Alkali Metals

Group 1 is the simplest column in the periodic table to describe and the hardest to keep in a bottle. Every member has a single electron sitting alone outside a completed noble-gas core, and essentially all of their chemistry follows from how loosely that one electron is held. Lose it and you get a cation with a rare-gas configuration, which is such a comfortable place to end up that none of these metals occurs uncombined anywhere on Earth.

That single-electron picture is not an approximation made for teaching. It shows up directly in the numbers: the first ionisation energy of an alkali metal is the lowest in its period, and the second ionisation energy is enormous — for sodium it jumps from 496 kJ/mol to over 4,500, because the second electron has to come out of a closed shell. No lattice or hydration energy anywhere in chemistry can repay a cost on that scale, so +1 is not merely the usual state here but the only one there has ever been anything to say about. There is no chemistry of Na(II) to argue about.

The one electron, and how far away it sits

Descending the column adds a complete shell of electrons between the nucleus and that outermost s electron. Those inner electrons screen the nuclear charge, so the effective pull felt by the valence electron barely rises even though the nuclear charge climbs steeply. What does change is distance, and distance wins.

First ionisation energies fall from 520 kJ/mol for lithium to 496 for sodium, 419 for potassium, 403 for rubidium and 376 for cesium. Each step makes the electron easier to remove, and since removing it is what group 1 chemistry consists of, the metals get more eager as they get heavier. This runs opposite to group 17, where the same shielding argument produces a falling reactivity — because gaining an electron and losing one respond to shielding in opposite directions.

The same weakening grip explains why these are such poor metals in the mechanical sense. A metallic lattice is held together by the pooled valence electrons, and one electron per atom spread over an increasingly large ion core is a feeble glue. Melting points fall the whole way down: cesium is liquid a few degrees above a warm room, and lithium, sodium and potassium are all less dense than water — lithium is the least dense solid element there is.

Relativity gets the last word on group 1

Extrapolate the column and you arrive at an obvious conclusion about francium, which turns out to be wrong. The reason is a preview of the relativistic effects that dominate the heaviest elements.

Relativistic corrections scale steeply with nuclear charge — negligible at lithium, and by element 87 large enough to bend a trend. An electron passing close to eighty-seven protons carries a mass appreciably above its rest mass, and that extra mass draws in every s orbital in the atom, the valence 7s among them. Francium's outer electron is therefore gripped harder than the slope of the column predicts, and the descent that has run cleanly for five elements stops at element 87 rather than bottoming out there. The measured figures, and the reason the break lands on the very last member, are set out under ionisation energy.

The consequence for the family is the single thing most references about group 1 get wrong: the column's defining property does not reach its extreme at the bottom of the column. Element 55 is where the trend ends, and element 87 is a footnote to it.

Lithium wins a contest it should lose

In aqueous solution the ranking inverts, and this is where group 1 catches out more students than any other part of the table. Lithium has the highest ionisation energy in the group but the most negative standard electrode potential, at −3.04 V, edging out cesium's −3.03 and beating sodium's −2.71 comfortably.

Electrode potential is not a measure of how readily a lone atom gives up an electron. It measures a whole process: breaking the metal lattice, ionising the atom, and then hydrating the ion. Li⁺ is tiny, so its charge is concentrated, so water molecules bind to it hard — its hydration enthalpy is roughly twice that of Cs⁺. That third term more than repays the extra cost of ionising it.

The practical consequence is that lithium is thermodynamically the strongest reducing agent in water while being visibly the least vigorous of the group when it meets water at all. Reactivity as observed is a rate; electrode potential is a position. They are allowed to disagree, and here they do.

The top of the column keeps its own rules

Every family has a member that breaks the pattern, and in group 1 it is lithium, for the same reason it is beryllium in group 2 and fluorine in group 17: the first element of a column is so much smaller than the second that its compounds behave differently in kind.

• Nitrogen is inert towards every member of this column except the first, which forms the nitride Li₃N without persuasion.
• Heat lithium's carbonate, nitrate or hydroxide and each breaks down to the oxide, where the corresponding sodium, potassium, rubidium and cesium salts merely melt.
• Lithium fluoride and lithium carbonate are sparingly soluble, while the heavier analogues dissolve freely — a reversal driven by lattice energy in a salt built from two small ions.
• In air, lithium gives the normal oxide Li₂O, sodium gives the peroxide Na₂O₂, and potassium, rubidium and cesium give superoxides of the form MO₂. Big, diffuse anions need big, diffuse cations to make a stable lattice, so the oxide the metal forms is a size argument, not a reactivity one.

