MathemagiciansLibraryExplorationsDeep Dives
← Physics · Upper Secondary
PhysicsUpper Secondary

The nuclear atom — Rutherford's experiment

PU6 — Modern Physics › PU6.1 — Atomic Structure and QuantumLecture 1 of 4~2370 words

How do you see inside something you cannot see?

An atom is about m across. No microscope of 1909 could show anything that small. Yet within two years, physicists in Manchester had found out what is inside an atom, and how big its centre is. They never saw it. They threw things at it and watched where the things went.

You may know the outline of the story already: alpha particles fired at gold foil, and a few came back. This lecture asks the physics questions behind the story. Why could a spread-out charge never turn an alpha particle round? How close does an alpha particle get to the centre of a gold atom? And what did the experiment really prove, and what did it not prove?

The model that was being tested

By 1909, physicists knew two facts about atoms. They contain electrons, which J. J. Thomson had found in 1897, and which are negative and very light. And whole atoms are neutral. So somewhere in each atom there must be positive charge that cancels the electrons.

Thomson's model, from 1904, put that positive charge everywhere. The atom was a ball of positive charge, spread evenly through the whole atom, with the electrons sitting inside it. It is usually called the plum pudding model: the electrons are the plums, and the positive charge is the pudding. (The name was given by others. It describes the picture well.)

The model makes a clear prediction. Remember that charge spread out over a large space makes only a weak force. So a fast, heavy, positive particle passing through such an atom should be pushed aside only a tiny amount. Every particle should come out of a thin foil going almost straight on.

The experiment

The probes were alpha particles. They are thrown out by some radioactive atoms at about m/s, roughly 7% of the speed of light. Each one has a charge of and a mass about 7300 times the mass of an electron. (A later lecture shows that an alpha particle is a helium nucleus.)

In 1909, Hans Geiger and Ernest Marsden, working in Ernest Rutherford's laboratory in Manchester, sent a narrow beam of alpha particles at thin metal foils. When an alpha particle hit a screen coated with zinc sulfide, it made a tiny flash of light. They counted the flashes by eye, through a microscope, in a dark room. The screen and microscope could be turned to any angle around the foil.

chamber with the air pumped outlead blockalpha beamgold foilscreen andmicroscopemostvery few
Seen from above. The screen and microscope turn around the foil, so the flashes can be counted at any angle, even behind the foil.

The results came in three parts.

Rutherford later said it was almost as surprising as firing a huge artillery shell at a piece of tissue paper and seeing it come back and hit you.

Stop and think

An atom contains electrons, and an alpha particle is charged. Could collisions with electrons have turned the alpha particles back?

Show the answer

No. An alpha particle has about 7300 times the mass of an electron. When a heavy object meets a very light one, the light one is pushed away and the heavy one hardly changes direction. Think of a moving car meeting a flying insect. The car does not turn round.

So the electrons could not be the cause. Whatever turned the alphas back had to be heavy, as well as charged. That is the first clue that the mass of the atom is in the same place as its positive charge.

Why a spread-out charge cannot turn an alpha back

This is the hard part of the lecture, and the most important. Spend time here.

The force between two charges is given by Coulomb's law, . The force grows as the distance shrinks. But in the plum pudding model, the alpha particle can never get close to a lot of charge at once. When it is outside the atom, all of the atom's positive charge pushes it, but from a distance of at least the atom's radius, about m for gold. Once it enters the atom, the positive charge is all around it, pushing from every side, and much of the push cancels. So the strongest push it can ever feel is at the edge of the atom.

Work out that strongest push for gold, with its 79 units of positive charge, all at the edge distance:

This force acts only while the alpha crosses the atom, which takes about s. Worked Example 2 shows that this can turn the alpha by about 0.01° at most. Passing through more than a thousand atoms in a foil, these tiny turns go in random directions and mostly cancel. Together they add up to about a degree. That matches what almost all the alpha particles did.

But it can never add up to 90°, let alone 180°. Turning a fast alpha particle round needs a push thousands of times bigger than any one plum pudding atom can give. The model has no way to give it.

Now suppose instead that all 79 units of positive charge are packed into a tiny region at the centre. An alpha particle heading straight at it can get very close, perhaps m away. At that distance the same charge pushes with a force more than 20 million times bigger, about 40 N on one tiny particle. That is enough to stop it and send it back.

That is it. That is the whole argument. A large force needs a small distance. A small distance from all the positive charge at once is only possible if the charge is concentrated.

Rutherford's nuclear model

The nuclear model of the atom (Rutherford, 1911)

Almost all the mass of an atom, and all its positive charge, is concentrated in a tiny nucleus at the centre.

The electrons are spread through the much larger space around the nucleus, so most of the atom's volume contains almost no mass.

A gold nucleus has a radius of about m. A gold atom has a radius of about m, about 20 000 times bigger.

Each part of the model explains one part of the result. Most alphas go straight through, because most of the atom contains nothing heavy enough to turn them. A few are turned through large angles, because they happen to pass close to a nucleus. Very few come back, because a nucleus is so small that a near head-on approach is rare.

