Dynamic bias latch-up resistant circuit for superconducting x-ray detector signal readout

Abstract

The application discloses a dynamic bias anti-latch circuit which can be used for signal reading of a superconducting X-ray detector, comprising three kinds of anti-latch circuits, namely, a direct current bias circuit, an output stage circuit and a superconducting niobium nitride nanowire detector device. The superconducting niobium nitride nanowire with a depth-width ratio exceeding 1:1 is prepared through a micro-nano processing technology, and after being packaged, the superconducting chip is assembled into the three kinds of anti-latch circuits. Circuit 1 is a gate bias circuit, a signal generator provides a gate trapezoidal bias voltage for the detector, circuit 2 is a front-end small resistance active quenching circuit, a 10Ω small resistance is connected in series at the front end of the Bias-Tee to control the circuit current, and circuit 3 is a detector parallel small resistance active quenching circuit, a 50Ω small resistance is connected in parallel at both ends of the nanowire device to discharge the superconducting current of the device. The application can solve the device latch problem caused by the excessively high photon energy in X-ray detection, realize high time precision measurement of X-rays, and can be applied to the field of X-ray detection research.

Description

TECHNICAL FIELD

[0001] The application relates to a dynamic bias anti-latch circuit capable of being used for weak signal reading of a superconducting X-ray detector and belongs to the technical field of superconducting single-photon detection. BACKGROUND

[0002] A superconducting nanowire single-photon detector (SNSPD) is a new type of single-photon detector that has arisen in the past 20 years, and was first prepared by Gol'tsman et al. on the basis of previous theories in 2001. In recent years, single-photon detection technology has shown broad application prospects in quantum secure communication, quantum computing, artificial intelligence, laser radar and other fields, making the single-photon detector a research hotspot at present. Traditional X-ray single-photon detection is mainly based on high-precision X-ray weak light detection of silicon drift (SDD) and microchannel plate (MCP), but it has the disadvantages of low time precision (1 ns) and high dark count rate (10 Hz). Therefore, in recent years, superconducting X-ray single-photon detectors have been developed at home and abroad. The superconducting X-ray single-photon detector is a high-sensitivity detector that uses a superconducting nanowire strip for photon detection. The photosensitive part of the detector is a high-aspect-ratio thin film nanowire structure. When working, the current on the nanowire is biased at a position slightly lower than the critical current. When the incident X-ray photon is absorbed by the nanowire, the superconducting state of the absorption region is temporarily destroyed and then automatically restored to the superconducting state. On the circuit, it appears as a rapid rise and then an exponential decay of an electrical pulse. By amplifying this pulse signal, a single photon can be identified.

[0003] Time precision (time jitter) is the most core technical index of the superconducting X-ray single-photon detector, which represents the offset of the time difference between the photon arrival time and the output electrical signal. For the superconducting nanowire single-photon detector, the performance of the time jitter greatly depends on the working current of the detector, which generally shows an exponential decrease with the increase of the working current. Therefore, to obtain superior time performance as much as possible, it is necessary to improve the working current of the device. However, unlike conventional infrared waves, X-rays have shorter wavelengths and contain more energy, which will cause the detector device to enter a latch state when working, seriously compressing the maximum working current of the device and leading to poor time jitter precision, which cannot meet the requirements of the detector index in the fields of X-ray pulsar navigation and biomedical imaging, and limits the development. In recent years, there has been an increasing demand for X-ray detectors with high time precision (ps), low dark count rate (mHz), short dead time and single-photon detection. SUMMARY

[0004] The application aims to solve the problem that the SNSPD is prone to entering the latch state when detecting X-rays, and provides a dynamic bias anti-latch circuit for weak signal reading of a superconducting X-ray detector.

[0005] The application provides a dynamic bias anti-latch circuit for weak signal reading of a superconducting X-ray detector.

[0006] The preparation process of the superconducting NbN nanowire detector device comprises the following steps:

[0007] The NbN nanowire detector device is made of a high aspect ratio NbN nanowire. The superconducting NbN nanowire is prepared by magnetron sputtering, electron beam exposure, and reactive ion beam etching. The detector chip grows an NbN film on a double-side polished Si3N4 / Si substrate. After optimization, a meandering NbN nanowire structure with an aspect ratio (L / S) of 80nm / 80nm is obtained at a film thickness of 100nm. The aspect ratio is more than 1:1, and the device can be used for X-ray single-photon detection. The electrode adopts a 50Ω coplanar waveguide structure to adapt to the radio frequency working frequency band of the device. The electrode wiring of the pixel device not only considers the connection between the nanowire structure and the readout circuit, but also facilitates wire welding and beam alignment. Unlike the infrared waveband superconducting nanowire detector, the device can be applied to X-ray absorption and response, and the photon energy range is 1keV-20keV.

