Scintillating-Fiber Tracking Calorimeter

• Scintillating-fiber tracking calorimeters are sampling detectors that combine dense absorbers with interleaved scintillating fibers to measure energy and reconstruct 3D shower profiles.
• They feature varied architectures—from tungsten and lead configurations to dual-readout systems—that enhance spatial localization and particle identification in cosmic-ray and collider experiments.
• Robust calibration strategies and optical readout methods (using PMTs or SiPMs) are employed to optimize energy resolution, dynamic range, and tracking accuracy in diverse application scenarios.

Scintillating-fiber tracking calorimeters are sampling calorimeters in which scintillating fibers or scintillating-fiber ribbons constitute the active medium inside or between dense absorbers, so that the same detector stack provides calorimetric measurement and spatial information on shower position, longitudinal development, and, in fine-grained implementations, incident direction. Reported realizations include tungsten/ribbon calorimeters for cosmic-ray nuclei, lead/scintillating-fiber three-dimensional electromagnetic calorimeters, Pb/ScFi collider prototypes, and dual-readout calorimeters that instrument scintillation and Cherenkov fibers in parallel (Yoon et al., 2010, Vecchi et al., 2012, Klest et al., 3 Apr 2025, Lee et al., 2017).

1. Architectural forms

The architecture appears in several distinct but related forms. The CREAM calorimeter is a 20 radiation length sampling calorimeter comprising 20 layers of tungsten interleaved with 20 layers of scintillating fiber ribbons, preceded by two graphite plates providing about $0.42$ proton interaction lengths; ISS-CREAM preserves the same tungsten/scintillating-fiber sandwich principle with a preceding carbon target to induce hadronic interactions (Yoon et al., 2010, Zhang et al., 2020). In both instruments, the fiber ribbons are segmented transversely and stacked longitudinally, so the calorimeter samples shower development layer by layer while also retaining shower-axis information.

A second family is exemplified by the AMS-02 ECAL, a 3-dimensional sampling calorimeter made of lead and scintillating fibers, with 18 samplings in the longitudinal direction and 72 samplings in the lateral direction. Its alternating superlayer orientations along $x$ and $y$ provide true 3D shower imaging rather than a purely projective energy measurement (Vecchi et al., 2012). A related lead/scintillating-fiber geometry appears in the Baby BCAL prototype for the future Electron-Ion Collider, where a 22.2 cm deep active volume, corresponding to $15.5\,X_0$ at normal incidence, is built from 0.5 mm corrugated lead sheets and 1 mm diameter scintillating fibers running parallel to the beam (Klest et al., 3 Apr 2025).

A third form is the dual-readout fiber calorimeter, in which scintillating fibers and clear Cherenkov fibers are embedded together in absorber material. In the reported Pb-fiber implementation, each tower contains equal numbers of plastic scintillating fibers and PMMA-based Cherenkov fibers, so the detector samples total charged-particle energy deposition and the relativistic electromagnetic component simultaneously (Lee et al., 2017). This is conceptually distinct from single-readout scintillating-fiber calorimetry because the event reconstruction explicitly uses the contrast between the two optical channels.

Ultracompact variants emphasize small radiation length and small Molière radius. The tungsten/scintillating-fiber electromagnetic calorimeter prototype developed for a high-rate muon $g-2$ experiment alternates 0.5-mm-thick tungsten plates and 0.5-mm-diameter plastic scintillating fiber ribbons, yielding a calculated radiation length of 0.69 cm and a calculated Molière radius of 1.73 cm (0910.0818). At the opposite end of the design space, the RADiCAL program uses alternating layers of very dense absorber and scintillating plates, with wavelength-shifting fiber or capillary extraction, to target ultracompact and radiation-hard electromagnetic calorimetry (Anderson et al., 2022).

2. Active media, segmentation, and absorber coupling

The defining functional element is the scintillating fiber itself. In CREAM and ISS-CREAM, each ribbon is 1 cm wide and made from 19 fibers of 0.5 mm diameter, with 50 ribbons per layer over a $50\ \mathrm{cm}\times 50\ \mathrm{cm}$ plane (Yoon et al., 2010, Zhang et al., 2020). Neighboring layers alternate orientation, so transverse sampling is orthogonal from plane to plane. This arrangement provides both calorimetry and what the CREAM and ISS-CREAM studies describe as shower position information or 3D shower-profile reconstruction.

