Fast quantum gates with first-order transitions via frequency-modulated tunable coupling element

Abstract

Techniques relate to operating a quantum processing device is provided. The device includes at least two fixed-frequency quantum circuits coupled to a frequency-tunable coupler. The frequency of the coupler can be modulated so as to drive at least two selectively addressable energy transitions in the quantum processing device. The method includes modulating the frequency of the coupler so as to drive two first-order energy transitions. This is done so as to transfer (at least partly) an excitation of one of the quantum circuits to at least another one of the quantum circuits, via the tunable coupler. Related quantum processing devices are also provided.

Claims

1. A method of operating a quantum processing device, the method comprising: modulating a frequency of a tunable coupler to drive two first-order energy transitions, wherein the quantum processing device comprises at least two fixed frequency quantum circuits coupled to the tunable coupler; and causing an excitation of one of the at least two fixed frequency quantum circuits such that the excitation is at least partly transferred to at least another one of the at least two fixed frequency quantum circuits, via the tunable coupler, wherein the frequency of the tunable coupler is modulated to implement a quantum gate operating on at least one of the at least two fixed frequency quantum circuits, wherein modulating the frequency of the tunable coupler is associated with a first frequency to drive a first transition associated with a first excitation of the one of the at least two fixed frequency quantum circuits and is associated with at least one second frequency distinct from the first frequency to drive at least one second transition.

2. The method according to claim 1, wherein the frequency of the tunable coupler is operable to be modulated to drive at least two selectively addressable energy transitions in the quantum processing device; and wherein the frequency of the tunable coupler is modulated at: the first frequency, so as to drive the first transition and thereby transfer, at least partly, the first excitation of the one of the at least two fixed frequency quantum circuits to the tunable coupler; and the at least one second frequency, so as to drive the at least one second transition and thereby further transfer, at least partly, the first excitation from the tunable coupler to the at least another one of the at least two fixed frequency quantum circuits, each of the first transition and the at least one second transition being a first-order energy transition.

3. The method according to claim 2, wherein the frequency of the tunable coupler is modulated at the first frequency to drive the first transition and is modulated at the at least one second frequency to drive the at least one second transition responsive to the first transition.

4. The method according to claim 2, wherein the frequency of the tunable coupler is concomitantly modulated at the first frequency and the at least one second frequency, so as to concomitantly drive the first transition and the at least one second transition.

5. The method according to claim 4, wherein modulating the frequency of the tunable coupler further comprises switching on drives at the first frequency and the at least one second frequency to generate an adiabatic holonomic gate transformation.

6. The method according to claim 2, wherein, to modulate the frequency of the tunable coupler, each of the first frequency and the at least one second frequency corresponds to a difference of energy between a pair of excited states of the quantum processing device, respectively, the pair of excited states involving a reference excited state which is representable as a tensor product of an excited state of the tunable coupler and a ground state of each of the at least two fixed frequency quantum circuits.

7. The method according to claim 6, wherein each of the first frequency and the at least one second frequency corresponds to a difference of energy between the reference excited state and another excited state, the another excited state representable as a tensor product involving a ground state of the tunable coupler and a state of the at least two fixed frequency quantum circuits, where the state is representable as a tensor product involving an excited state of only one of the at least two fixed frequency quantum circuits and a ground state of each of the remaining at least two fixed frequency quantum circuits.

8. The method according to claim 2, wherein the first-order energy transition is driven by applying harmonic microwave signals to the tunable coupler, the harmonic microwave signals modulated so as to modulate the frequency of the tunable coupler, in order to drive the first transition and the at least one second transition.

9. The method according to claim 2, wherein the quantum processing device further comprises at least three fixed frequency quantum circuits, each coupled to the tunable coupler; and wherein the frequency of the tunable coupler is concomitantly modulated at the first frequency, so as to drive the first transition, and at two or more second frequencies, so as to concomitantly drive two or more second transitions, respectively, whereby excitations are transferred, at least partly, from the tunable coupler to two or more other ones of the at least three fixed frequency quantum circuits, each of the two or more second transitions being a first-order energy transition.

10. The method according to claim 9, further comprising setting amplitudes and phases of signals applied to the tunable coupler, so as to achieve a multipartite entangled state of the at least three fixed frequency quantum circuits.

11. The method according to claim 1, wherein the quantum gate does not operate, effectively, on the tunable coupler.

12. A method of operating a quantum processing device, the method comprising: modulating a frequency of a tunable coupler to drive two first-order energy transitions, the quantum processing device comprising at least two fixed frequency quantum circuits coupled to the tunable coupler; and causing an excitation of one of the at least two fixed frequency quantum circuits such that the excitation is at least partly transferred to at least another one of the at least two fixed frequency quantum circuits, via the tunable coupler, wherein the frequency of the tunable coupler is modulated to implement a universal set of gates based on states of the at least two fixed frequency quantum circuits, wherein the frequency of the tunable coupler is operable to be modulated so as to drive at least two selectively addressable energy transitions in the quantum processing device.

