Flux qubit readout of TRANSMON qubits

Abstract

Embodiments of the present disclosure relate to flux qubit readout of TRANSMON qubits, and provide a method comprising: providing a data qubit and a flux qubit for measuring a state of the data qubit; exciting the data quantum bits to an excited state; biasing the flux qubits into a single well potential energy configuration; tuning the flux qubits such that photons from the excited state of the data qubits are transferred to the flux qubits; biasing the flux qubits containing the transferred photons into a dual well potential energy configuration; and raising a potential barrier between the first well and the second well of the dual-well potential energy configuration, where either the first well or the second well includes the transferred photons, and where the raised potential well prevents the transferred photons from leaking into adjacent wells of the dual-well potential energy configuration.

Description

[0001] This application is a divisional application of the invention patent application with the application date of November 27, 2019, application number 201980102613.7, and invention name “Flux Quantum Bit Readout of TRANSMON Quantum Bits”. Technical Field

[0002] This topic concerns the readout scheme of transmon qubits. Background Art

[0003] Large-scale quantum computers have the potential to provide rapid solutions to certain classes of difficult problems. Multiple challenges in the design and implementation of quantum architectures to control, program, and maintain quantum hardware have hindered the realization of large-scale quantum computing. Summary of the invention

[0004] This disclosure describes techniques for implementing a readout scheme for transmon qubits.

[0005] In general, one innovative aspect of the disclosed subject matter can be embodied in a detector for reading out a state of a quantum bit, the detector comprising a flux qubit and a flux bias generator, wherein the flux qubit comprises an inductor, a transistor comprising at least one Josephson junction, A method of manufacturing a quantum computer system comprising: providing ...

[0006] The foregoing and other implementations may each optionally include one or more of the following features, alone or in combination.

[0007] In some implementations, the detector further comprises a measurement unit configured to determine whether the flux qubit is in the first flux state or the second flux state and to output a signal depending on whether the flux qubit is in the first flux state or the second flux state.

[0008] In some implementations, the flux bias generator is configured to: generate a first value of a first flux bias such that the energies of a first flux state and a second flux state of a flux qubit are substantially the same; generate a first value of a second flux bias such that a potential barrier between the first flux state and the second flux state is minimized and a resonant frequency of the flux qubit is tuned to a frequency of interaction such that the flux qubit is coupled to a data qubit and the state of the data qubit is mapped to an energy state of the flux qubit; generate a second value of the first flux bias such that the energies of the first flux state and the second flux state of the flux qubit are different; and generate a second value of the second flux bias such that the flux qubit is decoupled from the qubit and the energy state of the flux qubit is mapped to a superposition of the first flux state or the second flux state.

[0009] In some implementations, in response to the first value of the second flux bias, the flux qubit is configured to couple to the qubit by tuning a resonant frequency of the flux qubit to resonate with the resonant frequency of the qubit.

[0010] In some implementations, in response to the second value of the second flux bias, a resonant frequency of the flux qubit differs from a resonant frequency of the qubit by more than 2 GHz.

[0011] In some implementations, the measurement unit includes a signal generator, a transmission line, and a power detector. The flux qubit is connected to the transmission line via a shunt line. The signal generator is configured to send a traveling wave to the power detector via the flux qubit through the transmission line. The measurement unit is configured to determine whether the flux qubit is in a first flux state or a second flux state based on an output of the power detector.

[0012] In some implementations, the measurement unit does not include a circulator, a parametric amplifier, and a high electron mobility transistor (HEMT).

[0013] In some implementations, the measurement unit includes a single flux quantum SFQ circuit arranged to measure a flux generated by a flux qubit and a discriminator. The discriminator is configured to determine whether the flux qubit is in a first flux state or in a second flux state based on an output of the single flux quantum SFQ circuit.

[0014] In some implementations, the capacitance of the capacitor is between 10 fF and 100 fF.

[0015] In some implementations, the area occupied by the SQUID ring is on the order of 1 μm 2 Up to 100 μm 2 between.

[0016] In some implementations, the flux qubit is arranged to form a barrier between the first flux state and the second flux state in response to the first value of the second flux bias such that tunneling between the first flux state and the second flux state is reduced.

[0017] In some implementations, the flux qubit is arranged such that a difference in energy of the first flux state and the second flux state is generated in response to the second value of the first flux bias.

[0018] In some implementations, the flux bias generator includes a current source configured to generate a current and a transducer arranged to convert the current into a magnetic field. The transducer is arranged such that the first flux bias and the second flux bias are provided by the magnetic field.

[0019] In some implementations, the transducer includes a first coil for generating a first flux bias and a second coil for generating a second flux bias.

[0020] In some implementations, the inductor includes a first gradient metric coil, and the first coil includes a second gradient metric coil, and the first gradient metric coil and the second gradient metric coil are configured such that the first flux bias is primarily coupled to the inductor and the second flux bias is reduced coupling from the second coil to the inductor.

[0021] Another innovative aspect of the disclosed subject matter can be embodied in a method of reading out a state of a data qubit, comprising providing a flux qubit, the flux qubit comprising an inductor, a SQUID ring comprising at least one Josephson junction, and a capacitor. The inductor, the at least one Josephson junction, and the capacitor are connected in parallel to each other, and the flux qubit is arranged to exhibit a first flux state and a second flux state. The method also includes applying a first value of a first flux bias through the flux qubit such that the energies of the first flux state and the second flux state of the flux qubit are substantially the same, applying a first value of a second flux bias through the SQUID ring such that a barrier between the first flux state and the second flux state is minimized and a resonant frequency of the flux qubit is tuned to the frequency of the interaction, tuning the resonant frequency of the data qubit to the frequency of the interaction such that the flux qubit is coupled to the data qubit and the state of the data qubit is mapped to an energy state of the flux qubit, applying a second value of the first flux bias such that the energies of the first flux state and the second flux state of the flux qubit are different, and applying a second value of the second flux bias such that the flux qubit is decoupled from the data qubit and the energy state of the flux qubit is mapped to a superposition of the first flux state or the second flux state.

[0022] The foregoing and other implementations may each optionally include one or more of the following features, alone or in combination.

[0023] In some implementations, the method also includes determining whether the flux qubit is in a first flux state or a second flux state, and outputting a signal based on whether the flux qubit is in the first flux state or the second flux state.

[0024] In some implementations, a first time interval between generating a first value for the second flux bias and generating a second value for the second flux bias is determined based on the degree of interaction such that the state of the qubit is fully mapped to the flux qubit.

[0025] In some implementations, the method further includes providing a data qubit and a measurement qubit for measuring a state of the data qubit, exciting the data qubit to an excited state, biasing the measurement qubit into a single well potential energy configuration, tuning the measurement qubit so that a photon from the excited state of the data qubit is transferred to the measurement qubit, biasing the measurement qubit containing the transferred photon into a double well potential energy configuration, and raising a potential barrier between a first well and a second well of the double well potential energy configuration. The first well or the second well includes the transferred photon. The raised potential well prevents the transferred photon from leaking into an adjacent well of the double well potential energy configuration.

[0026] In some implementations, tuning the measurement qubit so that photons from an excited state of a data qubit are transferred to the measurement qubit includes tuning the measurement qubit to resonate with the data qubit in an excited state.

[0027] In some implementations, biasing a measurement qubit containing a transferred photon into a double-well potential configuration includes tilting a potential curve of the measurement qubit so that an energy state of the measurement qubit containing the transferred photon is mapped to a first well and a second well of the double-well potential configuration.

[0028] In some implementations, the method also includes reading out the energy state of the measurement qubit.

[0029] In some implementations, reading out the energy state of the measurement qubit includes applying microwave reflectometry to the measurement qubit.

[0030] In some implementations, reading out an energy state of a measurement qubit includes: reading out a flux difference between a first energy state and a second energy state of the measurement qubit.

[0031] In some implementations, reading out the flux difference is performed using a single flux quantum (SFQ) measurement of the flux difference.

[0032] In some implementations, the data qubits are transmon qubits.

[0033] In some implementations, the measurement qubit is a flux qubit.

[0034] In some implementations, the data qubits are on a first substrate and the measurement qubits are on a second substrate bonded to the first substrate.