Lead/scintillating-fiber calorimeters often use denser lateral packing. The AMS-02 ECAL consists of lead, fibers, and glue in a volume ratio of $1.00:0.57:0.15$, with an average density of $6.8\ \mathrm{g/cm^3}$, an active volume of $68.5\times 68.5\times 16.7\ \mathrm{cm^3}$, and a total depth of $17\,X_0$ (Vecchi et al., 2012). The Baby BCAL prototype uses 1 mm diameter Kuraray SCSF-78MJ double-clad scintillating fibers bonded with optical epoxy to lead sheets, and it is segmented into four longitudinal layers in readout (Klest et al., 3 Apr 2025). The KLOE calorimeter block studied for DUNE reuse similarly employs grooved lead foils interleaved with blue-green scintillating fibers in a composite of density $x$0 (Alemanno et al., 2024).

The absorber choice sets the shower scale and sampling fraction. Tungsten is used in CREAM, ISS-CREAM, and the muon $x$1 prototype because it allows a short radiation length in a compact geometry (Yoon et al., 2010, Zhang et al., 2020, 0910.0818). Lead is used in AMS-02, Baby BCAL, and the dual-readout Pb-fiber calorimeter (Vecchi et al., 2012, Klest et al., 3 Apr 2025, Lee et al., 2017). A lead/bismuth/tin alloy with 80.5% absorber volume and 19.5% scintillating-fiber volume was used in a neutron-efficiency study, where the fibers ran parallel to the module length at the vertices of 2.1 mm squares (Anelli et al., 2010). That configuration illustrates a broader point: the tracking-calorimeter role of scintillating fibers is not restricted to electromagnetic calorimetry, because the fine fiber lattice can also sample neutron-induced secondaries and hadronic activity.

3. Optical transport, photosensors, and dynamic-range management

The optical chain is usually more elaborate than the absorber/fiber stack alone suggests. In CREAM, scintillation light from the ribbons is transmitted via light mixers and clear fibers and read by 73-pixel Hybrid Photodiodes, chosen for high uniformity, high channel count, and a linear dynamic range of about $x$2. To cover the much larger range of local energy deposits, each ribbon signal is split optically into low-, mid-, and high-range sub-bundles, with neutral-density attenuation filters of 50% transmission for the mid range and 16% for the high range, producing an overall dynamic range of $x$3 (Yoon et al., 2010). ISS-CREAM retains the same three-bundle HPD readout strategy (Zhang et al., 2020).

The AMS-02 ECAL uses Hamamatsu R-7600-00-M4 multianode PMTs, with high-gain and low-gain channels and an HG/LG conversion factor of about 33. The last dynode is also used for a fast standalone trigger (Vecchi et al., 2012). In Pb/ScFi collider prototypes, SiPM readout has become central: Baby BCAL reads out both ends with 40 acrylic light guides per side feeding Hamamatsu S12045 SiPM arrays, and reconstructs deposited energy from the geometric mean $x$4 (Klest et al., 3 Apr 2025).

Dual-readout systems introduce a second optical degree of freedom and a second set of sensor constraints. In the Pb-fiber dual-readout calorimeter, each tower provides completely independent scintillation and Cherenkov signals, each read out by a dedicated photomultiplier tube (Lee et al., 2017). In the later SiPM-based dual-readout demonstrator, each scintillating fiber and each Cherenkov fiber was read out individually by a SiPM, and the principal technical issue was optical crosstalk because the scintillation signal exceeded the Cherenkov signal by about a factor of 60. The measured cross-contamination into Cherenkov channels was $x$5, consistent with a laboratory LED result of $x$6, while the Cherenkov light yield was estimated at $x$7 detected photoelectrons per GeV after corrections (Antonello et al., 2018).

The choice between PMTs and SiPMs remains application-dependent rather than uniformly monotonic. In the KLOE lead/scintillating-fiber calorimeter study, SiPM arrays achieved efficiencies around $x$8 and timing resolutions of $x$9 to $y$0, while PMTs reached $y$1 to $y$2 efficiency and $y$3 timing for the scintillation signal; the paper concludes that no significant improvement in efficiency or timing is found in that application, and that coupling difficulties and lack of performance gain argue against replacing the installed PMTs (Alemanno et al., 2024). This is a recurrent design tension in scintillating-fiber calorimetry: magnetic-field compatibility and granularity favor solid-state sensors, but optical geometry, saturation, dark noise, and installed infrastructure may still favor vacuum photodetectors.

