SND@LHC: Hybrid Neutrino Detector

Updated 11 January 2026

SND@LHC is a hybrid neutrino experiment that uniquely studies high-energy neutrino interactions and searches for feebly interacting particles in the very-forward LHC region.
Its design integrates tungsten–emulsion targets, SciFi trackers, and a segmented muon system to achieve micron-level resolution and robust flavor identification.
Positioned 480 meters downstream from ATLAS, the detector enables precise flux measurements, forward charm production studies, and tests of Standard Model neutrino cross sections.

The Scattering and Neutrino Detector at the LHC (SND@LHC) is a compact, hybrid neutrino experiment situated 480 meters downstream of the ATLAS interaction point (IP1) in the TI18 tunnel. SND@LHC is optimized for the very-forward region ( $7.2 < \eta < 8.4$ ), where it accesses a previously unexplored laboratory for probing Standard Model (SM) neutrino processes, heavy-flavor production, and beyond-the-Standard-Model (BSM) physics. Its architecture combines tungsten–emulsion targets with electronic trackers and a segmented muon system, enabling reconstruction and identification of all three neutrino flavors. The experimental program is strongly focused on measuring the flux and interaction properties of high-energy ( $100\,\mathrm{GeV}$ – $1\,\mathrm{TeV}$ ) neutrinos produced in LHC $pp$ collisions, as well as searching for feebly interacting particles (FIPs) over an extensive parameter space.

1. Experimental Design and Instrumentation

SND@LHC employs a hybrid detector architecture specifically adapted for operation in a high-background, high-flux LHC environment (collaboration, 2023, Collaboration, 2023, Graverini, 2024, Collaboration, 2022). The main subsystems are:

Veto System: Two (later three) orthogonal planes of 42 $\times$ 6 $\times$ 1 cm $^3$ plastic-scintillator bars instrumented with SiPMs. The system is placed at the entrance and provides tagging of minimum-ionizing charged particles from IP1, achieving an inefficiency as low as $O(10^{-8})$ for muons.
Target and Vertexing Region: The core target comprises five “walls,” each formed of four Emulsion Cloud Chamber (ECC) bricks. Each ECC brick consists of alternating 1 mm-thick tungsten plates and nuclear emulsion films. The total target mass is $\sim830$ kg. The emulsion provides micron-scale spatial resolution, enabling identification of short-lived particles (e.g., $\tau$ decay kinks).
SciFi Trackers: Immediately after each ECC wall lies a SciFi tracking station—each station has two 39 $100\,\mathrm{GeV}$ 039 cm $100\,\mathrm{GeV}$ 1 planes (one horizontal, one vertical), built from six staggered layers of 250 μm polystyrene scintillating fibers. The spatial resolution per plane is $100\,\mathrm{GeV}$ 2150 μm; timing resolution for X–Y coincidences is $100\,\mathrm{GeV}$ 3250 ps.
Muon and Calorimeter System: Downstream, the muon system is composed of eight scintillator stations interleaved with iron. The first five (UpStream, US) employ coarse Y-directed bars; the last three (DownStream, DS) use fine X/Y bars for $100\,\mathrm{GeV}$ 41 cm spatial resolution. The full stack provides $100\,\mathrm{GeV}$ 511 $100\,\mathrm{GeV}$ 6 of interaction length. The interleaved structure enables both muon identification and hadronic calorimetry.
Data Acquisition: All sub-detectors are read out in a triggerless mode, time-clustered and filtered to reduce rates from $100\,\mathrm{GeV}$ 7(kHz) to $100\,\mathrm{GeV}$ 8(Hz) at full LHC luminosity.

The acceptance covers 7.2 < $100\,\mathrm{GeV}$ 9 < 8.4, corresponding to polar angles of $1\,\mathrm{TeV}$ 00.07–0.14 mrad relative to the beam axis.

2. Flux Characterization, Event Selection, and Muon Measurements

SND@LHC defines the particle flux per unit integrated luminosity and per unit area as

$1\,\mathrm{TeV}$ 1

where $1\,\mathrm{TeV}$ 2 is the reconstructed and efficiency-corrected number of particles (e.g., muons), $1\,\mathrm{TeV}$ 3 is the integrated luminosity, and $1\,\mathrm{TeV}$ 4 is the fiducial detector area (collaboration, 2023). Fiducial areas differ by subsystem:

Emulsion: 18 $1\,\mathrm{TeV}$ 518 cm $1\,\mathrm{TeV}$ 6
SciFi: 31 $1\,\mathrm{TeV}$ 731 cm $1\,\mathrm{TeV}$ 8
Downstream Muon: 52 $1\,\mathrm{TeV}$ 952 cm $pp$ 0

Muon tracking utilizes both Simple Tracking (ST) and Hough-Transform (HT) algorithms followed by a Kalman filter fit. Efficiency with HT tracking reaches $pp$ 1 for SciFi and $pp$ 2 for DS.

