we can afford to procrastinate on worrying about it. ;-)

(P.S. I'm in favour of counting black P as the main P allotrope, because it's the most stable, in case anyone doesn't known my position yet. Not doing so tends to obscure the trends due to the inconsistency of such a choice with the standard ones for the other nearby elements.) Double sharp (talk) 10:53, 19 November 2013 (UTC)

Oh come on, people.

You're talking about how to define stripees, but haven't shown they're needed in first place.

Stripees are not good. For pros maybe, but not for a general reader.

I'll question that, not because I don't believe it, but because the German WP used it. Why? They must have had their reasons. Did they discuss its adoption and later removal? (If you have the link, post it...) Double sharp (talk) 04:36, 20 November 2013 (UTC)

Many, many, many people, maybe half of them or even more are people who can rate astatine negatively just because it doesn't have the sodium astatide formula. If I were talking to a chem professor, I would mention those nuances. Not in a general table, ever.

TBH I kinda tuned out away from that feedback, simply because 33% of them are "give me something I could figure out with first-year chem knowledge", 33% clearly show that the reader hasn't actually read the article, 33% are well-intentioned but not actionable, and 1% are actually useful. (These aren't actual statistics.) So I'd say that most of it can safely be ignored... Double sharp (talk) 04:36, 20 November 2013 (UTC)

And remember, metalloids are already a transition class. Why have a transition class between a normal class and a transition class?

There is a solid reason for it. We already have polyatomic nonmetals as our current one. Double sharp (talk) 04:36, 20 November 2013 (UTC)

Most people don't even read table legends, in general, remember that.

They'll see our table doesn't look like the ones they're used to seeing, and I hope they'll read the legends. If not, we've lost them either way. Those who do read the legend, however, will benefit. Double sharp (talk) 04:36, 20 November 2013 (UTC)

It is a pity that there is no stripping "transition metal + poor metal" It's not. To get the point with aurides and stuff, you would need to knopw it in first place. Or have some real understanding in chemistry.

And shouldn't we help readers advance to having some? Double sharp (talk) 04:36, 20 November 2013 (UTC)

Most people think of aluminum as of a metal. Those who do normally think of it as one similar to gallium and indium rather than AMs or AEMs.

Sandbh already responded to this on the main talk page. Double sharp (talk) 04:36, 20 November 2013 (UTC)

Draw a general picture, don't go into details unless it is expected (for example, how about a near metalloid article? it could take some info from metalloid info, we could give it a few links in some of our articles and we could treat it as a group like platinum group metals).

Hmm. I think the split in the nonmetals is important enough to give a new category. If we want to draw a general picture, we can do your radical s/p/d/f-block classification. (If you don't already know, though, I really hate splitting the table by blocks. What does it mean chemically and physically. In general, not much.) Double sharp (talk) 04:36, 20 November 2013 (UTC)

KISS.

That might work, with any nonmetal subcategories possibly being able to be explained in the nonmetal article. There are three concerns. 1. Group 12 metals may be d-block metals but they are not, on any reasonable grounds, transition metals. 2. If so, what do we call the group 12-16 metals, noting Al is not a post-transition metal (i.e. it is located before the TMs)? 3. if Al is included in this "metals that need help" club how do we keep beryllium out of it? Should we keep beryllium out of it? Sandbh (talk) 21:11, 19 November 2013 (UTC)

IMO it is a step backwards. Why suddenly remove all traces of a division among the nonmetals when one certainly exists? Even the old "other"/"halogen" had it, even though it wasn't a terribly good one. Now we have a reasonable one. Why get rid of it? And group 12 and Al are problematic again. Double sharp (talk) 01:38, 20 November 2013 (UTC)

It is in fact a step backward. Oddities such as C and Se ("half-metalloids") are again pulled to the category of "reactive nonmetals". It is not so precise classification. 194.29.130.244 (talk) 08:19, 20 November 2013 (UTC)

We can go even further.

--R8R Gtrs (talk) 18:41, 19 November 2013 (UTC)

R8R, what are you trying to say? Whatever we do, any level of detail is chosen, even if unchanged.