[0035] Another innovative aspect of the disclosed subject matter can be embodied in a method that includes performing a quantum computing operation on a qubit to place the qubit in a first excited state of two energy states within a single-well potential energy configuration, biasing the qubit so that the two energy states are respectively mapped to two wells of a double-well potential energy configuration, and raising a potential barrier between a first well and a second well of the double-well potential energy configuration. The first excited state is mapped to the first well or the second well, and the raised potential well prevents the excited state from leaking into an adjacent well of the double-well potential energy configuration.

[0036] The foregoing and other implementations may each optionally include one or more of the following features, alone or in combination.

[0037] In some implementations, the method includes determining an excited state of the qubit using microwave reflectometry.

[0038] In some implementations, the method includes determining an excited state of a qubit using a SFQ circuit.

[0039] In some implementations, an array for performing quantum computing using a quantum error correction algorithm is provided. The array includes a plurality of qubits, and a plurality of detectors according to the aforementioned implementation of a detector for reading out the state of the qubit. The array is arranged so that each qubit has at least one detector as a nearest neighbor. According to surface code quantum computing as error correction, a plurality of qubits are used as data qubits, and a plurality of detectors are used as auxiliary qubits. The flux qubit of each detector is an auxiliary qubit for a quantum error correction algorithm.

[0040] In some implementations, a plurality of qubits are disposed on a first substrate and a plurality of detectors are disposed on a second substrate. Flux qubits of the plurality of detectors are capacitively coupled to the qubits via a vacuum gap formed between the first substrate and the second substrate.

[0041] In some implementations, the plurality of qubits are transmon qubits.

[0042] Another innovative aspect of the disclosed subject matter can be embodied in a method comprising determining bias conditions for a single well configuration and a double well configuration for a flux qubit such that the flux qubit is at an interaction frequency in the single well configuration. At the interaction frequency, the flux qubit resonates with a data qubit. The method further comprises determining a first bias condition for the data qubit for the interaction frequency, determining a second bias condition for the data qubit for a frequency away from the interaction frequency, applying a microwave pulse to the data qubit to prepare a state and tune to the interaction frequency, and measuring a state of the flux qubit to read out the state of the data qubit.

[0043] Another innovative aspect of the disclosed subject matter can be embodied in a method comprising: providing a data qubit and a flux qubit for measuring a state of the data qubit; exciting the data qubit to an excited state; biasing the flux qubit into a single-well potential energy configuration; tuning the flux qubit so that photons from the excited state of the data qubit are transferred to the flux qubit; biasing the flux qubit containing the transferred photons into a double-well potential energy configuration; and raising a potential barrier between a first well and a second well of the double-well potential energy configuration, wherein the first well or the second well includes the transferred photons, and wherein the raised potential well prevents the transferred photons from leaking into an adjacent well of the double-well potential energy configuration.

[0044] Another innovative aspect of the disclosed subject matter can be embodied in a method comprising: performing a quantum computing operation on a flux qubit to place the flux qubit in a first excited state of two energy states within a single-well potential energy configuration; biasing the flux qubit so that the two energy states are respectively mapped to two wells of a double-well potential energy configuration; raising a potential barrier between a first well and a second well of the double-well potential energy configuration, wherein the first excited state is mapped to the first well or the second well, and the raised potential well prevents the excited state from leaking into an adjacent well of the double-well potential energy configuration.

[0045] The details of one or more embodiments of the invention are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the invention will become apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a schematic diagram illustrating an exemplary embodiment of a dispersive measurement scheme for measuring the computational state of a transmon qubit.

[0047] Figure 2 is a schematic diagram illustrating an exemplary measurement scheme in which a detector qubit serves as a state detector of a data qubit.

[0048] Figure 3 is a schematic diagram showing an exemplary measurement scheme in which a phase qubit is used as a detector qubit for a transmon qubit.

[0049] Figure 4 is a schematic diagram showing an exemplary measurement scheme in which a flux qubit is used as a detector qubit for a transmon qubit.

[0050] Figure 5 is a flow chart illustrating a method of reading out the state of a transmon qubit using a flux qubit.

[0051] Figure 6 is a flow chart illustrating a method of calibrating a flux qubit to read out the state of a transmon qubit.

[0052] Figure 7a is a schematic diagram illustrating an exemplary embodiment of a flux bias generator.

[0053] Figure 7b is a schematic diagram showing an exemplary embodiment of a transducer for use with a flux qubit. DETAILED DESCRIPTION

[0054] Quantum computing requires coherently processing quantum information stored in the quantum bits (qubits) of a quantum computer. Superconducting quantum computing is a promising implementation of solid-state quantum computing technology, in which the quantum information processing system is formed in part by superconducting materials. In order to operate a quantum information processing system that employs solid-state quantum computing technology (such as superconducting qubits), the system is maintained at extremely low temperatures, such as tens of mK. Extreme cooling of the system keeps the superconducting material below its critical temperature and helps to avoid undesirable state transitions. In order to maintain such low temperatures, the quantum information processing system can be operated in a cryostat (such as a dilution refrigerator).

[0055] In some cases, large-scale quantum computers can be implemented using transmon qubits or their variants. A transmon qubit includes a large capacitor in parallel with one or more Josephson junctions, which improves the qubit's insensitivity to charge noise and allows the qubit to exhibit long coherence times. In addition, high-precision single and dual logic gate operations have been demonstrated using transmon qubits. When transmon qubits are used as data qubits for quantum computing operations, reading out the state of these qubits with high fidelity is an important part of the operation.

[0056] However, since the computational state of a transmon qubit cannot be distinguished by a single-shot flux or charge measurement, the state of a transmon qubit is measured based on the difference in energy rather than the difference in flux or charge. In addition, the energy difference between the two states of a transmon qubit is only one microwave photon. A microwave photon with a frequency of about 5 GHz has only 0.02 meV of energy, which is too low to be detected by conventional methods. To this end, a relatively complex and indirect dispersion measurement scheme is usually adopted: a traveling wave is sent via a transmission line into a resonator coupled to a qubit, and depending on the state of the transmon qubit, different degrees of phase shift or amplitude change are imparted to the traveling wave. This scheme requires special instruments with a relatively large volume within a cryostat and is known to result in non-idealities, which may impair the fidelity of the operations performed by the quantum computer, as will be discussed below.

[0057] An alternative strategy is to transfer the state of the transmon qubit to another qubit where flux or charge measurement is possible, i.e., a detector qubit. This reduces the need for amplification of small signals associated with attempting to measure a change in state, and the need to send relatively large numbers of photons to the transmon qubit. The detector qubit can be dynamically tuned to be resonant or non-resonant with the transmon qubit, so that the state of the transmon qubit is transferred as needed. This has been tested with phase qubits as detector qubits. However, a disadvantage of using phase qubits as detector qubits is their long dead time, which hinders fast operation, as will be discussed below. Figure 3 This is discussed in . In addition, when performing detection with a phase qubit, the phase qubit state jumps from a shallow potential well to a deep potential well. This results in the emission of dozens of microwave photons, which can propagate to other parts of the chip, compromising the quantum coherence of the qubit. This will also be discussed in the following Figure 3 This is discussed in .

[0058] To address these issues, the present disclosure involves using a flux qubit as a detector qubit, and capacitively coupling the flux qubit to a transmon qubit so that the state of the transmon qubit can be mapped to the flux qubit and detected by the difference in the self-flux of the flux qubit. The level of the self-flux of the flux qubit is large enough to be detected by techniques such as SFQ. In addition, the flux qubit does not emit any microwave photons.

[0059] Figure 1is a schematic diagram illustrating an exemplary embodiment of a dispersive measurement scheme for measuring the computational state of transmon qubit 101.

[0060] An input detection signal 121 is sent into the transmission line 120. The input detection signal 121 is a traveling wave, which may be one or more guided modes of the transmission line 120. The frequency of the input detection signal 121 may be a microwave frequency at or near the resonant frequency of the transmon qubit 101, so that the transmon qubit 101 may be resonantly or dispersively addressed by the input detection signal 121.

