RU238383U1 - Superconducting junction detector - Google Patents

Abstract

This utility model relates to instrumentation and measurement technology, specifically to cryogenic superconductor-based detectors that can be used to detect massive particles and X-ray photons, as well as to measure the energy of ultra-small signals 3-4 orders of magnitude below the X-ray range, including in highly sensitive microcalorimeters and contact thermodynamic control systems. The technical result of this utility model is the achievement of a low critical temperature by a superconducting junction detector through its suppression by a normal metal, as well as the achievement of a junction width less than 5% of the critical temperature, which ensures a high temperature sensitivity coefficient. The essence of the superconducting junction detector is that it consists of a sapphire substrate onto which thin metal films of hafnium and palladium are sequentially deposited, forming a sensitive element, as well as films forming electrodes. All films are deposited using an electron beam method. The electrodes can be produced by sequential deposition of titanium and aluminum using the electron beam method, the total thickness of which exceeds 200 nm, and the deposition of the sensitive element can be performed on a continuously cooled substrate. 2 c.p. phyl., 2 fig.

Description

The field of technology to which the utility model belongs

The utility model relates to the field of instrumentation and measurement technology, in particular to cryogenic detectors based on superconductors, which can be used both for recording massive particles and photons in the X-ray range, and for measuring the energy of ultra-small signals 3-4 orders of magnitude below the X-ray range, including in highly sensitive microcalorimeters, as well as in contact thermodynamic control systems.

State of the art

Tungsten has been considered as a basis for the production of films with a controlled critical temperature below 100 mK for quite a long time, and in this sense, tungsten-based microcalorimeters are an analogue of a utility model. The greatest progress in tungsten technology was demonstrated in the work (Roth, S., Ciemniak, C., Coppi, C. et al. Properties of Tungsten Thin Films Produced with the RF-Sputtering Technique. J Low Temp Phys 151, 216–222 (2008). https://doi.org/10.1007/s10909-007-9642-0).

Tungsten is unique in that it has several phases with distinct critical temperatures: alpha-tungsten with a transition temperature of approximately 15 mK, beta-tungsten with a transition temperature of 1 to 4 K, and gamma-tungsten with a critical temperature of up to 6 K. The difference in critical temperature is a consequence of their different crystallographic structures. Alpha-tungsten, with a critical temperature of 12-15 mK, is the most attractive for creating TES microcalorimeters. However, its limited use is due to the lack of cryostats with the appropriate temperature, as optimal TES operation requires a base temperature half that of the critical temperature, while commercial dilution cryostats rarely achieve temperatures below 10 mK. Therefore, pure alpha-tungsten is rarely used for TES. Increasing the critical temperature of alpha-tungsten requires the development of new technologies, such as those with beta-tungsten impurities. A disadvantage of this approach is poor reproducibility, meaning that a technology developed in one laboratory cannot be replicated elsewhere. For example, in the cited study, critical temperatures of the studied structures using alpha-tungsten were obtained in the range of 25 to 55 mK, and the superconducting transition width ranged from 1.5 to 4 mK. In the study (Abdelhameed, A.H., Bakhlanov, S.V., Bauer, P. et al. New limits on the resonant absorption of solar axions obtained with a Tm-containing cryogenic detector. Eur. Phys. J. C 80, 2020, p. 376), the critical temperature of tungsten films was stabilized at 23 mK. In the work (S. J. Hart, M. Pyle, J. J. Yen, B. A. Young, P. L. Brink, B. Cabrera, M. Cherry, N. Mirabolfathi, B. Sadoulet, D. Seitz, K. Sundqvist, and A. Tomada, Phase Separation in Tungsten Transition Edge Sensors, AIP Conference Proceedings 1185, 2009, p. 215), a technology was developed that makes it possible to obtain tungsten films for TES with a critical temperature of about 100 mK. In each case, the development of the technology took several years. An additional difficulty in obtaining stable tungsten films is that different modifications of the crystallographic structure not only have significantly different critical temperatures, but can also transform into each other during cooling and heating cycles of the films.