13. The method according to claim 12, wherein the universal set of gates are not based on states of the tunable coupler.

14. The method according to claim 12, further comprising setting amplitudes and phases of signals applied to the tunable coupler, so as to achieve one or more selected from the group consisting of: an iSWAP gate, a c-Phase gate, a Fredkin gate, and a square root of SWAP gate.

15. The method according to claim 12, wherein the frequency of the tunable coupler is modulated at: a first frequency, so as to drive a first transition and thereby transfer, at least partly, a first excitation of the one of the at least two fixed frequency quantum circuits to the tunable coupler; and at least one second frequency distinct from the first frequency, so as to drive at least one second transition, respectively, and thereby further transfer, at least partly, the first excitation from the tunable coupler to the at least another one of the at least two fixed frequency quantum circuits, each of the first transition and the at least one second transition being a first-order energy transition.

16. A method of operating a quantum processing device, the method comprising: modulating a frequency of a tunable coupler to drive two first-order energy transitions, wherein the quantum processing device comprises at least two fixed frequency quantum circuits capacitively coupled to the tunable coupler; and causing an excitation of one of the at least two fixed frequency quantum circuits such that the excitation is at least partly transferred to at least another one of the at least two fixed frequency quantum circuits, via the tunable coupler; wherein the frequency of the tunable coupler is modulated to implement a quantum gate operating on at least one of the at least two fixed frequency quantum circuits; wherein the frequency of the tunable coupler is modulated at a first frequency, so as to drive a first transition and thereby transfer, at least partly, a first excitation of the one of the at least two fixed frequency quantum circuits to the tunable coupler; wherein the frequency of the tunable coupler is modulated at at least one second frequency distinct from the first frequency, so as to drive at least one second transition.

17. The method according to claim 16, wherein the frequency of the tunable coupler modulated at the at least one second frequency distinct from the first frequency, so as to drive the at least one second transition further transfers, at least partly, the first excitation from the tunable coupler to the at least another one of the at least two fixed frequency quantum circuits.

18. The method according to claim 17, wherein each of the first transition and the at least one second transition is a first-order energy transition.

19. The method according to claim 16, wherein the frequency of the tunable coupler is operable to be modulated to drive at least two selectively addressable energy transitions in the quantum processing device.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a circuit diagram of an example of a multi-qubit coupling architecture, involving one or more frequency-tunable couplers, according to embodiments of the present invention;

(2) FIG. 2 shows an example geometry for a multi-qubit coupling architecture as in FIG. 1, using two transmon-type qubits coupled via a third transmon qubit that is configured as a frequency-tunable coupling element, according to embodiments of the present invention;

(3) FIG. 3 shows a circuit diagram of an example of a multi-qubit coupling architecture, involving one or more frequency-tunable couplers, according to embodiments of the present invention;

(4) FIG. 4 shows an example geometry for a multi-qubit coupling architecture as in FIG. 3, using four transmon-type qubits coupled via a qubit configured as a frequency-tunable coupling element, according to embodiments of the invention;

(5) FIG. 5 shows a circuit diagram of an example of a multi-qubit coupling architecture, involving one or more frequency-tunable couplers, according to embodiments of the present invention;

(6) FIG. 6 is a graphic depicting quantum states associated with two quantum circuits (e.g., computational qubits) coupled to a tunable coupler, as in FIG. 1 or 2, as well as first-order energy transitions, according to embodiments of the present invention;

(7) FIG. 7 schematically illustrates the modulation of the tunable coupler frequency, to which six transmon qubits are coupled in this example, and where harmonic microwave tones are simultaneously applied to the tunable coupler element via a single drive line, according to embodiments of the present invention;

(8) FIG. 8 is a graphic depicting quantum states associated to four fixed frequency qubits coupled to a tunable coupler, as in the configuration of FIG. 3, where transitions are driven simultaneously so as to produce an entangled state, according to embodiments of the present invention;

(9) FIG. 9 is a flowchart illustrating high-level steps of a method of operating a quantum processing device, where first-order energy transitions are concomitantly driven, according to embodiments of the present invention; and

(10) FIG. 10 is another flowchart, illustrating steps of a method similar to that of FIG. 9, except that the first-order energy transitions are sequentially driven, according to embodiments of the present invention.

(11) The accompanying drawings show simplified representations of devices or parts thereof, as involved in embodiments of the present invention. Similar or functionally similar elements in the figures have been allocated the same numeral references, unless otherwise indicated.

DETAILED DESCRIPTION

(12) In reference to FIGS. 1-10, an aspect of embodiments of the invention is first described, which relates to a method of operating a quantum processing device 1, 3, 4. Technical features depicted in FIGS. 2 and 4 are not necessarily to scale. This quantum processing device 1, 3, 4 is in all cases assumed to comprise at least two fixed frequency quantum circuits 11-14, 1n and a tunable coupler 20. The tunable coupler 20 acts as a frequency-tunable coupling element. The quantum circuits are, each, coupled to the tunable coupler 20. It is further assumed that the frequency of the tunable coupler 20 can be modulated, in order to drive selectively addressable energy transitions in the device 1, 3, 4, for the purpose of implementing the present methods, according to embodiments of the invention.

