Period 6

Period 6 is the longest completed row in the table, and the first in which four different kinds of subshell are filled. The 6s pair comes first, then fourteen 4f electrons, then ten 5d, then six 6p: thirty-two elements between cesium and radon. It is also the row in which the periodic table stops being explicable by non-relativistic physics.

The filling order follows the same n + l argument as the rows above. For 6s the sum is 6; for 4f, 5d and 6p it is 7 in every case, so the tie is broken by principal quantum number and the subshells go in the order 4f, 5d, 6p. The f-block appears here because this is the first row where an f subshell has dropped low enough in energy to be reached.

A row that will not fit on a page

Drawn honestly, period 6 is thirty-two columns wide, and so is period 7. A table that wide is unprintable on paper of any normal proportion, so the fourteen f-block elements are conventionally cut out and set below the main body as a detached footnote, with the main table reduced to eighteen columns.

That is a typographic decision, not a chemical one, and it has had lasting consequences for how the table is understood. Readers absorb the idea that the lanthanides are somehow apart from the table rather than being a contiguous stretch of one of its rows. Wide-format tables that keep the f-block in place exist and are perfectly correct; they are simply awkward to hang on a wall.

Where a small correction becomes the main term

Relativistic effects on electrons are present in every atom. What changes along this row is their size. The speed of an innermost electron scales roughly with nuclear charge, so where hydrogen's sits near 1/137 of light speed, gold's is at 79/137 — a little under three-fifths — and the mass increase that comes with it stops being a fourth-decimal correction to a calculation and starts setting the values a person can measure with a thermometer or see with an eye.

Two consequences follow, and they pull in opposite directions. The s orbitals contract. The d and f orbitals, which never penetrate to the nucleus in the same way and are better screened once those s electrons tighten, expand and rise in energy. Why an innermost contraction should reach the whole way out to the 6s, and what that does to an element with nothing but a 6s pair to bond with, is worked through step by step in why mercury is a liquid.

For reading the row, the point is that the effect is graded rather than switched on. It is barely worth a mention at cesium; it is the dominant term by the time the row reaches its heaviest metals.

Gold is yellow, and a car battery works

The illustration everyone reaches for is a colour. Gold and silver share an outer electron arrangement and are close chemical relatives, and they do not look remotely alike; the whole of the difference is that one of them sits in this row and the other in the row above. The electronvolt figure behind that belongs on gold's own page. What it establishes here is that the row's physics does not stay in a spectroscopist's fourth decimal place — it is visible across a jeweller's counter.

Two further oddities have the same root. Mercury stays liquid at room temperature because the contraction locks up a filled 6s² pair that would otherwise be doing the bonding; a note of its own sets that argument out in full. The refusal of thallium, lead and bismuth to spend their 6s pairs — the inert pair effect — is the same phenomenon acting in the p-block, and is treated with the post-transition metals.

The least expected consequence is in a car. A lead-acid cell delivers about two volts, and calculations published in 2011 by Rajeev Ahuja and colleagues found that roughly 1.7 of those volts come from relativistic contributions to the electronic structure of lead. Without them the cell voltage would be too low for the battery to be practical, and the internal-combustion vehicle would have needed a different starter technology. Very few results connect fundamental physics to ordinary machinery that directly.

The densest and the most refractory metals are all here

The fourteen f-block elements inserted into this row have a structural effect on everything that follows them. The 4f electrons shield the growing nuclear charge poorly, so the atoms have already been drawn in by the time the d-block picks up again at hafnium. A 5d metal is no bigger than the 4d metal directly above it and weighs close to twice as much, and the results are extreme.

Osmium and iridium are the two densest elements known, both above 22 g/cm³, and the long-running question of which one wins now has an answer: osmium, on lattice-parameter determinations precise enough that the 0.027 g/cm³ between them is larger than the uncertainty on either value. Reference tables still disagree, because rankings taken from weighed samples swapped order repeatedly through the twentieth century. The osmium page sets out the figures, and the low-temperature and high-pressure regimes where the order is not settled.

No metal melts higher than tungsten; rhenium is next, and nothing in the table boils higher than rhenium does. Platinum, iridium and gold are chemically among the least reactive metals there are.

That combination — very dense, very refractory, very unreactive — is not distributed around the table. It is concentrated in one stretch of one row, and it is concentrated there because of fourteen elements that were inserted upstream and shield badly.

Where does this row's d-block start?

The layout of period 6 hides an unresolved argument. Does the d-block begin at lanthanum, with the f elements running cerium to lutetium, or does it begin at lutetium, with the f elements running lanthanum to ytterbium? The question is the same as asking which element sits below yttrium in group 3.

Configuration alone does not settle it, because lanthanum's ground state has a 5d electron and no 4f electron at all, while lutetium has a filled 4f shell plus a 5d electron. An IUPAC project reported in 2021 in favour of lutetium, on the grounds of physical and structural trends, and many tables still print lanthanum. A third layout splits the difference by leaving group 3 with a fifteen-element footnote and no single element under yttrium, which satisfies nobody and is arguably the most honest of the three.

Nothing chemical depends on the answer. What depends on it is whether the row is drawn with its f-block starting at element 57 or element 58, and periodic tables printed in the same decade disagree.

The row runs out of stable isotopes

Period 6 ends with three consecutive elements that have none: polonium, astatine and radon. All three were found as members of decay chains rather than as constituents of minerals, which is a different kind of discovery from anything in the rows above, and they mark the boundary beyond which the table is entirely radioactive.

They also mark the point at which the row's chemistry becomes partly inferential. No one has ever accumulated enough astatine to see it, so much of what is said about the element is extrapolated down its column rather than measured, and this site flags that distinction on its page instead of presenting prediction as observation. Period 6 opens with cesium, whose properties are pinned to more decimal places than almost anything else in the table because the SI second is defined from one of them, and closes with three elements whose properties are partly guesses. That range, inside a single row, is peculiar to this one.