[0061] The transmon qubit 101 can be coupled to the transmission line 120 via the readout resonator 111. For example, the readout resonator 111 can be a quarter-wave coplanar waveguide resonator within a superconducting layer on which the transmon qubit 101 and the transmission line 120 are fabricated. The input probe signal 121 sent to the readout resonator 111 can acquire a phase shift and / or an amplitude change depending on the state of the transmon qubit 101. The transmon qubit 101 is dispersively coupled to the readout resonator 111. In other words, the resonant frequency of the transmon qubit 101 is detuned from the center frequency of the readout resonator 111. The transmon qubit 101 and the readout resonator 111 form a so-called dressed cavity state in which the resonant frequency of the readout resonator 111 changes depending on the state of the transmon qubit 101. In addition, due to the nonlinearity of the Josephson junction, the impedance of the transmon qubit 101 depends on the state of the transmon qubit 101. Therefore, if transmon qubit 101 is in first state 101-1, transmission line 120 outputs first output detection signal 122, and if transmon qubit 101 is in second state 101-2, transmission line 120 outputs second output detection signal 123. The state of transmon qubit 101 can be inferred by whether the output signal is first output detection signal 122 or second output detection signal 123.

[0062] Coupling to the environment via readout resonator 111 results in a reduction in the damping or radiative lifetime T1 of transmon qubit 101, which reduces the coherence of transmon qubit 101. This also imposes a limit on the measurement time of transmon qubit 101 and gives a minimum required operating speed. Therefore, in addition to detuning the resonance of readout resonator 111 from the resonant frequency of transmon qubit 101, the leakage rate or quality factor of readout resonator 111 can be optimized so that it allows the necessary measurement speed while maintaining the damping of transmon qubit 101 at an acceptable rate.

[0063] To further isolate the transmon qubit 101 from the damping of the external circuit, a Purcell filter 112 may also be placed between the transmission line 120 and the transmon qubit 101, for example, at the output side of the readout resonator 111, in series with the readout resonator 111. The Purcell filter 112 decouples the transmon qubit 101 from the external circuit over a frequency range including the resonant frequency of the transmon qubit 101 to suppress the damping of the transmon qubit 101. Photons near the resonant frequency of the transmon qubit 101 may be prevented from coupling to the external circuit. The Purcell filter may be implemented, for example, with a pair of symmetrical quarter-wave stubs connected to ground.

[0064] Even though Figure 1 The dispersion measurement scheme described in demonstrates fairly high accuracy, such as 98% accuracy for a single transmon qubit 101, but scaling this dispersion measurement scheme to systems with a large number of transmon qubits 101 may be complicated by several factors. For example, Figure 1 The dispersion measurement scheme described in may require a series of elements for amplifying weak signals, for example, quantum confined parametric amplifiers, low noise cryogenic amplifiers, such as high electron mobility transistors (HEMTs). The dispersion measurement scheme may also require magnetic non-reciprocal elements (such as circulators) to isolate the transmon qubit 101 from the noise generated by these amplifiers. These elements may occupy a large amount of millikelvin-level limited space in the cryostat, and the heat generated by each element may add up to exceed the cooling capacity of the cryostat. In addition, if the output signals 122, 123 are processed by demodulation electronics that perform heterodyne detection and threshold processing at room temperature, this may also require low-latency feedback conditioned on the measurement results.

[0065] Therefore, if Figure 1 The dispersion measurement scheme described in may not be suitable for scaling the operation to a large number of transmon qubits 101. Furthermore, although the signal-to-noise ratio of the dispersion measurement can be improved with the strength of the input probe signal 121, it has been observed that too many photons sent into the readout resonator 111 result in spurious qubit state transitions, which can compromise the fidelity of the operation.

[0066] As an alternative, a Josephson Photo-Multiplier (JPM) can be used as a microwave photon counter at the output of the transmission line 120 to detect the output signals 122, 123, which can be simplified slightly. Figure 1 The dispersive measurement scheme described in . The Josephson photomultiplier includes a single Josephson junction in an RF superconducting quantum interference device (SQUID) ring that is biased near the critical flux at which phase slip occurs. The transmon qubit 101 can still be dispersively coupled to the readout resonator 111. However, instead of using a circulator and a multi-stage amplifier, the difference in photon occupancy of the readout resonator 111 can be detected by a Josephson photomultiplier. Therefore, the state of the transmon qubit 101 can be directly detected and discerned by the Josephson photomultiplier at the millikelvin level in a cryostat. To date, the single measurement fidelity demonstrated using this scheme is 92%.

[0067] These relatively complex dispersion measurement schemes have been used because single photon detectors in the microwave frequency range are difficult to implement. The states of the transmon qubit 101 do not differ in the first order moments of their charge or flux wave functions, and differ only in energy by a single microwave photon. There are differences in higher order moments of charge and flux, but these are known to be difficult to measure and therefore are not feasible as a measurement scheme for the transmon qubit 101.

[0068] Figure 2 is a schematic diagram illustrating an exemplary measurement scheme in which detector qubit 210 is used as a state detector of data qubit 201 .

[0069] Data qubit 201 may be a superconducting qubit that is arranged to participate in quantum computation with other data qubits 201. Thus, data qubit 201 may be arranged to exhibit a level of nonlinearity required to perform quantum computation and to receive the excitation microwave pulses and flux bias required for quantum computation. For example, data qubit 210 may be a transmon qubit 101.

[0070] Data qubit 201 may be arranged to be coupled to detector qubit 210 via coupling element 220 .

[0071] The detector qubit 210 may be any type of superconducting qubit that includes one or more Josephson junctions. For example, the detector qubit 210 may be a phase qubit or a flux qubit. Figure 3 and Figure 4 Techniques for reading out phase qubits or flux qubits are described in . Alternatively, detector qubit 210 may include any other type of qubit, such as microwave transitions of quantum dots, diamond NV centers, or Rydberg atoms, which may resonate with the transition frequency of data qubit 201. Depending on the type of detector qubit 210, any suitable technique may be used to read out the state of detector qubit 210. Detector qubit 210 may allow the energy state of data qubit 201 to be measured based on, for example, flux, charge, or the amount of UV / visible / near IR photons.

[0072] Detector qubit 210 can be dynamically tuned to be in resonance or out of resonance with data qubit 201. When detector qubit 210 is tuned to be in resonance or close to resonance with data qubit 201, the state of data qubit 201 or the photons of data qubit 201 can be swapped with the state of detector qubit 210. In other words, data qubit 201 and detector qubit 210 can be coupled so that the interaction between data qubit 201 and detector qubit 210 can be in the form of virtual photons or excitons (which are excited quanta shared by data qubit 201 or detector qubit 210). This is because the state of data qubit 201 (which can be a superposition of the ground state and the first excited state of data qubit 201) can exhibit quantum coherent oscillations between data qubit 201 and detector qubit 210. For the remainder of the specification, it will be understood in the context that the term "exchanged photons" between data qubit 201 and detector qubit 210, i.e., photons, can refer to any intermediate quantum of interaction between data qubit 201 and detector qubit 210, and is not limited to isolated quanta of propagating light. For example, "exchanged photons" can also refer to delocalized excitons or virtual photons between data qubit 201 and detector qubit 210. In this sense, detector qubit 210 can act as a single photon detector for data qubit 201.

[0073] Coupling element 220 may include capacitive coupling, where data qubit 201 and detector qubit 210 are capacitively coupled to each other. For example, metal portions of data qubit 201 and detector qubit 210 may be placed in close proximity to allow capacitive coupling. For another example, a capacitor may be placed between data qubit 201 and detector qubit 210. Alternatively, coupling element 210 may include inductive coupling, where data qubit 201 and detector qubit 210 are inductively coupled to each other. For example, an inductive portion of data qubit 201 (such as a loop or elongated portion of data qubit 201 and detector qubit 210) may be positioned so that a magnetic flux generated by data qubit 201 may generate a current in detector qubit 210, and vice versa. Alternatively, coupling element 210 may include a combination of inductive coupling and capacitive coupling. Alternatively, coupling element 210 may include a transmission line, such as a coplanar waveguide. Alternatively, coupling element 210 may include a superconducting coupler qubit disposed between data qubit 201 and detector qubit 210. In this case, the resonant frequencies of data qubit 201 and detector qubit 210 do not need to be tuned to be resonant or non-resonant with each other, and only the superconducting qubit used as coupling element 210 may be controlled to adjust the coupling between data qubit 201 and detector qubit 210. Alternatively, coupling measure 210 may include a Josephson junction parametric amplifier or a Josephson junction parametric converter.