(13) Consistently, such methods generally revolve around modulating the frequency of the tunable coupler 20 (see steps S20, S20a, S20b in the flowchart of FIGS. 9, 10), in order to drive (S30, S30a, S30b) two first-order energy transitions. This is further carried out so that an excitation of one of the quantum circuits 11-14, 1n is at least partly transferred (S40, S40a, S40b) to at least another one of the quantum circuits 11-14, 1n, via the tunable coupler 20.

(14) To that aim, one may rely on a controller 32 (e.g., which can be realized as and/or include one or more control lines 30), to apply harmonic microwave signals (i.e., coherent microwave tones) to components of the system and trigger the desired transitions. This may notably require to adjust amplitudes and phases of the signals applied, in addition to modulating their frequency.

(15) Each of the quantum circuits 11-14, 1n is typically a non-linear quantum circuit, which can be used as a qubit 11-14, 1n. By non-linear, it is meant that at least two distinct energy levels (i.e., of different energies) of the quantum circuits can be addressed. However, the fixed frequency quantum circuits can also be harmonic oscillators, for instance.

(16) According to the present approach, the excitation of one of the quantum circuits 11-14, 1n is at least partly transferred to another quantum circuit, via the tunable coupler 20. The terminologies at least partly transferred and the like, as used herein, mean that the excitation may, in embodiments of the invention, possibly be fully transferred, in which cases the excitation is fully swapped from one quantum circuit to the other. However, in other embodiments of the invention, this excitation may also be transferred with a well-defined, predetermined probability, which is less than 1 (and yet greater than 0). Partial transfer of this excitation allows entangled states to be created, with excitations shared among the quantum circuits 11-14, 1n. In contrast, a full transfer allows an excitation to be swapped. Unless otherwise stated, excitation transfers evoked in the present description may correspond to partial or full transfers.

(17) The present methods all rely on first-order transitions, to enable a more efficient process. For instance, so-called red sideband energy transitions can be contemplated, as later described in detail.

(18) Such transitions may involve excited states of the quantum circuits that differ from each other, in energy, so as to make them more easily addressable. For example, in that case, the frequency of the tunable coupler 20 needs to be modulated at corresponding frequencies, so as to drive the desired transitions in the quantum processing device 1, 3, 4. To that aim, suitable frequencies need to be identified and applied, e.g., through a control line 30.

(19) In varied implementations, the excited states of the quantum circuits involved may have close or, even, identical energies. For example, distinct circuits may have excited states of close or identical energies. In that case, the energy transitions signals need to be selectively targeted using some other means. It may for instance be possible to have identical energies and drive all transitions at once. Typically, what happens in such a case is that the final state will oscillate between the initial state and the all-quantum-circuit superposition state, with weights given by the couplings of the individual circuits. To achieve a certain gate, for example, the couplings have to be either set in design or made tunable. In the present case, the amplitudes and/or phases of the applied signal can be chosen, so as to drive the desired first-order, energy transitions, as if distinct frequencies were used (although cross-talk issues might slightly impair the process in that case, it is not a problem).

(20) At least one tunable coupler 20 is involved. Still, two or more tunable couplers may possibly be needed, e.g., in an array of cells, where each cell contains one coupler and two or more quantum circuits, as in embodiments of the invention described later in reference to FIG. 5. In general, the present processing devices 1, 3, 4 may comprise a large array of quantum circuits and couplers. Such devices may, in general, be embodied as quantum enhanced sensing devices or quantum computing devices.

(21) The present quantum circuits 11-14, 1n are meant to be used as computational qubits (they are accordingly often referred to as qubit in the following). On the contrary, the tunable coupler 20 is not meant to be used as a computational qubit, even if the coupler 20 can be embodied as a qubit, as exemplified latter. Rather, the role of the coupler 20 is merely to couple the quantum circuits, to achieve the desired transitions.

(22) State-of-the-art methods are known, which rely on interactions between two qubits, or between one qubit and a resonator. However, the present architecture involves at least two qubits coupled to a tunable coupler according to embodiments of the invention. Yet, the present coupler 20 has a tunable frequency and is not meant to be used as a computational qubit by the quantum processing device 1, 3, 4. On the contrary, each of the computational quantum circuits (e.g., qubits) 11-14, 1n has a fixed frequency (hence not tunable). This, as one may realize, allows the device 1, 3, 4 to be substantially less sensitive to flux-noise. As such, longer coherence time of the fixed frequency devices can advantageously be exploited, e.g., to implement longer and/or more complex algorithms.