The second analogue is TES microcalorimeters made of iridium films (R. Hennings-Yeomans, C. L. Chang, J. Ding, A. Drobizhev, B. K. Fujikawa, S. Han, G. Karapetrov, Yu. G. Kolomensky, V. Novosad, T. O’Donnell, J. L. Ouellet, J. Pearson, T. Polakovic, D. Reggio, B. Schmidt, B. Sheff, V. Singh, R. J. Smith, G. Wang, B. Welliver, V. G. Yefremenko, J. Zhang; Controlling Tc of iridium films using the proximity effect. J. Appl. Phys. 21 October 2020; 128 (15): 154501. https://doi.org/10.1063/5.0018564). The disadvantages of this technical solution, as with tungsten, include the complexity of the technology, due to iridium's demanding deposition conditions. For iridium to superconduct, the substrate must be heated to temperatures of 600-800ºC during deposition. Furthermore, the critical temperature of pure iridium is quite high for the applications under discussion—110-130 mK. This, in turn, requires the deposition of additional layers of normal metals, such as gold-iridium-gold or iridium-platinum and iridium-palladium. The critical temperature of iridium in gold-iridium-gold trilayers ranges from 20 to 100 mK. The same result can be achieved with iridium-platinum bilayers. The superconducting transition width in this case is approximately 2 mK. Considering that iridium films are only superconducting at thicknesses greater than 100 nm, the total thickness of trilayers or bilayers exceeds 200 nm, increasing the volume of the sensing element and thereby reducing its sensitivity. Other disadvantages include the high cost and limited availability of iridium.

A more accessible and simpler material from a technological point of view is hafnium. A superconducting RFTES detector based on hafnium was proposed (patents RU2801920C1 and RU2801961C1) and implemented (A. V. Merenkov, V. I. Chichkov, A. B. Ermakov, A. V. Ustinov, S. V. Shitov; Superconducting RFTES Detector at MilliKelvin Temperatures. IEEE Trans. Appl. Supercond. 17 April 2018; 28 (7): 2100305. https://doi.org/10.1109/TASC.2018.2827981), consisting of a hafnium microbridge integrated into an RF range antenna. Such a detector can also be considered an analogue of a utility model, despite significant differences. The similarity between this analog and the claimed utility model lies in the use of hafnium as the sensing element material. The main difference is that the RFTES detector requires an antenna to heat a 2.5 x 2.5 µm, 50 nm thick hafnium microbridge. The antenna is made of a 100 nm thick niobium film. The published solution is designed to read the useful response not at a DC current, but in the 1.5 GHz frequency range. In this mode, it operates as a bolometer, in which the received signal changes the bridge impedance by heating the electronic subsystem. However, operating such a device requires rather complex and expensive microwave amplifiers, which is a drawback.

The second drawback of this analogue is related to a different hafnium deposition technology. In the cited work, hafnium is deposited using magnetron sputtering, which results in a coarse-grained structure and an elevated critical temperature of up to 380 mK.

There are several functional prototypes of the utility model, among which the closest prototype is a TES made of iridium and gold (Godehard Angloher, Michael Altmann, Matthias Buehler, Franz von Feilitzsch, Theo Hertrich, Paul Hettl, Jens Hoehne, Michael Huber, Josef Jochum, Rudolf Moessbauer, Johann Schnagl, Stefanie Waenninger, "Cryogenic microcalorimeters for high-resolution energy-dispersive x-ray spectrometry," Proc. SPIE 3765, EUV, X-Ray, and Gamma-Ray Instrumentation for Astronomy X, 22 October 1999; https://doi.org/10.1117/12.366543). This prototype is a square structure consisting of a thin film of iridium and gold deposited on a sapphire substrate using a specialized ultra-high vacuum setup. The total film thickness is 150 nm. The connecting lines for measuring the thin film are made of aluminum. The resulting structure has a critical temperature of 40 mK and a superconducting transition width of less than 2 mK.

The prototype's drawbacks include the transition width, which is 5% of the critical temperature of 40 mK, as well as the complexity of the iridium technology, discussed in the description of the analog. Furthermore, the prototype differs structurally from the proposed solution in that it includes a 1-μm-thick gold absorber deposited on top of the Ir/Au film. This absorber increases the microcalorimeter's heat capacity, although its area is significantly smaller than that of the heat-sensitive film. The absorber is necessary for applications requiring the absorption of X-rays with energies of 6 keV and higher. For detection of lower energies, this microcalorimeter configuration does not enhance detection, but rather complicates the technology and reduces the microcalorimeter's resolution due to the increased combined heat capacity of the heat-sensitive film and absorber. Furthermore, the presence of a second, thicker gold layer above the first thin layer locally lowers the critical temperature of the iridium film, creating nonuniformity in the critical temperature and thereby increasing the width of the superconducting transition. Therefore, an alternative technology without the above-mentioned limitations is needed.

Disclosure of the essence of the utility model

The objective of the utility model is to develop a technology for manufacturing a detector based on a superconducting junction with a high temperature sensitivity coefficient. The technology should ensure highly predictable and stable characteristics while being accessible and cost-effective.

