Measurements of the electron neutrino-argon differential cross section without pions in the final state in MicroBooNE

Using the full MicroBooNE dataset, this paper presents a new measurement of the electron neutrino-argon differential cross section without pions in the final state, revealing good agreement with generator predictions for lepton kinematics but notable discrepancies in hadronic system modeling, particularly regarding proton angles.

The Gist

Imagine you are trying to understand how a specific type of bullet (an electron neutrino) behaves when it hits a very dense, frozen wall made of argon (a type of noble gas).

This paper is the report from the MicroBooNE experiment, a giant, high-tech camera buried underground at Fermilab in the US. Their goal was to take a "snapshot" of these collisions to see exactly what happens when the bullet hits the wall, specifically looking for cases where no pions (a type of particle often created in these crashes) are left behind.

Here is the story of what they found, explained simply:

1. The Setup: A Tiny Needle in a Giant Haystack

Neutrinos are ghostly particles that rarely interact with anything. When scientists create a beam of them, it's mostly made of muon neutrinos (a different "flavor"). The electron neutrinos they are interested in are like finding a single red needle in a haystack of a billion blue needles.

• The Challenge: Because electron neutrinos are so rare in the beam, MicroBooNE had to collect a massive amount of data over several years to get enough "hits" to study.
• The Detector: Think of the MicroBooNE detector as a giant, 85-ton tank of liquid argon. When a neutrino hits an argon atom, it creates a tiny spark of light and a trail of electrons. The detector is so sensitive it can reconstruct a 3D movie of the crash, showing exactly where the particles went.

2. The Two Types of Crashes

The researchers looked at two specific scenarios for these electron neutrino crashes:

1. The "Naked" Crash: The neutrino hits, and the only things flying out are an electron and nothing else (no protons visible, no pions). It's like a silent, invisible hit.
2. The "Messy" Crash: The neutrino hits, and an electron flies out plus a proton (a piece of the argon nucleus) that is moving fast enough to be seen. It's like a crash where a piece of the wall flies off.

They wanted to measure the cross-section. In simple terms, this is just a fancy way of asking: "How likely is this specific type of crash to happen?"

3. The Big Question: Do Our Maps Match the Territory?

Scientists have computer programs (called Generators) that try to predict how these crashes should look. These programs are like GPS maps for particle physics. They are built on theories about how the universe works.

• The Problem: Most of these "maps" have been tuned using data from muon neutrinos (the common blue needles). No one had really checked if the maps were accurate for electron neutrinos (the rare red needles).
• The Analogy: Imagine you have a map of a city that was drawn perfectly for driving a truck. Now, you want to drive a motorcycle through the same city. You assume the map works, but maybe the motorcycle handles corners differently, or the traffic rules are different. You need to check if the map is still accurate for the motorcycle.

4. The Results: The Maps Are Mostly Good, But Have Holes

After analyzing their data, the MicroBooNE team compared their real-world "photos" to the computer "maps."

• The Good News: When they looked at the electron (the bullet itself), the maps were excellent. The predictions for the electron's energy and direction matched the real data very well. The "GPS" for the bullet was accurate.
• The Bad News: When they looked at the proton (the piece of the wall that flew off), the maps started to get fuzzy.
  • Some computer models predicted that protons would fly out at certain angles more often than they actually did in the real data.
  • Specifically, the models seemed to overestimate how many protons would fly straight forward.
  • One specific model (GiBUU) was way off the mark until the scientists added a "correction factor" (simulating how particles interact inside the nucleus), which made the map much better.

5. Why Does This Matter?

This might sound like a small detail, but it's huge for the future of physics.

• The Future: A massive new experiment called DUNE is being built. It will use the same technology (liquid argon) to study neutrinos from deep underground. DUNE hopes to solve mysteries like: Why does the universe have more matter than antimatter?
• The Risk: If the computer maps (theories) used to interpret DUNE's data are slightly wrong about how protons behave, scientists might misinterpret their results. They might think they found a new law of physics when they actually just had a bad map.
• The Solution: This paper provides a "calibration check." It tells the scientists building DUNE: "Hey, your maps for electron neutrinos are great for the bullet, but you need to tweak the part about how the wall pieces fly off."