(23) The present approach allows excitations to be swapped between two or more coupled qubits or partly transferred from one of the coupled qubits to the others, by suitably modulating the frequency of the tunable coupler. As first-order energy transition are involved, the present approach notably allows a number of fixed frequency qubits to be entangled with a single, short operation. This approach can notably be used to concomitantly transfer excitations (i) from one quantum circuit to the coupler and (ii) from the coupler to other quantum circuits, which has tremendous advantages in terms of operation speed. Doing this with suitable pulse shapes (e.g., having same amplitudes) results in transferring excitations from one qubit to the others, leaving the tunable coupler 20 unpopulated. Even more so, by simultaneously driving the desired transitions to other qubits, the two-step operation can be completed as fast as a single qubit-to-coupler transfer.

(24) By choosing appropriate frequencies, amplitudes, phases pulses of the drives, excitations of different multi-qubit states can be created. Not only is the high-coherence of fixed frequency qubits exploited, but, in addition, concomitant drives makes it possible to achieve high gate rates (e.g., in the 10-100 MHz range), with corresponding gate times in the 10 ns-100 ns range, i.e., as fast as single qubit gates. In other words, single entangling operations can possibly be achieved within 10 ns.

(25) Thus, the present approach can notably be used to enable shorter state preparation times. This, in turn, makes it possible for a quantum computer or a quantum sensor to perform with less error and to be scaled to a larger number of qubits.

(26) All these aspects and related advantages are discussed in detail in the following, in reference to particular embodiments of the invention.

(27) Referring more particularly to the flowcharts of FIGS. 9, 10, the frequency of the tunable coupler 20 is, in embodiments of the invention, modulated (S20, S20a, S20b) at two frequencies. That is, the coupler is, on the one hand, modulated at a first frequency, so as to drive (S30, S30a) a first transition. This results in transferring an excitation of one of the quantum circuits 11-14, 1n to the tunable coupler 20. On the other hand, the coupler 20 is modulated at one or more second frequencies (distinct from the first frequency), so as to drive (S30, S30b) one or more second transitions, respectively. The modulation frequencies should be at, or close to, the transition frequencies in practice. This way, the excitation transferred to the tunable coupler can be further transferred from the tunable coupler 20 to one or more of the remaining quantum circuits 11-14, 1n.

(28) Still, each of the first and the one or more second transitions relied on is a first-order energy transition, so as to ensure speed of operation. In contrast, higher-order energy transitions require longer times to complete and, this, with lower probabilities.

(29) As illustrated in FIG. 10, the process may involve two successive transitions. That is, the frequency of the tunable coupler 20 may first be modulated (S20a) at a first frequency (to drive (S30a) the first transition). Then, the coupler 20 is subsequently modulated (S20b) at one or more second frequencies, so as to drive (S30b) respective, second transitions after the first transition. Here, the two successive transitions involve a certain time gap in between. Such an approach makes it possible to easily control the applied signals, but it impacts the corresponding gate times.

(30) Now, surprisingly enough, using concomitant drives allows similar effects to be achieved, as present inventors have realized. Thus, in particular implementations (which may be preferred but are not a necessity) such as illustrated in FIG. 9, the two drives overlap, timewise. In other words, the modulations used to drive the two transitions are concomitantly applied, so as to reduce the operation time. Here, concomitant means that the time period during which the tunable coupler is modulated at the second frequencies at least partly overlaps with the time period during which tunable coupler is modulated at the first frequency. The modulations may even be simultaneously applied, in order to accelerate the transfers.

(31) Namely, the frequency of the tunable coupler 20 can be concomitantly (or, even, simultaneously) modulated (S20) at a first frequency and at one or more second frequencies, so as to concomitantly (or simultaneously) drive (S30) the first transition and the one or more second transitions. Such a mode of operation is particularly advantageous for creating fast multipartite entanglement of several quantum circuits 11-14, 1n. The concomitant (e.g., simultaneous) modulation at two or more frequencies may give rise to a collective, resonant interaction between the qubits and the tunable coupler 20.

(32) In practice, the operations (S20-S40 (or S20a-S40a)) depicted in FIGS. 9 and 10 are repeatedly performed (S50-S10), each cycle aiming at modifying a state of the quantum processing system. This is typically achieved by applying (S10) harmonic microwave signals (i.e., coherent tones) to the tunable coupler 20, e.g., via a controller 32 and a control line 30. The signals applied are themselves modulated, in order to modulate (S20, S20a, S20b) the frequency of the tunable coupler 20. This, in turn, makes it possible to drive (S30, S30a, S30b) the desired first-order transitions.

(33) Timers may be associated to the applied signals, at the level of the controller. At the end of a cycle, the signals applied are interrupted (S50) and a new cycle may then start (S10), as determined by the controller 32.