[0074] In order to transfer the state of the transmon qubit 201 to the detector qubit 210, the resonant frequency of the transmon qubit 201 or the resonant frequency of the detector qubit 210 can be dynamically tuned. In the case where the coupling metric 210 is another superconducting qubit, the coupling metric 210 can be dynamically tuned. The quantum states occupied in the data qubit 201 and the detector qubit 210 can be time-dependent and can exhibit quantum coherent oscillations. The temporal correlation of the states can be determined by the degree of interaction between the data qubit 201 and the detector qubit 210. By dynamically tuning to be resonant or non-resonant with each other, the state of the transmon qubit 201 can be mapped or transferred to the detector qubit 210. Depending on the duration of the interaction, the exchange of photons can be performed partially or completely. The state exchange between the data qubit 201 and the detector qubit 210 can be performed within the coherence time of the data qubit 201 and the detector qubit 210.

[0075] The concept of using another qubit as the detector qubit 210 has been tested with the phase qubit 310 as the detector qubit 210 .

[0076] Figure 3 It is a reference Figure 2 Schematic diagram showing an exemplary measurement scheme in which phase qubit 310 is used as detector qubit 210 to detect the state of transmon qubit 301 as data qubit 201 .

[0077] Transmon qubit 301 , used as data qubit 201 in this example, may include capacitor 302 and SQUID ring 303 including first Josephson junction 304 - 1 and second Josephson junction 304 - 2 . Figure 3 A schematic diagram of a potential energy curve 305 of a transmon qubit 301 is shown, which specifies the ground state 301-1 and the first excited state 301-2 of the transmon qubit 301. Due to the anharmonicity of the potential energy curve 305 or the nonlinearity of the transmon qubit 301, the energy levels of these states are not equidistant, and the microwave excitation that resonates with the transition from the ground state 301-1 to the first excited state 301-2 may be largely non-resonant with other transitions. Therefore, although there are higher excited states, these two states 310-1, 310-2 can be considered as the computational space of the transmon qubit 301. The greater the nonlinearity of the transmon qubit 301, the greater the difference between the transition frequency from the ground state 301-1 to the first excited state 301-2 and the transition frequency from the first excited state 301-2 to the second excited state, which is not shown in the potential energy curve 305.

[0078] The phase qubit 310 used as the detector qubit 210 in this example may include a Josephson junction 313, a capacitor 312, and an inductor 311. The inductor 311 may be inductively coupled to a line carrying a flux bias current. The potential energy curves 330, 340, 350 of the phase qubit 310 as a function of flux may include a double well structure. The wave function may be narrowly confined in each well, as shown in the rightmost potential energy curve 350. Since the bottom of each potential well can be approximated as a quadratic function, it is a resonant potential well, and the energy spacing between states near the bottom of the potential well can be substantially equal. Therefore, the anharmonic nature of the potential well is not obvious, and the phase qubit 310 can exhibit substantially linear behavior. In order to restore nonlinearity, a bias current or flux bias may be applied to introduce asymmetry in the potential energy curve, as shown in the leftmost potential energy curve 330 of the phase qubit 310.

[0079] The leftmost potential energy curve 330 shows that the left potential well, which is circled and shown in detail in the potential energy curve 340, becomes shallow, so that only the ground state 310-1 and the first excited state 310-2 can be confined in the left potential well. Under this flux bias condition, the computational space of the phase qubit 310 can be provided by the ground state 310-1 and the first excited state 310-2 in the left potential well.

[0080] In order to measure the state of the phase qubit 310, a short bias pulse can be provided to temporarily reduce the height of the barrier between the left well and the right well. As shown by the arrow in the potential energy curve 340, this can allow the first excited state 310-2 to tunnel out of the shallow left well and fall into the deep right well, but mainly prevent the ground state 310-1 from leaving the left well. Therefore, the barrier between the left well and the right well can be raised again, and the potential energy curve 350 can return to a symmetrical shape. Then, the ground state 310-1 and the first excited state 310-2 can be separated and trapped in the left well and the right well, respectively. In other words, the barrier between the left well and the right well can largely suppress the tunneling between the two wells. Then, two flux states corresponding to the wave functions narrowly confined in the left well and the right well can be detected and distinguished by magnetic flux. In this specification, the amount of flux generated by the phase qubit 310 in each flux state will be referred to as "self-flux". The self-flux of the phase qubit can be measured using a device such as a SQUID or SFQ (single flux quantum) circuit.

[0081] The transmon qubit 310 can be coupled to the phase qubit 330 via the coupling element 320. Figure 2As discussed in , coupling element 320 can be any of capacitive coupling, inductive coupling, or a combination of capacitive coupling and inductive coupling, a transmission line, a coupler superconducting qubit, a Josephson parametric converter, or a Josephson parametric amplifier. As a result of quantum computation in collaboration with other transmon qubits 301, transmon qubit 301 can carry a quantum state that is a superposition state between a ground state 301-1 and a first excited state 301-2. In order to detect the quantum state currently present in transmon qubit 301, phase qubit 310 can be tuned to resonate with transmon qubit 301. Alternatively, in order to detect the quantum state currently present in transmon qubit 301, transmon qubit 301 can be tuned to resonate with phase qubit 310. By being dynamically tuned to resonate or not resonate with transmon qubit 301, photons of transmon qubit 301 can be exchanged into phase qubit 310. In other words, the phase qubit 310 can receive the quantum state to the computational space of the phase qubit 330, which is a shallow left well as shown in the potential energy curve 340. In other words, the superposition state of the ground state 301-1 and the first excited state 301-2 of the transmon qubit 301 can be mapped to the superposition state of the ground state 310-1 and the first excited state 310-2 of the phase qubit 310. As described above, by applying a flux bias pulse to reduce the potential barrier between the left well and the right well, the ground state 310-1 and the first excited state 310-2 of the phase qubit 310 can be separated into two different flux states, as shown in the rightmost potential energy curve 350. The flux can be measured using a device such as a SQUID or SFQ circuit. Therefore, the phase qubit 310 can be used as a single photon detector or state detector for the transmon qubit 301. Using a very short pulse (about 10ns) to reduce the potential barrier, the quantum state can be distinguished with an accuracy of 90%.

[0082] However, there are several disadvantages to using the phase qubit 310 as the detector qubit 210. Once a single photon is exchanged into the phase qubit 310, in order to separate the two flux states 310-1, 310-2, the phase qubit 310 is biased so that the first excited state 310-2 is more likely to tunnel out of the shallow left well of the metastable state than the ground state 310-1. However, the tunneling rate difference between the ground state 310-1 and the first excited state 310-2 is only large enough to provide a maximum theoretical contrast of about 96%. In addition, the decay of the tunneled first excited state 310-1 to the bottom of the right well of the potential energy curve 330, 340, 350 is a dissipative process. It has been observed that the energy emitted in this process drives adjacent qubits to excited states, resulting in measurement crosstalk errors. The process of tunneling into the right-hand well also causes the phase qubit 310 to lose phase coherence due to the dissipative evolution of the wave function. Therefore, in order to be used as the detector qubit 210 again, the phase qubit 310 needs to be reset so that the ground state 310-1 and the first excited state 310-2 are reset in the left shallow well of the potential energy curves 330, 340, 350. The actual reset time of the known phase qubit 310 is tens to hundreds of microseconds, which may be very long compared to the coherence time of the currently available transmon qubit 301.

[0083] To solve Figure 1 Problems with the dispersion detection scheme described in Figure 3 In order to solve the problem of using phase qubit 310 as detector qubit 210 described in the specification, the present specification discloses using flux qubit 410 as detector qubit 210, so that the state of transmon qubit 401 can be mapped to flux qubit 410 coupled to transmon qubit and detected by the difference of self-flux of flux qubit 410.

[0084] Figure 4 It is a reference Figure 2 and Figure 3 Schematic diagram showing an exemplary measurement scheme in which flux qubit 310 is used as detector qubit 210 to detect the state of transmon qubit 301 as data qubit 201 .

[0085] Transmon qubit 401 used as data qubit 201 in this example may include capacitor 402 and SQUID ring 403 including first Josephson junction 404 - 1 and second Josephson junction 404 - 2 . Figure 4A schematic diagram of a potential energy curve 405 of a transmon qubit 401 is shown, which specifies the ground state 401-1 and the first excited state 401-2 of the transmon qubit 401. Although higher excited states exist, due to the above Figure 3 For the reasons discussed in , only these two states can be considered as the computational space of transmon qubit 401.

