single-photon detectors (original) (raw)

Acronym: SPD

Definition: photodetectors which can register single photons

Alternative term: photon counting detectors

Categories: light detection and characterization, quantum photonics

• photodetectors
  • infrared detectors
  • position-sensitive detectors
  • photodiodes
  • phototransistors
  • metal–semiconductor–metal photodetectors
  • velocity-matched photodetectors
  • photomultipliers
  • phototubes
  • pyroelectric detectors
  • photoconductive detectors
  • photoemissive detectors
  • solar-blind photodetectors
  • single-photon detectors
    * photomultipliers
    * Geiger-mode single-photon avalanche diodes
    * superconducting nanowire detectors
    * upconversion detectors
    * silicon photomultipliers
  • terahertz detectors
  • velocity-matched photodetectors
  • (more topics)

Related: photodetectorsphotonsphoton detection efficiencydark count ratequantum photonics

Page views in 12 months: 220

DOI: 10.61835/2wi Cite the article: BibTex BibLaTex plain textHTML Link to this page! LinkedIn

Content quality and neutrality are maintained according to our editorial policy.

📦 For purchasing single-photon detectors, use the RP Photonics Buyer's Guide — an expert-curated directory for finding all relevant suppliers, which also offers advanced purchasing assistance.

Contents

What are Single-photon Detectors?

Properties of Single-photon Detectors

Types of Single-photon Detectors

Vacuum Tubes

Geiger Mode Avalanche Photodiodes (SPAD)

Silicon Photomultipliers

Hybrid Photomultipliers

Superconducting Detectors

Nonlinear Upconversion Detectors

X-ray and Gamma-ray Detectors

Single-photon Imaging Detectors

Applications of Single-photon Detectors

Frequently Asked Questions

Summary:

This article provides a comprehensive overview of single-photon detectors, which are highly sensitive photodetectors capable of registering individual photons. It begins by defining these devices and distinguishing them from simple photon counters and photon-number-resolving detectors.

Key performance characteristics are thoroughly explained, including photon detection efficiency, dark count rate, dead time, timing jitter, and afterpulsing.

The text then details the various types of single-photon detectors, covering traditional vacuum-based devices like photomultiplier tubes (PMTs), semiconductor detectors such as single-photon avalanche diodes (SPADs) made from silicon or InGaAs, and advanced cryogenic superconducting detectors like superconducting nanowire single-photon detectors (SNSPDs) and transition-edge sensors (TES). Specialized types for X-ray detection and single-photon imaging are also discussed.

Finally, the article highlights the broad range of applications, including quantum photonics, time-resolved spectroscopy, LIDAR, medical imaging, astronomy, and high-energy physics.

(This summary was generated with AI based on the article content and has been reviewed by the article’s author.)

What are Single-photon Detectors?

Single-photon detectors are highly sensitive photodetectors that can respond to single photons of light and, more broadly, to single quanta of electromagnetic radiation up to the X-ray region. They typically produce a digital pulse rather than a continuously variable photocurrent. A fast electronic discriminator is often used so that an output pulse is generated whenever the raw signal exceeds a set threshold.

The term photon counting detector is frequently used with the same meaning as single-photon detectors, but strictly speaking should refer to devices that actually count individual photon events over some interval. Note also that most single-photon detectors are unable to distinguish a single-photon event from a multi-photon event, and are thus unable to count photons in Fock states, for example. Specialized photon-number resolving detectors (see below) and multiplexing schemes can provide limited number resolution. Unfortunately, imprecise use of such terms is common, which can cause confusion.

Existing detectors are not perfect in various respects. In particular, they have a finite quantum efficiency (missing some of the photons) and a nonzero dark count rate (false detections without incident light).

In essentially all common implementations, the detected photons are absorbed (i.e., destroyed), as is typical for photodetectors. In principle, one can perform quantum nondemolition measurements to detect the presence of photons without absorbing them, but such schemes are not industrially mature and mainly of research interest.

