UC Berkeley

High-energy neutrons produce secondary particles through spallation, which create a ubiquitous and prominent background in a wide range of rare-event particle and nuclear physics experiments. Above ground, the high-energy neutron energy-dependent flux has been measured, but with significantly varying results. Below ground, only two previous measurements have succeeded in observing these neutrons, and communicated their results in a fashion useful to others. In a separate effort, a model of the neutron energy-dependent flux was previously developed for measurements below 1000 m.w.e. No comparisons to this model and the measured neutron flux have been performed.

In an effort to provide new and independent measurements above and below ground, the Multiplicity and Recoil Spectrometer (MARS) was designed, constructed, and deployed to the Kimballton Underground Research Facility (KURF). MARS is a transportable 1 m^3 detector composed of plastic scintillator Gd based neutron detectors, and a lead spallation target. MARS uses neutron spallation in the lead to transform an incident high-energy neutron into many lower energy secondary neutrons. By recording the secondary neutron multiplicity over many incident neutron events, the incident neutron energy spectrum can be inferred. This multiplicity method employed by MARS represents a new approach in high-energy neutron spectroscopy, which requires a new algorithm to correct the observed signal into a neutron spectrum. A recently developed Markov Chain Monte Carlo (MCMC) inversion algorithm, with a calibrated Monte Carlo model of MARS, is used to perform this inversion.

Using this new multiplicity method, MARS performed measurements at the surface of Earth and at depths of 377 meters water equivalent (m.w.e.), 540 m.w.e., and 1450 m.w.e. Due to the transportable nature of MARS, minimal detector related systematic bias exists between these measurements. The minimal bias between these measurements at multiple depths will allow for the creation of a depth-dependent predictive model of the high-energy neutron energy-dependent flux.

This dissertation introduces the new multiplicity measurement approach, the MCMC inversion algorithm, the Monte Carlo model and associated calibrations, and presents results from the four measurements. Above ground, the MARS measurement results agree with most of the previous measurements in the energy range between 90 MeV and 250 MeV. Above 250 MeV the MARS results report slightly lower flux than most of the previous measurements, but are still within the spread of all previous measurements. Below ground, no direct comparison can be made to the MARS results at 377 m.w.e. The MARS result

at 540 m.w.e. appears to be in rough agreement with one of the previous below ground measurements, at the one measured energy where the results overlap. At 1450 m.w.e., the MARS results shows reasonable agreement with previous simulation predictions.

The rough agreement of the MARS results, at all relevant locations, with previous measurements and existing simulation where applicable, provide confidence that all MARS measurements have produced the correct high-energy neutron energy-dependent flux. Above ground, the new independent results strengthen the results of previous measurements. Below ground, the three measurements provide consistent results with minimal detector related bias between measurements due to the transportable nature of MARS, that will be used to produce a depth-dependent model of the high-energy neutron energy-dependent flux. For the rare-event particle and nuclear physics experiments affected by high-energy neutron backgrounds, this model will allow for the prediction of the high-energy neutron background at different measurement locations, a more robust instrumental design, and the ability to

estimate the high-energy neutron background contribution in their final measured data; the confidence in the results of these experiments will be improved.