Talk:Lanthanum

For GA
Look through the applications, make it read better (it's currently just a huge list), and see which ones are the really important ones. If going for FA later, expand the chemistry section (it's an okay overview, but there is more to say). Double sharp (talk) 03:22, 23 June 2016 (UTC)
 * (OTOH, since these are all real uses and Greenwood 1984 may not be up-to-date here, unlike for In, it may be fine. Awaiting the review, then.) Double sharp (talk) 03:40, 23 June 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 external links on Lanthanum. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20090321091620/http://www.1911encyclopedia.org/Lighting to http://www.1911encyclopedia.org/Lighting
 * Added archive https://web.archive.org/web/20090227062546/http://www.cobasys.com/pdf/tutorial/inside_nimh_battery_technology.pdf to http://www.cobasys.com/pdf/tutorial/inside_nimh_battery_technology.pdf
 * Added archive https://web.archive.org/web/20100923150541/http://www.cmc.dk/ to http:///

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 13:58, 11 May 2017 (UTC)

regarding the coverage of the old La/Lu controversy here
R8R removed the old paragraph describing it, which I restored. I tried to be balanced here: the first paragraph shows pretty much the main Lu argument (both La and Lu have the d differentiating electron), and then the La ones given in the literature in their own paragraph ("the differences are because La is bigger", "La has no f-involvement", and "eka-Y should be expected to be bigger than Y, and La is but Lu isn't"). So I restrained myself to giving the points made in the literature, and I also noted that Lu is more similar to Y (but not Sc). In general it is pretty difficult to find this material in the literature because it is usually considered a non-issue resolved in favour of La, so if you can find some more points in the actual literature (not just from us pontificating), please let me know, and I'll add them. Or if you also have any suggestions on how to make it look less biased.

But I notice my point doesn't seem to be quite understood, so here's a summary of how I think about it personally without the balancing in the article (it goes without saying that this is talk-page material, illustrating why I find your objections lack some force for me):


 * 1) We know that both La and Lu are trivalent.
 * 2) We know that both La and Lu have a 5d differentiating electron from the elements before them. (This and the previous point make them both suitable candidates for being eka-Y.)
 * 3) We know that Lu is more similar in behaviour to Y than La is. (But neither is very much like Sc, which is much more like Al.)
 * 4) From this difference we are led to the conclusion that since {Al, Sc, Y, La, Lu} are all trivalent, size is the key factor that determines similarity between them.
 * 5) It is not unusual for size to increase down the table. In fact it is the predominant trend throughout all of it. So if you want to claim that eka-Y should not be larger than Y you need a reason for group 3 to be special.
 * 6) *(If we were to use chemical similarity on the problem of group 2, we would be suggesting Be and Mg to be placed over Zn, so clearly chemical similarity is not the only factor.)
 * 7) Now it happens that there is such a reason: the lanthanide contraction. Groups 4 through 8, we are all agreed, come after the lanthanides (accepting that they might not be the only groups to do so), and because the 4f shell is terrible at shielding the nuclear charge we get a pattern that the 4d and 5d elements are very similar to each other and both quite distinct from the 3d element.
 * 8) So the question is whether the lanthanides precede or succeed group 3. If the former, then the lanthanide contraction succeeds as an explanation and we adopt Lu; but if the latter, it doesn't work, and then we have to adopt La.
 * 9) The defining property of lanthanide chemistry is 4f involvement. In other words, we should have some reasonable expectation that the Ln atoms and their ions should have 4f electrons. Now La doesn't have them, but Lu does.
 * 10) In fact, if you plot the chemically relevant Ln3+ electron configurations, they start at [Xe]4f0 for La and end at [Xe]4f14 for Lu. Now it seems much more reasonable to include the full f-shell than the empty one in the f-block, even though neither shows much f chemistry, just like we include Zn but not Ca in the d-block even though Zn really does not act like a transition metal.
 * 11) *Furthermore, since Yb is actually mixed divalent-trivalent as a metal due to a 4f-5d band overlap, Lu might be said to have "half a 4f" differentiating electron – an absent effect for La. (This is a side point anyway.)
 * 12) But most of all, orbitals undergo a sudden contraction and lowering of energy when they first become chemically active. For example, this happens at Z = 19 for 4s but only at Z = 21 for 3d, whence the 3d-4s problem resolved by Aufbau. Now if you look at the beginning of the sixth period, the collapse happens at Z = 55 for 6s, Z = 57 for 5d, and Z = 58 for 4f.
 * 13) These points strongly suggest that the beginning of 4f involvement is at Ce, just like the beginning of 3d involvement is at Sc.
 * 14) But if Ce is the first element to have significant 4f involvement as a valence shell, instead of just "supervalent hybridisation" like Ca, Sr, and Ba do with the d-shells that are about to collapse, then La cannot be an f-block element, any more than Ca preceding Sc is a d-block element.