(34) As mentioned earlier, red sideband transitions are preferably (but not necessarily) involved, whereby such transitions correspond to energy differences which are, each, smaller than the energy difference corresponding to the frequency of the tunable coupler. Each of the first and second transitions involved may be a first-order, red sideband transition, as assumed in FIGS. 6 and 8. First-order transitions allow improved operation times and process efficiency, compared to second-order or higher-order transitions. This can be achieved by leveraging specific excited states as described now. Namely, each frequency at which the tunable coupler 20 is modulated corresponds to a difference of energy between a pair of excited states of the quantum processing device. Yet, each pair of excited states may, in embodiments of the invention, involve a same excited state of the system constituted by the circuits 11-14 and the coupler 20. This same excited state, referred to as a reference excited state, arises from an excited state |1 of the tunable coupler 20. This reference excited state is noted |gg . . . , 1). In other words, the reference excited state is representable as a tensor product of the excited state |1 of the tunable coupler 20 (e.g., the first excited state thereof) and a ground state |g) of each of the quantum circuits 11-14.

(35) Each frequency may for instance correspond to a difference of energy between this reference excited state |gg . . . , 1 and another excited state, noted |ab . . . , 0, arising from an excited state |e of one (only) of the quantum circuits 11-14. In other words, the state |ab . . . , 0 is, this time, representable as a tensor product of the ground state |0 of the tunable coupler 20 and a state |ab . . . of the quantum circuits 11, 12, . . . and so forth. That is, |ab . . . , =|ab . . . .Math.0. Now, |ab . . . , 0 involves the excited state |e of exactly one of the quantum circuits 11, 12, . . . and so forth, and the ground state |g of each of the remaining quantum circuits 11, 12, . . . and so forth. In other words, only one of the states |a, |b, . . . and so forth corresponds to an excited state |e in the product |a.Math.|b.Math. . . . and so forth.

(36) Consider for example a simple architecture, based on two quantum circuits 11, 12 (computational qubits Q1, Q2), as in FIGS. 1 and 2. A corresponding level diagram is shown in FIG. 6. Each modulation frequency .sub.ge1 and .sub.ge2 corresponds to a respective energy difference, which nevertheless involves a reference excited state |gg, 1. The first modulation frequency .sub.ge1 corresponds to the difference between the energy of |gg, 1 and the energy of the excited state |eg, 0, the latter involving only one excited state e), i.e., the excited state of the first quantum circuit (circuit 11, or Q1). Consistently, the second modulation frequency .sub.ge2 corresponds to the difference between the energy of |gg, 1 and the energy of the excited state |ge, 0, which solely involves the excited state |e of the other quantum circuit (Q2). The frequency of the tunable coupler, .sub.tc, is determined by the difference between the energy of the excited state |1 and the ground state |0 of the coupler.

(37) More generally, when more than two circuits 11, 12, . . . and so forth are involved, the reference excited state involved is |gg . . . , 1, while the other excited states, i.e., |ge . . . , 0, |eg . . . , 0, . . . and so forth, arise from first excited states of the basis quantum circuits 11, 12, . . . and so forth yet involving one (and only one) excitation at a time. In other words, each element of the set {a, b, . . . } of sub-states of the quantum circuits corresponds to an excited state |e or a ground state |g of a respective quantum circuit. Because exactly one excited state of a given one of the quantum circuits is involved, the other quantum circuits are in their ground state.

(38) Based on excited states as described above and a suitable coupling of the quantum circuits, first-order, red sideband energy transitions can be achieved, which can easily be triggered by applying coherent microwave tones. Coherent microwave tones can for instance be applied thanks to a controller realized as a (coaxial, microwave) cable that creates a current and subsequently a magnetic field penetrating a squid loop, so as to modulate the frequency of the coupler, as latter discussed in reference to FIGS. 1-5.

(39) As defined earlier, that the first and second frequencies correspond, each, to a difference of energy means that such frequencies can be equal to such energy differences or, in variants, close to such differences of energy (and close enough, so as for the desired transitions to take place). That is, the driving of a possible transition can be resonant (i.e., the energy difference matches the drive frequency) or off-resonant (i.e., the energy is slightly off the drive frequency). Relying on slightly off-resonant frequencies, as in so called stimulated Raman adiabatic passage (STIRAP) schemes, results in the population of the reference excited state being avoided, because off-resonant frequencies are utilized for driving the transitions

(40) Beyond systems including a single pair of fixed frequency quantum circuits, the present methods can advantageously be applied to a quantum processing device 3 that comprises three, four, or more fixed frequency quantum circuits, as illustrated in FIG. 7 or 8. Again, each of the fixed frequency quantum circuits 11, 12, 13, . . . and so forth need to be coupled to a frequency-tunable coupler 20. Consistent with methods as described above, the frequency .sub.tc of the tunable coupler can be concomitantly modulated (S20), on the one hand, at a first frequency (so as to drive (S30) a first transition and transfer an excitation from one of the circuits to the tunable coupler) and, on the other hand, at two or more second frequencies, so as to concomitantly drive (S30) two or more second transitions, respectively. This way, excitations can be transferred (S40) from the tunable coupler 20 to two or more other ones of the quantum circuits. Again, each of said second transitions is a first-order energy transition and, preferably a red sideband transition as discussed above.

