Silicon Pixel Detector
A camera so fast it photographs a trillion collisions per year, each lasting less than a nanosecond.
Silicon pixel detectors are the innermost tracking layers of particle physics experiments like CMS and ATLAS at the Large Hadron Collider (LHC) at CERN. They surround the beam collision point and record precisely where each charged particle passes. By linking hits across multiple layers, computers reconstruct the particle's curved path in the magnetic field, measuring its momentum and identifying where it came from — including the decays of very short-lived particles like B mesons, whose decay 'vertex' can be just 100 micrometers from the collision point.
A silicon pixel sensor is a reverse-biased semiconductor junction. When a charged particle passes through the ~200–300 µm thick silicon, it knocks loose about 10,000 electrons per millimeter of path, which drift in an applied electric field to the pixel electrode above them. Each pixel (typically 100 µm × 150 µm in CMS Phase-1) has its own readout chip (bump-bonded to the sensor) that amplifies and digitizes the signal. At the LHC's innermost layer (~3 cm from the beam), particle hit rates reach 600 million particles per cm² per second, requiring specialized radiation-hard sensors and readout chips designed in 65 nm CMOS technology.
In plain terms
Imagine taking a photograph of fireworks using a camera with 124 million individual sensors, each smaller than a grain of sand. Each sensor lights up the instant a single spark passes through it. Four layers of these cameras stacked on top of each other let scientists draw a track through each spark's path — and from the curve of the track in a magnetic field, they can tell how heavy the particle was and where it came from.
Also called: pixel tracker, hybrid pixel detector, silicon vertex detector.
Why it matters
- The Higgs boson was discovered at the LHC in 2012 partly through silicon pixel detectors identifying the photons and leptons it decays into, reconstructed to within micrometers of the collision point
- Pixel detectors are the only way to distinguish the decay vertices of bottom quarks (B mesons) from primary collisions — essential for studying matter-antimatter asymmetry (CP violation)
- CMS's 124 million pixels can handle 600 million particle hits per cm² per second while maintaining >96% hit efficiency — an engineering achievement with no precedent in consumer electronics
- The radiation doses at the LHC inner layers (up to 10¹⁶ particles/cm²/year) would destroy ordinary electronics within hours; pixel detector R&D has driven advances in radiation-hard semiconductor design
Where it lives
- CMS Phase-1 pixel detector — 124 million pixels, 4 barrel layers + 3 forward disks, 100×150 µm pixel size (installed 2017, CERN)
- ATLAS ITk pixel detector — 5.1 billion pixels, 5 barrel layers, 13 m² total area (under construction for HL-LHC era, 2026–)
- ALICE Inner Tracking System 2 (ITS2) — monolithic CMOS pixels, 30×30 µm pixel size, 10 m² area; lightest large-area silicon tracker ever built
Key numbers
- CMS Phase-1 pixel count: 124 million (4 barrel layers + 3 endcap disks)
- Pixel size (CMS Phase-1): 100 × 150 µm (About 2 human hairs wide)
- Hit spatial resolution: ~10 µm (In x and y at the beam collision point)
- Hit rate at innermost layer: 600 million/cm²/s (At LHC design luminosity)
- Operating temperature: −20 °C (Required to reduce radiation damage effects)
Surprising facts
- CMS's 124 million pixel channels consume only about 50 microwatts each — yet the total heat load is still 7.5 kW, requiring the detector to be cooled to −20 °C to function
- The ATLAS ITk for the high-luminosity LHC will have 5.1 billion pixels across 13 m² of silicon — 55 times more channels than the current detector
- At 3 cm from the LHC beam, the inner pixel layer absorbs a particle flux equivalent to a radiation dose of several hundred megarads per year — roughly 1 million times a lethal human dose
- Pixel detectors achieve spatial resolution of ~10 µm — good enough to identify a particle that traveled only 100 µm from its birthplace before decaying
Open questions
- At the LHC, silicon pixel sensors must survive: At the LHC, silicon pixel sensors must survive radiation doses that would cause visible color changes and structural damage to ordinary glass — after years of operation, the bulk silicon develops a new type of charge-trapping defect called 'effective doping change' that inverts the detector's electrical polarity, requiring researchers to reverse the bias voltage just to keep it working
- The Lorentz angle: The Lorentz angle — the deflection of drifting charge carriers in the magnetic field — shifts hit positions by up to 10 µm, requiring a correction applied to every hit in every pixel; without this correction, the reconstruction of particle trajectories degrades significantly
A short history
Semiconductor strip detectors were first used in fixed-target particle physics experiments in the 1980s. The move to 2D pixel sensors was driven by the need to handle the high particle densities of the LHC, beginning with R&D in the 1990s through the RD19 collaboration at CERN. The first large pixel detector at a collider was NA62's pixel veto in the 1990s. CMS and ATLAS installed their first barrel pixel detectors for LHC Run 1 starting in 2008. The Phase-1 upgrade (2017) added a fourth barrel layer and improved radiation hardness. The Phase-2 all-silicon inner tracker (ITk) for the High-Luminosity LHC (HL-LHC) is currently in production.
Sources
- CMS Experiment — Silicon Pixels explainer (https://cms.cern/detector/identifying-tracks/silicon-pixels)
- CERN Courier — 'State-of-the-art tracking for high luminosities' (2021) (https://cerncourier.com/a/state-of-the-art-tracking-for-high-luminosities/)
- CERN Courier — 'Tracking the rise of pixel detectors' (2021) (https://cerncourier.com/a/tracking-the-rise-of-pixel-detectors/)
- arXiv — '3D pixel silicon sensors for CMS Phase-2 Inner Tracker' (2025) (https://arxiv.org/abs/2510.13304)