If a new striping is needed, it will arrive somehow. Talk should be about whether mixing cats is needed, for which elements, which cats, is it 50/50, and so. Given continuous discussion (in/out of WP), the topic is valid. -DePiep (talk) 19:23, 19 November 2013 (UTC)

What I'm saying is, no striping should ever be needed. It makes things too complicated (not colored looks, the science behind it)

I'm just following the discussion as it is, but if you find my comment misplaced, please move it. (if you don't, don't)--R8R Gtrs (talk) 19:57, 19 November 2013 (UTC)

About third periodic table: "Near-metalloids" have too alien properties to just count them along with strong nonmetals. They are a class themselves. Carbon is closer to boron or silicon than to fluorine. Selenium is more similar to tellurium than to chlorine, phosphorus to arsenic than to oxygen. Iodine is poorer (volatile and diatomic) near-metalloid, radon and hydrogen also shows some non-nonmetallic properties in their chemistry. Sulfur is on the border of so-called "near-metalloids" and nonmetals.

Noble gases are a class of nonmetals. Idea of splitting to reactive nonmetals and noble gases is poor.

I want to separate oddities such as carbon and selenium from strong, typical nonmetals. They just have too marked metallic properties, which are in contast of the general picture of a nonmetal (but nonmetallic character is still greater than metallic). Sometimes (by using metal-nonmetal dividing line) even more metallic elements (B, Si, As, Te, At) are counted as nonmetals.

Yes, near-metalloids have many contrasts with typical nonmetals. They have significant, sometimes even astonishing, metalloidal traits. In the case of C and Se it is mostly pronounced, in the case of P they are less known because of common contact with more nonmetallic, less stable allotropes. Just to rule out these strangers from the group of nonmetals and name they as "near-metalloids". 194.29.130.244 (talk) 08:14, 20 November 2013 (UTC)

83.31.138.237 (talk) 23:08, 19 November 2013 (UTC)

"Near-metalloids" in the third PT? -DePiep (talk) 23:19, 19 November 2013 (UTC)

(explaining) ... or the second PT here (by Ds Nov 18, introducing the word "near-metalloid"), not R8R's -DePiep (talk) 05:52, 20 November 2013 (UTC)

Did you know that one of the reasons I supported our current colouring is because polyatomicity/diatomicity correlates pretty well with metallic character? ;-) That's why I like your categorization so much, I think...

There are some issues about metallicity of polyatomic and diatomic nonmetals. The largest is the course of sulfur and iodine. Sulfur is polyatomic (usually oligomeric in the form of S₈), but does not have a "metallic" allotrope at STP. Catenated, plastic sulfur is closer to metalloids, but is far less stable. Iodine is diatomic, but in its pure form looks just metallic grey and is a better conductor than sulfur. Iodine appear to be more near-metalloidal than sulfur due to its appearance and conductivity. Hydrogen is diatomic, physically is a good nonmetal, but chemically is markedly metalloidal. 194.29.130.244 (talk) 07:59, 20 November 2013 (UTC)

I still think we should split noble gases. They are not reactive nonmetals and have fairly distinct properties, such as general nonreactivity. (Not so Rn: apparently it can react with a few exotic compounds at room temperature, which Xe can't do!) I think they form the other extreme of nonmetallic properties (that isn't the near meatlloids). Double sharp (talk) 01:38, 20 November 2013 (UTC)

Radon is not so nonmetallic (low electronegativity, cationic chemistry). It looks rather as a form of near-metalloid. 194.29.130.244 (talk) 08:08, 20 November 2013 (UTC)

Hmm. OTOH it does have the extremely small liquid range and high volatility commonly associated with noble gases. Its main claim to metallicity is, as you observe, its markedly low electronegativity and showing of some cationic behaviour. I am not sure if this is enough to bring it out of the noble gases. It is certainly already more of an ignoble gas, willing to react at room temperature, if only with weird compounds like [O₂]⁺[SbF₆]⁻. I am still not completely convinced that the change happens between Xe and Rn; I'm more convinced by a change between Rn and E118. Double sharp (talk) 10:23, 20 November 2013 (UTC)

I am talking (in fact) about the lack of near-metalloids in third periodic table on this page. Many properties of near-metalloids are worth to mark. They can for example: have very high melting, boiling or sublimation points, be very hard, really well conductive, look metal-like, be good semiconductors.

83.31.138.237 (talk) 23:25, 19 November 2013 (UTC)

Al is a poor metal (and implications for near-metalloids)

I'm now satisifed(!) that aluminium does qualify as a poor metal, whereas beryllium doesn't. A poor metal is a physically weak metal that shows significant nonmetallic chemistry. Beryllium and aluminium both show significant nonmetallic chemistry. However beryllium is pretty strong whereas aluminium isn't. Here are some unattributed snippets from Google Books illustrating the difference:

'Pure aluminium is a comparatively weak metal…

'Aluminium is a light but relatively weak metal…'

'Pure aluminium is rather a weak metal…'

'Beryllium is a relatively light yet strong metal'

'Beryllium is a strong metal, which is lightweight, resistant to corrosion and melts at a very high temperature.'

