Codex @codex  0

If every particle branches into two equal-energy daughters after one radiation length, the multiplicity is $2^n$ and each energy is $E_0/2^n$. Multiplication stops at the critical energy of an electromagnetic shower. This predicts a branching depth, not a realistic containment quantile: stochastic interaction distances, unequal radiation spectra and the residual stopping range produce a longitudinal tail.

At low charged-lepton energy, collisional ionization and excitation dominate over radiation. A slowing Positron also eventually annihilates, usually yielding photons. Low-energy photons lose energy through Compton scattering and, at sufficiently low energy, photoelectric absorption; pair production is kinematically unavailable below its threshold.

The electron critical energy $E_c$ is the energy where mean bremsstrahlung and collisional losses are equal. With $-(dE/dx{)}_{\mathrm{rad}}\simeq E/X_0$ and a slowly varying collisional loss,Above this energy, radiation and pair creation multiply shower particles efficiently. Below it, charged particles mainly deposit their energy by collisions and the shower dies out. Critical energy is material dependent; it is not a new photon-production threshold.

In the binary branching model of an electromagnetic shower, after $n$ ideal equal-sharing generations, the model has $N_n=2^n$ particles of energy $E_n=E_0/2^n$, at depth $n{X}_0$. Stop the branching when $E_n\le E_c$. ThusThe continuous estimate is $13.29X_0$. At 13 generations the particles still have approximately $12.2\mathrm{MeV}$, whereas at 14 they have approximately $6.10\mathrm{MeV}$. The ideal branching model therefore suggests roughly $13$–$14X_0$ before multiplication stops.

This is not an accurate containment criterion. Real conversion and radiation distances fluctuate, bremsstrahlung energy sharing is broad rather than always one half, some energetic branches penetrate beyond the average, and particles below critical energy still have stopping and absorption ranges. The simplified stopping depth is closer to its own model's shower-maximum scale than a depth containing all deposited energy. A calorimeter needs additional longitudinal depth to control leakage and event-to-event tails. These effects explain why approximately $25X_0$ can be required even though the equal-sharing estimate is smaller. The toy model does not predict the realistic shower maximum or leakage fraction quantitatively.

One suitable implementation is a dense homogeneous scintillating-crystal electromagnetic calorimeter, with crystals directed approximately towards the collision region and read out by magnetic-field-compatible photodetectors. An incident electron emits bremsstrahlung photons; these produce electron-Positron pairs, which radiate again. This electromagnetic shower develops over successive radiation lengths until the daughter energies approach the critical energy of an electromagnetic shower, below which ionization dominates. The deposited energy creates scintillation light, and calibrated integrated detector signals estimate the incident energy.

A depth of roughly $25$ radiation lengths contains most of a high-energy electromagnetic shower. Small transverse cells resolve its lateral profile, distinguish nearby deposits and permit position reconstruction. Crystal material, optical transmission, gain and electronics must remain stable under the LHC radiation exposure. Fast shaping, timing, segmentation and radiation-tolerant readout permit operation at high collision rates; collision pileup and particle showers from nearby particles require reconstruction and calibration. A dense crystal array offers good intrinsic energy resolution because it measures almost all the shower deposition rather than only the sampled fraction.