Periodic Reporting for period 3 - SPIAKID (SpectroPhotometric Imaging in Astronomy with Kinetic Inductance Detectors)
Reporting period: 2023-01-01 to 2024-06-30
The SPIAKID project, aims to build a spectro-imager based on superconducting sensors with kinetic inductance and to install it on the ESO New Technlogy Telescope (NTT) with a diameter of 3.58m located in La Silla at Chile by 2025.
These sensors, developed in the technology centre of the Pôle Instrumental of GEPI, could replace currently used devices (Charged Coupled Devices and Complementary metal–oxide–semiconductors) in many applications, in astronomy and also in other fields. Each pixel is an inductor-capacitor (LC) oscillating circuit manufactured with a supraconducting material. When cooled below the critical temperature of the material, the pixel can be set in oscillation, without any energy dissipation. In the superconductor the current is carried by couples of electrons that are quantum mechanically bound (Cooper pairs) and can move freely in the material. The arrival of a photon with energy larger than the biding energy of the Cooper pairs, breaks the bond of several pairs. This alters the inductance of the circuit resulting in a different phase, frequency and amplitude of the oscillation. By detecting this change in the oscillation properties the pixels can be operated in photon counting mode, also recording its arrival time. Furthermore, by suitably calibrating the change in the properties of the oscillation (phase, frequency, amplitude) one can deduce the energy of the detected photon. This allows to do spectrophotometry without using filters or dispersing elements. All the pixels are read simultaneously with a very high cadence, of the order of 100 μs, much smaller than the typical coherence time of the atmospheric turbulence, this opens up the possibility of achieving high angular resolution by sophisticated image reconstruction techniques.
The instrument will be developed with the primary scientific objective of studying the Ultra Faint Dwarf galaxies of the Local Group. These galaxies have an absolute luminosity below 1e5 solar luminosity. It will also address other scientific objectives such as the measurement of the redshift of quasars and galaxies, the characterization of the electromagnetic counterparts of gravitational wave sources or the characterization of small bodies in the solar system.
This demonstrator instrument has also several technological objectives: to demonstrate the maturity of this type of sensor for scientific use, the possibility of manufacturing and operating in multiplexed mode sensors of 2e4 pixels, the possibility of building mosaics with these sensors to achieve a larger field of view.
These sensors, developed in the technology centre of the Pôle Instrumental of GEPI, could replace currently used devices (Charged Coupled Devices and Complementary metal–oxide–semiconductors) in many applications, in astronomy and also in other fields. Each pixel is an inductor-capacitor (LC) oscillating circuit manufactured with a supraconducting material. When cooled below the critical temperature of the material, the pixel can be set in oscillation, without any energy dissipation. In the superconductor the current is carried by couples of electrons that are quantum mechanically bound (Cooper pairs) and can move freely in the material. The arrival of a photon with energy larger than the biding energy of the Cooper pairs, breaks the bond of several pairs. This alters the inductance of the circuit resulting in a different phase, frequency and amplitude of the oscillation. By detecting this change in the oscillation properties the pixels can be operated in photon counting mode, also recording its arrival time. Furthermore, by suitably calibrating the change in the properties of the oscillation (phase, frequency, amplitude) one can deduce the energy of the detected photon. This allows to do spectrophotometry without using filters or dispersing elements. All the pixels are read simultaneously with a very high cadence, of the order of 100 μs, much smaller than the typical coherence time of the atmospheric turbulence, this opens up the possibility of achieving high angular resolution by sophisticated image reconstruction techniques.
The instrument will be developed with the primary scientific objective of studying the Ultra Faint Dwarf galaxies of the Local Group. These galaxies have an absolute luminosity below 1e5 solar luminosity. It will also address other scientific objectives such as the measurement of the redshift of quasars and galaxies, the characterization of the electromagnetic counterparts of gravitational wave sources or the characterization of small bodies in the solar system.
This demonstrator instrument has also several technological objectives: to demonstrate the maturity of this type of sensor for scientific use, the possibility of manufacturing and operating in multiplexed mode sensors of 2e4 pixels, the possibility of building mosaics with these sensors to achieve a larger field of view.
The first task of the project was to assemble the project team, that now comprises five laboratories (three departements of Observatoire de Paris, Université PSL: GEPI, LESIA, LERMA; one of the Université de Bordeaux: LAB; one of Université Paris Cité: APC) and about 25 people. The first milestone achieved was the completion of the conceptual design of the SPIAKID instrument, that was validated by the Preliminary Design Review conducted in March 2022 by a panel composed by scientists and engineers of the European Southern Observatory and of the Institut Néel. On the basis of this conceptual design we could launch the tender and sign the contract for the procurement of the main subsystem of the SPIAKID instrument: a cryostat capable of cooling down to 100 mK and large enough to accomodate a focal plane comprising a mosaic of four detectors of 20 000 pixels each. Thanks to the optics that couples our instrument with the NTT telescope this mosaic will cover a square with a side of 2' on the sky.
