Skip to main content
Sandbox Physics
2026 NOBEL PRIZE IN PHYSICS

SNO: how heavy water distinguished disappearance from a change of flavor

Three reaction channels made it possible to ask whether the Sun’s neutrinos were missing or had arrived in a different detectable form.

Sandbox Physics · · Evidence & history

  1. 01Charged current

    Electron-flavor response

  2. 02Neutral current

    All-active-flavor response

  3. 03Elastic scattering

    A different flavor weighting

Reading guide · a conceptual comparison, not a plot of experimental records.

THE QUESTION

If an instrument sees too few neutrinos, is the source wrong—or the detector only sensitive to part of the flux?

A count depends on what the detector can see

The solar-neutrino problem grew from a persistent gap between predictions and measurements. Solar models, developed through work including John Bahcall’s, connected nuclear reactions to an expected neutrino flux. Raymond Davis Jr.’s radiochemical experiment measured a response to part of that flux. A low count did not immediately identify which part of the chain was wrong.

Every detector weights an incident spectrum through its cross sections, threshold, efficiency and target composition. Comparing raw counts from two instruments is therefore not enough. The counts must be translated through an explicit response model before they can be interpreted as fluxes. This is the key to understanding why another kind of target was so valuable.

Sources: [1] · [2]

One target, different questions

At the Sudbury Neutrino Observatory in Canada, heavy water provided deuterons: nuclei containing a proton and a neutron. A charged-current breakup channel responded to electron neutrinos in the solar energy range. A neutral-current breakup channel responded to all three active flavors. Elastic scattering provided another measurement with a different flavor weighting.

The comparison is not between a detector that works and one that fails. It is between deliberately different response functions. The charged-current result asks how much electron flavor arrives; the neutral-current result asks how much active flux arrives altogether. The elastic-scattering channel is sensitive to all active flavors but more strongly to electron neutrinos.

νe+d→p+p+e−νx+d→p+n+νx\nu_e+d\rightarrow p+p+e^- \qquad \nu_x+d\rightarrow p+n+\nu_x
The left reaction is charged current; the right is neutral current. Here x denotes any active neutrino flavor. The expressions describe reactions, not equally efficient event counters.

Sources: [3]

Read the fluxes with their assumptions

In its 2002 neutral-current paper, SNO reported an electron-flavor component around 1.76 and a total active flux around 5.09, in units of a million neutrinos per square centimetre per second. The difference required a non-electron component. The paper reported that component as 5.3 standard deviations above zero under its stated analysis assumptions.

These values are inferred fluxes, not three integer event counts. The analysis uses reaction cross sections, detector response and a specified boron-8 spectral shape; the reported uncertainties distinguish statistical and systematic contributions. Dividing the central values gives roughly one third, but that quotient is not a universal survival probability for every energy produced by the Sun.

Φe=1.76,Φactive=5.09[106 cm−2s−1]\Phi_{e}=1.76,\quad\Phi_{\mathrm{active}}=5.09\quad[10^6\,\mathrm{cm}^{-2}\mathrm{s}^{-1}]
Rounded central flux values from the 2002 analysis. Use the original paper for uncertainties, correlations, thresholds and spectral assumptions.

Sources: [3]

Why this changed the argument

If the Sun simply emitted fewer neutrinos of every flavor, both inferred components would decrease together. Instead, the total active flux was compatible with solar expectations while the electron component was smaller. The combined measurement showed that counting only electron neutrinos misses part of what arrives.

That is a stronger statement than “one experiment saw a deficit.” It tests a way for the flux to remain present while changing its interaction signature. Interpreting the transformation through oscillations connects to propagation and matter effects. It still does not mean that an individual neutrino was watched continuously changing flavor on its way from the Sun.

Sources: [3] · [2]

What to take into the next experiment

The Cherenkov Lab explains how a charged particle creates the optical pattern a water detector records. It does not simulate SNO’s full channel separation. In particular, neutrons, electrons, thresholds and background classes require different response models. A geometric ring demonstration cannot stand in for a neutrino-flux measurement.

Read this history beside Super-Kamiokande’s atmospheric result. SNO changed the response to flavor while studying a common source. Super-K compared propagation distances using different arrival directions. Their complementarity is scientific: two distinct ways of testing a transformation, rather than two pictures of the same measurement.

Sources: [4] · [5] · [2]

Try the measurement

Cherenkov rings

Primary sources & revision

  1. John Bahcall · Solving the mystery of the missing neutrinos
  2. Nobel Committee · Neutrino oscillations, scientific background (2015)
  3. SNO Collaboration · Direct Evidence for Neutrino Flavor Transformation (2002)
  4. Kamioka Observatory · Super-Kamiokande detector
  5. Super-Kamiokande Collaboration · Evidence for Oscillation of Atmospheric Neutrinos (1998)

First published and source-checked on 9 October 2026. Original explanatory text and diagrams by Sandbox Physics. Illustrations are schematic; no experimental event records are reproduced here. This is an independent educational publication, not an official Nobel or experiment collaboration publication.

Continue the story

Super-Kamiokande: using the Earth as a baseline Five Nobel chapters, five different neutrino questions Cowan and Reines: why two flashes were more convincing than one