Colleagues,
I seem to have stumbled on an organisational parallel between particle physics and chemistry, and would be grateful for your thoughts as to how far it holds.
In the attached image:
I’ve appended my further impressions.
René

Notes to imageA. The dashed line between the d-block metals (non-noble) and the noble metals denotes that the latter are a subset of the former while the dashed line between the biocentric nonmetals and the noble gases acknowledges the status of nitrogen as the original noble gas, before the discovery of the noble gases proper.B. Although d-block metals are on average slightly less electronegative than p-block metals, which in turn are slightly less electronegative than metalloid nonmetals, these differences are not chemically significant at the level of abstraction used here and are therefore treated as effectively zero.
My further impressions
In the case of the lower figure, my working presumption is that:
…amounts to a kind of carving nature at its joints.
I’m thinking of this in the sense that chemistry is structured by fundamental dualities (acid/base, cation/anion, oxidation/reduction), where paired relationships are not decorative but explanatory: they organise chemical behaviour and enable inference.
In the case of metals and nonmetals:
What is interesting is that if one then overlays average EN values on the eight types of metals and nonmetals (noting that EN wasn’t used to define the types in the first place), the within-pair contrasts can be organised into a small number of fairly clean, cross-cutting “bands” (large, intermediate, proximal, and quite small). This includes the vertical noble gas / noble metal contrast (Δχ ≈ 1), which intersects the middle horizontal contrast band (also Δχ ≈ 1).
Given how many atomic and chemical parameters correlate with EN, this alignment offers convergent support that the eightfold partition is tracking real regularities in chemical behaviour at a coarse-grained level, rather than being merely an aesthetic (symmetrical) rearrangement.
It also seems to me that, in one respect, the chemical “eightfold way” is the converse of the particle-physics Eightfold Way.
In physics, one begins with two coordinates (e.g. charge and strangeness), and the symmetry pattern reveals the multiplets.
For the elements of the periodic table, one begins with chemically salient classes (and their complementarities) and only then asks whether an independent quantitative proxy—EN, with its broad correlational reach—maps onto the resulting structure.
The fact that it does suggests that, in much the same way that the Eightfold Way in physics captured interrelationships among subatomic particles, the iterative pairing scheme is capturing something like “natural joints” in chemical behaviour at this level of abstraction.

--
You received this message because you are subscribed to the Google Groups "Periodic table mailing list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to PT-L+uns...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/PT-L/85E52487-8925-4F03-B951-01F7A280AFAF%40iinet.net.au.
Interesting. It reminds me of a color wheel with the mix of black and white, that is grey in the middle.
<image.png>Best Regards,V. "Larry" Tsimmerman
On Thu, Jan 15, 2026 at 6:28 AM 'René' via Periodic table mailing list <PT...@googlegroups.com> wrote:
Colleagues,
I seem to have stumbled on an organisational parallel between particle physics and chemistry, and would be grateful for your thoughts as to how far it holds.
In the attached image:
- the top figure, from particle physics, shows the Eightfold Way organisation of the lightest baryons, with strangeness (s) on one axis and electric charge (q) on the other;
- the lower figure organises the metals and nonmetals of the periodic table into a similar eightfold arrangement. In this case, the axes show differences in average EN, which serve as proxies for contrasts in bonding behaviour and metallic character.
I’ve appended my further impressions.
René
<PastedGraphic-2.png>
Notes to imageA. The dashed line between the d-block metals (non-noble) and the noble metals denotes that the latter are a subset of the former while the dashed line between the biocentric nonmetals and the noble gases acknowledges the status of nitrogen as the original noble gas, before the discovery of the noble gases proper.B. Although d-block metals are on average slightly less electronegative than p-block metals, which in turn are slightly less electronegative than metalloid nonmetals, these differences are not chemically significant at the level of abstraction used here and are therefore treated as effectively zero.
My further impressions
In the case of the lower figure, my working presumption is that:
- starting from the primary metal/nonmetal distinction;
- identifying the “noble metals / noble gases” pair; and
- continuing to iteratively pair the remaining types of metals and nonmetals…
…amounts to a kind of carving nature at its joints.
I’m thinking of this in the sense that chemistry is structured by fundamental dualities (acid/base, cation/anion, oxidation/reduction), where paired relationships are not decorative but explanatory: they organise chemical behaviour and enable inference.
In the case of metals and nonmetals:
- each kind spans a gamut from highly reactive through to chemically reluctant (“noble”); and
- several match-ups are fairly intuitive—beginning, as noted, with the noble metals and noble gases.
What is interesting is that if one then overlays average EN values on the eight types of metals and nonmetals (noting that EN wasn’t used to define the types in the first place), the within-pair contrasts can be organised into a small number of fairly clean, cross-cutting “bands” (large, intermediate, proximal, and quite small). This includes the vertical noble gas / noble metal contrast (Δχ ≈ 1), which intersects the middle horizontal contrast band (also Δχ ≈ 1).
Given how many atomic and chemical parameters correlate with EN, this alignment offers convergent support that the eightfold partition is tracking real regularities in chemical behaviour at a coarse-grained level, rather than being merely an aesthetic (symmetrical) rearrangement.
It also seems to me that, in one respect, the chemical “eightfold way” is the converse of the particle-physics Eightfold Way.
In physics, one begins with two coordinates (e.g. charge and strangeness), and the symmetry pattern reveals the multiplets.
For the elements of the periodic table, one begins with chemically salient classes (and their complementarities) and only then asks whether an independent quantitative proxy—EN, with its broad correlational reach—maps onto the resulting structure.
The fact that it does suggests that, in much the same way that the Eightfold Way in physics captured interrelationships among subatomic particles, the iterative pairing scheme is capturing something like “natural joints” in chemical behaviour at this level of abstraction.
On 19 Jan 2026, at 10:29 PM, Rene <re...@iinet.net.au> wrote:
On 16 Jan 2026, at 1:38 AM, Larry T. <ora...@gmail.com> wrote:
A