This thesis covers the development and the characterization of a single photon detector based on a superconducting microcalorimeter. The detector development is motivated by the Any Light Particle Search II (ALPS II) experiment at DESY in Hamburg, which searches for weakly interacting sub-eV particles (WISPs). Therefore, a detection of low-fluxes of 1064 nm light is required. The work is divided in three analyses: the characterization of a milli-kelvin (mK) cryostat, the characterization of superconducting sensors for single photon detection, and the determination of dark count rates concerning 1064 nm signals.Firstly, an adiabatic demagnetization refrigerator (ADR) is characterized, which allows to reach mK-temperatures. During commissioning, the ADR cryostat is optimized and prepared to stably cool superconducting sensors at 80 mK ± 25 μK. It is found that sensors can be continuously operated for ∼20 h before recharging the system in <2 h. Furthermore, the adiabatic system reaches a chance of success of ∼80 % for a recharge without technical problems.Secondly, superconducting sensors are analyzed. The focus is on microcalorimetric transition-edge sensors (TESs) based on 20 nm Tungsten (W) films fabricated by the U.S. National Institute of Standards and Technology (NIST). NIST TESs have a near unity detection efficiency for 1064 nm light (literature value). The energy resolution for 1064 nm signals is measured to be <8 %. The exponential falling time of a photon pulse is 1.5 μs. Furthermore, by determining TES parameters, it is found that the linear TES theory describes measured photon pulses well. The TES response is read out by a superconducting quantum interference device (SQUID) fabricated by Physikalisch-Technische Bundesanstalt (PTB). The system bandwidth is measured to be 0.9 MHz. Finally, the operation in the ADR cryostat as well as the ALPS II laboratory is optimized. This setup forms the ALPS TES detector.Thirdly, the background is measured to obtain a dark count rate for 1064 nm signals. The ALPS TES detector is calibrated by a 1064 nm single photon source and methods are developed to analyze signals. In long-term measurements, background events are measured by using different optical setups. By operating the TES without an optical link outside mK-environment, intrinsic background components are observed and classified. This results in an intrinsic dark count rate for 1064 nm signals of 1.0 · 10^-4 s^-1 . By operating a fiber-coupled TES, it is found that the dark count rate for 1064 nm signals is dominated by pile-up events of near-infrared thermal photons coming through the fiber from the warm environment. Considering a detection efficiency of ∼18 %, a dark count rate of 8.6 · 10^-3 s^-1 is determined for 1064 nm ALPS photons.Concerning ALPS II, this results in a sensitivity gain compared to the ALPS I detector. Furthermore, this thesis is the starting point of TES detector development in Hamburg, Germany.
This thesis covers the development and the characterization of a single photon detector based on a superconducting microcalorimeter. The detector development is motivated by the Any Light Particle Search II (ALPS II) experiment at DESY in Hamburg, which searches for weakly interacting sub-eV particles (WISPs). Therefore, a detection of low-fluxes of 1064 nm light is required. The work is divided in three analyses: the characterization of a milli-kelvin (mK) cryostat, the characterization of superconducting sensors for single photon detection, and the determination of dark count rates concerning 1064 nm signals.Firstly, an adiabatic demagnetization refrigerator (ADR) is characterized, which allows to reach mK-temperatures. During commissioning, the ADR cryostat is optimized and prepared to stably cool superconducting sensors at 80 mK ± 25 μK. It is found that sensors can be continuously operated for ∼20 h before recharging the system in <2 h. Furthermore, the adiabatic system reaches a chance of success of ∼80 % for a recharge without technical problems.Secondly, superconducting sensors are analyzed. The focus is on microcalorimetric transition-edge sensors (TESs) based on 20 nm Tungsten (W) films fabricated by the U.S. National Institute of Standards and Technology (NIST). NIST TESs have a near unity detection efficiency for 1064 nm light (literature value). The energy resolution for 1064 nm signals is measured to be <8 %. The exponential falling time of a photon pulse is 1.5 μs. Furthermore, by determining TES parameters, it is found that the linear TES theory describes measured photon pulses well. The TES response is read out by a superconducting quantum interference device (SQUID) fabricated by Physikalisch-Technische Bundesanstalt (PTB). The system bandwidth is measured to be 0.9 MHz. Finally, the operation in the ADR cryostat as well as the ALPS II laboratory is optimized. This setup forms the ALPS TES detector.Thirdly, the background is measured to obtain a dark count rate for 1064 nm signals. The ALPS TES detector is calibrated by a 1064 nm single photon source and methods are developed to analyze signals. In long-term measurements, background events are measured by using different optical setups. By operating the TES without an optical link outside mK-environment, intrinsic background components are observed and classified. This results in an intrinsic dark count rate for 1064 nm signals of 1.0 · 10^-4 s^-1 . By operating a fiber-coupled TES, it is found that the dark count rate for 1064 nm signals is dominated by pile-up events of near-infrared thermal photons coming through the fiber from the warm environment. Considering a detection efficiency of ∼18 %, a dark count rate of 8.6 · 10^-3 s^-1 is determined for 1064 nm ALPS photons.Concerning ALPS II, this results in a sensitivity gain compared to the ALPS I detector. Furthermore, this thesis is the starting point of TES detector development in Hamburg, Germany.
