Bolometer
A sliver of metal so thermally isolated it warms by a millionth of a degree when a single photon lands on it — and that is enough to measure.
Absorbs incident electromagnetic radiation across a broad spectral range — from near-infrared through far-infrared, submillimetre, microwave, and even soft X-rays — and registers the resulting temperature rise. The absorbed power is inferred from the change in electrical resistance of a thermally sensitive element. Because the detection principle is purely thermal, a single bolometer design can work from 300 GHz to 30 THz without modification, making it the workhorse detector for submillimetre astronomy, cosmic microwave background polarimetry, and terahertz spectroscopy.
An absorbing membrane — typically a spider-web mesh of metal or superconducting film — is weakly thermally coupled to a cold bath via thin silicon-nitride legs. When radiation is absorbed, the membrane temperature rises by ΔT = P/G, where P is the absorbed power and G is the thermal conductance to the bath. A thermometer element registers ΔT. In the most sensitive variant, the transition edge sensor (TES), the thermometer is a superconducting bilayer (e.g. aluminium–titanium or molybdenum–copper) biased at its sharp resistive transition at around 0.1–0.5 K. Because resistance changes steeply at the transition, even a picowatt of absorbed power produces a detectable current change, read out by a SQUID amplifier.
In plain terms
Imagine a tiny piece of metal floating in space, connected to the rest of the world by threads thinner than a spider's web. When even a little bit of heat arrives — from a microwave or infrared beam — the metal warms up just a tiny fraction of a degree. A bolometer measures that tiny warmth.
Also called: TES bolometer, transition edge sensor, thermal detector, microbolometer.
Why it matters
- Spectral blindness is a feature: a single bolometer array covers the entire millimetre and submillimetre sky without the narrow-band constraints of photon-counting detectors.
- Superconducting TES bolometers have achieved noise-equivalent power (NEP) below 10⁻¹⁷ W/√Hz — sensitive enough to detect the heat from a candle at a distance greater than the diameter of the solar system.
- Focal-plane arrays of thousands of TES bolometers, multiplexed via SQUID readout chains, enable wide-field CMB surveys; BICEP2 used 512 TES sensors at 150 GHz to search for primordial gravitational wave signatures.
- Room-temperature microbolometer arrays — uncooled vanadium-oxide or amorphous-silicon pixels — are the basis of every thermal imaging camera, from search-and-rescue drones to building energy audits.
Where it lives
- BICEP2 / Keck Array at South Pole — 512–2560 TES bolometers at 150–270 GHz, operating at 250–270 mK for CMB B-mode polarimetry
- ALMA submillimetre telescope — 66 antennas each carrying SIS and TES receiver sets across 84–950 GHz
- South Pole Telescope SPT-3G — 16,000 TES bolometers for CMB survey
- NIST / NASA JPL laboratory TES arrays — NEP ~ 10⁻¹⁷ W/√Hz prototypes for future space missions
- Commercial microbolometer focal-plane arrays (uncooled) — FLIR, Seek Thermal; pixel pitch down to 12 µm
Key numbers
- NEP (TES prototype): ~1.5 × 10⁻¹⁷ W/√Hz (NIST 256-pixel array at 300 mK bath; NIS-cooled)
- NEP (TES, measured): 0.8 aW/√Hz (arXiv:2309.07392; 1/f knee < 0.1 Hz)
- Operating temp (TES): 100–500 mK (Transition edge; set by bilayer material choice)
- Spectral range: Near-IR to microwave (30 THz – 100 GHz for typical TES; uncooled: 8–14 µm)
- BICEP2 detector count: 512 TES pixels (150 GHz, 26 cm aperture, South Pole)
- Langley's sensitivity (1878): 10⁻⁵ °C (Platinum-strip Wheatstone bridge; no cryogenics)
Surprising facts
- A NIST prototype 256-pixel TES bolometer array reaches an NEP of ~1.5 × 10⁻¹⁷ W/√Hz — registering power equivalent to the gravitational potential energy of a single bacterium falling one nanometre.
- Samuel Langley's original bolometer of 1878 could detect the heat of a cow standing a quarter of a mile (400 m) away, and was sensitive to temperature changes of 10⁻⁵ °C — impressive without a single transistor or semiconductor.
- TES bolometers are operated inside dilution refrigerators at 100–300 mK, where the thermal energy kT is only about 10⁻²³ joules — making the detector environment colder than any known naturally occurring object in the universe.
- Because bolometers are thermal sensors, they are inherently wavelength-agnostic: the same pixel design that measures CMB photons at 2 mm wavelength can detect X-rays at 0.1 nm — a factor of 20 million in photon energy.
Open questions
- Unexplained cosmic-ray coincidence bursts in TES arrays: Large TES bolometer arrays on CMB telescopes experience sudden simultaneous glitches across hundreds of pixels — far more correlated than Poisson statistics predict from individual cosmic-ray hits. The energy deposition pattern propagates through the silicon substrate faster than phonon diffusion models account for. The mechanism, possibly ballistic phonon propagation through perfect silicon at millikelvin temperatures, is not fully quantified; mitigation requires pixel-masking algorithms that discard valid data.
A short history
Samuel Pierpont Langley invented the bolometer in 1878 at the Allegheny Observatory in Pennsylvania, using two blackened platinum foil strips in a Wheatstone bridge to map the solar infrared spectrum — discovering previously unknown absorption lines beyond the visible. By 1880 the instrument was refined enough to detect a cow at 400 m. For nearly a century, bolometers used semiconductor thermistors (germanium or silicon) cooled to 4.2 K. The modern transition edge sensor era began in the 1990s when NIST and university groups demonstrated that superconducting bilayer films, operated at their midpoint transition, provided an order-of-magnitude improvement in NEP over semiconductor bolometers. CMB experiments — BOOMERANG (1997), BICEP2 (2010), SPT-3G — adopted TES arrays and demonstrated their scalability to thousands of pixels. The current generation of NIST and JPL TES devices approaches the fundamental thermal noise limit set by quantum fluctuations in the heat-exchange link.
Sources
- NIST — Prototype Thermal Design for a 256-Pixel TES Bolometer Array (https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=903249)
- Tay et al., arXiv:2309.07392 — High-sensitivity TES bolometers, NEP 0.8 aW/√Hz (https://arxiv.org/abs/2309.07392)
- BICEP2 / Keck Array — detector description, 512 TES bolometers at 270 mK (https://lweb.cfa.harvard.edu/~cbischoff/bicepkeck/)
- ESA Herschel — brief history of infrared astronomy; Langley bolometer 1878 (https://sci.esa.int/web/herschel/-/59550-a-brief-history-of-infrared-astronomy)
- NASA NTRS — Ultrasensitive TES based on electron-phonon decoupling (https://ntrs.nasa.gov/api/citations/20110015247/downloads/20110015247.pdf)