[0008] The superconducting NbN nanowire detector device is made of a high aspect ratio NbN nanowire. The superconducting NbN nanowire is prepared by magnetron sputtering, electron beam exposure, and reactive ion beam etching. The detector chip grows an NbN film on a double-side polished Si3N4 / Si substrate. After optimization, a meandering NbN nanowire structure with an aspect ratio (L / S) of 80nm / 80nm is obtained at a film thickness of 100nm. The aspect ratio is more than 1:1, and the device can be used for X-ray single-photon detection. The electrode adopts a 50Ω coplanar waveguide structure to adapt to the radio frequency working frequency band of the device. The electrode wiring of the pixel device not only considers the connection between the nanowire structure and the readout circuit, but also facilitates wire welding and beam alignment. Unlike the infrared waveband superconducting nanowire detector, the device can be applied to X-ray absorption and response, and the photon energy range is 1keV-20keV.

[0009] The power supply device is a gate voltage signal generator, which is connected to the DC end of the bias-Tee through the resistor element.

[0010] The gate voltage signal generator is triggered by an external signal from a laser.

[0011] The power supply device is a Keithley 2400 source, and the Keithley 2400 is connected to the DC end of the biasing device through the resistor element.

[0012] The power supply device is a Keithley 2400 source, and the resistor element is two, that is, a first resistor element and a second resistor element, the Keithley 2400 source is connected to the DC end of the biasing device through the first resistor element, one end of the second resistor element is connected to the DC end of the biasing device, and the other end is grounded, and the nanowire detector device is connected in parallel with the second resistor element.

[0013] The Keithley 2400 is powered in a constant voltage source mode.

[0014] The output stage circuit includes a radio frequency amplifier (AMP) and a high-speed oscilloscope, the IN port of the radio frequency amplifier is connected to the RF end of the biasing device, and the OUT port is connected to the high-speed oscilloscope.

[0015] One end of the nanowire detector device is connected to the RF&DC end of the biasing device, and the other end is grounded.

[0016] The application provides a readout circuit design based on a signal generator and a low-value resistor, which includes three circuits, and the three circuits can effectively solve the device latch problem.

[0017] Advantages: Compared with the prior art, the application has the following remarkable advantages:

[0018] The application provides three schemes for solving the latch problem caused by the excessively high X-ray energy when the superconducting X-ray detector works, greatly improves the device working current, avoids the superfluid being compressed, and exhibits more excellent high-time-precision measurement performance. The application has the advantages of simple structure, simple readout circuit, easy preparation, easy operation, no need for special personnel training, low cost and high working efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 It is a simulation circuit electrothermal diagram of a conventional SNSPD detection circuit and detection of X-rays.

[0020] Figure 2 It is a measurement result diagram of a conventional SNSPD detection circuit and detection of device superconducting current and time jitter.

[0021] Figure 3 It is a superconducting nanowire X-ray detector device layout.

[0022] Figure 4 It is a superconducting nanowire X-ray detector application system diagram.

[0023] Figure 5 The gate bias anti-latch-up circuit diagram provided for the embodiment 1 of the present application;

[0024] Figure 6 The gate bias anti-latch-up circuit measurement result diagram provided for the embodiment 1 of the present application;

[0025] Figure 7 The active quenching anti-latch-up circuit diagram provided for the embodiment 2 of the present application with a small resistance in parallel with the detector;

[0026] Figure 8 The active quenching anti-latch-up circuit measurement result diagram provided for the embodiment 2 of the present application with a small resistance in parallel with the detector;

[0027] Figure 9 The active quenching anti-latch-up circuit diagram provided for the embodiment 3 of the present application with a small resistance in series with the front end;

[0028] Figure 10 The active quenching anti-latch-up circuit simulation circuit thermal map when detecting X-rays provided for the embodiment 3 of the present application with a small resistance in series with the front end;

[0029] Figure 11 The active quenching anti-latch-up circuit measurement result diagram provided for the embodiment 3 of the present application with a small resistance in series with the front end; DETAILED DESCRIPTION

[0030] The technical solutions of the present application will be further described below with reference to the accompanying drawings.