The technical result of the utility model is the achievement of a low critical temperature of a detector based on a superconducting transition by suppressing it with a normal metal, as well as obtaining a transition width of less than 5% of the critical temperature, which ensures a high temperature sensitivity coefficient.

An additional technical result is ensuring the repeatability of the parameters of manufactured samples, as well as their stability during thermal cycling and long-term storage.

The said technical result is achieved due to the fact that in a detector based on a superconducting junction, consisting of a sapphire substrate onto which thin metal films are successively deposited, forming a sensitive element and electrodes, what is new is that the sensitive element, the thickness of which is less than 105 nm, consists of hafnium as the first film, and palladium as the second, the deposition of which is carried out using an electron beam method.

The electrodes can be produced by sequential deposition of titanium and aluminum using the electron beam method, the total thickness of which exceeds 200 nm, and the deposition of the sensitive element can be performed on a continuously cooled substrate.

Hafnium is used as the first layer due to the fact that it is a superconductor with a sufficiently low critical temperature of 130 mK, which does not require layers of normal metals with thicknesses several times greater than the thickness of the base superconductor to suppress it.

Using palladium as a second layer allows us to suppress the critical temperature of hafnium below 100 mK and create a non-oxidized surface for subsequent electrode deposition. An advantage of palladium over other noble metals is that it has the highest density of electron states at the Fermi level, requiring a thinner layer thickness to achieve the same critical temperature suppression effect.

The total thickness of the sensitive element, not exceeding 105 nm, ensures low film roughness and homogeneity of its properties, which ultimately ensures a narrow superconducting transition width.

Using the electron beam deposition method, films with an amorphous structure are grown, which in the case of hafnium helps to achieve a technical result due to the fact that amorphous films have a lower critical temperature compared to crystalline ones.

All distinctive features taken together are sufficient to achieve the specified technical result.

Brief description of the drawings

Fig. 1 shows a general view of the detector and a diagram of the layers with the relative scale maintained.

The following elements are designated by numbers: 1 - a two-layer sensitive element made of a superconductor and a normal metal, 2 - aluminum electrodes, 3 - a sapphire/silicon substrate, 4 - the first layer of the sensitive element (Hf), 5 - the second layer of the sensitive element (Pd), 6 - a titanium sublayer under the aluminum electrodes to improve adhesion.

Fig. 2 shows the results of measured resistances of detectors on sapphire substrates as a function of temperature, demonstrating a range of critical temperatures and transition widths less than 5% of the critical temperature for detectors from three different chips:

Chip 1

200 µm - superconducting transition width 1.8% (0.8 mK) of the critical temperature 43.7 mK;

Chip 1

500 µm - superconducting transition width 0.2% (0.1 mK) of the critical temperature 45.95 mK;

Chip 2

500 µm - superconducting transition width 4.5% (2.6 mK) of the critical temperature 57.5 mK;

Chip 2

1000 µm - superconducting transition width 5.0% (3 mK) of the critical temperature 59.6 mK;

Chip 3

100 µm (triangles) - superconducting transition width 0.7% (0.3 mK) of the critical temperature 42.7 mK;

Chip 3

100 µm (circles) – superconducting transition width 1.8% (0.8 mK) of the critical temperature 43.5 mK;

Chip 3

500 µm – the width of the superconducting transition is 0.8% (0.4 mK) of the critical temperature of 48.2 mK.

Implementation of a utility model

Modern superconducting microcalorimeter technologies utilize superconducting films with a sub-Kelvin critical temperature. Transition-edge-of-superconducting (TES) microcalorimeters have achieved high energy resolutions for detecting signals with energies on the order of kiloelectronvolts (keV). However, detecting signals of one electronvolt (eV) and below remains out of reach, requiring increased sensitivity and reduced device noise.

In general, the energy resolution TES of a microcalorimeter depends mainly on three parameters: heat capacity C, the logarithmic derivative of resistance with respect to temperature α and the temperature TES T ₀ :

Thus, increasing the resolving power of a TES microcalorimeter can be achieved by reducing its size, sharpening the resistance-temperature dependence R(T), and lowering the operating temperature _T0 , which is also the critical temperature of the TES. The most effective method is to lower _T0 . However, as calculations show, lowering the critical temperature alone is insufficient to achieve energy resolutions below 1 eV. To achieve this, all three of these parameters must be optimized.

The main limitation that has prevented microcalorimeters with the required parameters from being developed so far stems from the difficulty of controlling the parameters of superconducting films with low critical temperatures. Film production technologies for the most popular materials, such as tungsten and iridium, face serious limitations in the temperature range below 100 mK.

