Talk:Electron configurations of the elements (data page)

Reference by Haire
This newly added reference seems to be the only source for the undiscovered elements, Z > 118. Could we please have more information on what it contains. Perhaps a sentence such as This article contains predicted electron configurations for the elements 119-168, based on approximate relativistic Hartree-Fock-Dirac calculations with configuration interaction and spin-orbit coupling by X and Y (journal reference). Note that my version is probably incorrect because I have not seen the article and I am guessing, but it illustrates the level of detail which I believe should be added to the mention of Haire's article. Dirac66 (talk) 15:04, 3 December 2012 (UTC)
 * ✅ Double sharp (talk) 06:10, 4 December 2012 (UTC)

Don't worry, i will go and do some research things. Hubbiety (talk) 09:16, 22 December 2016 (UTC)

Chromium and other exceptions
To Bobdolan22 and everyone else who keeps trying to "correct" chromium: The point is that the Madelung rule (1s, 2s, 2p, 3s, 3p, 4s, 3d, etc.) is only an approximation, and there are lots of exceptions of which chromium is the first. Yes, the rule would say that Cr is 4s23d4, but the experimental atomic spectrum shows that it is actually 4s13d5. This article gives the true experimental configurations, even when they disagree with the Madelung rule. For more discussion see Electron configuration. Dirac66 (talk) 01:06, 25 March 2013 (UTC)
 * Perhaps interesting cases and exceptions to important "rules", like that which you mentioned, can be pointed out on the article in some way? JonathanHopeThisIsUnique (talk) 04:07, 7 January 2016 (UTC)
 * Good idea. I have now added a second short paragraph in the introduction to the table, to explain what the rule is and to point out that there are numerous exceptions. Dirac66 (talk) 18:11, 7 January 2016 (UTC)

I should also point out that these "irregularities" are an artifact of insisting on ground-state gas-phase electron configurations. The d and f elements have lots of configurations very close in energy and difference in chemical environments can easily change which happens to be the ground state, so the Madelung exceptions you often see actually have very little relevance to real chemistry. In particular, the Madelung-predicted configurations like [Ar]3d44s2 are always at least close to the ground state for the first 118 elements, well within the range of chemical bond energies. The ref and paragraph I added points out this chemical irrelevance of the irregularities. Double sharp (talk) 07:59, 28 May 2020 (UTC)

New property for WikiData
Please vote for new property of WikiData - "Electron_configuration" creation Васин Юрий (talk) 18:32, 22 August 2019 (UTC)

Nefedov
Hi - comparing and contrasting this with Periodic Table I see Nefedov et al's results are not examined here - http://www.primefan.ru/stuff/chem/nefedov.pdf. They are quite different – FLYING CHRYSALIS  💬 20:37, 27 December 2019 (UTC)
 * There have not been complete and conclusive results from E123 onwards: there are many disagreements in the sources (the two Fricke papers do not even agree with each other 100% IIRC). I suppose we should mark this as just one possibility, and present the others as well, and do the same Extended periodic table. That would be more in line of what the other data pages do, e.g. Boiling points of the elements (data page). Double sharp (talk) 04:58, 28 December 2019 (UTC)
 * I definitely support that; we'd then have to add additional rows for each undiscovered element and label them as distinct predictions from different sources (or some other substantial reformatting). ComplexRational (talk) 13:22, 28 December 2019 (UTC)
 * Perhaps there should be several tables, labelled something like "Configurations determined from experimental spectra", "Configurations predicted for known elements" and "Configurations predicted for undiscovered elements." Dirac66 (talk) 21:46, 28 December 2019 (UTC)
 * Sure, that would be all right. What's the last element for which the configuration is actually determined from experimental spectra? AFAIK that for Lr is only inferred from volatility experiments, for example, not directly measured. Double sharp (talk) 06:38, 29 December 2019 (UTC)
 * According to the articles and ref Silva, it seems the last directly confirmed configuration is Fm. ComplexRational (talk) 14:26, 29 December 2019 (UTC)
 * OK, so I guess we would have to split into H through Fm (determined from spectra); Md through Hs (inferred through chemical experiments); Mt through Og (calculations only); E119 and up (calculations only, elements not even discovered). Double sharp (talk) 15:21, 29 December 2019 (UTC)
 * Yes, and we can add notes to the element entries to deal with special cases, as for Nickel in the table now.
 * Also note that there is some information on the reliability of the result for each element in the References section, especially in the WebElements subsection. But the table should still be divided up to make the point more clearly. Dirac66 (talk) 01:55, 30 December 2019 (UTC)
 * The g shell configurations should be regarded as tentative, and I'm not sure that it even makes sense to assign a single configuration. In my (unpublished) DHF/CI calculations on single configurations 5gj 6fk 7dl 8sm 8pn, there are often near-degeneracies, e.g. 6f3 8p1 and 6f2 8p2, and even indeterminate 5g occupations around Z=129-131. So these are very multi-reference states with many possible contributing configurations. KGDyall (talk) 20:46, 11 September 2020 (UTC)
 * Hey, thanks for telling us this! I've added a brief note about the tentativeness beyond E120 to the article, citing the Nefedov article above.
 * You might like to talk to Droog Andrey too: he's also a computational chemist (I'm not), and he has some great ideas about how post-120 elements should best fit into the periodic table. Double sharp (talk) 13:19, 12 September 2020 (UTC)