[0031] Preparation of a superconducting niobium nitride nanowire detector device:

[0032] A high aspect ratio superconducting niobium nitride nanowire is prepared in this embodiment, with a depth of 100 nm, an aspect ratio equal to 1:1, and a line width / interval (L / S) of 100 nm / 100 nm. The preparation method of this embodiment includes the following steps:

[0033] (1) A NbN thin film is grown on a double-side polished Si3N4 / Si / Si3N4 substrate using a direct current magnetron sputtering process, with a sputtering rate of 1.25 nm / s and a growth time of 80 s, to prepare a 100 nm thick NbN thin film. The magnetron sputtering system is a DE500 from DE Company.

[0034] The room temperature square resistance of the prepared 100 nm thick NbN thin film is measured to be 23.6 Ω, and the superconducting transition temperature Tc is measured to be 11.2 K in a liquid helium Dewar, and the superconducting transition width ΔT is 0.24 K.

[0035] (2) Drop LOR10B stripping glue on the chip, set the spin coater to run at a low speed of 300 r / min for 6 s, then increase the speed to 4000 r / min and run for 60 s. Place the chip with the spin-coated LOR10B stripping glue on a baking table at 150°C for about 5 min;

[0036] (3) Spin AZ1500 photoresist on the chip, operate similarly to (2), and place it on a baking table at 100°C for 2 min to solidify the glue;

[0037] (4) After coating the glue, install the chromium mask plate with the electrode pattern on the ultraviolet photolithography machine, place the chip on the photolithography machine tray, vacuum fix it, adjust the position of the electrode pattern and the chip, and expose it to ultraviolet light for 7 s;

[0038] (5) Wash away the AZ1500 photoresist and the corresponding area of LOR10B glue below with positive resist developer, and then rinse with deionized water. Develop for about 18 s, and then place the developed chip on a baking table at 100°C for 2 min for post-baking;

[0039] (6) Use magnetron sputtering to plate gold on the chip;

[0040] (7) After plating gold, wash away the LOR10B glue layer with positive resist developer. The gold plated on the glue will fall off the chip, leaving an Au electrode on the chip;

[0041] (8) Prepare a nanowire structure with a target line width / interval (L / S) of 100 nm / 100 nm, and design the pattern L / S to be 150 nm / 150 nm.

[0042] (9) Spin AR-P 6200.13 positive electron beam resist on the surface of the NbN film, with the spin coater speed at 64000 r / min, spin for 60 s, the resist thickness is about 400 nm, and pre-bake at 150°C for 60 s to remove the organic solvent;

[0043] (10) Use the electron beam exposure system EBPG5200 with a maximum accelerating voltage of 100 kV to perform 4 times of scanning exposure, with an exposure beam current of 100 pA and an exposure dose of 90 μC / cm 2 , 85 μC / cm 2 , 90 μC / cm 2 , 95 μC / cm 2 , and 100 μC / cm 2 ;

[0044] (11) Develop the exposed chip with AR600-546 developer at 20°C for 6 s, fix it with deionized water for 60 s, and solidify the glue at 130°C for 60 s to enhance its etch resistance;

[0045] (12) After developing, the exposed pattern is transferred to the NbN film by RIE (Samco International), and the etching discharge power is set to 80 W respectively; the etching gas is CF4, the gas flow is 30 sccm, the chamber gas pressure is set to 2 Pa respectively, and the etching time is 2 min+30 s;

[0046] (13) The residual etch resist is removed by using AR600-71 stripping solution;

[0047] (14) Then, a slicing process is performed, and a protective glue is spin-coated before slicing, and the spin-coating process parameters are selected as AZ1500 600 rmp / 2000 rmp 6 s / 40 s, the glue baking process parameters are selected as baking the glue at 100°C for 2 min, the thickness is reserved to 0.2 mm during slicing, and the slicing speed is set to 2 mm / s;

[0048] (15) The device is adhered to the metal base by using low-temperature glue, the gold electrode of the device is connected to the SMA coaxial signal output interface by wire bonding, and packaging is completed.

[0049] The NbN nanowire device with an L / S of 100 nm / 100 nm prepared in the embodiment can be applied to a high-energy single-photon detection experiment subsequently.