In parallel to the development of the SPIAKID instrument we are developing the detectors, exploring different pixel designs and also discovering some unexpected phenomena. The capacitor in the LC circuit can be manufactured either by deposing several superconducting strips on the substrate material (e.g. sapphire), in this case we speak of interdigitated capacitor, or by manufacturing two parallel plates of superconduting material separated by an insulator, in this case we talk about parallel plate capacitor or MIM (Metal-Insulator-Metal). The parallel plate capacitor has several attractive features, mainly the possibility of manufacturing smaller pixels, for a given value of the capacity, and a simpler design. However our tests indicated that the insulator was not totally blocking the possibility of a charge flow between the two plates. For this reason we decided to develop a pixel with a parallel plate capacitor and the ultimate insulator: vacuum. This proved successful, although the manufacture of such a pixel is difficult and was in fact object of a patent. For SPIAKID we shall use the simpler interdigitated capacitors and not parallel plate vacuum pixels, that are however well adapted to instruments working in the domain of the microwaves. Yet the SPIAKID project stimulated a major progress in our understanding of the MKID devices and in their manufacture. In the attempt to increase the quantum efficiency of the detector we explored the effect of placing a reflecting surface below the inductance (Nicaise et al. 2022), we found that the transparent material (Al2O3) used to space the reflector from the inductor introduces a parasitic capacitance and increses the noise. We are going to further develop this design to try to minimise the parasitic capacitance and the noise.
During our study of the detectors we came across an unexpected new phenomenon, that manifests as large ``inverse response'' of the detector, in the sense that the frequency of the oscillator shifts towards higher frequencies, while the normal response of the MKID is to shift the frequency to lower frequencies. This phenomenon is observed only at temperatures that are about 1/100
of the critical temperature and we interpret as the interaction of phonons, produced by light absorption, with the energy levels of the substrate (Two-Level-Systems). This result was published on a refereed paper (Hu et al. 2021) and we are exploring the possibility of using this inverse response as an alternative way to achieve photon detection.
Three major milestones in the detector development have been achieved during this period: the manufacture of a detector with 20 000 pixels, the manufacture of TiN thin films with sub-stoichiometric composition displaying a critical temperature around 1 K, and the operation of a detector in photon counting mode.
In parallel to the development of the SPIAKID instrument we are developing the detectors, exploring different pixel designs and also discovering some unexpected phenomena. The capacitor in the LC circuit can be manufactured either by deposing several superconducting strips on the substrate material (e.g. sapphire), in this case we speak of interdigitated capacitor, or by manufacturing two parallel plates of superconduting material separated by an insulator, in this case we talk about parallel plate capacitor or MIM (Metal-Insulator-Metal). The parallel plate capacitor has several attractive features, mainly the possibility of manufacturing smaller pixels, for a given value of the capacity, and a simpler design. However our tests indicated that the insulator was not totally blocking the possibility of a charge flow between the two plates. For this reason we decided to develop a pixel with a parallel plate capacitor and the ultimate insulator: vacuum. This proved successful, although the manufacture of such a pixel is difficult and was in fact object of a patent. For SPIAKID we shall use the simpler interdigitated capacitors and not parallel plate vacuum pixels, that are however well adapted to instruments working in the domain of the microwaves. Yet the SPIAKID project stimulated a major progress in our understanding of the MKID devices and in their manufacture. In the attempt to increase the quantum efficiency of the detector we explored the effect of placing a reflecting surface below the inductance (Nicaise et al. 2022), we found that the transparent material (Al2O3) used to space the reflector from the inductor introduces a parasitic capacitance and increses the noise. We are going to further develop this design to try to minimise the parasitic capacitance and the noise.
During our study of the detectors we came across an unexpected new phenomenon, that manifests as large ``inverse response'' of the detector, in the sense that the frequency of the oscillator shifts towards higher frequencies, while the normal response of the MKID is to shift the frequency to lower frequencies. This phenomenon is observed only at temperatures that are about 1/100
of the critical temperature and we interpret as the interaction of phonons, produced by light absorption, with the energy levels of the substrate (Two-Level-Systems). This result was published on a refereed paper (Hu et al. 2021) and we are exploring the possibility of using this inverse response as an alternative way to achieve photon detection.
Three major milestones in the detector development have been achieved during this period: the manufacture of a detector with 20 000 pixels, the manufacture of TiN thin films with sub-stoichiometric composition displaying a critical temperature around 1 K, and the operation of a detector in photon counting mode.
The manufacture of parallel plate vacuum capacitor pixels is development that is beyond the state of the art. Also the manufacture and characterisation of pixels with a relfector is breaking unexplored ground and further investigation is necessary to see if this can improve the quantum efficiency of the detectors.
A mosaic of MKIDs has never been operated. Our design has shown that it is possible, the experimental validation of this will have to wait until the main cryostat arrives and we can integrate the detectors, the electronics and the optics. Ultimately we expect to bring the instrument to the telescope demonstrating the maturity of this technology.
A mosaic of MKIDs has never been operated. Our design has shown that it is possible, the experimental validation of this will have to wait until the main cryostat arrives and we can integrate the detectors, the electronics and the optics. Ultimately we expect to bring the instrument to the telescope demonstrating the maturity of this technology.