[0086] The flux qubit 410 used as the detector qubit 210 in this example may include a SQUID ring 413 and an inductor 411, the SQUID ring 413 including a first Josephson junction 413-1 and a second Josephson junction 413-2. Importantly, the flux qubit 410 is connected in parallel with a capacitor 412. The capacitance of capacitor 412 can be set large enough so that it can behave like a transmon qubit under a certain range of flux biases. Although the larger capacitance of capacitor 412 can increase the phase coherence of the flux qubit 410, it may reduce nonlinearity. Higher coherence prolongs the time window for interaction between the transmon qubit 401 and the flux qubit 410. Since the flux qubit 410 is used as the detector qubit 210, the capacitance can be set to extend the coherence time, rather than keeping the nonlinearity within the range that will be used as the data qubit 201. In this case, the capacitance of capacitor 412 can be in the range of 10fF to 100fF. Alternatively, the capacitance of capacitor 412 may be determined so that flux qubit 410 may be used interchangeably between data qubit 201 and detector qubit 210. In this case, the capacitance of capacitor 412 may be in the range of 1 fF to 50 fF.

[0087] The inductance of the inductor 411 can be determined to be as large as possible to minimize the effects of magnetic flux noise. However, there are practical constraints, including the fact that the coil-wound inductor has self-resonance when the size of the inductor 411 is too large. Generally, this can be overcome by increasing the inductance without increasing the coil length using additional Josephson junctions 413-1, 413-2.

[0088] The potential energy curves 430, 440, 450 of the flux qubit 410 as a function of flux can include a double well structure including two wells separated by a potential barrier, where each of the wells corresponds to a different discrete flux state, namely a left flux state 410-3 and a right flux state 410-4. When the potential barrier between the two wells is high enough, the wave function is narrowly confined in each well, as shown by the rightmost potential energy curve 450. In other words, the potential barrier between the left well and the right well can be high enough to maximally suppress the tunneling of the left flux state 410-3 and the right flux state 410-4 between the two wells. For example, for a flux qubit 410 in which the self-resonance of the inductor 411 and the capacitor 412 is at about 20 GHz, a 2 meV potential barrier can suppress the tunneling rate to 1 Hz.

[0089] The two flux states 410-3, 410-4 can be detected and distinguished by the flux qubit 410 depending on the magnetic flux generated by the flux state. As in the phase qubit 310, in this specification, the amount of flux generated by the flux qubit 410 in each flux state 410-3, 410-4 will also be referred to as "self-flux". The difference in self-flux between the left flux state 410-3 and the right flux state 410-4 can be as large as a single flux quantum. The self-flux of the flux qubit 410 can be measured using an equivalent device such as a SQUID or SFQ circuit or other device capable of measuring self-flux. The difference in self-flux between the left flux state 410-3 and the right flux state 410-4 can deviate slightly from a single flux quantum or Φ0. This is because the parabolic potential energy of the inductor 411 may cause the double well to deviate from the ideal periodicity of the junction potential energy corresponding to a single flux quantum. In the case of a SQUID 413 where superconductivity is used for the inductor 411 or the Josephson junctions 413-1, 413-2, these non-idealities may further deviate from a single flux quantum from the difference in flux.

[0090] The shape of the potential energy curve can be controlled with a first flux bias across the entire circuit of the flux qubit 410 and a second flux bias across the SQUID loop 413 of the flux qubit 410. By separately and dynamically controlling these two flux biases, mapping the state of the transmon qubit 401 to the flux qubit 410 and subsequent measurement of the flux state of the flux qubit can be performed, as explained in more detail later.

[0091] By applying a first flux bias across the entire circuit of the flux qubit 410 or primarily across the inductor 411 of the flux qubit 410, the potential energy curves 440, 450 of the flux qubit 410 can be "tilted", in other words, an energy asymmetry is introduced between the two discrete flux states 410-3, 410-4. In the example shown in the potential energy curves 440, 450, the energy of the left flux state 410-3 is lower than the energy of the right flux state 410-4.

[0092] The tilt of the potential energy curves 440, 450 localizes the hybridized energy states 410-1, 410-2 of the flux qubit 410 to the two flux states 410-3, 410-4 of the flux qubit 410 corresponding to the states confined in the two wells. When the potential energy curve is not tilted, both flux states of the flux qubit 410 occupy the ground state of each well with substantially the same energy. When the barrier height is finite, the two flux states 410-3, 410-4 form two hybridized states 410-1, 410-2 spread over the two wells, as shown in the potential energy curve 430. The potential energy curve 430 corresponds to the case where the barrier between the two wells is minimized. Even if there is a finite barrier between the two wells, the hybridized states 410-1, 410-2 will be spread over the two wells via tunneling through the barrier. The two hybrid states form two different energy levels, a ground state 410 - 1 and a first excited state 410 - 2 , as shown in the potential energy curve 430 .

[0093] In the absence of tilt, when the barrier is raised, each energy state 410-1, 410-2 will map to either the left flux state 410-3 or the right flux state 410-4 with equal probability. Therefore, the energy states 410-1, 410-2 cannot be distinguished based on self-flux.

[0094] The critical current of the Josephson junctions 413-1, 413-2 can be controlled by applying a second flux bias through the SQUID loop 413 of the flux qubit 410. This thus changes the height of the barrier between the two flux states 410-3, 410-4 and also changes the resonant frequency of the flux qubit 410.

[0095] Since the sloped potential energy curves 430, 440, 450 make the left well lower energy than the right well, as the barrier height increases adiabatically, the ground state 410-1 will be guided to the left flux state 410-3 and the first excited state 410-2 will be guided to the right flux state 410-4.

[0096] In particular, the area of ​​the SQUID ring 413 of the flux qubit 410 can be arranged to be large enough so that the second flux bias through the SQUID ring 413 can be applied largely independently of the first flux bias. For example, the area can be 1 μm 2 To 100μm 2 In some implementations, the area of ​​the SQUID ring 413 can be approximately 40 μm 2 Although such a large area of ​​SQUID ring 413 may make flux qubit 410 more sensitive to flux noise or stray capacitance, the ability to control barrier height and resonant frequency largely independent of the tilt of the potential energy may be important when using flux qubit 410 as detector qubit 210, as will be explained below.

[0097] Figure 5 Shown is a reference Figure 1 , Figure 2 and Figure 4 A flow chart of a method for reading out the state of a transmon qubit 401 using a flux qubit 410 is shown.

[0098] In step 510, a first value of a first flux bias may be applied to flux qubit 410 such that potential energy curves 430, 440, 450 of flux qubit 410 are largely symmetrical. In this state, the resonant frequency of transmon qubit 401 may be detuned away from the resonant frequency of flux qubit 410 such that the interaction between transmon qubit 410 and flux qubit 410 is not significant. For example, the resonant frequency of transmon qubit 410 may be detuned from the resonant frequency of flux qubit 410 by 2 GHz or more.

[0099] In some cases, a second value of a second flux bias may be applied to the detuned condition, as will be explained in more detail in step 550 .

[0100] In step 510, transmon qubit 401 may be excited, thereby preparing the state of transmon qubit 401.

[0101] In step 520, a first value of a second flux bias may be applied to flux qubit 410. At the first value of the second flux bias, the barrier height is minimized, as shown in potential energy curve 430, and flux qubit 410 is brought into an interaction frequency at which flux qubit 410 will interact with transmon qubit 401 and the state of transmon qubit 410 will be mapped to flux qubit 410. At the first value of the second flux bias, flux qubit 410 may be arranged to receive photons from transmon qubit 401 so that the quantum state may oscillate between transmon qubit 401 and flux qubit 410. As described above, the second flux bias applied by SQUID ring 413 controls the barrier height between the two wells and the resonant frequency of flux qubit 410. Under such conditions, the phase coherence of flux qubit 410 is maximized, which provides a time window for coherent interaction with transmon qubit 410. The T1 time of the flux qubit 410 may be approximately 30 μs. In step 520 , the first value of the first flux bias applied in step 510 may be maintained such that the potential energy curves 430 , 440 , 450 of the flux qubit 410 remain symmetrical.

[0102] In step 530, transmon qubit 401 may be tuned to an interaction frequency that resonates with flux qubit 410, such that if transmon qubit 401 is excited, photons are exchanged into flux qubit 410. The superposition state of ground state 401-1 and first excited state 401-2 may be mapped to the superposition state of ground state 410-1 and first excited state 410-2 of the hybrid energy state of flux qubit 410.

[0103] In some implementations, steps 520 and 530 may be performed simultaneously.

[0104] In some implementations, if transmon qubit 401 and flux qubit 410 are tuned to resonance in step 520 by applying the first value of the second flux, step 530 may be omitted.

