Homestake Chlorine Experiment

Objective

Measure the flux of solar electron neutrinos above 0.814 MeV through radiochemical detection of the reaction νe + ³⁷Cl → ³⁷Ar + e⁻.

Method

615 tons of tetrachloroethylene (C₂Cl₄) held in a tank 1,478 m underground. Every 60–100 days the tank was flushed with helium to extract produced ³⁷Ar atoms, which were cryogenically trapped, loaded into a miniature proportional counter, and counted over subsequent months for the characteristic 2.82 keV Auger-electron decays of ³⁷Ar.

Key results

• 1968: first report of ~2.5 solar-neutrino units, about one-third of the Standard Solar Model prediction.
• Consistent deficit maintained through thirty years of operation across numerous runs and systematic checks.
• Established the solar neutrino problem, which drove neutrino-oscillation physics for three decades.
• Confirmed by subsequent gallium experiments (SAGE, GALLEX) at lower energies and water-Cherenkov experiments (Kamiokande, Super-K) at higher energies.

Significance

The first experiment to detect neutrinos from an astrophysical source. Homestake's consistent factor-of-three deficit was the prototype experimental anomaly that ultimately forced the discovery that neutrinos oscillate and have mass. Ray Davis received the 2002 Nobel Prize in Physics for this work.

A chemistry experiment in a gold mine

Ray Davis began the Homestake experiment in 1965, building on a 1946 proposal by Bruno Pontecorvo that ${}^{37}$Cl could serve as a neutrino target. Tetrachloroethylene — used commercially as a dry-cleaning fluid — was selected for its high chlorine content, low cost, and radiological cleanliness. 390,000 liters (615 tons) were loaded into a tank in the 4,850-foot level of the Homestake Gold Mine in the Black Hills of South Dakota.

The 1,478 m rock overburden suppressed cosmic-ray muon flux to the point where cosmogenic ${}^{37}$Ar production in the tank was below the solar signal.

The measurement cycle

Each measurement consisted of:

1. Exposing the tank for ~60–100 days to accumulate ${}^{37}$Ar
2. Flushing with helium gas, which carries dissolved argon out of the tank
3. Cryogenically trapping the argon on activated charcoal at liquid-nitrogen temperatures
4. Loading the extracted argon (a few tens of cm³ at STP) into a 1-cm³ proportional counter
5. Counting the 2.82 keV Auger-electron decays of ${}^{37}$Ar over 200–300 days (half-life 35 days)

The expected signal was a few atoms of ${}^{37}$Ar per run. Backgrounds — chiefly cosmogenic production during surface handling — had to be controlled at the single-atom level.

The deficit

Homestake reported an average of about 2.5 solar-neutrino units (SNU) — where 1 SNU corresponds to $10^{-36}$ captures per target atom per second — compared to John Bahcall’s Standard Solar Model prediction of roughly 8 SNU. The factor-of-three deficit persisted through three decades of measurement refinement.

The chlorine experiment was primarily sensitive to ${}^8$B solar neutrinos (via the reaction threshold of 0.814 MeV) with a small contribution from ${}^7$Be. It could not resolve spectrum shape or timing.

The thirty-year wait

The persistent deficit sustained several competing hypotheses:

1. The Standard Solar Model was wrong — perhaps the solar core was cooler than believed
2. The Davis experiment had a subtle systematic error
3. Solar ${\nu }_e$ transformed into something the detector could not see

Davis treated hypothesis 2 with decades of care: extensive cross-checks, independent calibrations, proof-of-principle with argon-recovery-efficiency tests using ${}^{36}$Ar and ${}^{37}$Ar spikes. The systematic bound was tightened to the few-percent level.

Hypothesis 1 was disfavored as helioseismology and neutrino spectroscopy progressively constrained the solar core conditions.

Hypothesis 3 — neutrino oscillation — was vindicated by SNO in 2001–2002 and by KamLAND in 2003. The Homestake deficit is now understood as the ${\nu }_e$-survival probability of about 0.35 in the MSW-LMA oscillation regime, applied to the ${}^8$B-dominated energy range of the chlorine target.

Legacy and shutdown

The experiment was decommissioned in 1994 when the Homestake mine’s ongoing operations became uneconomical. The tank was emptied and the equipment removed. The mine site has since been reborn as the Sanford Underground Research Facility, which now hosts the LZ dark-matter experiment and future neutrino projects.

Davis received the 2002 Nobel Prize in Physics at age 88, sharing it with Masatoshi Koshiba of Kamiokande and Riccardo Giacconi (X-ray astronomy). His thirty-year vigil at Homestake is a canonical example of how patient experimental work can set the stage for later fundamental discovery.