Period 3

Period 3 is the row that teaching syllabuses are built on, and it earns the position honestly. It has eight members, one from each main group; it runs from an unmistakable metal to an unmistakable noble gas with a semiconductor exactly in the middle; and unlike period 2 it does not spend the whole row disagreeing with the columns it belongs to. Sodium behaves like potassium. Chlorine behaves like bromine. What you learn here transfers.

Electronegativity crosses almost the full width of the Pauling scale in eight steps, from 0.93 at sodium to 3.16 at chlorine, and everything else in the row follows that single gradient. What makes the row worth studying is that its most striking trend is not about the atoms at all.

A melting-point graph that is really a graph of structure

Chart the melting points of these eight elements and the line does something no smooth atomic property could produce. It climbs from sodium through magnesium and aluminum, jumps to a sharp peak at silicon, and then collapses — phosphorus, sulfur, chlorine and argon all melt below the temperature at which sodium does.

The shape has nothing to do with atomic radius or nuclear charge. It is a graph of three different kinds of solid.

• Sodium, magnesium and aluminum are metallic lattices. Each contributes one, two and three electrons respectively to a delocalised sea, and each cation carries a correspondingly higher charge, so the electrostatic attraction holding the lattice together strengthens across the three.
• Silicon is a giant covalent network. Every atom is bonded to four others in a continuous three-dimensional structure, and melting it means breaking covalent bonds throughout the crystal. Nothing else in the row comes close.
• Phosphorus, sulfur, chlorine and argon are simple molecular solids. Their covalent bonds are strong, but those bonds are inside the molecules and are not broken on melting. What has to be overcome is the weak dispersion force between one molecule and the next.

The order within that last group is the detail worth noticing, because it does not follow the row. Sulfur melts higher than phosphorus even though phosphorus comes first, because an S₈ ring has far more electrons than a P₄ tetrahedron and is therefore far more polarisable. Molecular size, not position in the period, sets the ranking — which is the clearest possible warning against reading physical properties straight off the table.

Oxides: eight steps from strongly basic to strongly acidic

The chemical trend across period 3 is best seen in what each element does with oxygen, because the change is monotonic and it changes character at exactly the point the metal-to-nonmetal walk does.

Sodium oxide and magnesium oxide are ionic and basic; the first dissolves to give a strongly alkaline solution, the second only sparingly, producing the mildly alkaline suspension familiar from indigestion remedies. Aluminum oxide is the hinge: it reacts with acids as a base and with hot concentrated alkali as an acid, which is what amphoteric means, and it sits precisely where the metals end.

Past that point the oxides are acidic and increasingly so. Silicon dioxide is a giant covalent solid and only weakly acidic, reacting with strong alkali and not much else. Phosphorus(V) oxide and sulfur trioxide are molecular and give phosphoric and sulfuric acid respectively. Dichlorine heptoxide gives perchloric acid, one of the strongest acids there is.

Three things change together across that sequence: the bonding goes ionic, then giant covalent, then simple molecular; the melting points of the oxides fall accordingly; and the acid-base character inverts. They change together because they are the same phenomenon described three ways — the metal is losing its grip on its electrons less and less willingly, and eventually stops trying.

Chlorides and the polarising cation

The chlorides make the same point more sharply, because the transition from ionic to covalent happens over fewer elements.

Sodium chloride is a high-melting ionic solid that dissolves to a neutral solution. Magnesium chloride is still ionic but its solution is faintly acidic, because the hydrated Mg²⁺ ion is polarising enough to weaken an O–H bond in its own water of hydration. Aluminum chloride abandons the ionic model altogether: it sublimes rather than melting cleanly, exists in the vapour as an Al₂Cl₆ dimer, and reacts with water to give an acidic solution and hydrogen chloride fumes. Silicon and phosphorus chlorides hydrolyse completely.

The underlying rule is Fajans': a cation that is small and highly charged distorts a neighbouring anion's electron cloud, and once the distortion is severe enough the bond is better described as covalent. Across sodium, magnesium and aluminum the charge rises from +1 to +3 while the ionic radius shrinks, so polarising power increases far faster than either factor alone. By aluminum the ionic description has stopped being useful, and the row has not yet reached its halfway point.

A shell with room for eighteen and a row of eight

Period 3 is the first row where the number of elements and the capacity of the shell come apart. For n = 3 the angular momentum quantum number may be 0, 1 or 2, so 3s, 3p and 3d orbitals exist, and eighteen electrons would be required to fill them. The row is eight elements long.

The missing ten are not missing; they are late. The 3d subshell lies higher in energy than 4s, so electrons occupy 4s first, and 3d does not begin filling until the following row. That single inversion is why the periodic table has a wide middle section at all, and why this row is eight long while the next one is eighteen.

The consequence for period 3 chemistry is that its later members can form more bonds than their period-2 counterparts — phosphorus pentachloride, sulfur hexafluoride, the perchlorate ion — while their d orbitals remain empty in the ground state. These atoms are larger and can simply accommodate more neighbours than a second-row atom of the same group can.

Four elements out of eight are what the ground is made of

By mass the Earth's crust is a little under half oxygen — 46 per cent on the standard handbook figures. Silicon takes 28 per cent and aluminum 8.2, so those two plus oxygen already account for more than four-fifths of everything above the mantle.

It is worth getting the order right after that, because this row is routinely credited with more of the list than it holds. Fourth place is iron at 5.6 per cent and fifth is calcium at 4.2 — both period 4. Sodium is sixth at 2.4 and magnesium seventh at 2.3, with potassium eighth. So of the eight commonest elements in the crust, four are period 3 — silicon, aluminum, sodium, magnesium — which is half the members of a row eight elements long, and those four between them are about 41 per cent of the crust by mass. The two elements that break up the run at the top are not one interloper but two, and magnesium does not make the leading five at all.

Why a row with no heavy metals in it should dominate the accessible planet comes down to where each element went when the core separated out. Silicon, aluminum, sodium and magnesium are lithophile: their affinity is for silicate rock, and they stayed in it. Iron's is for molten metal, and most of the Earth's iron is now several thousand kilometres below the crust — iron is around a third of the whole planet by mass but only 5.6 per cent of its outermost shell. Period 3 is over-represented up here partly because it contains nothing that sank.

Feldspars make the point in one formula. They are aluminosilicates of sodium, potassium or calcium, they are the most abundant mineral group in the crust, and three of the five metals and semimetals that can appear in their formulae come from this row.

Argon closes the period at 3s²3p⁶, and although it is nearly one per cent of the atmosphere by volume, it is the least remarked-upon common substance on Earth precisely because a filled p subshell gives it nothing to do. One final measurement caution: the atomic radius quoted for argon looks larger than chlorine's and appears to break the contraction trend. It does not — argon's figure is a van der Waals radius, measured between non-bonded atoms, while chlorine's is a covalent radius measured within a bond. They are different quantities and the comparison is meaningless.