Properties of Single-photon Detectors

Core figures of merit include:

• Photon detection efficiency (PDE): This is the probability that an incident photon produces a registered count (assuming that the detector is ready for detection, not e.g. in the dead time after a previous detection event). It is not the same as the quantum efficiency if the latter is defined as the probability of a photon generating an electronic carrier, since the probability of a carrier creating an avalanche can be significantly below unity, depending on operation conditions.
• For some technologies (e.g. nanowires), the photon detection efficiency also depends on the polarization of light since the absorption probability depends on it.
• Calibration methods (e.g. the Klyshko two-photon technique) are used in quantum optics to measure detection efficiency.
• Dead time, recovery, and maximum count rate: After a detection, the detector is briefly unable to register a new event (dead time). Related is the recovery time back to nominal sensitivity. These limit the usable count rate and can introduce saturation or pile-up in time-resolved methods like TCSPC (time-correlated single-photon counting). Some devices use an adjustable hold-off time to mitigate afterpulsing (see below).
• Dark count rate: This is the average rate of spurious counts with no incident light. It sets the minimum detectable signal and is reduced by cooling, careful shielding from stray light, and in some cases by using a smaller active area. For array devices, the dark count rate is often quoted per pixel and per unit area.
• Timing jitter: This is the uncertainty in the registered arrival time of a photon, typically reported as FWHM or root mean square (r.m.s.). Low jitter is crucial for coincidence measurements in quantum optics, time-of-flight LIDAR and TCSPC.
• Latency: A fixed delay between photon absorption and the output pulse (e.g. tens of nanoseconds in some photomultipliers) can result from the transit time. It is relevant when matching channels in coincidence experiments.

Other often relevant device- and system-level properties:

• Afterpulsing: Additional false counts can occur after a detection, e.g. from charge traps in avalanche photodiodes. This can be mitigated by cooling, optimized materials, and hold-off times.
• Active area and input coupling: As there is a trade-off between large area (easier alignment/collection) and higher dark count rate and capacitance, the active area often needs to be minimized. Some detectors receive photons from free space (possibly through an AR-coated optical window) and/or microlenses), while others are fiber-coupled.
• For small active areas, one often uses light delivery through single-mode fibers. When light from multimode fibers needs to be detected, one can use photonic lanterns for sending it to multiple single-mode detectors.
• Optical/electrical crosstalk: Particularly in arrays (e.g. SPAD arrays, SiPMs), an avalanche in one microcell can spuriously trigger neighbors.
• Linearity and saturation: The measured count rate is ideally proportional to the true photon rate, but may be saturated in the presence of dead time; one distinguishes paralyzable and non-paralyzable detectors.
• Photon-number resolution: This is the ability to distinguish 0/1/2/… photons in a pulse. In a few detector types (e.g. TES, VLPC, MKID), this can be achieved intrinsically, otherwise via multiplexing (splitting light across time bins or multiple detectors).
• Output format and readout: Pulse shape/height, logic standard (TTL, NIM, ECL), and compatibility with time-taggers or TDCs matter for combination with other electronics.
• Operation conditions and practicality: Required bias voltage, cryogenic temperature, magnetic-field tolerance, size and power consumption are additional factors which can be relevant for applications.

Types of Single-photon Detectors

Different technologies can be used for single-photon detection:

Vacuum Tubes

Plain vacuum phototubes are not ideal for single-photon detection because single-electron photocurrents are hard to discriminate. Photomultipliers (PMTs), however, use a dynode chain for large internal gain, producing robust current pulses that are easy to threshold.

Characteristic strengths of photomultiplier tubes include high speed, large active area, low jitter (especially in microchannel-plate PMTs), and a wide choice of photocathodes covering near-IR to UV with varying efficiency and dark rates. Limitations include poor quantum efficiency in the longer-wavelength infrared, magnetic-field sensitivity, and possible ion-feedback afterpulsing.

See the article on photomultipliers for more details.

Geiger Mode Avalanche Photodiodes (SPAD)

Avalanche photodiodes (APDs) can be operated in the Geiger mode for photon counting. They are then called single-photon avalanche diodes (SPADs). Here, the applied reverse voltage is slightly above the avalanche breakdown voltage. An electron avalanche can then be triggered by a single photon, and must be stopped by lowering the voltage for a short time interval, which determines the dead time.

Depending on the operation wavelength, the quantum efficiency can be well above 50%. The dark count rate can be strongly reduced by cooling the diode, but this can increase the rate of afterpulsing caused by trapped electrical carriers.

Silicon-based SPADs are used between roughly 350 and 1050 nm and can reach dark count rates of only a few hertz. A typical r.m.s. timing jitter is some tens of picoseconds.

A special type of silicon SPADs, called visible-light photon counters (VLPCs), is used only for advanced scientific instrumentation in some research laboratories. It uses a thin gain layer of arsenic-doped silicon, compensated with an acceptor like boron, where the impurity band sits ≈50 meV below the conduction band. Such devices require cryogenic temperatures (<10 K) and provide high quantum efficiency for visible light and (due to low multiplication noise) a potential for photon number discrimination.