Measured muon fluxes (all in units of $pp$ 3) are (collaboration, 2023):

Emulsion: $pp$ 4
SciFi: $pp$ 5
DS: $pp$ 6

Systematic uncertainties arise from ATLAS luminosity calibration (2.2%), local tracking efficiency variations (2.2%-2.9%), and tracking algorithm choice (2.0%-4.8%). Combined total systematics are $pp$ 7– $pp$ 8.

Monte Carlo chain (DPMJET $pp$ 9FLUKA $\times$ 0GEANT4) underpredicts the measured fluxes by $\times$ 1– $\times$ 2, a plausible deficit considering uncertainties in hadron production, decay kinematics, transport, and rock propagation.

Subdetector	Fiducial area	$\times$ 3 [ $\times$ 4]
Emulsion (ECC)	18 $\times$ 518 cm $\times$ 6	$\times$ 7
SciFi	31 $\times$ 831 cm $\times$ 9	$\times$ 0
DownStream (DS)	52 $\times$ 152 cm $\times$ 2	$\times$ 3

These values benchmark the muon-induced background for the neutrino program and validate detector performance (collaboration, 2023).

3. Neutrino Flavor Identification, Event Yields, and Topologies

SND@LHC’s hybrid design separates neutrino interaction channels and flavors (Collaboration, 2023, collaboration, 2024, collaboration, 2024, Collaboration, 2022):

$\times$ 4 CC: Identified by a long, penetrating muon track traversing all eight muon layers, with corresponding hadronic shower in the target.
$\times$ 5 CC: Manifest as electromagnetic showers in SciFi/ECC, no penetrating muon.
$\times$ 6 CC: Kink topology in emulsion, with short secondary (decay) vertex; identification requires sub-micron resolution and secondary-track reconstruction.

Charged-current (CC) and neutral-current (NC) event rates are, for full Run 3 ( $\times$ 7250 fb $\times$ 8):

Total CC: 1690 events (comprising 72% $\times$ 9, 23% $^3$ 0, 5% $^3$ 1)
Total NC: 555 events

Observed collider $^3$ 2 CC events (8 candidates, 0.076 $^3$ 30.031 background) correspond to a $^3$ 4 significance for the forward region, signifying first direct observation of LHC-produced neutrinos in this pseudorapidity range (Collaboration, 2023). Recent analysis isolated 9 “muon-less” events (dominated by $^3$ 5CC and NC), with $^3$ 6 expected background ( $^3$ 7 significance) (collaboration, 2024).

Inclusive CC cross section is approximately $^3$ 8, validated using GENIE and full detector simulation.

4. Hadronic Calibration, Energy Resolution, and Reconstruction Strategy

The total visible hadronic energy in $^3$ 9N interactions is reconstructed from combined signals in the SciFi target and upstream hadronic calorimeter (HCAL), optimized via calibration with $O(10^{-8})$ 0– $O(10^{-8})$ 1 hadron beams (Collaboration, 2 Apr 2025). Event-by-event shower tagging utilizes a hit-density algorithm: a $O(10^{-8})$ 2-ch sliding window in the SciFi identifies showers with $O(10^{-8})$ 3 in-time hits.

Total reconstructed energy is modeled as

$O(10^{-8})$ 4

where $O(10^{-8})$ 5 (SciFi amplitude) and $O(10^{-8})$ 6 (calorimeter amplitude) are calibrated separately for each target wall. For tungsten targets, expected performance is:

Energy resolution: $O(10^{-8})$ 720% at $O(10^{-8})$ 8 GeV, $O(10^{-8})$ 910% above $\sim830$ 0 GeV
Linearity: within $\sim830$ 1 over $\sim830$ 2– $\sim830$ 3 GeV
Systematic uncertainty on reconstructed neutrino energy: $\sim830$ 45%

Depth resolution for the hadronic shower origin is $\sim830$ 510 cm, with overall $\sim830$ 690% assignment purity for 100–300 GeV showers.

5. Physics Program: Heavy Flavor, PDF Constraints, and Lepton Universality

SND@LHC’s acceptance ( $\sim830$ 7) provides unique access to collider neutrinos produced predominantly via charm-hadron decays. The yield of forward $\sim830$ 8 and $\sim830$ 9 CC interactions is directly sensitive to $\tau$ 0 at $\tau$ 1, inaccessible to central detectors (Graverini, 2024, Collaboration, 2022).