(41) As illustrated in FIG. 7, this approach can notably be used to entangle N fixed frequency superconducting transmon qubits with a single short operation. The common tunable coupling element may for instance be embodied as a frequency-tunable, superconducting qubit, i.e., a SQUID loop. As further depicted in FIG. 7, an entangled state between the computational qubits can be created in a single step by simultaneously applying coherent microwave tones (wave arrows) to the tunable coupler via a single drive line.

(42) For example, and as illustrated in FIG. 8, a red sideband transition can be induced by modulating the frequency of the tunable coupler with a coherent microwave tone set at the difference frequency between the excited state |gggg,1 of the tunable coupler and the excited state |eggg,0 of qubit 1. This transition transfers an excitation from the qubit to the tunable coupler with, for example, a (time-dependent) rate Q1. By simultaneously modulating the tunable coupler with additional coherent microwave tones set at the difference frequencies between |gggg,1 and the first excited states arising from the remaining qubits, this excitation can then be shared between all qubits. The resulting quantum state is a multipartite entangled state, where entanglement is shared between multiple qubits.

(43) In the example of FIG. 8, the tunable coupling element is assumed to capacitively couple to all computational, transmon qubits. As a consequence, the gate rates i of the red sideband transitions will be on the order of the (capacitive) couplings between the transmon qubits and the tunable coupler. This, given that first-order transitions are relied on, makes it possible for the gate rates to be as high as 100 MHz.

(44) By driving all red sideband transitions simultaneously (for qubit 1, qubit 2, qubit 3, and qubit 4 in FIG. 8), a multipartite entangled state can therefore be realized within 10 ns, i.e., as fast as a single qubit operation. In light of the high coherence of the fixed frequency transmon qubits, such entangled multi-qubit states are also generated with very high fidelities (approaching unity). As a result, the present approach can be used to mitigate problems arising from sequential state preparation (as discussed in the background section), according to embodiments of the invention. This approach can thus allow longer and more complex algorithms to be implemented within the coherence time of the fixed frequency qubits.

(45) Interestingly, the approach proposed here is applicable for arbitrary frequency differences between the tunable coupler and the transmon qubits. This is particularly appealing, in as much as larger frequency differences (>1 GHz) between the tunable coupler and transmon qubits would be essential or used to address the specific problem of frequency crowding, which is particularly prominent in the frequency range below 1 GHz.

(46) As discussed earlier, gate operations as contemplated herein may further require suitably setting amplitudes and phases of the signals applied to the tunable coupler 20, in addition to modulating frequencies, to trigger the desired first-order transitions. With carefully selected amplitudes and phases, one may simultaneously drive at the difference frequencies between the remaining N 1 qubits and the coupler, so as to populate additional excited states arising from the remaining qubits. This way, an initial excitation can be shared among all qubits, which can be exploited to achieve (S20-S40) a multipartite entangled state.

(47) Practically, operations described herein (e.g., steps S20, S20a, S20b, in FIGS. 9, 10) can be carried out so as to implement quantum gates. Various types of quantum gates can be envisioned, in the context of embodiments of the invention. At a minimum, the present approach can be exploited to implement a quantum gate reversibly operating on at least one of the quantum circuits 11-14, 1n. In other words, this gate can be described by a unitary matrix. More generally yet, quantum gates can be implemented, which operate on states of any subset, or even all of the quantum circuits 11-14, 1n. However, because the tunable coupler is, here, not meant to be operated as a computational qubit, such a quantum gate will not operate, effectively, on the tunable coupler 20.

(48) That a gate operates on states of one or more of the quantum circuits but not on the coupler means that the frequency of the tunable coupler 20 is modulated, so as to transfer excitations from one circuit to the other(s). The state of the tunable coupler 20 may possibly be altered in the process. However, after the gate operation, the tunable coupler is back to its initial state (identity operation). Thus, the gate does, effectively, not operate on the tunable coupler.

(49) The frequency of the tunable coupler 20 may notably be modulated (S20, S20a, S20b) so as to implement (S20-S50) a universal, finite set of gates operating on states of two or more of the quantum circuits 11-14, 1n, e.g., in combination with single qubit gates. For example, this universal set of gates enables a set of operations to be performed with the quantum processing device 1, 3, 4, where such operations can be reduced (or, possibly, approximated) by a sequence of gates from this finite set. In other words, operations enabled by the device 1, 3, 4 can be translated into a finite sequence of gates from this universal set.

(50) Implementing such a universal set of gates may further require suitably setting amplitudes and phases of the signals applied to the coupling elements 20, in addition to merely modulate its frequency, as discussed earlier. However, none of these gates will effectively be based on states of the tunable coupler 20. In particular, appropriate frequencies, amplitudes and phase pulses of the drives can be chosen, so as to achieve an iSWAP gate, a c-Phase gate, a Fredkin gate (also known as CSWAP gate, or a square root of SWAP gate (commonly denoted by {square root over (SWAP)}). Such gates are known per se. Interestingly here, such gates can be operated on states of one or more of the fixed frequency quantum circuits, to exploit their longer coherence time. Any combination of the examples of gates cited above can in fact be contemplated, e.g., to form a universal, finite set of gates.