4. Tracking and imaging functionality

The “tracking” component of a scintillating-fiber tracking calorimeter is generally shower tracking rather than precision charged-particle tracking in the spectrometer sense. In ISS-CREAM, perpendicular scintillating-fiber layers provide a 3D profile of the shower and enable reconstruction of incident particle direction and core location (Zhang et al., 2020). In AMS-02, the fine-grained lead/fiber structure allows accurate energy and direction determination for photons and supports positron/proton separation through longitudinal and lateral shower topology (Vecchi et al., 2012). The CREAM calorimeter similarly uses 1 cm transverse segmentation and 20 longitudinal samplings to reconstruct shower profiles, and its best signal-to-noise ratio is obtained by summing over five ribbons per layer centered at the shower core because the Molière radius is only 1.12 cm (Yoon et al., 2010).

A notable example of calorimeter-only directional reconstruction is the Neighbor Cell Deposited Energy Ratio method developed for the AMS-02 ECAL. In each layer, the ratio of deposited energies in neighboring cells is related exponentially to the hit position within the central cell, and straight-line fits through the reconstructed layer positions produce the incident angle. The reported angular resolution for photon direction is

$y$4

compared with

$y$5

for the commonly adopted center-of-gravity method (Li et al., 2013). The same study reports nearly 100% reconstruction efficiency for NCDER and no bias for large incident angles, whereas the full lateral-fit method is more computationally expensive and reconstructs only about 60% of events.

Dual-readout fiber calorimeters occupy an intermediate position. The hadron-detection study states explicitly that the device is not a true tracking detector, yet its fine fiber segmentation and parallel fiber geometry offer spatial localization of shower development, and, combined with external wire chambers, can provide transverse impact point information at millimeter precision (Lee et al., 2017). A common misconception is therefore that any scintillating-fiber calorimeter with fine pitch is automatically a standalone tracker. The reported literature supports a narrower statement: fine granularity improves shower-axis reconstruction and image-based particle identification, but standalone track measurement depends on geometry, readout segmentation, and the availability of external reference detectors.

5. Calibration strategies and measured performance

Calibration is typically channel-by-channel, simulation-assisted, and geometry-aware. In CREAM, low-range equalization used a 150 GeV electron beam scanned across the calorimeter; for each ribbon, events centered on that ribbon were compared with Monte Carlo energy deposits to derive an ADC-to-MeV constant, with $y$6 given as an example for a typical ribbon. Lead bricks were placed before the calorimeter to shift the shower maximum upward for calibration of top layers, and the stack was rotated for bottom-layer calibration; mid- and high-range channels were then inter-calibrated through overlapping signal regions (Yoon et al., 2010).

ISS-CREAM extended this procedure with position, energy, angle, and high-voltage scans, and addressed light attenuation explicitly. The measured ribbon intensity was modeled as

$y$7

with a short attenuation length of a few centimeters and a long attenuation length of about 250 cm. The corresponding correction,

$y$8

removes position dependence along the ribbon after calibration (Zhang et al., 2020). The same paper reports repeated scans with RMS/mean at the per-mil level.

Representative performance figures span a broad range of beam energies and physics targets.

Detector or class	Selected results	Context
CREAM calorimeter	$y$9 at 150 GeV; slope $15.5\,X_0$0; energy uniform within $15.5\,X_0$1	Electron calibration beam (Yoon et al., 2010)
ISS-CREAM calorimeter	Electron response linear from 50–175 GeV; slope $15.5\,X_0$2; $15.5\,X_0$3 at 150 GeV; pion resolution $15.5\,X_0$4	After attenuation correction (Zhang et al., 2020)
AMS-02 ECAL	Energy resolution better than $15.5\,X_0$5 for $15.5\,X_0$6; linearity deviation $15.5\,X_0$7 from 8 to 180 GeV; proton rejection $15.5\,X_0$8–$15.5\,X_0$9 at 90% electron efficiency	ECAL-only performance (Vecchi et al., 2012)
Pb-fiber dual-readout calorimeter	$g-2$0 for single hadrons; jets $g-2$1	Electron-calibrated hadron and jet reconstruction (Lee et al., 2017)
Baby BCAL Pb/ScFi prototype	Electron resolution $g-2$2 at 4–10 GeV; constant term $g-2$3; longitudinal profile improves pion rejection by a factor of 1.6–1.8	EIC-oriented Pb/ScFi benchmark (Klest et al., 3 Apr 2025)
W/SciFi $g-2$4 prototype	Measured intrinsic sampling term $g-2$5	1.5–3.5 GeV electrons (0910.0818)

These measurements also delimit the performance envelope of different subtypes. Conventional scintillating-fiber sampling calorimeters can be highly linear for electromagnetic showers but remain non-compensating for hadrons, as illustrated by the ISS-CREAM pion response and resolution (Zhang et al., 2020). Dual-readout systems attempt to mitigate precisely that limitation by measuring the electromagnetic shower fraction event by event; in the Pb-fiber case the reconstruction uses

$g-2$6

where $g-2$7 and $g-2$8 are the scintillation and Cherenkov signals and $g-2$9 for the reported device (Lee et al., 2017).

