Single-Photon Detector Efficiency Testing: Detection Efficiency Verification Methods and Services

Single-photon detector efficiency testing verifies the detection efficiency, dark count rate, and timing performance of advanced photodetectors such as SPADs, SNSPDs, SiPMs, and PMTs used in quantum communication, lidar, and photonics research. Accredited third-party laboratories perform these measurements using calibrated reference photodiodes and attenuated photon sources traceable to national metrology institutes, in accordance with relevant IEC, ISO, and GB/T standards. This article explains the technical principles, testing items, measurement methods, and typical applications of single-photon detector efficiency verification.

Technical Overview

A single-photon detector (SPD) is a photosensitive device capable of registering the arrival of individual photons, representing the ultimate limit of optical sensitivity. Because the output signal of an SPD is inherently probabilistic, its most fundamental performance metric is the detection efficiency: the probability that a photon incident on the detector aperture generates a registered count within the measurement gate window. Detection efficiency directly determines link budgets in quantum key distribution (QKD), signal-to-noise ratio in photon-starved lidar, and sensitivity limits in fluorescence and astronomical measurements.

Efficiency verification is necessary because manufacturer datasheet values are typically specified under ideal laboratory conditions that may not reflect the device's behavior at the customer's operating wavelength, temperature, bias condition, or gating configuration. Independent efficiency testing performed by accredited third-party laboratories provides objective, metrologically traceable characterization that supports device selection, incoming inspection, research validation, and quality control in production.

Key efficiency-related quantities include the photon detection efficiency (PDE), defined for devices such as SiPMs as the product of fill factor, quantum efficiency, and avalanche trigger probability; the system detection efficiency (SDE), which includes coupling and optical losses; and the spectral detection efficiency curve, which describes efficiency as a function of wavelength. These parameters must be measured together with dark count rate (DCR), afterpulsing probability, and timing jitter, because efficiency can only be meaningfully specified at a stated DCR and overbias level.

Testing Samples

Single-Photon Avalanche Diodes (SPADs)

Discrete SPAD chips and packaged modules operating in Geiger mode, based on silicon (typically 350–900 nm), InGaAs/InP (900–1700 nm), or extended-wavelength material systems. Samples may be free-running or gated-mode devices with fiber-coupled or free-space optical interfaces.

Superconducting Nanowire Single-Photon Detectors (SNSPDs)

Cryogenically cooled nanowire detectors in closed-cycle or dip-stick cryostat configurations, covering visible to mid-infrared wavelengths. Testing includes system detection efficiency at the fiber input plane under specified bias current and operating temperature.

Silicon Photomultipliers (SiPMs)

Multi-pixel Geiger-mode APD arrays used in scintillation detection, PET, and lidar. Verification covers photon detection efficiency versus wavelength and overvoltage, along with correlated noise characterization.

Photomultiplier Tubes (PMTs) and Hybrid Detectors

Vacuum-tube photon counting devices with photocathode input windows, tested for quantum efficiency and counting plateau behavior at specified high-voltage settings.

Detector Modules and Integrated Systems

Complete detection subsystems including active quenching circuits, counting electronics, and control software, evaluated at the system level for end-to-end detection efficiency.

Testing Items

Typical efficiency verification programs include the following measurable parameters:

• Photon detection efficiency (PDE) and system detection efficiency (SDE) at specified wavelengths
• Spectral detection efficiency curve across the operating wavelength range
• Dark count rate (DCR) at nominal operating conditions
• Afterpulsing probability and correlated noise characterization
• Timing jitter (single-photon temporal resolution) and timing walk
• Dead time and maximum count rate capability
• Count rate linearity and dynamic range
• Detection efficiency dependence on bias voltage, overvoltage, gate width, and temperature
• Optical crosstalk (for SiPM multi-pixel devices)
• Background illumination and stray-light suppression performance

Testing Methods

Efficiency measurements are performed by illuminating the device under test (DUT) with a photon flux of known magnitude and comparing the registered count rate against the calculated incident photon rate. All measurements are conducted in light-tight enclosures with temperature stabilization, and laboratory competence is maintained in accordance with ISO/IEC 17025 requirements.