For longer wavelengths in the near-infrared region, which silicon detectors cannot reach efficiently, devices based on indium gallium arsenide (InGaAs) and indium phosphide (InP) or germanium (Ge) are used. Their quantum efficiency is lower than that of silicon devices in the visible spectrum, but at least higher than for infrared photomultipliers. Count rates are typically limited to a few megahertz, or more for silicon APDs.

There are also infrared APDs exploiting impurity-band conduction in heavily arsenic-doped silicon layer; they are called blocked impurity band (BIB) detectors. These are suitable for a wide wavelength range, e.g. 2 μm to 30 μm, but need cryogenic temperatures.

A special variant for infrared applications are III–V nanowire detectors, which are built from vertical (or occasionally lateral) III–V semiconductor nanowires — typically using a narrow-bandgap absorber such as InGaAs or GaAsSb, and a wider-bandgap multiplication region (e.g., GaAs or InP). Specific advantages result from such device architectures: The tiny avalanche volume reduces effects of detrimental charge traps (causing afterpulsing) and increases speed and timing accuracy through an ultralow capacitance. III–V nanowires can also be selectively grown or integrated on silicon with good material quality.

See the article on avalanche photodiodes for more details.

Silicon Photomultipliers

Silicon photomultipliers (SiPMs) are solid-state single-photon detectors made from thousands of Geiger-mode single-photon avalanche diodes (SPAD microcells). They combine single-photon sensitivity with moderate photon number resolution and fast timing in a compact, low-voltage device. They can be tiled into large arrays to reach active areas comparable to those from photomultiplier tubes (PMTs), so that they can replace those in many applications (restricted to their sensitive wavelength range of roughly 400 nm to 900 nm).

See the article on silicon photomultipliers for more details.

Hybrid Photomultipliers

Hybrid photomultipliers (see the article on photomultipliers) essentially consist of a vacuum tube with an integrated avalanche diode; they offer the combination of some beneficial features of photomultipliers and avalanche diodes, in particular a high speed, excellent pulse-height resolution (useful for limited photon-number discrimination) and more compact form than ordinary photomultipliers.

Superconducting Detectors

Different types of single-photon detectors are based on superconductor technology:

Superconducting Nanowire Single-photon Detectors (SNSPD)

Here, a meandered ultrathin nanowire (made of NbN, NbTiN, WSi, MoSi, etc.) is biased near its critical current. It momentarily becomes resistive upon photon absorption, producing a fast voltage pulse. SNSPDs offer interesting performance features:

• high photon detection efficiencies in a wide wavelength range (including at 1310/1550 nm, sometimes >80%)
• very low dark count rate (noise only from blackbody radiation)
• short dead time (10–20 ns)
• low timing jitter (tens of picoseconds)
• high count rates

Disadvantages include the need for cryogenic operation (typically below 4 K) and polarization dependence unless specially engineered.

Transition-edge Sensors (TES)

Transition-edge sensors are operated at the sharp superconducting-to-normal transition, where small temperature rises from photon absorption produce large resistance changes. The crucial feature of TES devices is that they can be photon-number resolving with high energy resolution. However, they are substantially slower than SNSPDs and require very low operating temperatures (≈100 mK).

Microwave Kinetic Inductance Detectors (MKID)

These contain superconducting resonators whose kinetic inductance changes upon photon absorption. They offer intrinsic energy resolution and large multiplexed arrays (useful in astronomy) at sub-Kelvin temperatures.

Nonlinear Upconversion Detectors

For longer wavelengths where efficient single-photon detectors are scarce, sum-frequency generation in a nonlinear crystal (often a waveguided PPLN device) converts photons to the visible, enabling detection with efficient silicon SPADs or with superconducting detectors.

Pump-induced noise (e.g. Raman backgrounds, fluorescence) must be minimized to obtain a low dark count rate.