[0105] In the case where transmon qubit 401 is excited to have a state prepared within transmon qubit 401 in step 510 , steps 520 and 530 may be performed immediately after the excitation of transmon qubit 401 .

[0106] In step 540, a second value of the first flux bias may be applied to the flux qubit 410, causing the potential energy curves 430, 440, 450 to tilt. Then, the energies of the first flux state 410-3 and the second flux state 410-4 may become different. Figure 4 As explained in , this allows the energy states 410 - 1 , 410 - 2 to be mapped to the left flux state 410 - 3 and the right flux state 410 - 4 when the potential barrier is adiabatically raised in step 550 .

[0107] The first value of the second flux bias applied in step 520 may be maintained so that the height of the potential barrier is kept to a minimum at this stage. In some implementations, steps 530 and 540 may be performed simultaneously. In some implementations, the second value of the first flux bias may be applied to the flux qubit 410 throughout the process so that the potential energy curves 430, 440, 450 of the flux qubit 410 are always tilted. This is conditioned on the fact that the exchange of states in steps 520 and 530 is not affected by the tilt of the potential energy curves 430, 440, 450 of the flux qubit 410.

[0108] In step 550, a second value of a second flux bias may be applied to the flux qubit 410 so that the potential barrier between the two wells is raised to "lock" the flux states 410-3, 410-4 so that tunneling between the left flux state 410-3 and the right flux state 410-4 may be substantially suppressed. As described above, in some implementations, the height of the potential barrier may be about 2 meV. The transition from the first value to the second value of the second flux bias may be adiabatic (in other words, gradual) to minimize the probability of the flux qubit changing state. The flux states 410-3, 410-4 may remain in the wells for a sufficiently long time before measurement. By having a large potential barrier, the qubit will stay in the corresponding well for a longer time without tunneling, removing the need to measure immediately as in some current systems.

[0109] The second value of the second flux bias may also de-resonate flux qubit 410 with transmon qubit 401 by tuning away from the interaction frequency. Measurement of the state of flux qubit 410 may be performed when flux qubit 410 is significantly detuned from transmon qubit 401. The detuning may be, for example, 2 GHz or more.

[0110] When microwave reflectometry is used for state detection, the second value of the first flux bias applied in step 530 may be maintained to keep the potential energy curves 430, 440, 450 asymmetric, which will be discussed in more detail later.

[0111] The resonant frequency of transmon qubit 401 can also be tuned away from the interaction frequency so that the interaction between transmon qubit 401 and flux qubit 410 is insignificant.

[0112] The time interval between steps 520, 530 for resonating the transmon qubit 401 and the flux qubit 410 and steps 540, 550 for measuring the state from the flux qubit 410 may be determined in consideration of quantum coherent oscillation between the transmon qubit 401 and the flux qubit 410, so that when the mapping of the state of the transmon qubit 401 is completed, the flux qubit 410 is decoupled from the transmon qubit 401. For example, in order to determine the time interval of maximum transfer efficiency between the transmon qubit 401 and the flux qubit 410, the transmon qubit 401 may be excited to have a predetermined quantum state at step 510. After resonating the transmon qubit 401 and the flux qubit 410 by performing steps 520 and 530, a first time interval T may be introduced before steps 540 and 550 may be performed to measure the state transferred to the flux qubit 410. When the first time interval T is changed, these steps 510, 520, 530, 540, 550 may be repeated. The duration of the first time interval T for maximizing the transfer efficiency between transmon qubit 401 and flux qubit 410 may be determined, which gives a maximum detection probability at flux qubit 410 . Figure 6 Shown is a reference Figure 4 and Figure 5 Flowchart illustrating a method of calibrating flux qubit 410 to read out the quantum state of transmon qubit 401.

[0113] At step 610, a first value of the second flux bias (i.e., a condition for a single well configuration) and a second value of the second flux bias (i.e., a condition for a double well configuration for locking flux states 410-3, 410-4 of the flux qubit 410) may be determined. The first value of the second flux bias may be determined such that the resonant frequency of the flux qubit is at the interaction frequency.

[0114] At step 620 , a bias condition of transmon qubit 401 that causes the resonant frequency of transmon qubit 410 to reach the interaction frequency may be determined.

[0115] At step 630, the resonant frequency of transmon qubit 401 can be tuned away from the interaction frequency, and the resonant frequency away from the interaction frequency can be spectroscopically probed by sending microwave pulses while sweeping the frequency of the pulses. A first value of a second flux bias is applied to flux qubit 410 to lower the barrier so that photons can be exchanged from transmon qubit 401 into flux qubit 410 once transmon qubit 401 enters the interaction frequency.

[0116] The remainder of the process involves determining the time interval of maximum transfer efficiency between transmon qubit 401 and flux qubit 410, which has been discussed above.

[0117] At step 640, a microwave pulse may be sent into transmon qubit 401 to prepare a quantum state within transmon qubit 401. This step may be performed at step 510 described above.

[0118] Shortly after each pulse is sent to prepare the quantum state in transmon qubit 401, in other words, within a time scale much shorter than the coherence time of transmon qubit 401, the resonant frequency of transmon qubit 401 can be tuned to the interaction frequency. This can be achieved by following steps 520 and 530 described above.

[0119] After transmon qubit 401 and flux qubit 410 resonate, a first time interval T may be introduced.

[0120] During a first time interval T, the prepared quantum state of transmon qubit 401 can be transferred to flux qubit 410 .

[0121] At step 650, the state of transmon qubit 401 may be read out by measuring the state of flux qubit 610 following steps 540 and 550 described above.

[0122] By repeating steps 640 and 650 while changing the first time interval T, the duration of the first time interval T for maximizing the transfer efficiency between the transmon qubit 401 and the flux qubit 410 can be determined. Thus, the readout condition of the transmon qubit 401 can be established.

[0123] Steps 640 and 650 may be performed within the coherence time of the transmon qubit 401 from the moment of the microwave pulse to prepare the quantum state. In other words, the first time interval T may vary within the coherence time of the transmon qubit 401.

[0124] There are at least two ways to measure the state of flux qubit 410, as described below.

[0125] Microwave reflectometry can be used to distinguish left flux state 410-3 and right flux state 410-4. Flux qubit 410 can be biased so that the left and right wells behave like classical resonators with different energy level spacings. This can be achieved, for example, by adjusting the second flux bias so that each well becomes deeper and the bottom of each well can be approximated as a resonant potential and adjusting the first flux bias so that the asymmetry between the two wells is large enough to be detected by the difference in spacing between the ground state and the first excited state of each well. This frequency difference can be detected using microwave reflectometry, similar to Figure 1 However, since a relatively large strength input detection signal 121 can be used to detect the frequency difference, an amplifier and a circulator may not be necessary to detect the output signals 122, 123.

[0126] Alternatively, the self-flux of the flux qubit 410 can be measured directly. The magnetic flux of the left flux state 410-3 and the right flux state 410-4 can differ by a magnetic flux quantum, which can be detectable by a SFQ (single flux quantum) circuit or a SQUID magnetometer. For example, a QFP (quantum flux parametron) can be coupled to the flux qubit 410, and a SFQ pulse train can be sent to the QFT to read out the QFP state, which provides a state readout of the flux qubit 410.

[0127] Figure 7a is a schematic diagram illustrating an exemplary embodiment of a flux bias generator 760 .

[0128] The flux bias generator 760 includes a current source 761 configured to generate a current and a transducer 762 arranged to convert the current into a magnetic field. The transducer 762 may be arranged to perform the above Figure 5 and Figure 6 The method described herein generates a first flux bias and a second flux bias within a desired range.

[0129] Figure 7b It is a reference Figure 4 A schematic diagram of an exemplary embodiment of a transducer 762 for use with a flux qubit 710 is shown.