[0050] As known in the art, if the current recovers faster than the temperature, the current in the nanowire has started to rise and is higher than the self-heating current value before the resistance in the nanowire disappears, and the nanowire enters a self-heating state: the resistance Joule heat is balanced with the heat dissipation, and a stable resistance region is maintained. This process is called the latch-up effect. After considering the Joule heat generated by the resistance on the nanowire, the thermal model of the nanowire corresponds to the active heat conduction equation. There are two different materials in the nanowire, namely the resistance state and the superconducting state, and the related thermal parameters of the material are closely related to the state, and therefore, the equation has nonlinearity. From the perspective of the thermal model, the heat flow distribution on the object is For a point on the NbN At time t, the heat flow per unit time per unit volume out of the microelement is:

[0051]

[0052] The heat balance equation of the microelement is:

[0053]

[0054] In the above formula, The heat generation per unit time per unit volume in the microelement. ρ, c p The density and specific heat capacity at constant pressure of the microelement, respectively.

[0055] In the superconducting one-dimensional nanowire model, the heat flow is divided into two parts, one part is the heat conduction of the NbN nanowire along the length direction:

[0056]

[0057] Another part is the heat dissipation of the unit volume at the junction between the nanowire and the substrate, A is the interface surface area, v is the volume, and d is the film strip thickness:

[0058]

[0059] The Joule heat generated by the unit volume in the resistance region:

[0060]

[0061] Where J is the current density, and p is the resistivity. c

[0062] The temperature change of the nanowire strip located at x is:

[0063]

[0064] Where c is the specific heat capacity of NbN per unit volume.

[0065] Therefore, the heat equation in the one-dimensional direction is:

[0066]

[0067] From the perspective of the circuit model, the SNSPD can be equivalent to an inductance in series with a resistance, where the inductance is a dynamic inductance, and the current through the nanowire can be solved by the following equation:

[0068]

[0069] According to the above formula, for a conventional superconducting SNSPD readout circuit, the electrical and thermal state of the nanowire during operation is simulated using Comsol software, and the circuit diagram and simulation results are shown in Figure 1 . Figure 1 is a conventional SNSPD detection circuit and the electrical and thermal simulation results when detecting X-rays, wherein a is the simulation of the conventional detection circuit diagram, b is the simulation of the I n , I o , R n variation diagram, and c is the simulation of the heat distribution diagram after the nanowire loses superconductivity. Figure 1 It can be found that when the Bias-Tee inductance is 5 μH and the capacitance is 1 μF, the current I n ​​Stable time continues to maintain about 1000 muA, the temperature on line cannot be reduced to normal temperature state, the recovery time is long, I o The readout pulse signal can be read out, and it can be found that the nanowire is in a 100 ohm resistance state of continuous flux loss, and the next detection of X-ray photons cannot be performed. Figure 2 As shown in the figure, a is a superconducting nanowire device I-V graph, and b is a time jitter measurement result graph. Due to the latch problem, the highest working current of the superconducting nanowire detector device is 650 muA, the time jitter is 196 ps, the superconducting current is always serious, the time precision is low, and it can be seen that the conventional detection circuit cannot meet the detection requirement of X-rays.

[0070] In order to further reduce the recovery time, it is necessary to eliminate the latch effect, such as selecting a substrate with a higher interface heat transfer coefficient than NbN to accelerate heat dissipation, or using an active quenching circuit on the external circuit to actively reset the detector.

[0071] The application is a kind of anti-latch circuit design for weak signal readout of superconducting X-ray detector, including three kinds of anti-latch circuit schemes, including a direct current bias circuit, an output stage circuit and a superconducting niobium nitride nanowire detector device with a depth-width ratio of more than 1:1 connected in sequence.

[0072] The superconducting niobium nitride nanowire detector device is prepared on the basis of a 100 nm thick high depth-width ratio NbN nanowire process. The superconducting niobium nitride nanowire single-photon detector device includes multiple channels, and the detection area of each chip is a square, including four pixels. Figure 3 As shown in the figure, each device has a square detection area, and the photosensitive surface size is 50*50 mu m 2 In the application, the detector works in the RF frequency band, so the electrode is designed to adopt a 50 ohm coplanar waveguide structure. At the same time, the electrode wiring of the 4-pixel device needs to consider the connection between the nanowire structure and the readout circuit, and also needs to facilitate bonding and beam alignment. The electrode adopts L-EDIT software to design the electrode pattern, uses electromagnetic field theory simulation to calculate the electrode impedance, and then modifies the design according to the calculation result. Repeat the above steps to finally obtain the corresponding chip layout. The electrode preparation adopts optical exposure, electron beam evaporation and other technologies and equipment to obtain the corresponding chip electrode pattern.