Since you (evidently) disagree with my conclusion, then the issue must be with (at least) one of these statements. So where do you find a problem?

One possible objection may be how much weight you apply to the Aufbau principle. If you consider that to be sacrosanct, then indeed it would trump the smaller differences and suggest Lu under Y. The problem is that I don't believe that the Aufbau principle is sacrosanct. I, like many actual chemists, believe that it is supposed to be a rough guide to the periodic table in terms of actual electron configurations, which influence but don't dictate actual chemistry, and this may sound anti-reductionist (it isn't) but we don't understand completely where the Aufbau comes from, only that it is a trend that you can see from the subshell collapses. So I think that as long as we don't have a coherent theory that pops out Aufbau from first principles instead of just showing it up as a cool pattern, I think that the choice between La and Lu needs chemical justification like any other.

But, of course, this is just how I think, and in the article we have to reflect what is in the literature. This may be a bit of a tall order because the main arguments offered by the few Lu-advocates I see are Aufbau, greater similarities of Lu than La to Y, and commonalities in trends with groups 4 through 8, all of which are covered. (Though I notice that it's mostly the La advocates who note that while Lu is very similar to Y, it certainly isn't as similar to Sc, which is different because it's even smaller.) There seem to be a few more La arguments around (since this has finally gotten enough spotlight that a few La advocates have responded). But please do suggest some improvements on the talk page; I daresay it would be more useful to our readers than deleting the entire thing, which is all found in reliable sources. Double sharp (talk) 02:26, 20 November 2017 (UTC)