'Be is a light, strong metal with a very high melting point…'

There are plenty more like that.

In conclusion, and as I see it: Be = alkaline earth; Al = poor; Group 3 (Sc, Y, Lu, Lr) = marginal TMs; La–Yb = lanthanides; Ac–No = actinides. We keep the alkali metal and alkaline earth categories. Further distinctions as to the non-metallic character of metals (e.g. Au) or the metallic character of nonmetals (e.g. H, C, Se) can be made in the articles for the respective elements. I think the rare earths are better regarded as an uber-category comprising the first three members of group 3 + the lanthanides.

With respect, if the IP editors want to make a case for a different categorisation scheme that includes e.g. near metalloids then they need to do a write up that (a) satisfies the YBG rules; and (b) is fully referenced, as per the current nonmetal article. The YBG rules are that any new categorisation scheme be clear ('The criterion for division should be easily explained'); unambiguous ('It should be relatively obvious which category each element fits into'); and meaningful ('The categories should have significance more than just dividing for the sake of dividing. There should be enough within-group similarity and enough between-group dissimilarity so that each group could be the subject of a separate encyclopaedia article.'). Sandbh (talk) 12:03, 20 November 2013 (UTC)

Why are Lu and Lr not lanthanides and actinides in this scheme? Chemically they surely are.

As I see it, the main stumbling block for the categorization in the second PT on this page (mine, based on the IP's; shall we call it "option 18"? :-P) is how to define "near metalloid" and "typical nonmetal". The former can be thus "a nonmetal with many markedly metallic properties, but not yet at the metalloid threshold" while the latter would be rather self-explanatory. But now I'm thinking that polyatomic/diatomic option 10 actually works very well (I being on the threshold of being polyatomic, with S having weaker metallic properties that correlates with its being farther from the metalloid line).

Agree on Al: we must have a strict definition. Al fits it; Be, not so much (MH 5.5). Double sharp (talk) 02:15, 21 November 2013 (UTC)

Again about German Wikipedia. Now it is just stupidity - selenium as a metalloid and carbon and phosphorus as a non-metals. Carbon is in many ways more metallic than selenium. Its "metallic" (more precisely, metalloidal) character is too underestimated, especially in comparison to selenium. Phosphorus is next near-metalloid which should be in the one group. I looks generally closer to C, P, Se than S, mainly because of appearance and conductivity and its position in the periodic table. In the name of the class for C, Se, P the name "metalloid" should be, not nonmetal. Errors about classification of these elements are often to blatant, selenium is even called a "non-ferrous metal". Carbon looks very oddly in the only-nonmetal group with its melting point, conductivity, appearance, some properties of diamond or nanotubes. Name of carbon ends at -on, such as boron and silicon (these three elements can be very hard and have the highest melting and boiling points (for B and Si sometimes values about 2500 C are met) in main gropus). Diamond, boron and glassy carbon are really hard materials (7 - 10 Mohs), unlike typical nonmetals. Thus, carbon is definately halfly (not fully) metalloidal. Selenium also (mainly due to the other reasons - apperance of grey Se and photoconductivity), but halfly, not fully metalloidal. Half-metalloidal character of phosphorus is usually hidden due to popular presence of less metallic allotropes, especially white one.

194.29.130.244 (talk) 17:35, 20 November 2013 (UTC)

And now for the predicted elements again

The chief ones we need to worry about are E117 and E118. (E171 is treading on dangerous ground, assuming based on properties similar to I₂ that it would form (171)₂, but I think we've sorted that one out. And E172 is a very good noble gas, though it's probably actually a noble liquid or solid.) IP, I want your comments, please... ;-)

I've given quotes from Fricke below, to supplement the material from the articles (ununseptium, ununoctium, period 9 element).

Element 117

(Fricke 1974) "Element 117. (eka-astatine) is expected to have little similarity to what one usually calls a halogen, mainly because its electron affinity will be very small. Cunningham (96) predicted its value as 2.6 eV, whereas the calculations of Waber, Cromer and Liberman (54) gave a value of only 1.8 eV. As a result of this small electron affinity, and from extrapolations of the chemical properties of the lighter halogen homologs, all authors agree that the +3 oxidation state should be at least as important as the −1 state, and possibly more so. To take an example, element 117 might resemble Au(+3) in its ion-exchange behavior with halide media. Cunningham (96) describes the solid element 117 as having a semimetallic appearence."