[0073] (1) Test system preparation

[0074] The whole test system is shown in Figure 4 .

[0075] a. Detector installation

[0076] The superconducting nanowire device is loaded into the low-temperature refrigerator, the refrigerator is sealed, the power supply is connected, the vacuum pump group is turned on, the refrigerator is vacuumed for more than 30 minutes, and after the vacuum degree reaches 10 -3 mbar, the compressor and water chiller are started to cool down. After about 7 hours, the temperature reaches about 2.5K, the vacuum valve is closed, and the pump group is closed.

[0077] b. Preparation of femtosecond X-ray source

[0078] The power supply of the femtosecond laser is turned on and preheated for 1 hour. After focusing and leveling the target surface, the vacuum pump group is used to vacuum the target chamber and the membrane rotating chamber to below 0.05 mbar.

[0079] c. Test system preparation

[0080] Turn on the current source meter and oscilloscope and adjust to the appropriate range; connect the ultrafast diode, Bias Tee, LNA-650 radio frequency amplifier (AMP), coaxial cable, etc. in turn; turn on the stepper motor and double-axis translation table controller, turn on the membrane rotating controller, and determine that each mechanism of the target chamber and the membrane rotating chamber is operating normally.

[0081] (2) I-V curve test steps

[0082] The superconducting critical current (I c ) is a unique phenomenon of superconducting devices, which is a straight line segment with zero voltage and rising current in the I-V curve. The size of I c depends on the size of the nanowire (cross-sectional area), the quality of the film (current density j c ), and the working temperature (T). Therefore, if each pixel in the superconducting nanowire array detector has similar superfluid I c , the width, thickness, and film quality of the nanowire in each pixel of the superconducting nanowire array detector are uniform.

[0083] (3) Time jitter test steps

[0084] The three schemes of the circuit provided by the application can solve the latch problem, so that the detector can work at a higher bias current, and the power supply circuit has been introduced above. In the time jitter measurement, the three circuits can continuously provide high bias current for the detector. Next, the time jitter test steps are introduced. For a photodetector, the time resolution is the minimum time interval that can distinguish between two consecutive arriving photons. In the measurement of the angular position of a pulsar, the main measurement is the timing accuracy of the time of flight, that is, the application uses a high-speed oscilloscope to test and calibrate the time resolution.

[0085] The femtosecond laser outputs a wavelength of 800 nm femtosecond pulse laser, which is divided into two paths by a beam splitter B. One path propagates through a path BP and is output as a reference signal (synchronization signal) generated by a superfast photodetector P (superfast photodiode) to a readout device (time correlation counter or oscilloscope); the other path propagates through a path BT (including a focusing system and a film transfer mechanism), focuses the femtosecond laser on a target (copper target, aluminum target, etc.) to radiate X-rays, and the X-rays reach the superconducting detector pixel through the optical path BS1 to obtain an X-ray single-photon detection signal and output to a single-channel readout device (time correlation counter or oscilloscope).

[0086] When the time difference between the X-ray superconducting detector (SNSPD) and the superfast photodetector (PD) is used to estimate the timing accuracy of the detector, the result is affected by the timing error. The time difference measurement value contains the internal noise of the superconducting detector itself, and is also affected by the photodetector error, the target transfer error, the visible light and X-ray optical path error, and the clock bias error. Since the reference light path and the detection light path are output by the same femtosecond laser, the time delay error of the femtosecond laser can be completely eliminated. At the same time, the photodetector and the superconducting detector are read out by the same local clock, so the clock bias in the time difference can be ignored.

[0087] Test steps:

[0088] (1) Connect the four sets of amplifiers and bias-tees through SMA adapters to the corresponding devices in the pulse tube cryocooler.

[0089] (2) Turn on the power of the four sets of amplifiers and adjust the current and voltage to appropriate values.

[0090] (3) Turn on the power of the four sets of keithley 2400 source tables, adjust the protection current range and the appropriate voltage range.

[0091] (4) Set the device bias current to the corresponding working current for measuring time jitter through the source table.

[0092] (5) Four-channel signals are connected to the power combiner through high-frequency low-loss coaxial lines.

[0093] (6) The combined signals are input to the 3-channel of the oscilloscope through high-frequency stabilized coaxial lines.

[0094] (7) Part of the laser is input to the ultrafast photodiode to generate a synchronization signal through a 92:8 beam splitter film.

