Kamiokande

Objective

Search for proton decay in 3 kt of water with a Cherenkov imaging array. Subsequently extended to real-time solar-neutrino detection and supernova monitoring.

Method

3 kilotons of purified water in a cylindrical tank 1,000 m underground, instrumented with 948 20-inch photomultiplier tubes. Charged particles above Cherenkov threshold produce light cones whose pattern reconstructs direction, energy, and particle type.

Key results

• No proton decay observed, setting limits below the original SU(5) predictions.
• 1987: detection of 11 neutrinos from Supernova 1987A, confirming the core-collapse neutrino mechanism.
• Real-time observation of solar ⁸B neutrinos through elastic scattering on electrons — confirming the Homestake deficit.
• Angular and energy measurements of atmospheric neutrinos, establishing the atmospheric anomaly that Super-Kamiokande would resolve.

Significance

Pioneered the large-volume water Cherenkov technique that Super-Kamiokande and Hyper-Kamiokande inherit. First real-time solar-neutrino detector; first detection of astrophysical neutrinos outside our solar system. Masatoshi Koshiba shared the 2002 Nobel Prize for this work.

From proton-decay to neutrino observatory

Kamiokande (Kamioka Nucleon Decay Experiment) was built in the early 1980s to search for proton decay, then a key prediction of grand unified theories. The experiment began operation in 1983 with 3 kilotons of water viewed by 948 photomultipliers, giving 20% photo-coverage.

The search for proton decay came up empty, progressively excluding the simplest SU(5) predictions and driving the field toward larger detectors (Super-Kamiokande, and eventually Hyper-Kamiokande). But the detector was well-suited to a new science target — observing MeV-range astrophysical neutrinos in real time through elastic scattering on atomic electrons: ${\nu }_x+e^{-}\to {\nu }_x+e^{-}$

Supernova 1987A

On 23 February 1987, a supernova exploded in the Large Magellanic Cloud 51 kpc away. Kamiokande-II (an upgraded configuration with lower threshold) detected 11 electron-recoil events in a 13-second burst coincident with the optical discovery. The IMB detector in Ohio saw 8 events in the same time window, and the Baksan detector in the USSR saw 5.

The 24 total supernova neutrino events were the first detection of neutrinos from outside our solar system. Their energy distribution and timing validated the thermal core-collapse neutrino-emission mechanism, and the ~4-hour separation from the optical peak bounded deviations from luminal propagation and from mass-induced dispersion.

Solar neutrinos

Beginning in 1987, Kamiokande measured solar ${}^8$B neutrinos in real time, confirming the Homestake deficit: the measured event rate was about half the Standard Solar Model prediction. Directionality — the pointing back to the Sun, visible in the angular distribution of elastic-scattering recoils — provided the first real-time identification of solar neutrinos.

The atmospheric anomaly

Kamiokande also measured atmospheric neutrinos produced by cosmic-ray air showers. The ratio of muon-like to electron-like events was significantly below the expected $\sim$2:1, an anomaly reported in the late 1980s. At the time, systematic uncertainties prevented a conclusive interpretation; Super-Kamiokande’s much larger statistics resolved the anomaly as oscillation a decade later.

Decommissioning

Kamiokande was shut down in 1996 and replaced by Super-Kamiokande in the same mine. The experience with the smaller detector — photocathode coverage, water purification, PMT calibration, reconstruction algorithms — directly informed the Super-Kamiokande design.

Masatoshi Koshiba received the 2002 Nobel Prize in Physics for the Kamiokande work, sharing it with Ray Davis (Homestake) and Riccardo Giacconi (X-ray astronomy). Takaaki Kajita, Koshiba’s student and Super-Kamiokande’s later atmospheric-oscillation lead, would share the 2015 Nobel Prize.