Thomson's atomic model

The problem Thomson was solving. Before 1897, the atom was assumed to be a featureless, indivisible sphere — the word "atom" literally means "uncuttable." Thomson's cathode-ray experiments showed that atoms contain much lighter, negatively charged particles (electrons) that can be pulled out of them. Any workable atomic model now had to answer two questions at once: where do the electrons sit, and what carries the positive charge that keeps the atom neutral overall?

The postulates (1904). Thomson's answer had four parts.

• The atom is a sphere of positive charge spread uniformly throughout its volume, with a radius of about $10^{-10}\ \text{m}$.
• Negatively charged electrons are embedded inside this sphere, distributed through it the way plums sit in a pudding or seeds sit in a watermelon — giving the model its nicknames, the "plum-pudding model" and the "watermelon model."
• The total positive charge exactly equals the total negative charge of all the embedded electrons, so the atom carries no net charge.
• The mass of the atom is assumed to be spread uniformly over the whole sphere, along with the positive charge.

What it got right. The model was not a wild guess — it explained the one fact every atomic model had to explain: atoms are electrically neutral even though they contain charged electrons. It also gave atoms a definite size and treated the electron as a genuine constituent of ordinary matter, not a laboratory curiosity of the discharge tube.

What it left open. The postulates say nothing about where inside the sphere the electrons sit, whether they move, or why the positive charge would stay smoothly spread rather than clumping. That vagueness about the positive charge's distribution is exactly the feature that scattering experiments would go on to test — and fail.

The goal is to see, with nothing more than the idea of spread-out charge, why Thomson's model makes a definite and testable prediction about what happens when fast, heavy, positively charged particles are fired at an atom — and why that prediction is the reason the model was eventually abandoned.

Step 1. Start from the postulate that matters most for this question: the positive charge of an atom, roughly $10^{-10}\ \text{m}$ across, is spread evenly through the entire volume of the sphere, not concentrated anywhere. An approaching charged particle therefore never meets a small, dense region of charge — at every point along its path, the repulsive push it feels is weak, because only a thin, diffuse "cloud" of charge is nearby at any instant.

Step 2. Bring in a second postulate: the atom's mass, too, is spread uniformly over the same volume. A fast, heavy alpha particle is thousands of times more massive than an electron, so the light embedded electrons cannot deflect it noticeably — and because the positive charge (which could in principle push back) is spread just as thinly as the mass, there is no dense "wall" anywhere inside the atom for the particle to collide with or bounce off.

Step 3. Combine the two. Since the repulsive electric field is weak and gradually varying everywhere inside the sphere, and there is no concentrated mass or charge to act like an obstacle, the model predicts that an incoming alpha particle should sail through the atom suffering only a small, gentle deflection — and every particle should be deflected by roughly the same small amount, since it experiences essentially the same diffuse charge distribution wherever it enters.

Step 4. This is a sharp, falsifiable prediction: no atom modelled this way can turn a fast alpha particle back through a large angle, let alone straight back the way it came, because doing so needs a strong, sudden force from a small, dense, charged region — precisely what Thomson's uniform sphere does not have. When Rutherford's gold-foil experiment found that a small fraction of alpha particles were scattered through large angles, and a few bounced almost straight back, this prediction failed outright, and the plum-pudding model had to be replaced by a model with a small, dense, charged nucleus.

Question. A beam of fast alpha particles (positively charged, several thousand times heavier than an electron) is fired at a thin gold foil. Using Thomson's model, in which each gold atom's positive charge and mass are spread uniformly through a sphere of radius about $10^{-10}\ \text{m}$, predict what the beam should do after passing through the foil, and state whether this prediction matches what Rutherford actually observed.

Step 1. Identify what each alpha particle "feels" travelling through a Thomson-model atom. Because the positive charge is spread over the whole $10^{-10}\ \text{m}$ sphere rather than concentrated at a point, the electric field at any position inside the atom is comparatively weak, and it changes smoothly as the particle moves through — there is no location where the field spikes.

Step 2. Apply the same reasoning to mass. The atom's mass is likewise spread through the full volume, so there is no small, hard region able to act like a wall and reverse the particle's momentum; a fast alpha particle simply pushes through the diffuse cloud.

Step 3. Predict the outcome for the whole beam. Nearly every alpha particle should pass straight through the foil with, at most, a small deflection, and the deflections should be small and roughly similar from particle to particle — no particle should experience a violent, large-angle deflection, because the model has no mechanism to produce one.

Step 4. Compare with Rutherford's actual result. The gold-foil experiment (Rutherford, Geiger and Marsden) found that most alpha particles did pass through with little deflection, matching the prediction — but a small fraction, roughly $1$ in $8000$, were scattered through large angles, and a handful bounced back almost the way they came. This large-angle scattering directly contradicts Thomson's prediction, and it is why the plum-pudding model was rejected in favour of Rutherford's nuclear model, which places the positive charge and nearly all the mass in a tiny, dense nucleus.

Competitive stretch

If asked to justify why the mismatch is so decisive rather than just noting it, the reasoning is: a large deflection requires a large force acting over a short distance, and Coulomb repulsion from a diffuse charge cloud can never supply that — only a concentrated charge, packed into a volume far smaller than $10^{-10}\ \text{m}$, can produce a field strong enough close-up to turn a fast alpha particle back. The rare back-scattered particles are therefore direct evidence for a small, dense, positively charged core inside the atom, which is exactly the feature Thomson's model lacked.

• Calling Thomson's model the "nuclear model" — it has no nucleus; the nucleus is Rutherford's contribution, introduced specifically to replace this model.
• Thinking Thomson's own experiments disproved his model — it was Rutherford's gold-foil scattering (with Geiger and Marsden) that disproved it, years after Thomson proposed it.
• Reading "plum-pudding" as a claim about electron motion — the analogy describes the positive charge distribution, not whether electrons are fixed or moving; the testable, and ultimately wrong, feature is the smeared-out positive charge.
• Forgetting the model's genuine success — it correctly explained why atoms are electrically neutral, by balancing total positive and total negative charge.

Ask the tutor: want a side-by-side comparison of Thomson's postulates against Rutherford's nuclear model and its scattering results?