On this page
In 1919 Ernest Rutherford found that fast alpha particles could knock protons out of nitrogen. But his method, counting flashes on a screen, could not show what happened to the alpha particle. Did it bounce off, or was it swallowed? Answering that took a young ex-naval officer, a cloud chamber, and tens of thousands of photographs. This post tells how Patrick Blackett did it, checks the energy balance of the reaction, and draws out what his patience teaches.
Who was Patrick Blackett?
His Nobel facts page gives his dates as 18 November 1897 to 13 July 1974, both in London. He won the 1948 Nobel Prize in Physics "for his development of the Wilson cloud chamber method, and his discoveries therewith in the fields of nuclear physics and cosmic radiation."
His Nobel biography says he trained as a naval officer and fought at the battles of the Falkland Islands and Jutland in the First World War, then resigned to study physics under Rutherford at Cambridge. Later, with Giuseppe Occhialini, he built a cloud chamber triggered by Geiger counters, so that cosmic rays "take their own photographs", and confirmed the positron. The Royal Society's biographical memoir notes his "deep interest in Indian affairs", which it traces to family links with India.
Why use a cloud chamber?
A cloud chamber, invented by C. T. R. Wilson, fills with supersaturated vapour. A charged particle passing through leaves a line of ions, droplets form on them, and you can photograph the track. Our post on C. T. R. Wilson and the cloud chamber explains how it works.
In his 1925 paper, Blackett explains the choice: the scintillation method can follow only the proton, while the Wilson chamber "provides the obvious and perhaps the only certain way" of seeing the motion of the other particles. Using Rutherford and Chadwick's data, he estimated that about 20 protons are ejected per million fast alpha particles. Rare events need many pictures.
How many photographs did it take?
In his Nobel lecture, Blackett describes an automatic chamber only 6 cm across and 1 cm deep, taking a photograph every 15 seconds. Production "was started in earnest in 1924 and 23,000 photographs were taken within a few months. With an average of 18 tracks a photograph these gave over 400,000 tracks, each of which had to be scrutinized". On good days the apparatus took 1,200 photographs.
"Eight forked tracks were found" that looked different from ordinary collisions. Each showed only two tracks leaving the collision point: a long thin proton track and a short thick one. If the alpha particle had bounced off, there would have been three. So the alpha particle had been captured, and, by conservation of charge and mass, the new nucleus was a heavy isotope of oxygen:
¹⁴N + ⁴He → ¹⁷O + ¹H
Blackett notes that oxygen-17 was not yet known, and was found shortly afterwards in band spectra. What had been called "disintegration" of nitrogen was, in his words, "in reality an 'integration' process."
Worked example 1: how rare was each event?
Total tracks. 23,000 photographs × 18 tracks = 414,000 tracks.
Fraction that transmuted. 8 / 414,000 = 1.9 × 10⁻⁵, about one track in 52,000.
Time for 1,200 photographs. 1,200 × 15 s = 18,000 s = 5 hours of non-stop running.
Every track had to be inspected by eye. At this rate, finding the eight forks meant looking at around 50,000 ordinary tracks for each discovery.
Worked example 2: does the reaction release or absorb energy?
Use atomic masses from NIST for nitrogen, helium, oxygen and hydrogen.
Before. 14.003074 + 4.002603 = 18.005677 u.
After. 16.999132 + 1.007825 = 18.006957 u.
Mass change. 18.005677 − 18.006957 = −0.001280 u. The products are heavier.
Energy. The NCERT Class 12 Nuclei chapter gives 1 u = 931.5 MeV/c². Q = −0.001280 × 931.5 = −1.19 MeV.
A negative Q means the reaction absorbs energy. The alpha particle must bring in more than 1.19 MeV of kinetic energy, which the fast alpha particles from radioactive sources could supply. Electron masses cancel because both sides have 9 electrons.
Which mistakes does this help you avoid?
Mixing nuclear and atomic masses. Use atomic masses on both sides and the electrons cancel. Mixing the two gives errors of about 0.0005 u per electron.
Losing the sign of Q. Write "before minus after" every time.
Forgetting to balance. Check mass numbers (14 + 4 = 17 + 1) and charges (7 + 2 = 8 + 1).
The JEE (Main) 2026 syllabus lists atomic masses, mass defect and binding energy, and our nuclear binding energy problems give more practice. Check the syllabus for your exam year.
To check each step of a Q-value calculation against a full chain of reasoning, the Fermi Problem Library has worked problems with a step-by-step solver that explains the reasoning rather than printing an answer.
What can a student take from Blackett's way of working?
Rare events need volume. Eight results came from over 400,000 tracks. Some questions are only answered by patient, careful repetition.
Test the ordinary case first. Before hunting forks, Blackett checked that ordinary collisions conserved energy and momentum, so he knew what normal looked like.
Count the tracks. Two outgoing tracks instead of three settled the question.
Next step: balance and find Q for ⁹Be + ⁴He → ¹²C + ¹n using NIST masses, the reaction behind the neutron's discovery.