The experiment will:

Measure the inclusive $\tau$ 2 up to TeV neutrino energies, probing nucleon structure functions, charm production (forward gluon PDF), and lepton flavor universality (LFU) via ratios $\tau$ 3 and $\tau$ 4.
Provide first collider-based tagged samples of $\tau$ 5, enabling comparison of $\tau$ 6, $\tau$ 7, and $\tau$ 8 cross sections and SM universality tests at high $\tau$ 9.
Constrain the forward charm contribution, critical for modeling the prompt atmospheric neutrino background for neutrino telescopes.

For Run 3, statistical uncertainties on $100\,\mathrm{GeV}$ 00-tagged cross sections are $100\,\mathrm{GeV}$ 015% (stat), systematic uncertainties on flux and reconstruction dominate total errors ( $100\,\mathrm{GeV}$ 0210%–35% depending on process and flavor) (Graverini, 2024, Zaffaroni, 2023).

6. Nonstandard Interactions, BSM Probes, and Upgrades

SND@LHC is equipped for model-independent searches for FIPs via scattering or decays in the detector. Run 3 sensitivities extend to:

Light dark matter (LDM) via “leptophobic portal” vector mediator: can probe $100\,\mathrm{GeV}$ 03– $100\,\mathrm{GeV}$ 04 for $100\,\mathrm{GeV}$ 05– $100\,\mathrm{GeV}$ 06 GeV, reaching $100\,\mathrm{GeV}$ 07– $100\,\mathrm{GeV}$ 08 signal events for elastic/NC-CC ratio signatures (Boyarsky et al., 2021, Biswas, 6 Jan 2026).
Portal decays (dark photon, scalar, heavy neutrino): $100\,\mathrm{GeV}$ 09 signal event thresholds correspond to $100\,\mathrm{GeV}$ 10–$100\,\mathrm{GeV}$11 or $100\,\mathrm{GeV}$ 12– $100\,\mathrm{GeV}$ 13 sensitivity below $100\,\mathrm{GeV}$ 14 GeV.
Nonstandard neutrino interactions (NSI), especially in the charm sector: current SND@LHC configuration has $100\,\mathrm{GeV}$ 15– $100\,\mathrm{GeV}$ 16 sensitivity to best-fit charm NSI/LUV, rising to $100\,\mathrm{GeV}$ 17 for an upgraded detector exploiting full HL-LHC statistics (Bhattacharya et al., 2024).
Time-of-flight discrimination (200 ps) can in principle separate TeV-scale FIPs from SM neutrinos (Collaboration et al., 2020).

Planned HL-LHC upgrades (“AdvSND”) involve a magnetized iron/silicon calorimeter for $100\,\mathrm{GeV}$ 18 charge identification, silicon microstrip vertex detectors (timing $100\,\mathrm{GeV}$ 1920 ps), and larger acceptance. This enables:

Full separation of $100\,\mathrm{GeV}$ 20/ $100\,\mathrm{GeV}$ 21, including first direct observation of $100\,\mathrm{GeV}$ 22
Tagged neutrino beams via ultra-fast timing in coincidence with ATLAS, critical for precision studies of $100\,\mathrm{GeV}$ 23 sources (“Pontecorvo’s concept”) (collaboration, 31 Mar 2025, Graverini, 2024)
O(10⁵⁾ neutrino interactions, enabling percent-level cross section measurements, small- $100\,\mathrm{GeV}$ 24 PDF determination, and BSM searches with an order-of-magnitude increase in sensitivity.

7. Summary and Context within the LHC Neutrino Program

SND@LHC is the first experiment to deliver high-significance collider neutrino observations in the very-forward LHC region, complementing FASER $100\,\mathrm{GeV}$ 25 (on-axis). Its hybrid ECC+SciFi+calorimetry architecture supports flavor tagging, precision energy reconstruction, and high-resolution vertexing, supporting both Standard Model and BSM science. The combination of unique acceptance, flavor-sensitivity, and accessible energy regime opens domains of forward QCD, heavy-flavor production, and neutrino cross-section measurements at energies relevant to cosmic neutrino detection.

Compared to central and on-axis detectors, SND@LHC’s acceptance preferentially samples charm-induced neutrino flux and provides a kinematic lever arm on the proton structure at previously untested $100\,\mathrm{GeV}$ 26 and $100\,\mathrm{GeV}$ 27. The planned HL-LHC upgrade will markedly enhance both SM and new-physics reach, including the possibility of tagged charm events and first direct $100\,\mathrm{GeV}$ 28 identification (collaboration, 31 Mar 2025, Graverini, 2024).