Several of those quirks are shared with magnesium rather than with sodium — the diagonal relationship, which arises because moving one place right and one place down leaves charge density roughly unchanged.

Ash, electricity, and two elements found in spa water

The family name comes from the Arabic al-qalī, the ash of saltwort plants, which is where soda and potash were obtained long before anyone knew they were compounds of metals. The metals stayed hidden because no chemical reducing agent available in 1800 was strong enough to prise the electron back.

Humphry Davy broke that deadlock in 1807 with electricity rather than chemistry, decomposing molten potash to get potassium and, days later, molten soda to get sodium. He extended the method through 1808 to the group 2 earths. Lithium arrived by a different route: Johan August Arfwedson identified it in the mineral petalite in 1817 while working in Berzelius's laboratory, and the metal itself was not isolated until a few years later.

Cesium and rubidium have the most distinctive origin story in the table. In 1860 Robert Bunsen and Gustav Kirchhoff were testing their new spectroscope on the mineral water of Dürkheim and found two blue lines belonging to nothing known. The following year lepidolite gave them a pair of deep red lines. Both elements were named for the colour of those lines — caesius, sky blue, and rubidus, darkest red — and they were the first elements ever discovered by spectroscopy rather than by chemical separation. The technique that found them went on to find helium in the Sun.

Francium closed the column in 1939, when Marguerite Perey at the Curie Institute identified it in a minor branch of actinium-227's decay. It was the last element to be discovered in nature rather than made, and there is thought to be less than about thirty grams of it in the entire crust at any instant. Its longest-lived isotope lasts twenty-two minutes.

Almost nothing here is used as a metal

The industrial story of group 1 is mostly a story about ions. Potassium is mined as potash and spread on fields at tens of millions of tonnes a year as chloride or sulfate; virtually none of it is ever reduced to the element. Sodium reaches the world as chloride, carbonate and hydroxide, in quantities that make it one of the largest tonnage commodities in chemistry, and only a small fraction is made into metal.

Where the metal is wanted, electrolysis is still the only route, because the ion sits at the bottom of a very deep energy well. The Downs cell electrolyses molten sodium chloride with calcium chloride mixed in to bring the melting point down by roughly two hundred kelvin, taking sodium off the cathode and chlorine off the anode. Demand for sodium metal fell sharply once leaded petrol was phased out, since making tetraethyl lead had been its largest single outlet.

Lithium is the exception and the reason this column is now a strategic commodity. It is won either from hard-rock spodumene, mostly Australian, or by pumping brine from salt flats in Chile and Argentina into evaporation ponds and waiting months for the sun to concentrate it. Both routes end in a carbonate or hydroxide destined for cathode material.

Cesium's contribution is stranger and more fundamental. Since 1967 the SI second has been fixed by counting a stated number of cycles of the microwave radiation emitted by a particular hyperfine transition in cesium-133 — a definition that replaced one based on the Earth's motion, and that made a property of the loosest electron shell in the periodic table the reference every clock in the world is ultimately calibrated against.

Biologically, the group runs the electrical system of every animal. The sodium-potassium pump maintains opposite gradients of the two ions across cell membranes and consumes a substantial share of the body's resting energy budget — in neurons, estimates run to more than half. Potassium is the dominant intracellular cation and sodium the dominant extracellular one, which is a distinction worth noticing given how similar their chemistry looks on paper.

Four things this column is regularly marked wrong for

1. Naming francium as the most reactive metal. The relativistic objection is above, but there is a second and more basic problem with the claim: nobody has ever had enough francium in one place to watch it do anything. Experiments work with laser-trapped clouds numbering a few hundred thousand atoms. Bulk reactivity here is not merely mispredicted — it is unmeasured.
2. Treating electrode potential and observed reactivity as the same quantity. Lithium tops one table and comes last on the other, and both results are correct.
3. Putting hydrogen in this family. It shares the ns¹ configuration and nothing else. It is a nonmetal, it forms a diatomic gas, and its ionisation energy is nearly three times cesium's.
4. Assuming every group 1 oxide is M₂O. Only lithium's is. The heavier members give peroxides and superoxides, and being asked to write the formula for the product of cesium in air is a standard way of finding out who has memorised rather than understood.