Transition-edge sensors (TESs) are among the most promising technologies for single-photon detection, offering high quantum efficiency, low dark count rates, and energy-resolving capabilities. However, optimal TES performance typically requires ultra-low temperatures below 100 mK, necessitating bulky dilution refrigerators or adiabatic demagnetization refrigerators, which limits their practical deployment. In this work, we demonstrate a titanium-based TES optimized for 1064 nm photon detection that achieves high performance at an elevated bath temperature of 220 mK—approximately 90% of its superconducting transition temperature, making it compatible with more compact sorption refrigerators. Through the integration of an optical cavity, a detailed complex impedance and noise characterization, and a systematic investigation of dark count sources, the detector achieves a system detection efficiency exceeding 97%, an energy resolution of 240 meV, and an intrinsic dark count rate as low as 2 × 10−4 cps. These findings significantly expand the applicability of TES single-photon detectors to practical photon-counting systems, such as quantum information and infrared astronomy.
Strong interests on optical quantum based metrology, quantum information and particularly in quantum cryptography are continuously growing. The main limitations to the developments in these fields are due to non-ideal devices: both single photon sources and single photon detectors. In these field of applications, detectors require to be able to resolve the number of photons in a light pulse. Presently state of the art indicates that classical semiconductor light detectors (i.e. avalanche photodiode or single photon avalanche diode) are not able to discriminate the number of photon arriving at the same time. In the meanwhile, superconducting devices have shown the possibility to resolve single photon pulses. One of the most promising superconducting detectors is the Transition-Edge Sensor (TES): a microcalorimeter that takes advantage of the sharp transition (few millikelvin) from the superconducting to the normal phase; for this reason it is sometimes called Superconductive Phase Thermometer (SPT). In the ultraviolet (UV) to infrared (IR) wavelength range, the photons are absorbed directly by the superconductive thin film and the absorbed energy induces an increase of the TES resistance. Thanks to the applied bias voltage, which maintains the device in the transition region the photon absorption induces a decrease of the TES current, measured by a dc-SQUID amplifier, and the pulse integral of the bias power reduction corresponds to the absorbed energy. This means that TESs have the very interesting properties to be able to detect single photons with an intrinsic energy resolution, without filters or gratings, that limit the quantum efficiency. By contrary of classical detectors, if monochromatic light irradiates TESs, as usually happens in communication systems, they show the photon-number resolving (PNR) capability and due to the good signal to noise ratio TESs are almost free from dark counts. Moreover, in the superconducting detector family, TESs are the only true photon-number resolving detectors operating in the VIS-NIR range. Together with quantum information science, the PNR property results useful even for optical radiometry too. In the optical community, the candela - the International System (SI) unit for the luminous intensity - has not a common consensus whether its present definition fully satisfies the current and future needs of growing associated technology. Furthermore, actually there are substantial efforts directed toward a new definitions of four base SI units: the proposal wants to link the SI units to fundamental constants, leaving f.i. material artifact. Considering the recent advances in optical radiometry and in quantum technologies, for the candela world it means redefine its unit linking to the Planck constant and consequently expressing the luminous intensity unit in terms of photon number rather than optical power. This challenge has been accepted by several national metrology institutes to demonstrate the feasibility of redefining the candela. Inside this research project called `qu-candela', the TES PNR capability has been considered to build the bridge between the quantum and classical world of radiometry: i.e. the detector possibility to measure optical powers from one single photon per second to the lower limits of cryogenic radiometry, 104 photons per second. The theme of this work of thesis is to investigate both optical and electrical characterization of different kind of TESs based on a titanium/gold multilayer film, produced and developed at the National Institute of Metrological Research (INRIM) of Torino. Thanks to the proximity effect, the multilayer allows to lower with continuity the critical temperature from that of the Ti bulk (Tc ~ 390 mK) to those of interest: ~ 300 mK and ~ 100 mK. Detectors with higher Tc have shown a faster response pulse with a relaxing time constant of the order of 200 ns, while for the lower Tc sensors, the time constant is about 10 µs. By contrary to the response time, the detector intrinsic energy resolution is proportional to its film critical temperature. Our sensors work to discriminate incident photon from UV wavelengths to those typical of the telecommunications, 1310 nm and 1550 nm. Irradiating a TES with an active area of 10x10 µm^2 by incident photons of 0.79 eV (corresponds to a wavelength 1570 nm), the best energy resolution obtained has been 0.18 eV. Detectors with higher active area 20x20 µm^2 have a worse energy resolution, because it is also proportional to the material film heat