X-ray and Gamma-ray Detectors

Mainly for fundamental research, there are various types of detectors for X-rays and γ-rays which can detect single photons, partly with energy-resolving capability:

• Scintillator-based detectors: X-ray or gamma-ray photons interact with a scintillator material (e.g. sodium iodide, cesium iodide, bismuth germanate (BGO), GOS ceramics), causing it to emit visible or UV light. This light is then detected by a photomultiplier tube (PMT), silicon photodiode, or a silicon photomultiplier. Such detectors are widely used for gamma-ray spectrometry, medical imaging (CT, PET), and astronomy.
• Semiconductor detectors: High-energy photons can interact within a semiconductor (silicon, germanium, cadmium telluride/CdTe or CZT), creating electron-hole pairs, which are separated and collected by an externally applied electric field, generating an electronic pulse. They can be made as pixel arrays and application-specific integrated circuits (ASICs) for energy and spatial resolution. High sensitivity and low noise can be achieved. Silicon is suitable for soft X-rays (<50 keV), but becomes less efficient at higher photon energies. Germanium, CdTe, CZT are used for higher energies due to their higher atomic number, giving higher stopping power.
• Gas-filled detectors: These include the well-known Geiger–Müller tubes and proportional counters (with energy resolution). Photons interact with gas atoms, producing ionization. The resulting charge is collected to produce an electrical signal.
• Calorimeter detectors: Here, the total absorbed energy of the photon heats a material, causing a tiny temperature rise measured electrically. Such detectors are used mainly in astronomy research, hardly in other fields due to high complexity.
• Compton telescopes: For MeV-range gamma rays, there are two-layer detectors, with a first layer of low-Z material (often silicon) for initial Compton scattering, and a second layer of high-Z material for photoelectric absorption. Measuring positions and energies from both events allows imaging and spectroscopy via application of the Compton formula.

Single-photon Imaging Detectors

Single photon detection is also possible for some imaging detectors:

• SPAD arrays:: Monolithic 2D arrays of SPADs with per-pixel quenching and timing electronics enable single-photon imaging and per-pixel time-of-flight (depth) or lifetime (FLIM) measurements. Their main issues are crosstalk, limited fill factor, and per-pixel dark count rate.
• Image intensifiers: Image intensifiers are used in intensified cameras, registering single photons. Some devices allow for nanosecond-scale time gating.
• EMCCD and sCMOS in photon-counting mode: While conventional CCD and CMOS sensors cannot be used on the single-photon level, essentially because of too high noise, some optimized modern variants are highly sensitive. Electron-multiplying CCD sensors can achieve effective photon detection via thresholding at low flux. Modern scientific CMOS (cCMOS) sensors offer extremely low noise, though true single-photon discrimination is more limited. For more details, see the article on image sensors.
• X-ray/γ photon counters: In the high-energy regime, semiconductor spectrometers (e.g. Si drift, HPGe, CdTe/CdZnTe) routinely register individual photons with energy resolution. In computed tomography, photon-counting detectors enable spectral imaging.

The term single-photon imaging is sometimes used, meaning imaging with single-photon detectors; of course, a single photon cannot provide an image.

Applications of Single-photon Detectors

Single photon counters are used in various areas of science and technology:

• Quantum photonics and quantum information processing: quantum cryptography (QKD), entanglement experiments, heralded single-photon sources, measurements of second-order correlation, quantum computing photonic gates, quantum teleportation, and field-deployable free-space/fiber quantum links.
• Time-resolved spectroscopy & microscopy: TCSPC for fluorescence lifetime imaging microscopy (FLIM), single-molecule spectroscopy, smFRET, Raman spectroscopy at ultra-low flux, super-resolution techniques (e.g. PALM/STORM), and biophotonics assays.
• Sensing and ranging: LIDAR (including long-range and automotive ToF), bathymetry, atmospheric sensing, and single-photon optical time-domain reflectometry (OTDR) for long, lossy fibers.
• Medical imaging: Photon detection across the spectrum — from X-ray photon-counting CT to PET scintillation readout (increasingly with SiPMs), and time-of-flight PET for better localization.
• Astronomy and space: Ultra-faint optical/UV photometry, intensity interferometry, adaptive optics wavefront sensing in low-flux regimes, single-photon deep-space laser communications, and large cryogenic arrays (TES/MKID) for ground- and space-based observatories.
• High-energy physics: Cherenkov and scintillation light readout (PMTs/SiPMs) with single-photon sensitivity, precise timing for time-of-flight systems.
• Metrology & calibration: Absolute detector calibration (e.g. Klyshko methods), radiometric standards at the single-photon level.

Frequently Asked Questions

This FAQ section was generated with AI based on the article content and has been reviewed by the article’s author (RP).

What is a single-photon detector?

A single-photon detector is a highly sensitive photodetector that can register an individual photon. It typically produces a digital output pulse for each detected photon, unlike conventional detectors that generate a continuously variable photocurrent.