Summary

In short, MicroBooNE took a giant dataset of rare particle collisions, acted like a forensic investigator, and checked the computer simulations against reality. They found that while our understanding of the "bullet" (electron) is solid, our understanding of the "debris" (protons) needs a little bit of tuning. This ensures that the next generation of experiments will be able to read the universe's secrets without getting lost on a bad map.

Technical Summary

1. Problem and Motivation

• Scientific Context: Neutrino oscillation experiments, particularly the upcoming Deep Underground Neutrino Experiment (DUNE), rely on precise measurements of electron neutrino ($\nu_e$) appearance in a muon neutrino ($\nu_\mu$) beam. A critical challenge is the accurate modeling of $\nu_e$ interactions on argon nuclei, as current interaction generators are primarily tuned to muon neutrino data.
• The Gap: There is a scarcity of exclusive $\nu_e$ cross-section measurements on argon. Existing measurements often combine neutrinos and antineutrinos or use lighter nuclear targets (e.g., carbon, oxygen). Furthermore, theoretical models often fail to account for the mass difference between electrons and muons, which significantly affects kinematics, especially at low energies and in the forward region.
• Specific Goal: This paper presents a new measurement of the charged-current (CC) $\nu_e$ cross section on argon with no pions in the final state (CC0$\pi$). It utilizes the full MicroBooNE Booster Neutrino Beam (BNB) dataset to explore the relationship between the leptonic (electron) and hadronic (proton) systems, specifically distinguishing events with and without visible protons.

2. Methodology

A. Dataset and Detector

• Experiment: MicroBooNE, a Liquid Argon Time Projection Chamber (LArTPC) located at Fermilab.
• Beam: The Booster Neutrino Beam (BNB), an 8 GeV proton beam producing a neutrino beam peaked at 700 MeV. The beam is predominantly $\nu_\mu$ with a small intrinsic $\nu_e$ component (0.51%) and negligible $\bar{\nu}_e$ content.
• Statistics: The analysis uses the full operational dataset of 13.09 $\times$ 10$^{20}$ protons on target (POT), representing a 1.9$\times$ increase over previous MicroBooNE $\nu_e$ measurements.
• Detector: An 85-ton LArTPC with a 273 V/cm electric field. It reconstructs 3D particle tracks and showers. A Cosmic Ray Tagger (CRT) was used to veto cosmic backgrounds.

B. Event Selection and Signal Definition

The analysis targets two exclusive topologies:

1. 1eNp0$\pi$: Events with an electron and at least one visible proton (Kinetic Energy $KE_p > 50$ MeV).
2. 1e0p0$\pi$: Events with an electron and no visible protons ($KE_p < 50$ MeV).

Selection Criteria:

• Fiducial Volume: Defined within the active TPC volume.
• Kinematic Cuts:
  • Electron Energy ($E_e$) > 30 MeV (true) / > 0.5 GeV (selected for 1e0p0$\pi$).
  • Electron angle relative to beam ($\cos \theta_e$) > 0.6 (for 1e0p0$\pi$).
  • No neutral pions ($\pi^0$) allowed.
  • Proton visibility threshold set at 50 MeV (where reconstruction resolution is <9%).
• Background Rejection:
  • Cosmic Rays: Vetoed using the CRT and topological cuts (vertical tracks).
  • $\pi^0$ Background: Separated from electrons using shower topology (conversion distance) and $dE/dx$ at the shower start.
  • Muon Neutrinos: Rejected using Boosted Decision Trees (BDTs) trained on kinematic variables.
• Purity/Efficiency: The final sample achieves ~57% purity (14% efficiency) for 1e0p0$\pi$ and ~67% purity (19% efficiency) for 1eNp0$\pi$.