 * Sure, a discussion it is, then.
 * Let me make one point clear: for the purposes of this article, I am not pro arguing for that the idea of lanthanum being a d element is wrong or right. I think that the article should not pick sides at all and the text I deleted definitely did pick a side. (That especially outraged me given that the subsections on bulk properties and isotopes both took less space than a discussion on this frankly minor issue anyway. I wouldn't be surprised to see a shorter subsection on the issue even in a full-sized FA. Group 3 element would be much more appropriate for a detailed discussion on the matter.) The text gave one opinion without fully giving the other one. I could explain how so sentence by sentence if you wish.
 * I am unfortunately not ready at the moment to discuss the "truth" on the matter at the moment (typing this on my cellphone). On a quick note, though, point 5 is clearly biased: if you are to truly determine something, you should pick any position as default to fall back to just in case. You even acknowledge this when you say that Aufbau is just a general guideline and deviations are possible. Also, it would make more sense for me to pick the other version as default because that's the norm for the d block but that's irrelevant anyway because you shouldn't pick any. Also, 9 is words with no support behind them: you can use analogies for illustrating your point but not making one. Why are we talking about cations and neutral atoms anyway? This is not a universal criterion throughout the PT. Also, why compare in many aspects to group II but never to IV? 12 ignores (not debunks; ignores) the point the -Lu-Lr supporters make: La is an exception like copper.
 * I am not arguing that your thesis is flawed or otherwise. I am only saying some of your arguments are and they sometimes don't even try to acknowledge the other position. I think we therefore better have no discussion rather than a flawed one especially given that we probably don't need this discussion in this article anyway.--R8R (talk) 09:39, 20 November 2017 (UTC)
 * I've hidden this section for now, since I'm rather swayed by your point that this article isn't the place for it any more than those for Lu, Ac, and Lr are. We might conceivably salvage part of it for the discussion at group 3 element, though it strikes me that I have internalised -La-Ac too well, so perhaps we should get you to write some of it too. ^_^
 * I'm also short of time, so here are some quick replies:
 * 5. It seems to me that even if you say that the other version is the norm for the d-block, you still need to answer why it is the norm there, and nowhere else. And in fact I find the "lanthanide contraction" explanation a little unconvincing, because it is dependent on where group 3 comes after. It seems to me that you cannot argue in favour of La as eka-Y on those grounds, because it would mean that you don't accept that the lanthanides precede group 3, which means you are in fact begging the question. And the same for Lu as eka-Y, so this is a lame argument and thank you for spotting it, and I shall try to patch up the hole when I have more time to think. I think we can instead have something like:
 * "The predominant trend in the early d-block groups 4 through 8 is for the 4d element to be very similar in size to the 5d element, because of the poor shielding effect of the 4f subshell in the lanthanides that precede them. The predominant trend in the s-block groups is for an increase in size from the 5s element to the 6s element because of the lack of this inner 4f subshell. So the question of La vs Lu can be reduced to the equivalent question of which of them is an f-block element."
 * What do you think of that?
 * 9. Seems to me that the electron configurations are pretty much universal because of how well they reflect chemistry, and when they appear otherwise it's in period one where everything goes topsy-turvy because 1s2 is already a full shell and there's no 1p, making a vast difference from everything else. But I think I'm not phrasing my point very clearly, which is that if La is 4f0 everywhere it doesn't seem to be very convincing as a 4f element, especially since Ac is also 5f0 everywhere.
 * 12. The question of exceptions is also related with why I didn't cover group IV: because only in II and III is the idea of the differentiating electron not helpful. I mean, if you want to compare Ti/Zr/Ce to Ti/Zr/Hf, you just have to see that Ti/Zr/Ce has differentiating electrons 3d/4d/4f from the preceding Sc/Y/La, while Ti/Zr/Hf has differentiating 3d/4d/5d from Sc/Y/Lu, so only the second one gives a match: there are exceptions, of course, but when you consider the question of which orbitals are actually chemically active the point stands with the exceptions as a minor anomaly (like I guess we would all accept for Cu). It's only in II and III that we have a totally consistent set of anomalies: going from group I (A or B) to II (A or B) the differentiating electron is always s, and going from group II (B or "C") to III (B or "C") the differentiating electron is always d, and neither Ca/Sr/Ba | La/Ac nor Zn/Cd/Hg | Lu/Lr show much involvement of the inner subshell. I think my thought process is mostly that if the two situations are this analogous, they should be treated the same way, nebulous as that sounds, because I just find the thought of a 4f element with no 4f involvement at all hard to stomach.
 * Now about the exceptions. I think the point of the exceptions is that they still behave like their supposed blocks by Aufbau if the exceptions weren't there: Cu still acts like a d-block element in how it displays variable oxidation states, and Th still acts like an f-block element, Th3+ being [Rn]5f1. And you can express this more precisely by looking at the energy levels of the electron shells and seeing which ones are drowned in the core, which ones are too high to be usable in chemistry (rather than in physics), and which ones are at a happy medium (all three being reasonably far apart and distinguishable), which takes care of all the exceptions. Cu shows 3d and 4s as valence subshells, Pd shows 4d and 5s, and so on, so while the details are a bit off the general picture of what is being filled is right. And it is very right for the beginnings of chemical activity of a subshell, since the falls from the heights into the medium-range are very steep, even if the descent into the core is slow (graph of 3d, 4s, 4d, 5s at doi 10.1021/ed8001286). But still La and Ac show up with 4f and 5f respectively being up in the heights and not ready to collapse till Ce and Th, and it's that point that clinched it for me, TBH. Double sharp (talk) 15:07, 20 November 2017 (UTC)
 * Sorry to keep you waiting. I read your response yesterday and formulated a reply on that very day. However, as soon as I came home, I found myself too tired and sleepy to write a long meaningful message. Same today even though it's much earlier today. I know what I mean to write back and will as soon as I can.--R8R (talk) 16:03, 21 November 2017 (UTC)
 * Yes, the new 5 is okay.
 * The problem in your thinking is that you've clearly chosen a perspective from which you would look at the problem and don't try to do the same from the other position; you'd otherwise see most, if not all, of your arguments can be mirrored. I am not unsympathetic to this as I was once a heavy -Lu-Lr supporter. The point that got me there for sure was exactly that Lu added an electron to an already complete shell (which precisely mirrors your killing argument, making neither really "killing"). I'm changing my ways, though. During our last discussion at WT:ELEM, I did suggest this was only a problem of personal liking; however, quite to my surprise, we have found some uniquely pro-La-Ac and pro-Lu-Lr points. I didn't investigate on them due to lack of free time and you and Sandbh were quite okay not to, about which I was genuinely surprised. (That being said, if I recall all of that correctly.) I suggest you ensure being right by assuming the opposing position and see if they can compete with your argument.
 * I understand your 12. Okay then. Still, if you want to look at the differentiating electron, again, look at the other side and investigate group III: if it is okay to use Be-Mg-Ca/Zn and Ca-Sr-Ba/Hg to support Sc-Y-La, then it is just as okay to use B-Al-Sc/Ga and Ga-In-Lu/Tl to support Sc-Y-Lu, which exactly mirrors your point.
 * Yes, there is indeed some unique validity in that a series begins when behavior common for that series starts. This must be a uniquely pro-La-Ac point, since Yb still shows some а behavior, so a Lu showing none of it well parallels Zn showing none of d. That is noted. Perhaps it would be a good idea to revisit that old discussion to see all of the unique arguments on both sides.
 * For the sake of completeness, your 9, however, isn't just flawed; it contradicts chemistry as I know it. We don't judge about an element's location by properties of its ions; we judge it by properties of the neutral element. For instance, cesium is an s-block element alright despite it being 6s0 basically on every occasion. Cesium is an s-block element because its differentiating electron is 6s. Interestingly, you already referred to the differentiating electron; keep your consistency at it.--R8R (talk) 19:59, 22 November 2017 (UTC)
 * I've looked at this and will give an actual response soon; still need to think about it a bit more. Off the cuff I'd say that what I meant is that La and Ac didn't have f-electrons to begin with and they're not using these orbitals even to remove electrons from or put more in, like Cs getting rid of its 6s electron it already had or Th bumping its electrons down into 5f as you ionise it, but I still need to write it up properly. Double sharp (talk) 09:09, 25 November 2017 (UTC)

