Light is widely used for communications, carrying phone conversations and video signals through fiber optic cables around the world in pulses composed of many photons. Single photons are the weakest signal that can be transmitted and they have a wide range of applications including implementing an encryption technique called quantum key distribution and creating high-resolution maps. Single photons of light typically have very little energy, making them difficult to detect.
Light sensors provide detection of photons. Some sensors are optimized for narrow-band operation to detect light in a narrow wavelength region. Other sensors detect light over a broad wavelength range. Typical sensors detect multiple photons and operate continuously by biasing the sensor to detect all incident photons at the sensor within a wavelength range that the sensor can detect, with a maximum detection limit at a saturation value or a damage threshold. Many sensors cannot detect single photons. To fill this gap, specialized sensors were developed to detect single photons but many of them detect a limited number of single-photon events or fail to detect random arrival of a single photon at the specialized sensor.
A common type of detector based on indium-gallium-arsenide semiconductors has been catching individual photons for years and it is widely used in quantum cryptography research because it can detect photons at the particular wavelengths (colors of light) that travel through fiber. Unfortunately, when the detector receives a photon and outputs a signal, sometimes an echo of electronic noise is induced within the detector. Traditionally, to reduce the chances of this happening, the detector must be disabled for some time after each detection, limiting how often it can detect photons. Recent research has shown that the noisy echo can also be suppressed by activating the detector only for very short times, effectively creating a “gate” in the detector that only opens briefly to accept signals. Unfortunately, this also means that the photons must arrive only during those short intervals.
Individual photons of light now can be detected far more efficiently using a device that overcomes longstanding limitations with one of the most commonly used types of single-photon detectors. The method detects the photons that arrive when the gates are either open or closed. The method is a highly sensitive way to read tiny signals from the detector, and is based on electronic interferometry, or the combining of waves such that they cancel each other out.
The approach allows readout of tiny signals even when the voltage pulses that open the gate are large; these large pulses allow the detector to be operated in a new way. The pulses turn on the detector during the gate as usual. But in between gate openings, the pulses turn the detector off so well that signals produced by absorbing a photon can linger for a while in the device. Then the next time the gate opens, these lingering signals can be amplified and read out.
The new detector can count individual photons at a very high maximum rate — several hundred million per second — and at higher than normal efficiency while maintaining low noise. Its efficiency is at least 50 percent for photons in the near infrared — the standard wavelength range used in telecommunications. Commercial detectors operate with only 20 to 30 percent efficiency.
The device consists of a multiplication region, a photon absorption region, a punch-through voltage range, a breakdown voltage, and a source in electrical communication with the photon detector that provides the primary waveform. This source comprises a first voltage that is less than a maximum value of the punch-through voltage range, or effective to maintain a charge carrier in the absorption region; a second voltage that is greater than the breakdown voltage; and a reference member in electrical communication with the source that provides a reference waveform in response to receiving the primary waveform.
The added ability to detect photons that arrive when the gate is closed increases the detector’s efficiency, an improvement that would be particularly beneficial in applications in which photons could arrive at any moment, such as atmospheric scanning and topographic mapping.