Atomic Structure

Ask anyone to draw an atom and you get a small ball with rings around it and dots on the rings. That picture was proposed in 1913, was known to be inadequate within about a decade, and is still what most people carry around. It is not useless — it gets the nucleus right, and it gets the idea of energy levels right — but almost everything it suggests about where electrons are is wrong, and the corrections are what the rest of chemistry is built on.

This page is about what an atom is actually like. How to count the particles in a specific atom is a separate skill, covered under protons, neutrons and electrons.

A marble in a stadium

Start with the emptiness, because nothing else about atomic structure makes sense until the scale is real to you.

A typical atom has a radius of about 10⁻¹⁰ m. The nucleus of an iron atom has a radius of about 4.6 × 10⁻¹⁵ m. The atom is therefore roughly twenty thousand times wider than its nucleus.

Scale the nucleus up to a marble one centimetre across and sit it on the centre spot of a stadium. The nearest electrons are then about two hundred metres away, out past the far stands. Between the marble and the stands there is nothing whatsoever — no air, no smaller particles, no substance of any kind.

Because volume goes as the cube of radius, the nucleus occupies something like one part in 10¹³ of the atom's volume, while holding more than 99.9% of its mass. Solid objects feel solid not because atoms are full, but because their electron clouds repel each other long before any nucleus gets near another.

Three particles, one of which barely weighs anything

Everything in the atom is built from three particles.

• The proton carries a charge of +1.602 × 10⁻¹⁹ C and sits in the nucleus. The number of protons is what makes an atom an atom of a particular element, and nothing else does.
• The neutron has almost exactly the proton's mass and no charge at all. It also sits in the nucleus, where it changes the mass and the stability of an atom without changing its chemistry.
• The electron carries the same size of charge as a proton but negative, and is about 1/1836 as heavy.

That mass ratio is the number to remember. It is why atomic mass is effectively a count of nuclear particles, why the electron's contribution can be neglected in a molar mass, and why electrons — being so light — behave in ways that are unmistakably quantum-mechanical while the nucleus can usually be treated as a stationary point.

Chemistry happens entirely in the outer part of the electron cloud. The nucleus is a spectator that supplies charge; change it and you have done nuclear physics, not chemistry.

Why Thomson's atom had to go

The modern story starts in 1897, when J. J. Thomson showed that cathode rays were streams of particles far lighter than any atom and identical whatever the source. Atoms had internal parts, which Dalton's indivisible spheres had ruled out.

Thomson's replacement, from 1904, distributed electrons through a diffuse sphere of positive charge — the "plum pudding". It was a reasonable model: it accounted for the electrons, it was electrically neutral, and it was stable.

It made one testable prediction. Anything fired at such an atom should pass through with, at most, a small deflection, because the positive charge is spread thinly everywhere. When Hans Geiger and Ernest Marsden measured the scattering of alpha particles by thin gold foil between 1909 and 1911, most did pass almost straight through — but roughly one in eight thousand came back at more than ninety degrees. Rutherford's remark about it, that it was as surprising as firing a fifteen-inch shell at tissue paper and having it bounce back, is quoted so often that it is easy to miss the argument underneath: a large deflection requires a concentrated charge, and a rare large deflection requires that concentration to be tiny.

Rutherford's atom should have collapsed in picoseconds

The nuclear atom that replaced the pudding had a catastrophic problem, and Rutherford knew it.

Classical electromagnetism says that an accelerating charge radiates energy. An electron in a circular orbit is accelerating continuously, so it must radiate, lose energy and spiral inward. Working the numbers through gives a lifetime of order ten picoseconds. Matter would not exist.

The model had a second failure that was equally damning and easier to observe. Heated hydrogen does not emit a continuous smear of colours; it emits a handful of sharp lines at fixed wavelengths — 656 nm, 486 nm, 434 nm and 410 nm in the visible range. Nothing about a classical orbit explains why only certain wavelengths appear.

What Bohr fixed, and what he could not

Niels Bohr's 1913 model asserted, without derivation, that electrons may occupy only certain orbits, that they do not radiate while in one, and that light is emitted or absorbed only when an electron jumps between two.

As a piece of prediction it was stunning. It reproduced hydrogen's line wavelengths to the accuracy of the measurements and gave the right ionisation energy, 13.6 eV, from first principles plus one quantisation rule.

As a piece of physics it was a patch, and its limits appeared quickly:

• it fails for helium — two electrons and it no longer works at all, let alone for anything larger;
• it predicts nothing about the relative brightness of the lines it locates;
• it cannot account for the fine structure, the splitting of single lines into close pairs;
• it offers no reason why the quantisation rule should hold.

The resolution, from Schrödinger and Heisenberg in 1925–26, abandoned the orbit entirely. Electrons are described by a wavefunction whose square gives the probability of finding the electron at each point in space, and the regions where that probability is high are called orbitals. There is no path, no orbital period and no definite position between measurements — which is what the orbitals and subshells guide unpacks.

James Chadwick completed the particle inventory in 1932 by identifying the neutron, which explained at last why atomic mass exceeds the mass of the protons and why an element can come in several isotopes.

Who actually found the nucleus, and who found the neutron

Both discoveries carry one famous name and involved several people, and it is worth being precise about it.

The gold foil measurements were made by Geiger and Marsden. Rutherford designed the enquiry and, crucially, did the mathematics that showed a point-like charge was the only thing that could produce the observed distribution of large-angle deflections — so the interpretation is genuinely his, while the observations are not. Marsden was an undergraduate at the time.

The neutron has a messier history. In 1930 Walther Bothe and Herbert Becker reported an unusually penetrating, uncharged radiation from beryllium exposed to alpha particles, and took it for gamma rays. In early 1932 Irène Joliot-Curie and Frédéric Joliot found that this radiation could eject protons from paraffin wax — an extraordinary result they also attributed to gamma rays. Chadwick saw immediately that no gamma ray could transfer that much momentum to a proton, and that a neutral particle of roughly the proton's mass would do it exactly. He published within weeks and took the 1935 Nobel Prize alone.

The Joliot-Curies had the observation and missed the interpretation. That pattern — the credited discoverer being the person who understood the data rather than the person who took it — recurs throughout this history, and it is worth knowing that "discovered by" in a textbook is usually compressing a more interesting story.