I've been discussing this with on his talk page, and looking around at various sources, and it looks to me like we should really prune this table because there is not much agreement between sources past 121 in fact. (Nefedov in fact makes something close to the point did above; there is so much configuration interaction that a single configuration may well start becoming a little bit nonsensical.) Therefore I am thinking that this table should be pruned to either element 118 (predictions up there) as the heaviest known element, or element 121 (three more predictions) as the heaviest element for which all contemporary sources seem to agree on the configuration; and to leave what lies ahead to Extended periodic table (where the table should be edited to show all the possibilities that have been floated, I guess). What do you think? Double sharp (talk) 13:37, 11 February 2021 (UTC)


 * Thanks for telling me about the article Extended periodic table of which I was not aware. The last section Extended periodic table seems to be a good presentation of the predictions for the undiscovered elements, so I don't think it is necessary to repeat the information in this article. Therefore I now think that this article should end with Oganesson (E118), at least until more elements are actually discovered. This can be followed by a link explaining that predicted configurations for undiscovered elements to E173 are in the other article.
 * And I still think it would be a good idea to indicate the type of data for each element Z = 1-118. We could divide the table into three parts as discussed above (H-Fm, Md-Hs, Mt-Og), but this division could become awkward if in the future the elements in each category are no longer consecutive. Suppose for example that in 2025, someone reports chemical experiments which indicate the configuration of Flerovium (E114) only. Do we then put it in Table 2 even though Mt-Nh are still in Table 3 (calculations only)? A more flexible presentation allowing for new data might be to retain one table, but to add after the name of each applicable element words such as "based on chemical behavior" (for Md-Hs now) and "predicted from theoretical calculations" (for Mt-Og now). Dirac66 (talk) 21:22, 11 February 2021 (UTC)
 * One table sounds the best to me for those reasons. So, we'll revert to when this stopped at Og, and add new elements as predictions only when they're discovered. Double sharp (talk) 07:16, 12 February 2021 (UTC)
 * P.S. It does not seem like the sources are making much of the distinction between configurations from direct measurement and configurations from inferences – see for example NIST which just gives everything till Hs without comment. So, I have stuck with the lede statement that elements beyond Hs are just predicted and a two-way division between experimentally known and experimentally unknown. Double sharp (talk) 10:09, 12 February 2021 (UTC)


 * I think that it is much better now. I have added a See also link to the extended table Z = 119-173 so the information will be accessible. However by putting it in a separate article it seems clearer that the configurations are not to be taken as seriously for undiscovered elements. Dirac66 (talk) 16:36, 12 February 2021 (UTC)

Electron configurations of monovalent ions
Data from https://www.nist.gov/pml/atomic-reference-data-electronic-structure-calculations/atomic-reference-data-electronic-8. Data for elements beyond U are predictions.

The elements having irregular electron configurations for their atoms also have irregular electron configurations for monovalent ions. V, Co, Ni, Y, Lu have regular electron configurations for their atoms but their monovalent ions are irregular.

The elements having only one (n-1)d electron (Y, Lu, Ac, Pa, U, Np, Cm) tend to lose its (n-1)d electron before ns when forming monovalent ions. This does not apply to Sc and Gd, which lose an ns electron before (n-1)d. La and Ce even lose two ns electrons and get one more (n-1)d electron. Here Lu (and not La) behaves as the real homologue of Y.


 * I made some changes to your table, per NIST (up to Hs) and the calculations in (Mt-Cn). Double sharp (talk) 10:39, 3 May 2024 (UTC)
 * P.S. for divalent ions (once the g-shells come in, these will probably be close to the ground state in many cases, though I won't be surprised if they're not exactly the ground state). Double sharp (talk) 10:42, 3 May 2024 (UTC)
 * Thanks! 129.104.241.193 (talk) 21:09, 5 May 2024 (UTC)
 * Thanks to the link you provided, I could list the exceptions for divalent ions as follows:
 * {| class="wikitable" style="text-align:center"