[0095] (8) The synchronization signal is input to the channel 1 of the oscilloscope through high-frequency coaxial lines.

[0096] (9) The time delay between the two channels is adjusted to the appropriate position through the oscilloscope.

[0097] (10) Set the appropriate trigger level of the oscilloscope, and obtain the time difference between the synchronization signal and the device response signal through the oscilloscope measurement function.

[0098] (11) The data is automatically collected by the Python script control oscilloscope, and 100000 count points are read.

[0099] (12) The time jitter σ of the superconducting nanowire detector can be obtained after processing.

[0100] (13) Turn off all equipment to ensure the safety of equipment and personnel.

[0101] Circuit 1:

[0102] As shown in Figure 5 , the embodiment is a gate bias anti-latch circuit, which includes a DC bias circuit 1, an output stage circuit 2 and a nanowire detector device 3; wherein the DC bias circuit 1 is composed of a gate voltage signal generator, a resistor element R1 and a bias tee (Bias-Tee); the gate voltage signal generator is triggered by an external signal from a laser, and after passing through the resistor R1, it is connected to the DC end of the Bias-Tee; the output stage circuit 2 is composed of an LNA-650 radio frequency amplifier (AMP) and a high-speed oscilloscope; the IN port of the AMP is connected to the RF end of the Bias-Tee, and the OUT port is connected to the high-speed oscilloscope; one end of the nanowire detector device 3 is connected to the RF&DC end of the Bias-Tee, and the other end is grounded.

[0103] The working principle of this example is:

[0104] In the actual test experiment, the R1 resistance value is 10kΩ, the gating signal is triggered by an external signal from a laser, the gating signal is adjusted to be a trapezoidal gating signal with a rising edge of 1.5μs, a falling edge of 1.5μs, a pulse width of 0.5μs and an amplitude of 5V, and the signal delay is adjusted so that the output signal of the synchronous photodetector is at the highest amplitude of the gating signal, thereby ensuring that the superconducting X-ray single-photon detector can respond to photons at the highest amplitude of the gating signal. The computer is connected to the signal generator through a GPIB communication line, and a python program is used to control the source table to gradually change the output voltage of the signal generator (first forward and then reverse scanning), and after the scanning is completed, the I-V curve of the pixel can be generated, and then other pixels are tested in turn. The time jitter test procedure is as shown above.

[0105] When the X-ray hits the nanowire and responds, a thermal island is generated on the nanowire according to the thermal island effect and gradually grows, the nanowire loses superconductivity and enters a resistance state. After the direct current voltage source is changed to a gating voltage, the current flowing through the nanowire will be forced to decrease due to the action of the gating voltage. According to the Joule effect, when the current decreases, the heat generation will also decrease. At this time, the heat dissipation speed on the nanowire is greater than the heat generation, and the thermal island gradually decreases until it returns to the superconducting state. In addition, the various modules of the circuit are connected by coaxial cables, and the output signal is processed and amplified by a radio frequency amplifier and then sampled and detected by a high-speed oscilloscope.

[0106] The measured results of the embodiment are shown in Figure 6 Figure 6 The measured results of the gating bias anti-latch circuit provided in Embodiment 1 of the present application are shown in FIG. 1, wherein a is a superconducting nanowire device I-V graph, and b is a time jitter measurement result graph. Figure 6 As can be seen from the above, the superconducting current of the device is increased from 650μA to 1650μA, and the time jitter is increased from 196ps to 21.18ps. It can be seen that the gating bias anti-latch circuit can effectively solve the device latch problem and improve the time precision performance of the device.

[0107] Circuit 2:

[0108] As shown in Figure 7 ​As shown, the example is a detector parallel small resistance active quenching anti-latching circuit, comprising a direct current bias circuit 1, an output stage circuit 2 and a nanowire detector device 3; wherein the direct current bias circuit 1 is powered by a keithley 2400 source table, and is composed of resistance elements R1, R2 and a bias tee (Bias-Tee); the keithley 2400 adopts a constant voltage source mode and is connected to the DC end of the Bias-Tee after passing through the resistance R1; one end of the resistance R2 is connected to the DC end of the Bias-Tee, and the other end is grounded; the output stage circuit 2 is composed of an LNA-650 radio frequency amplifier (AMP) and a high-speed oscilloscope; the IN port of the AMP is connected to the RF end of the Bias-Tee, and the OUT port is connected to the high-speed oscilloscope; one end of the nanowire detector device 3 is connected to the RF&DC end of the Bias-Tee, and the other end is grounded.

