New results from the DANSS experiment

TL;DR

The DANSS experiment collected extensive data over six years to search for sterile neutrinos, setting new exclusion limits and observing reactor anti-neutrino rates consistent with theoretical models, but without definitive evidence for sterile neutrinos.

Contribution

This work provides new exclusion regions for sterile neutrino parameters and improves understanding of reactor anti-neutrino flux over a long-term data collection.

Findings

Increased significance of sterile neutrino signal from 1.3σ to 2.35σ.

Excluded a large parameter space including the BEST best-fit point.

Measured reactor anti-neutrino rate dependence on fuel composition, consistent with models.

Abstract

There are several experimental indications of sterile neutrinos with a mass in the 1 eV ballpark and many experiments are trying to clarify the situation. During 6 years the DANSS experiment collected more than 6 million Inverse Beta Decay (IBD) events and measured the background level during 4 reactor-off periods. Data were collected at 3 distances (10.9 m, 11.9 m, and 12.9 m) from the center of the core of a 3.1 GW$_{th}$ reactor with event rate up to 5 thousand per day. The detector position was changed frequently usually 2-3 times a week. Therefore many systematic uncertainties were canceled out in the analysis. After collection of additional 0.7 million IBD events the significance of the best-fit point in the 4$\nu$ case increased from 1.3$\sigma$ to 2.35$\sigma$. This is still not statistically significant and we present the exclusion area that covers a very interesting range of the sterile neutrino parameters up to $\sin^22\theta_{ee} < 0.004$ in the most sensitive point. In particular DANSS excludes a large fraction of sterile neutrino parameters preferred by the recent BEST results including the BEST best-fit point. The IBD rate dependence on the fuel composition was measured. It agrees with predictions of the Huber-Mueller model. During almost 6 years the reactor power was measured with $\approx1.8\%$ accuracy in 2 days using the anti-neutrino event rate normalized to the reactor power at the initial period.