[0130] As above Figure 4As discussed in , the flux qubit 710 includes an inductor 711, a capacitor 712, and a SQUID ring 713, the SQUID ring 713 including a first Josephson junction 713-1 and a second Josephson junction 713-2. As described above, since the flux qubit 710 includes a large parallel capacitance, in some implementations, the capacitor 712 is in the form of a paddle, which includes a first capacitor pad 712-1 and a second capacitor pad 712-2 on each side of the SQUID ring 713. When the capacitors 712, 712-1, 712-2 are in the form of a paddle, the capacitor can be increased by increasing the area of ​​the paddle. The first capacitor pad 712-1 and the second capacitor pad 712-2 are electrically connected to two terminals formed between the first Josephson junction 713-1 and the second Josephson junction 713-2 along the SQUID ring 713, respectively. The first capacitor pad 712-1 and the second capacitor pad 712-2 are connected to the inductor 711 via the first wire 714-1 and the second wire 714-2, respectively. The first ends of the first wire 714-1 and the second wire 714-2 may originate from the SQUID ring on each side of the first Josephson junction 713-1, and the first wire 714-1 is directly connected to the first capacitor pad 712-1 and the second wire 714-2 is directly connected to the second capacitor pad 712-2 via the SQUID ring 713. The second ends of the first wire 714-1 and the second wire 714-2 are electrically connected to the two terminals of the inductor 711.

[0131] In some implementations, the inductor 711 may include a gradiometric coil. Figure 2 As shown, starting from the two terminals connected to the first wire 714-1 and the second wire 714-2, the inductor 711 forms two loops adjacent to each other, so that when current flows into the inductor 711, the magnetic fields generated at the two loops are in opposite directions to each other. Therefore, when a magnetic field (e.g., a second flux bias) is applied across the entire area of ​​the inductor 711, the effects of the magnetic fields of the two loops formed in the inductor 711 largely cancel each other. When magnetic fields of opposite directions are coupled into the two loops formed in the inductor 711, the generation of the inductor current may be efficient.

[0132] The transducer 762 includes a first coil 762-1 and a second coil 762-2. Figure 4As discussed in , the shape of the potential energy curve can be controlled by a first flux bias through the entire circuit of the flux qubit 410, 710 or primarily through the inductor 411, 711 of the flux qubit 410, 710 and a second flux bias through the SQUID ring 413, 713 of the flux qubit 410, 710. By separately and dynamically controlling these two flux biases, mapping the state of the transmon qubit 401 to the flux qubit 410, 710 and subsequent measurement of the flux state of the flux qubit can be performed.

[0133] The first coil 762-1 is used to apply a first flux bias through the inductor 711 of the flux qubit 710, so that the potential energy curves 440, 450 of the flux qubits 410, 710 can be "tilted", in other words, an energy asymmetry is introduced between the two discrete flux states 410-3, 410-4.

[0134] When the inductor 711 of the flux qubit 710 is configured as a gradient metric coil as described above, the first coil 762-1 can also be configured as a gradient metric coil so that the magnetic flux from the first coil 762-1 is only efficiently coupled to the inductor 711, and less efficiently coupled to other parts of the flux qubit 710, such as the SQUID ring 713.

[0135] In some implementations, the first conductive line 714-1 and the second conductive line 714-2 may be arranged to cross, in other words, to be above each other at at least one position without being electrically connected to each other, so that parasitic coupling from the second coil 762-2 to the second flux bias in the inductor 711 is reduced. For example, Figure 7b It is shown that the first conductive line 714-1 and the second conductive line 714-2 are arranged to cross once between the SQUID ring 713 and the inductor 711. However, the number of crossings between the first conductive line 714-1 and the second conductive line 714-2 is not limited to once.

[0136] The second coil 762-2 is used to apply a second flux bias through the SQUID ring 413, 713 of the flux qubit 410, 710 by controlling the critical current of the Josephson junction 713-1, 713-2. This thereby changes the height of the barrier between the two flux states 410-3, 410-4 and also changes the resonant frequency of the flux qubit 410, 710.

[0137] When the first flux bias is applied via an inductor in the form of a gradient metric coil, the second flux bias generated from the second coil 762-2 may not be efficiently coupled to the inductor 711 of the flux qubit 710. Therefore, highly independent control of the first flux bias and the second flux bias can be achieved.

[0138] Thus, the magnetic flux through the entire circuit of the flux qubit 710 and the SQUID ring 713 can be independently controlled using the first coil 762 - 1 and the second coil 762 - 2 .

[0139] Figure 7b The example given in is only one embodiment of the transducer 762. Other designs of the transducer 762 can be used to generate the first flux bias and the second flux bias. In some implementations, the transducer 762 can be disposed on a substrate separate from the substrate containing the flux qubit 710. For example, the transducer 762 can be disposed on a surface of the substrate that can be close to the surface of the substrate containing the flux qubit 710. The arrangement of the first coil 762-1 and the second coil 762-2 can be such that when the two substrates are aligned laterally, the first coil 762-1 and the second coil 762-2 can be close to the inductor 711 and the SQUID ring 713, respectively, so that the first flux bias and the second flux bias can be provided.

[0140] A flux qubit 410, 710 connected in parallel with a relatively large capacitor can be used as a single photon detector for the transmon qubit 201. If a flux qubit 410, 710 can resonate with a potential energy curve that defines an energy level comparable to that of the flux qubit 410 when the barrier height is minimized, it can be used as a single photon detector for any other qubit.

[0141] Since the flux qubit 410, 710 allows negligible error rates when detecting the two flux states 410-3, 410-4, the measurement accuracy can be improved, which will provide high fidelity of operation. The reset time or cycle time of the flux qubit 410, 710 can be determined by the speed at which the barrier height is increased in step 550. The speed should be low enough to ensure adiabaticity of the process, but high enough to allow reasonable detection and operation speeds.

[0142] The total footprint of the flux qubits 410, 710 on the chip can be compatible with the two-dimensional grid of transmon qubits 301, 401. The flux qubits 410, 710 remove the need for a parametric amplifier HEMT circulator and alleviate the corresponding heat dissipation on the chip. Figure 1 The dispersion scheme described in suffers from spurious transitions due to the high photon numbers.

[0143] In order to achieve practical large-scale quantum computing involving a large number of qubits, the error rate of the qubits constituting the quantum computer should be below an acceptable threshold. A common error correction scheme is the so-called "surface code" quantum computer, which includes a two-dimensional array of data qubits and auxiliary qubits or measurement qubits, in which the nearest neighbors can be coupled to each other. The data qubits and auxiliary qubits can form an interlaced grid of two sub-grids.

[0144] In surface codes, data qubits and auxiliary qubits are entangled together using a series of physical qubit CNOT operations, followed by measurements of the entangled state which provides a means of error correction and error detection. The auxiliary qubits do not participate directly in the computation, but can be coupled to the data qubits to monitor the state of the data qubits, thereby detecting, for example, errors in the data qubits. In surface codes, a set of data qubits and auxiliary qubits entangled in this way are used to define a logical qubit.

[0145] Furthermore, a specific sequence of entanglement operations on pairs of data qubits and auxiliary qubits, so-called stabilizers, can be used to stabilize the state of the data qubits, since it suppresses spurious flips of the qubit state through a set of measurements. By repeatedly measuring qubits within a logical qubit using a complete set of commuting stabilizers, the logical qubit collapses into a simultaneous and unique eigenstate of all stabilizers. The stabilizers can be measured without disturbing the system of logical qubits. Spurious flips of the data qubits can be detected when the measurement results change, corresponding to one or more qubit errors, and the quantum state is projected onto different stabilizer eigenstates through these measurements.

[0146] The number of physical qubits required to define a logical qubit depends largely on the error rate of the physical qubits and the arrangement of the auxiliary qubits and data qubits.

[0147] In some implementations, transmon qubits 401, 701 can be used as both data qubits and auxiliary or measurement qubits in a surface code quantum computer. Flux qubits 410, 710 described in this specification can be used to read out the state of an auxiliary qubit after a parity measurement is completed between a data qubit and an auxiliary qubit.

[0148] In some implementations, the flux qubits 410, 710 described herein can be used as auxiliary qubits or measurement qubits in a surface code quantum computer. When the auxiliary qubit is measuring the parity of the data qubit, for high coherence, the auxiliary qubit can be biased to a "transmon mode" where the barrier height is minimized. After the parity measurement is completed, the flux qubits can be biased to a "double well mode" so that their states can be easily read out with superconducting electronic devices. This implementation may be advantageous because the scheme does not require any amplifiers or other microwave circuits to read out the auxiliary qubits.

[0149] The two-dimensional array of data qubits (transmon qubits 301, 401) and auxiliary qubits (flux qubits 410, 710) can be implemented on the surface of a single substrate. Alternatively, the flux qubits and transmon qubits can be arranged on two separate substrates 780, 790, and the surfaces of the two substrates can be close so that the flux qubit can be coupled to the transmon qubit by, for example, vacuum capacitance.