[0109] The working principle of the example is as follows:

[0110] In the actual test experiment, R1 has a resistance of 10kΩ, R2 has a resistance of 50Ω, and the source table adopts a constant voltage source mode. The computer is connected to the keithley 2400 source table through a GPIB communication line, and a python program is used to control the source table to gradually change the output voltage of the signal generator (first forward and then reverse scanning). After scanning is completed, the I-V curve of the pixel can be generated, and then other pixels are tested in turn. The time jitter test steps are as shown above.

[0111] When the device nanowire is not lost, the device resistance is approximately 0Ω, and the resistance of the whole loop is approximately 10kΩ. When the X-ray hits the nanowire, the nanowire loses superconductivity and is equivalent to a high resistance resistor, with a resistance of kΩ order. However, since a 50Ω small resistance is connected in parallel at both ends, the resistance of the whole loop is still approximately 10kΩ. Under the action of the constant voltage source, the overall loop current remains almost unchanged. However, according to the shunt principle, the resistance of the nanowire after losing superconductivity is much larger than the 50Ω parallel resistance, so most of the current flows through the parallel resistance to the ground, and the current flowing through the nanowire decreases rapidly, thereby reducing the generation of joule heat, so that the nanowire quickly recovers to the superconducting state. After the nanowire recovers to the superconducting state, its equivalent resistance decreases to 0, and the current returns to the normal working state, waiting for the next response. In addition, the various modules of the circuit are connected by coaxial cables, and the output signal is processed and amplified by the radio frequency amplifier, and then sampled and detected by the high-speed oscilloscope.

[0112] The measured results of the example are as shown in Figure 8 , Figure 8 The measured results of the active quenching anti-latching circuit of the detector parallel small resistance provided in Example 2 are shown in the figure, wherein a is a superconducting nanowire device I-V graph, and b is a time jitter measurement result graph. From Figure 8As can be seen, the superconducting current of the device increased from 650μA to 1650μA, and the timing jitter also improved from 196ps to 21.18ps. This shows that the gated bias anti-latch-up circuit can effectively solve the latch-up problem of the device and improve the timing accuracy performance of the device.

[0113] Circuit 3:

[0114] like Figure 9 As shown, this example is a detector front-end series small resistor active quenching anti-latch-up circuit, including DC bias circuit 1, output stage circuit 2, and nanowire detector device 3. The DC bias circuit 1 is powered by a Keithley 2400 source meter, and consists of a resistor R1 and a biaser (Bias-Tee). The Keithley 2400 uses a constant voltage source mode and is connected to the DC terminal of the Bias-Tee after passing through resistor R1. It is worth noting that R1 here is a small 10Ω resistor. The output stage circuit 2 consists of an LNA-650 RF amplifier (AMP) and a high-speed oscilloscope. The IN port of the AMP is connected to the RF terminal of the Bias-Tee, and the OUT port is connected to the high-speed oscilloscope. One end of the nanowire detector device 3 is connected to the RF and DC terminals of the Bias-Tee, and the other end is grounded.

[0115] The working principle of this example is as follows:

[0116] In the actual test experiment, the resistance of R1 was 10Ω, and the source meter was in constant voltage source mode. The computer was connected to the Keithley2400 source meter via a GPIB communication line, and the source meter was controlled by a written Python program. The output voltage of the gradient signal generator was scanned (first forward and then backward). After the scan was completed, the IV curve of the pixel was generated. Then, other pixels were tested in sequence. The time jitter test steps are as described above.