An example of a specific embodiment of the claimed utility model is the detector shown in Fig. 1, which is a sensitive element consisting of a two-layer thin-film structure with hafnium and palladium layers of 100 and 5 nm thickness, respectively, and two 200 nm thick aluminum electrodes, providing the device with a critical temperature of approximately 40 mK. The detector is fabricated on a sapphire substrate, which has good thermal conductivity and helps prevent overheating.

Before the spraying process, the substrate surface is pre-cleaned with argon plasma, which removes organic contaminants and oxide layers, thereby improving the adhesion of the applied materials. Using electron beam evaporation, four layers are deposited sequentially with the following parameters:

First layer: material - hafnium, thickness - 100 nm, pressure in the working chamber - 5.9·10 ^-8 mm Hg, deposition rate - 2 Å/s, substrate temperature - 0°C, pre-sputtering - 3 min;

Second layer: material - palladium, thickness - 5 nm, pressure in the working chamber - 2.8 10 ^-8 mm Hg, deposition rate - 0.2 Å/s, substrate temperature - -5°C, pre-sputtering - 20 s.;

Third layer: material - titanium, thickness - 5 nm, pressure in the working chamber - 8.5 10 ^-8 mm Hg, deposition rate - 0.5 Å/s, substrate temperature - 21°C, pre-sputtering - 2 min;

Fourth layer: material - aluminum, thickness - 200 nm, pressure in the working chamber -5.7·10 ^-8 mm Hg, deposition rate - 2 Å/s, substrate temperature - 21°C, pre-sputtering - 3 min.

Cooling the substrate during hafnium and palladium deposition is not mandatory. However, experiments have shown that when depositing these materials through a resist mask, cooling the substrate simplifies the post-deposition resist explosion process, which can change its properties and deform due to the high temperatures of the deposited metals.

The operation of the utility model is carried out as follows.

A constant voltage is applied to the detector. When external energy is applied to the sensitive element, which consists of a bilayer of hafnium and palladium, its temperature increases. A small change in temperature causes a sharp change in resistance, which is ultimately recorded as a change in current through the detector, which can be read, for example, by a DC superconducting quantum interferometer (SQUID).

The key feature of the utility model is the use of hafnium, a material previously unused for similar applications. Hafnium's melting point (2227°C) is lower than that of both materials used in similar devices and the prototype: iridium's melting point is 2446°C, while tungsten's melting point is 3422°C. This melting point allows hafnium to be evaporated in an electron-beam deposition system, while tungsten deposition requires only a magnetron sputtering system. Iridium is also deposited primarily using a magnetron, although electron-beam deposition has been studied. Electron-beam deposition systems differ from magnetron deposition systems in that they have smaller targets and require lower material consumption, reducing sample fabrication costs. Hafnium can be deposited using either a magnetron or an electron beam. However, electron beam deposition allows the creation of hafnium structures with an amorphous surface and a lower critical temperature (130 mK), while magnetron sputtering leads to an increase in surface roughness and the formation of large crystallites [A. I. Mardare, C. M. Siket, A. Gavrilović-Wohlmuther, C. Kleber, S. Bauer, A. W. Hassel, “Anodization Behavior of Glassy Metallic Hafnium Thin Films”, Journal of The Electrochemical Society, v. 162, 2015, pp. E30–E36], which increases the critical temperature to 300 mK. In addition, the hafnium surface deposited by electron beam is less susceptible to oxidation, which ensures the stability of superconducting properties over time.

The proposed technology for manufacturing thermistor detectors based on a bilayer film of hafnium and palladium enables low critical temperatures and a high temperature sensitivity coefficient. Electron beam deposition significantly simplifies the process and makes the technology more cost-effective than prototypes and similar materials using iridium and tungsten. This solution also boasts high parameter repeatability and stability during thermal cycling.

Claims (3)

1. A detector based on a superconducting junction, consisting of a sapphire substrate onto which thin metal films are successively deposited, forming a sensitive element and electrodes, characterized in that the sensitive element, the thickness of which is less than 105 nm, consists of hafnium as the first film, and palladium as the second, the deposition of which is carried out using an electron beam method. 2. A detector based on a superconducting junction according to paragraph 1, characterized in that the electrodes are made by sequential deposition of titanium and aluminum using an electron beam method, the total thickness of which exceeds 200 nm. 3. A detector based on a superconducting junction according to paragraph 1, characterized in that the deposition of the sensitive element is carried out on a constantly cooled substrate.