See the paths for yourself in the simulation below. The forces in it follow Coulomb's law exactly.

Simulation — alpha particles passing a gold nucleus

Choose how far from the line of the nucleus the alpha particle starts (its aim), and its energy. Press Fire. Try aims of 0, 20 and 150 fm. Then press Show many to see a spread of aims at once.

Energy:

1 fm (femtometre) = 10⁻¹⁵ m. The nucleus is held still, which is close to true for gold, about 49 times heavier than an alpha particle. The electrons are left out.

The figure shows the paths of 7.7 MeV alpha particles aimed at different distances from the line of a gold nucleus. The table gives the angle each one is turned through.

alphas arrive from the left →100 fm
Paths of 7.7 MeV alpha particles past a gold nucleus (the dark dot, drawn to scale). The further from the line an alpha is aimed, the less it bends. No alpha gets inside the dashed circle, 29.5 fm from the centre.
Aim (fm)Turned through
5≈ 143°
20≈ 73°
40≈ 41°
80≈ 21°
160≈ 11°

Two things should stand out. First, the paths bend smoothly. The alpha particle never touches the nucleus; it is pushed away by the electric force long before it arrives. Second, the alpha has to be aimed very close to the line of the nucleus to be turned back. At 7.7 MeV it must pass within about 15 fm. That is why so few came back.

How close does an alpha get? An upper limit for the nucleus

The simulation used a number you can find for yourself: the distance of closest approach for a head-on alpha particle. The idea is energy. As the alpha moves towards the nucleus, the repulsion slows it, and its kinetic energy is turned into electric potential energy. At the closest point it stops for an instant. All its kinetic energy has become potential energy.

Step through — the distance of closest approach

Press Next to add one step. Read the reason before you go on.

Each step gives one line of the calculation and the reason for it. The alpha particle has 7.7 MeV of kinetic energy and heads straight at a gold nucleus.

  1. Far away: kinetic energy Ek = 7.7 MeV, potential energy 0. Because the electric potential energy of two charges is zero when they are very far apart.
  2. Convert: Ek = 7.7 × 10⁶ × 1.60 × 10⁻¹⁹ J = 1.23 × 10⁻¹² J. Because 1 eV = 1.60 × 10⁻¹⁹ J.
  3. At the closest point d, the alpha stops for an instant, so Ek = 0 and U = kq₁q₂/d. Because energy is conserved and the nucleus hardly moves.
  4. So kq₁q₂/d = 1.23 × 10⁻¹² J, and d = kq₁q₂ ÷ (1.23 × 10⁻¹² J). Because all the kinetic energy has become potential energy.
  5. kq₁q₂ = 8.99 × 10⁹ × (2 × 1.60 × 10⁻¹⁹) × (79 × 1.60 × 10⁻¹⁹) = 3.64 × 10⁻²⁶ J m. Because the alpha has charge +2e and the gold nucleus +79e.
  6. d = 3.64 × 10⁻²⁶ ÷ 1.23 × 10⁻¹² ≈ 2.95 × 10⁻¹⁴ m ≈ 30 fm. Because this is the division in step 4.

So the alpha gets no closer than about 3 × 10⁻¹⁴ m. All the nucleus's charge must be inside that distance, so the radius of the nucleus is less than about 3 × 10⁻¹⁴ m.

Notice what kind of answer this is. It is an upper limit. The alpha particle turned back at 30 fm, so the nucleus can be no bigger than that. But it could be much smaller, and the experiment cannot tell. Rutherford's own 1911 paper gave the same kind of limit, about m, from a slightly different calculation. Later experiments, with faster particles that get closer, showed that a gold nucleus has a radius of about m.

Worked examples

Worked Example 1 — How empty is an atom?

A gold nucleus has radius m. A gold atom has radius m. What fraction of the atom's volume does the nucleus fill? If the nucleus were 1 cm across, how wide would the atom be?

Volumes. Both are spheres, so each volume is . In the ratio, the cancels:

The nucleus fills about one ten-millionth of a millionth of the atom.

Scale model. The atom is times wider than the nucleus. So a 1 cm nucleus would sit in an atom about 21 000 cm, or about 200 m, across. That is two football fields, with a marble in the middle and a few specks of dust (the electrons) somewhere around it.

Check: cube the ratio of widths: , and . ✓

Worked Example 2 — The plum pudding's best effort

A 7.7 MeV alpha particle (mass kg) crosses a plum pudding gold atom of radius m. Use the largest possible force, N, to estimate the largest angle it could be turned through.

Speed. From : m/s.

Time inside the atom. It crosses a width of about m, so s.

Sideways momentum. Impulse = force × time. Even if the largest force pushed sideways the whole time, kg m/s.

Angle. The forward momentum is kg m/s. For a small turn, the angle in radians is about , which is about 0.01°.

This is a generous estimate, because the real force is smaller for most of the crossing. Even so, a single atom can turn the alpha by only about a hundredth of a degree. To turn it by 90°, the plum pudding would need a push thousands of times bigger. It cannot give it.