[0117] In practical applications, when the nanowire is not quenched, the device resistance is approximately 0Ω, and the resistance of the entire circuit is on the order of ohms. When X-rays strike the nanowire, after the nanowire loses quench, it is equivalent to a high-resistance resistor on the order of kΩ. Under the action of a constant voltage source, the voltage remains constant, but the resistance increases significantly. Therefore, according to Ohm's law, the current flowing through the nanowire decreases dramatically, resulting in much less Joule heat generation. At this point, heat dissipation exceeds heat generation, and the nanowire can quickly recover to the superconducting state. After recovering to the superconducting state, the resistance decreases, and the current increases, returning to the high bias operating current state, ready for the next response. This embodiment uses the electrothermal equations described above as the theoretical basis for software simulation, such as... Figure 10 As shown, Figure 10 This is a simulation circuit electrothermal diagram of the active quenching anti-latch-up circuit with a small front-end series resistor provided in Embodiment 3 of the present invention when detecting X-rays, where a is the simulated circuit I. n I o Rn The change diagram, b is the heat distribution diagram of the nanowire after losing superconducting state obtained by simulation. The inductance of the Bias-Tee circuit element used in simulation is 500nH, the capacitance is 10pF, and a small resistance of 10Ω is connected in series at the front end. It can be found that after the device loses superconducting state for 40ns, the working current on the nanowire also returns to the normal state, according to the Joule heat effect, the heat generated is also reduced, and the device quickly cools down to recover to the superconducting state. In addition, each module of the circuit is connected by a coaxial cable, and the output signal is processed and amplified by a radio frequency amplifier, and then sampled and detected by a high-speed oscilloscope.

[0118] The measured results of the embodiment are shown in Figure 11 , Figure 11 The measured results of the active quenching anti-latch-up circuit provided by the embodiment 3 are shown in the figure, wherein a is the I-V diagram of the superconducting nanowire device, and b is the time jitter measurement result diagram. Figure 11 It can be known that the superconducting current of the device is increased from 650μA to 1650μA, and the time jitter is increased from 196ps to 21.18ps. It can be seen that the gate bias anti-latch-up circuit can effectively solve the device latch problem and improve the time precision performance of the device.

[0119] The three kinds of superconducting X-ray anti-latch-up readout circuit schemes can effectively solve the problem that the superconducting nanowire device is trapped in a latch state and the working current is seriously compressed due to too large photon energy in X-ray detection. In the embodiment, three kinds of anti-latch-up circuits are used to provide working voltage and signal readout for the superconducting nanowire X-ray single-photon detector, which can effectively increase the superconducting current of the device from 650μA to 1650μA. When the working current is 650μA, the time jitter is measured to be 196ps, and when the working current is increased to 1650μA, the time jitter is measured to be about 20ps. The time precision performance of the device is greatly improved, which has important breakthrough significance in the fields of X-ray pulsar navigation, biological medical imaging and the like.

Claims

1. A dynamic bias anti-latch circuit useful for weak signal readout of a superconducting X-ray detector, characterized by, The anti-latching circuit comprises a direct current bias circuit, an output stage circuit and a superconducting niobium nitride nanowire detector device; wherein the direct current bias circuit comprises a power supply device, a resistance element and a biaser, one end of the superconducting niobium nitride nanowire detector device is connected to the RF&DC end of the biaser, and the other end is grounded, and the RF end of the biaser is connected to one end of the output stage circuit; The anti-latching circuit is one of the following three structures: Circuit 1: gate bias anti-latching circuit, the power supply device is a gate voltage signal generator, the gate voltage signal generator is connected to the DC end of the biaser through a resistance element, and the gate voltage signal generator is triggered by an external signal from a laser to supply power; Circuit 2: detector parallel small resistance active quenching anti-latching circuit, the power supply device is a keithley 2400 source, the keithley 2400 source is connected to the DC end of the biaser through a first resistance element, one end of a second resistance element is connected to the DC end of the biaser, and the other end is grounded, and the nanowire detector device is connected in parallel with the second resistance element; Circuit 3: detector front-end series small resistance active quenching anti-latching circuit, the power supply device is a keithley 2400 source, and the keithley 2400 source is connected to the DC end of the biaser through a resistance element with a resistance of 10Ω. The preparation process of the superconducting niobium nitride nanowire comprises: growing a niobium nitride film on a Si3N4 / Si substrate through a micro-nano processing process, preparing a niobium nitride nanowire with a depth-width ratio of more than 1:1, a line width / interval of ≤100nm / 100nm, and two ends connected to 50Ω coplanar waveguide electrodes.

2. The dynamic bias anti-latch circuit for superconducting X-ray detector weak signal readout according to claim 1, characterized in that, The keithley 2400 source adopts a constant voltage source mode for power supply.

3. The dynamic bias anti-latch circuit for superconducting X-ray detector weak signal readout of claim 1, wherein, The output stage circuit comprises a radio frequency amplifier and a high-speed oscilloscope.

4. The dynamic bias anti-latch circuit for superconducting X-ray detector weak signal readout of claim 3, wherein, The IN port of the radio frequency amplifier is connected to the RF end of the biaser, and the OUT port is connected to the high-speed oscilloscope.

Images