(51) Another interesting feature of the present approach is that the drives at the first and second frequencies may be switched on (when modulating the frequency of the tunable coupler) at paces chosen so as to generate an adiabatic holonomic gate transformation. In such a transformation, populating the excited state of the tunable coupler is strongly suppressed during and after applying the pulses, assuming the pulses are slow and not applied simultaneously. A non-adiabatic holonomic gate, in contrast, uses fast pulses simultaneously. In this case, the excited state is populated during, but not after the pulses. Also, the adiabatic transfer implies a modulation frequency that is slightly detuned from the transition frequencies, whereas in embodiments of the invention as otherwise described herein, the modulation frequencies should in general be at or close to the transition frequencies.

(52) Referring to FIGS. 1-5 and 7, another aspect of embodiments of the invention is now discussed, which relates to quantum processing devices 1, 3, 4. Main aspects of such devices have already been discussed in respect of the present methods. Therefore, such devices are only briefly described in the following.

(53) At a minimal, such devices 1, 3, 4 comprise two fixed frequency quantum circuits 11-14, 1n, in addition to a tunable coupler 20, to which the quantum circuits are coupled. Again, the quantum circuits are meant to be used as computational qubits, contrary to the coupling element 20. The present devices, however, may comprise more than two quantum circuits and more than one coupler 20, as in FIG. 5. In addition, such a quantum processing device 1, 3, 4 includes a controller 30 (of control system 32). The controller 30 is configured in the processing device 1, 3, 4 to modulate a frequency of signals applied to the tunable coupler 20, so as to drive first-order energy transitions in the quantum processing device 1, 3, 4. As explained earlier, this makes it possible for an excitation of one of the quantum circuits to be transferred (at least partly) to one or more of the remaining circuits, via the tunable coupler 20, in operation.

(54) The controller/control system 32 will typically include (or be implemented as) a control line 30 (i.e., a physical wire), to apply the signals. The controller allows multi-frequency signals to be applied and, if necessary, to suitably set amplitudes and phases of the applied signals. The control system 32 is configured to apply the signals discussed herein via one or more control lines 30 as understood by one skilled in the art.

(55) In particular, the controller having control lines 30 may be configured to modulate the frequency of the coupler 20 at a first frequency (to drive a first transition and thereby transfer an excitation from one quantum circuit to the coupler 20) and at one or more second frequencies, each distinct from the first frequency. This way, one or more second transitions can be driven, to further transfer this excitation from the coupler 20 to one or more of the remaining quantum circuits, consistently with principles described earlier. Again, each of the first transition and second transitions is a first-order energy transition, e.g., a red sideband transition.

(56) The controller having control lines 30 may notably be configured to impose a certain time gap between the various modulations applied. Preferably (but not a necessity), though, the controller may concomitantly modulate signals applied to the tunable coupler 20, so as to concomitantly drive first and second transitions, as explained earlier.

(57) As illustrated in FIGS. 3, 5 and 7, embodiments of the present quantum processing devices 3, 4 may involve more than two, e.g., three, four, six or more fixed frequency quantum circuits 1n, where each circuit is coupled to the tunable coupler 20. The controller having control lines 30 is further adapted to concomitantly modulate the frequency of the coupler 20 at a first frequency (to drive a first transition) and at two or more second frequencies, so as to concomitantly drive two or more second transitions. Thus, excitations can be transferred from the tunable coupler 20 to at least two of the remaining quantum circuits 1n, in operation.

(58) Preferably (but not a necessity), the present quantum circuits 11-14, 1n are embodied as superconducting quantum circuits, forming the basis computational elements. In particular, the present circuits can be fixed frequency, transmon-type quantum circuits, realized as single-Josephson junction superconducting qubits. This way, the processing devices 1, 3, 4 take advantage of long coherence times of the fixed frequency transmons. Note that the coupler 20 too may be realized by a transmon, although the coupler 20 will not form part of the basis computational elements 11-40, 1n.

(59) In all cases, the tunable coupler 20 may be capacitively coupled to each of the quantum circuits 11-14, 1n. For example, and as illustrated in FIG. 1, a processing device 1 may involve two fixed frequency, single-Josephson junction superconducting qubits 11, 12 with long coherence times (e.g., transmon-type devices) as the basis computational elements and a tunable coupler 20. In this example, the tunable coupler 20 is laid out as a non-harmonic oscillator circuit, such as a qubit of the transmon type. Yet, the coupler is given an extra degree freedom, by way of an additional Josephson junction () having an X symbol, which forms a superconducting quantum interference device (SQUID) loop. In turn, the SQUID loop can be used to tune the coupler's frequency. In the example of FIG. 1, the modulation of the tunable coupler frequency is realized by altering the magnetic flux (t) penetrating the SQUID loop formed by the two Josephson junctions via a current I(t) passed through an inductively coupled flux bias line 30. The modulation of the flux induces a modulation of the coupling strength J(t)=J cos(t). Each qubit 11, 12 is capacitively coupled () (having a parallel lines symbol) to a readout resonator 41, 42 (i.e., also denoted by R1 and R2 in FIG. 1) and a charge bias line (not shown) for single qubit gate operations.