What is the difference between photon detection efficiency (PDE) and quantum efficiency (QE)?

Quantum efficiency (QE) is the probability that a photon generates an electronic carrier. Photon detection efficiency (PDE) is the overall probability that an incident photon produces a registered count, which can be lower than the QE if not every generated carrier successfully triggers a detection event.

What are dark counts?

Dark counts are spurious detection events that occur even when no light is incident on the detector. The dark count rate sets the minimum detectable signal and is a crucial figure of merit, especially for low-light applications.

Can single-photon detectors count the exact number of photons in a light pulse?

Most standard single-photon detectors cannot distinguish between one and multiple photons arriving simultaneously. However, specialized photon-number-resolving detectors, such as transition-edge sensors (TES), or multiplexing schemes can provide this capability to a limited extent.

How does a photomultiplier tube (PMT) detect single photons?

In a PMT, an incident photon strikes a photocathode, releasing an electron. This electron is then accelerated through a series of electrodes (dynodes), creating a large avalanche of electrons that results in a measurable electrical pulse.

What is a single-photon avalanche diode (SPAD)?

A SPAD is an avalanche photodiode operated in Geiger mode, with a reverse bias voltage set above its breakdown voltage. A single photon-generated carrier can trigger a large, self-sustaining avalanche, which is then detected as a pulse.

What are superconducting nanowire single-photon detectors (SNSPDs)?

An SNSPD consists of an ultrathin superconducting nanowire biased near its critical current. When a photon is absorbed, it creates a small resistive hotspot, generating a fast voltage pulse that signals the detection.

What are the main advantages of SNSPDs?

SNSPDs offer a combination of high photon detection efficiency over a wide wavelength range, very low dark count rates, short dead times, and excellent timing jitter of only tens of picoseconds, though they require cryogenic cooling.

What is single-photon imaging?

Single-photon imaging uses arrays of single-photon detectors, such as SPAD arrays or image intensifiers, to build an image from individually detected photons. This allows for imaging at extremely low light levels and can be combined with timing information for 3D or lifetime imaging.

What are the main applications of single-photon detectors?

They are used in many advanced fields, including quantum photonics (e.g., quantum cryptography), time-resolved spectroscopy, LIDAR, medical imaging (e.g., PET scans), astronomy, and high-energy physics experiments.

Suppliers

Sponsored content: The RP Photonics Buyer's Guide contains 36 suppliers for single-photon detectors. Among them:

⚙ hardware

The Hamamatsu Photonics MPPC (Multi-Pixel Photon Counter) is a device called SiPM, which is a photon counting device that is a multi-pixelized Geiger mode APD. While it is an optical semiconductor device, it has an excellent detection ability, so this device can be used in a variety of applications to detect very low-level light at the photon counting level.

Hamamatsu Photonics SPAD (Single Photon Avalanche Diode) is an element with a structure of a single pixel that combines a Geiger mode APD and a quenching resistor into one set. It is an optical semiconductor element that enables photon counting.

⚙ hardware

SPAD 23 is a photon-counting array with 23 hexagonally packed single-photon avalanche diodes (SPADs) with best-in-class performance. The system software enables photon counting and time tagging and can be accessed through TCP/IP for easy integration into LabVIEW, MATLAB or Python. See our data sheet.

SPAD 512 is a camera integrating a 512×512 SPAD image sensor. Up to 100,000 fps in 1-bit mode enable high-speed imaging (photon-counting). Fine time gating enables the study of time-varying samples. See our data sheet.

SPAD Lambda is a linear detector with a 320×1 SPAD array. The detector is capable of both time gating and time stamping for the ultimate control over time-varying signals of interest. This arrangement is ideal for spectral detection applications. Thanks to microlenses and state-of-the-art production facilities, this detector offers high detection efficiency. See our data sheet.

Bibliography

(Suggest additional literature!)

Questions and Comments from Users

Here you can submit questions and comments. As far as they get accepted by the author, they will appear above this paragraph together with the author’s answer. The author will decide on acceptance based on certain criteria. Essentially, the issue must be of sufficiently broad interest.

Please do not enter personal data here. (See also our privacy declaration.) If you wish to receive personal feedback or consultancy from the author, please contact him, e.g. via e-mail.

By submitting the information, you give your consent to the potential publication of your inputs on our website according to our rules. (If you later retract your consent, we will delete those inputs.) As your inputs are first reviewed by the author, they may be published with some delay.