[0150] In some implementations, both the data qubits and the auxiliary qubits may be flux qubits 410, 710 as described in this specification. During coherent operation, all flux qubits 410, 710 may be biased to a "transmon mode" where the barrier height is minimized. Then, for state measurement of the auxiliary qubits, the auxiliary qubits may be biased to a "double well mode".

[0151] The implementation of quantum themes and quantum operations described in this specification can be implemented in a suitable quantum circuit, or more generally, in a quantum computing system (also referred to as a quantum information processing system, including the structures disclosed in this specification and their structural equivalents or a combination of one or more of them). The terms "quantum computing system" and "quantum information processing system" may include, but are not limited to, a quantum computer, a quantum cryptography system, a topological quantum computer, or a quantum simulator.

[0152] The terms quantum information and quantum data refer to information or data carried, stored or stored by a quantum system, wherein the smallest non-trivial system is a quantum bit, such as a system that defines a unit of quantum information. It should be understood that the term "quantum bit" includes all quantum systems that can be appropriately approximated as a two-level system in the corresponding context. Such a quantum system may include a multi-level system, for example, having two or more levels. For example, such a system may include atoms, electrons, photons, ions, or superconducting quantum bits. In some implementations, the computational ground state is identified by the ground state and the first excited state, however, it should be understood that other arrangements in which the computational state is identified by a higher-level excited state are also possible. It is understood that a quantum memory is a device capable of storing quantum data for a long time with high fidelity and high efficiency, such as a light-matter interface for transmission and a substance for storing and preserving quantum features (such as superposition or quantum coherence) of quantum data.

[0153] Quantum circuit elements (also referred to as quantum computing circuit elements) include circuit elements for performing quantum processing operations. That is, quantum circuit elements are configured to perform operations on data in a non-deterministic manner using quantum mechanical phenomena such as superposition and entanglement. Specific quantum circuit elements (such as quantum bits) can be configured to represent and operate information in more than one state at the same time. Examples of superconducting quantum circuit elements include quantum LC oscillators, quantum bits (such as flux quantum bits, phase quantum bits, or charge quantum bits), superconducting quantum interference devices (SQUIDs) (such as RF-SQUIDs or DC-SQUIDs), etc.

[0154] In contrast, classical circuit elements generally process data in a deterministic manner. Classical circuit elements can be configured to collectively execute the instructions of a computer program by performing basic arithmetic, logic, and / or input / output operations on data, where the data is represented in analog or digital form. In some implementations, classical circuit elements can be used to send data to and / or receive data from quantum circuit elements via electrical or electromagnetic connections. Examples of classical circuit elements include circuit elements based on CMOS circuits, rapid single flux quantum (RSFQ) devices, reciprocal quantum logic (RQL) devices, and ERSFQ devices, which are energy-efficient versions of RSFQ that do not use bias resistors.

[0155] The manufacture of quantum circuit elements and classical circuit elements described herein may require the deposition of one or more materials (such as superconductors, dielectrics and / or metals). Depending on the selected materials, these materials can be deposited using deposition processes (such as chemical vapor deposition, physical vapor deposition (e.g., evaporation or sputtering) or epitaxial techniques, etc.). The processes described herein for manufacturing circuit elements may require the removal of one or more materials from the device during manufacture. Depending on the material to be removed, the removal process may include, for example, wet etching techniques, dry etching techniques, or stripping processes. The materials forming the circuit elements described herein can be patterned using known photolithography techniques (e.g., optical photolithography or electron beam photolithography).

[0156] During operation of a quantum computing system using superconducting quantum circuit elements and / or superconducting classical circuit elements (such as the circuit elements described herein), the superconducting circuit elements are cooled in a cryostat to a temperature that allows the superconductor material to exhibit superconducting properties. Superconductor (alternatively, superconducting) materials can be understood as materials that exhibit superconducting properties at or below the superconducting critical temperature. Examples of superconducting materials include aluminum (superconducting critical temperature of about 1.2 Kelvin), indium (superconducting critical temperature of about 3.4 Kelvin), NbTi (superconducting critical temperature of about 10 Kelvin), and niobium (superconducting critical temperature of about 9.3 Kelvin). Therefore, superconducting structures (such as superconducting traces and superconducting ground planes) are formed by materials that exhibit superconducting properties at or below the superconducting critical temperature.

[0157] Although this specification contains a number of specific implementation details, these should not be interpreted as limitations on the scope of the claimed protection, but rather as descriptions of features specific to a particular implementation. Specific features described in the context of separate implementations in this specification may also be implemented in combination in a single implementation. On the contrary, the various features described in the context of a single implementation may also be implemented separately in multiple implementations or in any suitable sub-combination. In addition, although features may be described above as working in a particular combination, and even initially claimed as such, one or more features from the claimed combination may be deleted from the combination in some cases, and the claimed combination may be directed to a variant of a sub-combination or a sub-combination.

[0158] Similarly, although operations are described in a particular order in the accompanying drawings, this should not be understood as requiring that these operations be performed in the particular order shown or sequentially, or requiring that all of the operations shown be performed, in order to obtain the desired results. For example, the actions stated in the claims can be performed in a different order and still obtain the desired results. In some cases, multitasking and parallel processing may be advantageous. In addition, the separation of various components in the above implementations should not be understood as requiring such separation in all implementations.

[0159] Several embodiments of the present invention have been described. However, it should be understood that various modifications can be made without departing from the spirit and scope of the present invention. Therefore, other embodiments are within the scope of the following claims.

Claims

1. A method comprising: providing a data qubit and a flux qubit for measuring a state of the data qubit; Exciting the data qubits into excited states; biasing the flux qubit into a single-well potential configuration; Tuning the flux qubit causes photons from an excited state of the data qubit to transfer to the flux qubit; biasing a flux qubit containing the transferred photons into a double-well potential configuration; as well as A potential barrier between a first well and a second well of a double well potential configuration is raised, wherein the first well or the second well includes the transferred photons, and wherein the raised potential well prevents the transferred photons from leaking into an adjacent well of the double well potential configuration.

2. The method according to claim 1, wherein: Tuning the flux qubit so that photons from an excited state of the data qubit are transferred to the flux qubit includes: tuning the flux qubit to resonate with the data qubit in an excited state.

3. The method according to claim 1 or 2, wherein: Biasing a flux qubit containing transferred photons into a double-well potential energy configuration includes tilting a potential energy curve of the flux qubit so that an energy state of the flux qubit containing transferred photons is mapped to a first well and a second well of the double-well potential energy configuration.

4. The method according to any one of claims 1 to 3, further comprising: Reading out measures the energy state of the quantum bit.

5. The method according to claim 4, wherein: Reading out the energy state of the flux qubit, comprising: applying microwave reflectometry to the flux qubit.

6. The method according to claim 4, wherein: Reading out the energy state of the flux quantum bit includes: reading out a flux difference between a first energy state and a second energy state of the flux quantum bit.

7. The method according to claim 6, wherein: Reading out the flux difference is performed using a single flux quantum (SFQ) measurement of the flux difference.

8. The method according to any one of claims 1 to 7, wherein: The data qubit is a transmon qubit.

9. The method according to any one of claims 1 to 8, wherein: The data qubits are on a first substrate and the flux qubits are on a second substrate bonded to the first substrate.

10. A method comprising: performing a quantum computing operation on the flux qubit to place the flux qubit in a first excited state of two energy states within the single well potential energy configuration; biasing the flux qubit so that the two energy states are respectively mapped to the two wells of the double-well potential energy configuration; A potential barrier between a first well and a second well of a double-well potential energy configuration is raised, wherein a first excited state is mapped to the first well or the second well, and the raised potential well prevents the excited state from leaking into an adjacent well of the double-well potential energy configuration.

11. The method according to claim 10, comprising: Use microwave reflectometry to determine the excited states of flux qubits.

12. The method according to claim 10, comprising: Use the SFQ circuit to determine the excited states of the flux qubit.

13. A method comprising: Determining bias conditions for a single well configuration and a double well configuration of a flux qubit such that the flux qubit is at an interaction frequency in the single well configuration, wherein at the interaction frequency, the flux qubit resonates with the data qubit; determining a first bias condition for the data qubit with respect to the interaction frequency; determining a second bias condition for the data qubit for a frequency away from the interaction frequency; applying microwave pulses to the data qubits to prepare the state and tune to the interaction frequency; and The state of the flux qubit is measured to read out the state of the data qubit.

Images