(60) The coupling is performed by modulating the frequency of the tunable coupler 20 near or at the frequency corresponding to first-order energy transitions, e.g., red sideband transition as illustrated in FIG. 6. This, in turn, allows a two-qubit quantum gate implementation.

(61) The coupler can be regarded as formed by one or more electrodes, as well as the SQUID loop (the SQUID loop containing the two Josephson junctions). The tunable coupler 20 may for instance couple to the two (or more) transmons 11, 12, via separate electrodes (the latter separated by the SQUID loop) or via a single electrode, with the SQUID loop coupling to the ground. FIG. 2 shows example geometry for two transmon-type qubits 11, 12 coupled via a third transmon 20 acting as a frequency-tunable coupling element. External magnetic flux applied via the inductively-coupled bias line 30 threads the SQUID loop at the center of the tunable coupler 20 and modifies its frequency by modifying the effective Josephson inductance. The symmetric placement of the flux-bias line 30 inhibits the decay of the coupling transmon 20 by minimizing the capacitive coupling to the flux-line 30.

(62) The tunable coupler 20 can be designed to couple capacitively to more than two qubits, as shown in FIG. 3. The example architecture of FIG. 3 is essentially similar to that of FIG. 1, except that an additional pair of qubits 13, 14 (i.e., Q3, Q4) now mirrors the first pair 11, 12 (i.e., Q1, Q2). Each additional qubit 13, 14 is again capacitively coupled to a readout resonator 43, 44 (i.e., R3 and R4 in FIG. 3) and a charge bias line (not shown). In this example too, the modulation of the tunable coupler frequency can be achieved by altering the magnetic flux (t) penetrating the SQUID loop via a current I(t) passed through the inductively coupled flux bias line 30. This, in turn, induces a modulation of the coupling strength, which is exploited to trigger first-order energy transitions, according to the present methods. Similarly, the example geometry of FIG. 4 is similar to that of FIG. 2, except for the additional pair of qubits 13, 14. In addition, each qubit is now rotated with respect to a previous qubit, and the geometry of the coupler is adapted to the four-qubit configuration.

(63) Another aspect of embodiments of the invention is now described in reference to FIG. 5, which relates to a quantum processing chip 4. The quantum processing chip 4 comprises a plurality of cells 60, each comprising two or more fixed frequency quantum circuits 1n, as well as a tunable coupler 20, to which the two or more quantum circuits 1n are coupled. In the example architecture of FIG. 5, the cells form a 2D lattice. Yet, the central square delimited by the four inner qubits does not correspond to a cell in this example, as they are not coupled to a same tunable coupler. Each of the four inner qubits bridges one cell to a contiguous cell, such that the cells effectively comprise, each, three qubits on average. Yet, the depicted architecture allows four qubits to be operated per cell. Each cell 60 is otherwise similar to the four qubit arrangement depicted in FIG. 3. In the architecture of FIG. 5, all nearest neighbor qubits can be coupled (except for qubit pairs on the central square). This architecture, however, can be modified to include a coupler at the center to increase the connectivity between the qubits.

(64) In FIG. 5, a controller system (is, includes, and/or is connected to a control line 30 in a loop) is used to modulate a frequency of signals applied to each tunable coupler 20, e.g., via respective control lines 30, so as to drive first-order energy transitions in selected cells. In other words, cells can be independently operated, via a respective control line 30. However, the operation of the cells follows a logic implemented by the controller system (such as the control system 32 in FIG. 7), e.g., so as to enable a sequence of gates, which in turn allows more complex operations to be performed by the quantum processing chip 4.

(65) For each operated cell 60, an excitation of one of the quantum circuits 1n can be transferred to one or more of the remaining circuits 1n of the cell, via a respective tunable coupler 20, in operation.

(66) While the embodiments of the present invention have been described with reference to a limited number of variants and the accompanying drawings, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the scope of embodiments of the present invention. In particular, a feature (device-like or method-like) recited in a given embodiment of the invention, variant or shown in a drawing may be combined with or replace another feature in another embodiment of the invention, variant or drawing, without departing from the scope of embodiments of the present invention. Various combinations of the features described in respect of any of the above embodiments of the invention or variants may accordingly be contemplated, that remain within the scope of the appended claims. In addition, many minor modifications may be made to adapt a particular situation or material to the teachings of embodiments of the present invention without departing from its scope. Therefore, it is intended that description not be limited to the particular embodiments of the invention disclosed herein, but that the will include all embodiments of the invention falling within the scope of the appended claims. In addition, many other variants than explicitly touched above can be contemplated.