Calibrated Reference Photodiode Method

A continuous-wave source, typically a wavelength-tunable laser or lamp-monochromator system, is split into a reference arm and a measurement arm. The reference arm is measured with a calibrated photodiode whose responsivity is traceable to a national metrology institute, establishing the absolute photon flux. The DUT count rate, corrected for dark counts and afterpulsing, is divided by the incident photon flux to yield detection efficiency. This method is preferred for continuous-wave illumination and spectral sweeps.

Attenuated Pulsed Laser Method

A pulsed laser is attenuated with calibrated optical attenuators until the mean photon number per pulse is well below one, delivering approximately 0.1 photons per pulse to eliminate multi-photon events. The detection efficiency is computed from the ratio of detected single-shot probabilities to the known incident photon probability distribution, assumed to follow Poisson statistics. This approach is standard for gated InGaAs/InP SPADs used in fiber-based quantum communication at 1310 nm and 1550 nm.

Substitution and Comparison Methods

The DUT is alternated with a calibrated reference detector at the identical optical plane, canceling beam geometry and coupling uncertainties. This substitution technique is applied when highest accuracy is required, particularly for fiber-coupled SNSPD and SPAD modules.

Correlated Noise and Timing Characterization

Afterpulsing probability is measured from the time histogram of dark events following avalanche detections, while timing jitter is measured using a time-correlated single-photon counting (TCSPC) arrangement with a femtosecond or picosecond pulsed source referenced to an electrical sync signal. Dead time and count rate linearity are evaluated by stepping the illumination level and recording the saturation behavior of the output count rate.

Results are reported with a comprehensive measurement uncertainty budget covering source power calibration, attenuator linearity, wavelength accuracy, statistical counting error, and correction factors, in line with the guidelines of the ISO Guide to the Expression of Uncertainty in Measurement (GUM).

Applications

Efficiency verification of single-photon detectors supports a wide range of scientific and industrial fields:

• Quantum key distribution and quantum communication networks, where link budget calculations depend on verified detector efficiency at telecom wavelengths
• Quantum computing and photonic quantum information experiments requiring calibrated single-photon sensitivity
• Single-photon lidar and quantum lidar for long-range three-dimensional imaging
• Fluorescence lifetime spectroscopy, time-resolved spectroscopy, and single-molecule detection
• Positron emission tomography (PET) and other medical imaging modalities using SiPM-based scintillation readout
• Astronomical observation, adaptive optics, and satellite laser ranging
• Fiber optic distributed sensing and Brillouin/Raman scattering instrumentation
• ICCD and scientific detector production quality control, incoming inspection, and research calibration

FAQ

What is the difference between photon detection efficiency and system detection efficiency?

Photon detection efficiency (PDE) describes the probability that a photon reaching the active region generates a count, excluding external coupling losses. System detection efficiency (SDE) is measured at the detector system input, such as a fiber connector, and therefore includes coupling, filtering, and window losses. Test reports should clearly state at which plane the efficiency was measured.

Why must dark count rate be reported together with detection efficiency?

Detection efficiency can always be increased by raising the bias voltage, but this simultaneously increases dark count rate and afterpulsing. A meaningful efficiency specification therefore includes the operating point: efficiency is quoted at a stated DCR, overvoltage, temperature, and gate width.

Which wavelengths can be tested?

Typical coverage ranges from the ultraviolet through the visible and near-infrared to telecom and shortwave infrared bands, commonly from 400 nm to 1700 nm, with cryogenic SNSPD measurements extendable to longer wavelengths depending on the laboratory's source and calibration capabilities.

How is measurement traceability ensured?

Reference photodiodes and optical power meters are calibrated against national metrology institute standards, optical attenuators are step-calibrated, and every reported result includes a documented uncertainty budget. Testing is performed by accredited third-party laboratories under quality management systems conforming to ISO/IEC 17025.

What sample preparation is required before testing?

Customers typically provide the detector chip or module, its operating conditions (bias voltage, quenching circuit, gating parameters, or cryogenic temperature setpoint), target wavelengths, and any drive electronics. Free-space devices should have a defined optical aperture, while fiber-coupled devices should be terminated with a specified connector type for repeatable coupling.

How long does efficiency testing take?

Standard single-wavelength efficiency verification with DCR and jitter measurement is usually completed within a few working days. Full spectral efficiency mapping, temperature sweeps, or cryogenic SNSPD characterization requiring cool-down cycles may require additional time depending on the number of test conditions.