capacity. In the meanwhile due to the same reason these kind of sensors present a bigger saturation energy. This has allowed to investigate on the TES capability to discriminate up to 29 incident photons simultaneously. Until now, such count represents the bigger amount of photons discriminated by single photon detectors, without reaching the device saturation, with a linear behaviour. From this count it has been estimated 12 photons on average, per pulse, at 9 kHz repetition rate; this results in a photon flux of about 105 photons/s, demonstrating the possibility of having a detector able to work from low flux regime to 1 photon/s to flux measurable by conventional semiconductor device (f.i. single photon avalanche detector SPAD). An innovative absolute calibration technique for PNR detector has been demonstrated. The absolute technique is based on the Klyshko's efficient solution to measure detection efficiency in photon counting rate and well know for common click-no-click detector. In fact, exploiting the recent developments in quantum state world, it is possible to work with quasi single photon state, by using a parametric down conversion heralded single photon source, and calibrate PNR detectors without requiring reference standards. The best detection efficiency, of ca. 50%, has been reached by coupling the smaller active area detectors with a 9 µm core optical fiber, single mode at telecom wavelengths
Single-photon detectors operating at visible and near-infrared wavelengths with high detection efficiency and low noise are a requirement for many quantum-information applications. Superconducting transition-edge sensors (TESs) are capable of detecting visible and near-infrared light at the single-photon level and are capable of discriminating between one- and two-photon absorption events; however these capabilities place stringent design requirements on the TES heat capacity, thermometry, and optical detection efficiency. We describe the fabrication and evaluation of a fiber-coupled, photon-number-resolving TES detector optimized for absorption at 1550 and 1310 nm wavelengths. The measured system detection efficiency at 1556 nm is 95 % +/- 2 %, which to our knowledge is the highest system detection efficiency reported for a near-infrared single-photon detector.
Superconducting photon detectors have emerged as a powerful new option for detecting single photons. System detection efficiency that incorporates the quantum efficiency of the device and system losses is one of the most important single-photon detector performance metrics for quantum information applications. Superconducting transition-edge sensors (TESs) are microcalorimeters that have the ability of distinguishing single photons with negligible dark counts. In addition, TESs are capable of directly measuring the photon number in a pulse of light. We have achieved near-unity system detection efficiency with TESs at particular wavelengths in the near-infrared by using multilayer structures that enhance the absorption of light into the active device material. We describe the design of the multilayer structure enabling high detection efficiency TESs including issues and requirements for obtaining detection efficiency values higher than 99 %. We describe the device fabrication and finally, show recent results of devices optimized for high detection efficiency using the multilayer structure.
Transition-Edge Sensors (TESs) are microcalorimeters that measure the energy of incident single photons by the resistance increase of a superconducting film biased within the superconducting-to-normal transition. TES are able to detect single photons from IR to X-ray with an intrinsic energy resolution and photon-number discrimination capability. Metrology, astronomy and quantum \ncommunication are the fields where these properties can be particularly useful. In this work, we report about characterization of different TESs based on Ti films. Single photons have been detected from 200nm to 800 nm working at transition temperature Tc ∼ 100 mK. Using a pulsed laser at 690nm we have demonstrated the capability to resolve up to five photons.
There is increasing interest in using superconducting optical photon detectors in a variety of applications. These applications require detectors that have extremely low dark count rates, high count rates, and high quantum efficiency. I will describe our work on two types of superconducting detectors, the Superconducting Nanowire Single Photon Detector (SNSPD or nSSPD) and superconducting Transition-Edge Sensor (TES). An SNSPD is an ultra-thin, ultra-narrow (nm scale) superconducting meander that is current biased just below its critical current density. When one or more photon is absorbed, a hot spot is formed that causes the superconductor to develop a resistance and consequently a voltage pulse. At NIST and JPL, we have been developing nanowire detectors using an amorphous alloy of tungsten-silicide. An example of this detector is shown in Fig 1(a). In this case, there are two tungsten silicide layers separated by a dielectric layer to optimize detection of light for any polarization[1]. Fig 1(b) is an example of a TES detector with gold cooling fins to enhance timing of the detector[2]. By exploiting the sharp superconducting-to-normal resistive transition of tungsten at 100mK, TES detectors give an output signal that is proportional to the cumulative energy in an absorption event. This proportional pulse-height enables the determination of the energy absorbed by the TES and the direct conversion of sensor pulse-height into photon number. I will discuss our progress towards developing both types of detectors with quantum efficiencies approaching 100%.