6. Applications, limitations, and current directions

The application space is correspondingly broad. CREAM and ISS-CREAM were built for direct measurements of high-energy cosmic-ray nuclei, with the former targeting nuclei from H to Fe over $50\ \mathrm{cm}\times 50\ \mathrm{cm}$0 to 1000 TeV and the latter measuring elemental spectra for $50\ \mathrm{cm}\times 50\ \mathrm{cm}$1 to 26 from the International Space Station (Yoon et al., 2010, Zhang et al., 2020). AMS-02 uses its lead/scintillating-fiber ECAL to measure cosmic rays up to a few TeV and to separate positrons from protons, while also providing direct photon energy and direction measurement (Vecchi et al., 2012). Baby BCAL benchmarks the Pb/ScFi component of the future Barrel Imaging Calorimeter in ePIC, where even four longitudinal layers already improved electron–pion separation through shower-profile information (Klest et al., 3 Apr 2025).

Hadronic and neutral-particle applications expose different strengths. The dual-readout Pb-fiber calorimeter reconstructs proton and pion beam energies to within a few percent at all tested energies after electron calibration, and its leakage counters significantly narrow non-Gaussian tails (Lee et al., 2017). The lead/bismuth scintillating-fiber calorimeter used for neutron studies showed efficiency per active centimeter of $50\ \mathrm{cm}\times 50\ \mathrm{cm}$2 at a 10 MeV electron-equivalent threshold, compared with $50\ \mathrm{cm}\times 50\ \mathrm{cm}$3 for a bulk organic scintillator slab under the same neutron spectrum; the paper attributes the enhancement to neutron interactions in the high-$50\ \mathrm{cm}\times 50\ \mathrm{cm}$4 absorber producing secondary charged particles that reach the fibers (Anelli et al., 2010). This suggests that scintillating-fiber tracking calorimetry can be tuned not only for electromagnetic imaging but also for compact neutron-sensitive sampling structures.

Current development directions move in two partially convergent directions: higher integration of precision layers, and higher robustness in extreme environments. The hybrid electromagnetic calorimeter module that integrates silicon pixel layers into a longitudinally segmented scintillating-fiber sampling calorimeter reports maximum achievable improvements of 56% for position resolution and 26% for time resolution, together with a 16% boost in signal significance for $50\ \mathrm{cm}\times 50\ \mathrm{cm}$5 in the $50\ \mathrm{cm}\times 50\ \mathrm{cm}$6 channel (Fei et al., 22 Sep 2025). RADiCAL, by contrast, couples dense scintillating plates to wavelength-shifting capillaries or fibers and targets ultracompact, radiation-hard modules with simultaneous position, energy, and timing measurement; the reported module dimensions are $50\ \mathrm{cm}\times 50\ \mathrm{cm}$7, with energy resolution of approximately $50\ \mathrm{cm}\times 50\ \mathrm{cm}$8, timing resolution below $50\ \mathrm{cm}\times 50\ \mathrm{cm}$9–$1.00:0.57:0.15$0, and position resolution at the few-millimeter level (Anderson et al., 2022).

A persistent technical question is therefore not whether scintillating-fiber calorimeters can provide tracking information, but how much tracking functionality is worth extracting inside the calorimeter itself. The reported literature supports several answers rather than one. For cosmic-ray and space detectors, coarse but reliable shower-axis reconstruction can be sufficient (Yoon et al., 2010, Vecchi et al., 2012). For collider calorimetry, finer longitudinal segmentation and topology-aware reconstruction directly improve particle identification (Klest et al., 3 Apr 2025). For high-rate or radiation-hard environments, ultracompact modules and fast timing may be equally central design criteria (0910.0818, Anderson et al., 2022). A plausible implication is that the modern scintillating-fiber tracking calorimeter is increasingly best understood not as a single detector type, but as a family of sampling calorimeters that trade absorber density, optical topology, longitudinal segmentation, and readout modality against specific requirements in shower imaging, compensation, timing, and system integration.