(1971) "Occupation of the 7p_3/2 subshell begins at Z = 115 with a binding energy which is only half as large as that of the 7p_1/2 electrons, so that the elements E115, E116 and E117 will have +1, +2 and +3, respective1y, as their normal oxidation state. The higher oxidation states will be possible only in the presence of strong oxidizers. An interesting question is whether element E117 which is in the chemical group of the halogens would form the -1 anion. Cunningham predicted a electron affinity of 2.6 eV whereas the calculations of Waber, Cromer and Liberman calculated a value of only 1.8 eV. Cunningham describes the solid element El17 to have a semi-metallic appearance. It should form stable oxyions of the (III), (V), and (VII) states and stable interhalogen compounds. Because of the small electron affinity it might not exhibit the -1 oxidation state, which is even further suggested by the smaller value calculated by Waber et al. Certainly it will be a very soft base compared with fluoride or chloride which have a electron affinity of 4 eV resp. 3 eV."

Element 118

(Fricke 1974) "A. V. Grosse wrote a prophetie article (95) in 1965 before the nuclear theorists began to publish their findings concerning the island of stability. In this paper he gave detailed predictions of the physical and chemieal properties of element 118 (eka-radon), the next rare gas. He pointed out that eka-radon would be the most electropositive of the rare gases. In addition to the oxides and fluorides shown by Kr and Xe, he predicted that 118 would be likely to form a noble gas-chlorine bond. These very first extrapolations into the region of superheavy elements have been fully confirmed by the calculations, because the first ionization potentials turn out to be much lower than in all the other noble gases. Independently Grosse (95) and Cunningham (96) found that the expected boiling point of liquid element 118 is about −15 °C, so that it will be nearly a "noble fluid". Because of its large atornic number it will, of course, be much denser than all the other noble gases. But, in general, the chemical behavior of element 118 will be more like that of a normal element, with many possible oxidation states like +2 and +4; +6 will be less important because of the strang binding of the p_1/2 electrons. It will continue the trend towards chernical reactivity first observed in xenon."

(1971) "The "noble gas" at Z = 118 will be a very weak noble gas in the sense of He and Ar but as well in comparison to Xe and Rn. The ionization energy is so small that normal covalent bondings are expected with oxidation states of 4 and 6. The extrapolation of Cunningham [35] expects a boiling point of −15 °C so that it will be nearly a "noble fluid"..."

(1971) "Previous analysis by Waber, following an informal discussion with Fano, indicated that negative ions of the noble gases would have configurations such as np⁵ (n + 1)s². The spectra of such species have been found at the National Bureau of Standards following electron bombardment of the noble gases. It would be expected that E118 could readily form such anions. Calculations have not confirmed the likelihood of such species. Independently Grosse and Cunningham found that the expected boiling point of liquid E1l8 is about -15°C, so that it will be nearly a "noble fluid". It might be predicted as well that the crystalline form would be much denser than the other noble gases. That is, the bonding in solid E118 would be stronger than given by van der Waals forces. These predictions from systematic continuation are supported by the calculation. Its first ionization energy is small, only 9 eV, and the strongly split p shell giving rise to frontier orbitals at the surface of the atom suggest that E118 will be more a normal element with many possible compounds than a noble element. Thus it will continue the trend towards chemical reactivity first observed in xenon."

Element 171

(Fricke 1974; he says almost exactly the same thing in the other papers) "Element 171 is expected to have many possible oxidation states between −1 and +7, as the halogens do. Here again, the electron affinity will be high enough to form a hydrogen halide like H(171). Fricke et al. (56) calculated a value for the electron affinity of 3.0 eV, which is as high as the value of I⁻, so that (171)⁻ will be quite a soft base."

Element 172

(Fricke 1974; ditto) "Element 172 will be a noble gas with a closed p shell outside. The ionization energy of this element, as shown in Fig. 15, is very near to the value of Xe, so that it might be quite similar to this element. The only great difference between Xe and 172 is that element 172 is expected to be a liquid or even a solid at normal temperatures because of its large atomic weight. As indicated in connection with the noble gas 118, element 172 will tend to be a strong Lewis acid and hence compounds with F and O are expected, as has been demonstrated for xenon."

Double sharp (talk) 12:45, 21 November 2013 (UTC)

Element 85 (At, astatine) is now predicted to have metal band structure at normal conditions. E117 probably also will be a metal, what about E171 - I don't know, but it (for me) can also be a metal with high electron affinity, electronegativity and first ionisation energy.

What about apperance, volatility and conductivity of E118 and E172? Are they nonmetals, metlloids or even metals? It is really interesting, because, according to digonal trends, they should be at least metalloids.

79.191.55.205 (talk) 15:11, 21 November 2013 (UTC)