Worked Example 3 — How dense is a nucleus?

A gold nucleus has mass kg and radius m. Find its density. What would a teaspoon (5 cm³) of nuclear matter weigh?

A teaspoon is m³, so its mass would be about kg. That is about a thousand million tonnes.

Check: gold metal has a density of about kg/m³. The nucleus is about times denser, which matches Worked Example 1: the same mass is squeezed into a volume about times smaller. ✓

Trap 1 — "The alpha hits the nucleus and bounces off"

It never touches it. A 7.7 MeV alpha turns round about 30 fm from the centre of a gold nucleus, while the nucleus has a radius of about 7 fm. The bounce is caused by the electric repulsion, acting across empty space, like two magnets pushing apart.

Trap 2 — "The experiment measured the size of the nucleus"

It gave an upper limit only. The closest approach shows that the nucleus is no bigger than about m. It cannot show how much smaller it is. When a question asks what the experiment shows about size, say "less than".

Trap 3 — Saying more than the evidence shows

The experiment showed a tiny, massive, positive centre, and mostly empty space. It did not show that electrons travel in orbits like planets, and it said nothing about protons or neutrons. Rutherford's 1911 paper did not even use the word "nucleus". Keep each claim tied to the result that supports it.

Try it yourself

Work each one on paper first. Open a hint only when you are stuck. Use N m²/C² and C.

  1. Explain, using the three results of the experiment, why each part of Rutherford's model is needed: (a) mostly empty space, (b) a positive centre, (c) a centre that is tiny and heavy.

    Hint 1

    Match each result (most straight through, a few turned, very few back) with one part of the model.

    Hint 2

    For (c), think about what "very few" says about how hard the centre is to hit, and about the electrons' mass.

    Full answer

    (a) Almost all alphas pass almost straight through, so most of each atom has nothing that can turn them. (b) Alphas are positive and are pushed away, not pulled in, so the centre must be positive (repulsion sends them back towards the source). (c) Only very few come back, so the target that can turn them back is tiny. It must also be heavy, because light electrons could never turn a heavy alpha round. Check: each part of the model is linked to a different measured result, with none left over. ✓

  2. An alpha particle with 5.0 MeV of kinetic energy heads straight at a gold nucleus. Find its distance of closest approach. Compare it with the 7.7 MeV answer.

    Hint 1

    Set the kinetic energy equal to .

    Hint 2

    J m, as in the step-through. Convert 5.0 MeV to joules.

    Full answer

    J. Then m, about 45 fm. The slower alpha turns back further away. Check: is inversely proportional to , so fm. ✓

  3. In 1913, Geiger and Marsden tested Rutherford's model by changing one thing at a time. Predict what happens to the number of alphas turned through large angles in each case. (a) The alphas are made slower. (b) The foil is made of a metal with a smaller charge on its nucleus. (c) The foil is made twice as thick (but still thin).

    Hint 1

    Large turns need a large push. What makes the push stronger, or the alpha easier to turn?

    Hint 2

    For (c), think about how many nuclei an alpha passes near.

    Full answer

    (a) More large turns. A slower alpha has less kinetic energy, so the same repulsion turns it more, and its closest approach is further out (Question 2). (b) Fewer large turns. A smaller nuclear charge gives a smaller force at every distance. (c) About twice as many. Each large turn comes from one close pass by one nucleus, and a foil twice as thick has twice as many nuclei to pass. Geiger and Marsden found results of this kind, which is why the model was accepted. Check: in all three, the answer follows from and the chance of a close pass, with nothing else needed. ✓

  4. A big estimate. A gold foil is m thick and has atoms per m³. A 7.7 MeV alpha is turned through more than 90° only if it is aimed within about m of the line of a nucleus. Estimate the fraction of alphas turned through more than 90°.

    Hint 1

    Each nucleus offers a small target: a circle of radius m, facing the beam.

    Hint 2

    Find the number of nuclei behind each square metre of foil (atoms per m³ × thickness). Multiply by the area of one target circle.

    Full answer

    Nuclei behind each square metre: . Area of one target circle: m². Fraction of each square metre covered by targets: , about 1 in 60 000.

    This is the same kind of tiny number that was measured: Rutherford quoted about 1 in 20 000 for a gold foil of this thickness, turned through about 90° or more. For an estimate this simple, being within a factor of a few is a strong result. Check: the targets cover so little of the foil that they almost never hide behind each other, so adding their areas is fair. ✓

The bigger idea

Nobody in 1911 had seen a nucleus. Rutherford worked out its existence, its charge and a limit on its size from the paths of particles he could not see either, counted as flashes on a screen. This is still how physicists study things too small to see. Fire particles whose behaviour you understand at a target, measure where they go, and reason backwards to the structure that turned them.

The same method, with far faster particles, showed the size of the nucleus, then the protons and neutrons inside it, and later the quarks inside those. Today's particle colliders are, in this sense, Geiger and Marsden's flashing screen, made enormous.

Key concepts in this lesson