! Z !! Symbol !! Electron configuration by Aufbau principle !! Actual electron configuration
 * 57 || La || [Xe] 4f1 || [Xe] 5d1
 * 64 || Gd || [Xe] 4f8 || [Xe] 4f7 5d1
 * 71 || Lu || [Xe] 4f14 5d1 || [Xe] 4f14 6s1
 * 89 || Ac || [Rn] 5f1 || [Rn] 7s1
 * 90 || Th || [Rn] 5f2 || [Rn] 5f1 6d1
 * 91 || Pa || [Rn] 5f3 || [Rn] 5f2 6d1
 * 103 || Lr || [Rn] 5f14 6d1 || [Rn] 5f14 7s1
 * 104 || Rf || [Rn] 5f14 6d2 || [Rn] 5f14 7s2
 * 105 || Db || [Rn] 5f14 6d3 || [Rn] 5f14 6d2 7s1
 * 106 || Sg || [Rn] 5f14 6d4 || [Rn] 5f14 6d3 7s1
 * 107 || Bh || [Rn] 5f14 6d5 || [Rn] 5f14 6d4 7s1
 * 108 || Hs || [Rn] 5f14 6d6 || [Rn] 5f14 6d5 7s1
 * 109 || Mt ||[Rn] 5f14 6d7 || [Rn] 5f14 6d6 7s1
 * 110 || Ds || [Rn] 5f14 6d8 || [Rn] 5f14 6d7 7s1
 * 111 || Rg || [Rn] 5f14 6d9 || [Rn] 5f14 6d8 7s1
 * 112 || Cn || [Rn] 5f14 6d10 || [Rn] 5f14 6d8 7s2
 * }
 * It is quite irregular that Y2+, Ce2+, Hf2+, U2+, Np2+, and Cm2+ are regular. It is even more astonishing that Th3+ might be regular (5f1). 129.104.241.193 (talk) 21:22, 5 May 2024 (UTC)
 * Some time ago I wrote a note (h) in Periodic table about this kind of thing. There's a gradual transition between the Madelung order (roughly correct at zero charge) and simply filling in order of n (roughly correct for almost-bare nuclei). Within the chemically relevant ionisations for metals (roughly +2 to +6), this amounts to a mild correction to Madelung for the most part, in which (n−1)d and (n−2)f get drowned below ns and np. The 6d metals are exceptions for relativistic reasons. Double sharp (talk) 08:48, 6 May 2024 (UTC)
 * 106 || Sg || [Rn] 5f14 6d4 || [Rn] 5f14 6d3 7s1
 * 107 || Bh || [Rn] 5f14 6d5 || [Rn] 5f14 6d4 7s1
 * 108 || Hs || [Rn] 5f14 6d6 || [Rn] 5f14 6d5 7s1
 * 109 || Mt ||[Rn] 5f14 6d7 || [Rn] 5f14 6d6 7s1
 * 110 || Ds || [Rn] 5f14 6d8 || [Rn] 5f14 6d7 7s1
 * 111 || Rg || [Rn] 5f14 6d9 || [Rn] 5f14 6d8 7s1
 * 112 || Cn || [Rn] 5f14 6d10 || [Rn] 5f14 6d8 7s2
 * }
 * It is quite irregular that Y2+, Ce2+, Hf2+, U2+, Np2+, and Cm2+ are regular. It is even more astonishing that Th3+ might be regular (5f1). 129.104.241.193 (talk) 21:22, 5 May 2024 (UTC)
 * Some time ago I wrote a note (h) in Periodic table about this kind of thing. There's a gradual transition between the Madelung order (roughly correct at zero charge) and simply filling in order of n (roughly correct for almost-bare nuclei). Within the chemically relevant ionisations for metals (roughly +2 to +6), this amounts to a mild correction to Madelung for the most part, in which (n−1)d and (n−2)f get drowned below ns and np. The 6d metals are exceptions for relativistic reasons. Double sharp (talk) 08:48, 6 May 2024 (UTC)
 * 111 || Rg || [Rn] 5f14 6d9 || [Rn] 5f14 6d8 7s1
 * 112 || Cn || [Rn] 5f14 6d10 || [Rn] 5f14 6d8 7s2
 * }
 * It is quite irregular that Y2+, Ce2+, Hf2+, U2+, Np2+, and Cm2+ are regular. It is even more astonishing that Th3+ might be regular (5f1). 129.104.241.193 (talk) 21:22, 5 May 2024 (UTC)
 * Some time ago I wrote a note (h) in Periodic table about this kind of thing. There's a gradual transition between the Madelung order (roughly correct at zero charge) and simply filling in order of n (roughly correct for almost-bare nuclei). Within the chemically relevant ionisations for metals (roughly +2 to +6), this amounts to a mild correction to Madelung for the most part, in which (n−1)d and (n−2)f get drowned below ns and np. The 6d metals are exceptions for relativistic reasons. Double sharp (talk) 08:48, 6 May 2024 (UTC)
 * Some time ago I wrote a note (h) in Periodic table about this kind of thing. There's a gradual transition between the Madelung order (roughly correct at zero charge) and simply filling in order of n (roughly correct for almost-bare nuclei). Within the chemically relevant ionisations for metals (roughly +2 to +6), this amounts to a mild correction to Madelung for the most part, in which (n−1)d and (n−2)f get drowned below ns and np. The 6d metals are exceptions for relativistic reasons. Double sharp (talk) 08:48, 6 May 2024 (UTC)