C. Cross Section Extraction

• Unfolding: The measured distributions are "unfolded" from reconstructed space to true space using the Wiener-SVD (WSVD) method to correct for detector resolution and migration effects.
• Formula: The differential cross section is calculated as:
  $$\langle \frac{d\sigma}{dx} \rangle_i = \frac{\sum_j R^{-1}_{ij} (N_j - B_j)}{N_{target} \times \phi \times \Delta x_i}$$
  Where $R$ is the response matrix, $N_j$ is observed events, $B_j$ is predicted background, $\phi$ is flux, and $N_{target}$ is the number of argon nuclei.
• Systematics: Uncertainties include flux (6%), detector response (5–12%), interaction model reweighting, final state interactions (FSI), and POT counting (2%). Total systematic uncertainties range from 9–15%.

3. Key Contributions

• Full Dataset Utilization: This is the first measurement using the complete MicroBooNE BNB dataset, significantly improving statistical power over previous partial analyses.
• Exclusive Topology: It provides the most detailed exclusive $\nu_e$ CC0$\pi$ cross-section measurements on argon to date, separating events based on proton visibility.
• Lepton-Hadron Correlation: It uniquely explores the correlation between the electron and proton kinematics (specifically the opening angle), providing constraints on models that describe the interplay between the leptonic and hadronic systems.
• Model Comparison: The results are compared against a wide array of modern neutrino interaction generators (GENIE v3.0.6 & v3.6.0, NuWro, NEUT, GiBUU) with various model configurations (FSI, MEC, QE).

4. Results

A. Data vs. Models

• Lepton Kinematics: The data shows good agreement with most models regarding electron energy ($E_e$) and electron angle ($\cos \theta_e$). P-values are generally high (>0.88) for GENIE v3.6.0 (AR23) and GENIE v3.0.6 without the MicroBooNE tune.
• Hadronic System (Proton Kinematics):
  • There are notable discrepancies in the modeling of the hadronic system.
  • GiBUU Model: The GiBUU model without in-medium nucleon-nucleon corrections significantly underpredicts the data in the 1e0p0$\pi$ bin (p-value = 0.008). Including in-medium corrections improves agreement (p-value = 0.215).
  • Proton Visibility: Models that agree well with visible protons (1eNp0$\pi$) do not necessarily agree well with invisible protons (1e0p0$\pi$), highlighting the difficulty in modeling low-energy proton emission.
• Angular Correlations:
  • Proton Angle ($\cos \theta_p$): All models tend to overpredict the cross-section at large cosines (forward direction) compared to data. The tuned GENIE model (G18T) has a low p-value (0.029) here.
  • Electron-Proton Opening Angle ($\cos \theta_{ep}$): Similar overprediction is observed in the forward region, though less severe than for proton angle alone.

B. Model Sensitivity

• FSI Models: Within GENIE, the hA18 (empirical) FSI model provides better agreement than the hN18 (cascade) or G4Bertini models.
• MEC/QE Models: Variations in Quasi-Elastic (QE) and Meson Exchange Current (MEC) models (e.g., SuSAv2 vs. Valencia) showed less impact on the overall fit than the FSI models, though SuSAv2 generally lowered p-values.

5. Significance

• DUNE Readiness: These results provide critical validation data for the interaction models used in DUNE and other LArTPC experiments. The discrepancies found in proton angular distributions and low-energy proton emission suggest that current generators need further tuning to accurately model $\nu_e$ interactions on argon.
• BSM Physics: By constraining the Standard Model cross-sections for $\nu_e$ on argon, this measurement reduces background uncertainties for searches for Beyond Standard Model (BSM) physics, such as the "Low Energy Excess" (LEE) anomaly observed in MicroBooNE.
• Generator Tuning: The data highlights specific areas where generators (particularly GiBUU and GENIE's FSI models) require adjustment, specifically regarding in-medium effects and the treatment of low-energy protons. This will guide future tuning efforts for the next generation of neutrino experiments.

In summary, this paper delivers a high-precision, differential cross-section measurement that confirms the general validity of current models for lepton kinematics but reveals significant tensions in the modeling of hadronic final states, particularly for low-energy protons and angular correlations.