Unfortunately real life intervened after that post and I forgot about this discussion, but in the meantime this issue got another airing in the WT:ELEM archives. That latest discussion came very close indeed to convincing me back to Sc-Y-Lu-Lr. ^_^ I do agree now that group 3 element is the proper place for this longstanding argument and that any discussion here ought to be short and perhaps just limited to saying that there is an argument, because regardless of which side you pick, you still need to do comparative chemistry with Ac and the rare earths anyway. So I think it's fine that this section is no longer in the article, but that the lede mentions that La's exact classification is under some debate and both La and Lu have sometimes been assigned the 5d1 slot (and there's not much we could do in the body except for repeating tha statement again without going off-topic, so I think we might be able to invoke IAR for its absence from the body). Double sharp (talk) 13:07, 7 December 2018 (UTC)
 * I absolutely agree about the part on the lead section. As for mentioning that in the body, I suggest considering a mention in History, something similar to Thorium. History would gain from some expansion anyway.--R8R (talk) 13:40, 8 December 2018 (UTC)

Thankfully, I finally learned enough to be able to tell where exactly this argument fails: points 8 and 11. Lanthanum has f-involvement, and it continues until only ytterbium. Therefore, the case doesn't work, and so I changed my mind to -Lu-Lr (as most sources focusing on this point agree with, even if textbooks are slow to update). Fortunately, IUPAC has already published the provisional report, so this mistake (as Bernd T. Mathias put it) could finally be corrected. Although it still has to get into an actual report, the IUPAC periodic table, and finally out in the wild. Double sharp (talk) 13:08, 8 February 2021 (UTC)

Use in high-temperature superconductors
I am missing a mention of the use of La in High-temperature superconductors (HTS) and more broadly the Lanthanum hydrides. Unfortunately, I'm not sure where to put it, since HTS are not really 'applications' yet. Any thoughts on this?

Another good source for this is this very recent paper.

Aypac (talk) 15:27, 10 January 2019 (UTC)

World annual production?
No information in the article. I found few other sources, but they are not accurate, or outdated. Around 2011 I see sources say 12.5 kt per year. 2015 estimates say 54 kt (oxide). More recent articles say that China alone is producing about 90 kt of La and Ce (together) per year (while world usage is about 50kt per year of La and Ce together). Unclear if this is metal, but more probably oxide. 81.6.34.169 (talk) 16:35, 22 June 2024 (UTC)


 * Thanks. Without good sources we should say nothing. ✅ Johnjbarton (talk) 16:50, 22 June 2024 (UTC)