Beam positioning and control for quantum computing
By dividing the ion array into multiple groups and using optical tweezers and excitation beams to isolate adjacent groups, the problem of heating and coupling motion modes in trapped ion systems is solved, enabling flexible quantum computing operations and efficient qubit coupling.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- QUANTUM TECH UG GMBH
- Filing Date
- 2024-10-29
- Publication Date
- 2026-06-05
AI Technical Summary
In trapped ion systems, as the number of ions in the array increases, the heating of ion motion modes caused by electric field noise increases rapidly, and the frequency spacing between adjacent motion modes becomes closer, making it difficult to resolve and drive specific modes for high-fidelity qubit coupling.
By dividing the ion array into multiple groups of adjacent ions and using reconfigurable optical tweezers and excitation beams to isolate the ion groups and limit the interaction between adjacent groups, the intensity and position of the beam are modulated using optical devices such as multi-channel acousto-optic modulators and spatial light modulators to achieve flexible quantum computing operations.
It effectively isolates adjacent ion groups, reduces heating of motion modes, improves the resolvability of qubit coupling and the flexibility of computational operations, and supports the realization of large-scale quantum computing.
Smart Images

Figure CN122162144A_ABST
Abstract
Description
[0001] Cross-reference to related applications This application claims the benefit of U.S. Provisional Patent Application 63 / 595,349, filed November 2, 2023, which is incorporated herein by reference.
[0002] field This disclosure relates to the field of quantum computing, and more specifically to the control and manipulation of qubits in quantum computer systems.
[0003] background Quantum computers apply the principles of quantum physics to solve computational problems and have the potential to perform certain calculations more efficiently than existing digital computers. The basic building block of a quantum computer is the qubit. Quantum computers use qubits and gates that operate on them (including single-qubit, two-qubit, and multi-qubit gates) to perform digital quantum computations and analog quantum simulations.
[0004] In trapped ion systems, individual atomic ions act as qubits, and these systems hold promise as a scalable and reliable platform for quantum computing and quantum simulation. In these systems, individual atomic ions are typically trapped by electromagnetic fields in an ultra-high vacuum and cooled to their motional ground state. The internal electronic energy levels of the ions and their motion within the trap are controlled with high precision using laser, microwave, or radio frequency (RF) fields. To perform digital quantum computing, gates are applied to the internal and motional states of the atomic ions by driving fields of appropriate frequency, amplitude, and duration.
[0005] In trapped ion systems, entanglement gates are typically generated by driving ions with electromagnetic fields, which produce phonon-mediated qubit-qubit interactions. For example, aspects of driving multi-qubit gates in trapped ion arrays are described in PCT International Publication WO 2023 / 105434, the disclosure of which is incorporated herein by reference.
[0006] In this specification and claims, the term “amplitude” as used with respect to a radiation beam refers to complex amplitude, which includes the amplitude, frequency and phase of the radiation.
[0007] Overview The embodiments of the present invention described below provide improved systems and methods for performing quantum computing using trapped ion arrays.
[0008] Therefore, according to embodiments of the present invention, an apparatus for quantum computing is provided, comprising an ion trap configured to hold a first array of ions at corresponding positions along an array axis. A radiation source is configured to emit a second array of beams of radiation, the beams comprising a first beam and a second beam, the first beam having a corresponding first intensity and a frequency selected to excite selected internal transitions of ions, the second beam having a second intensity at least ten times greater than any of the first intensities, and the radiation source being configured to switch the corresponding positions of the first and second beams within the second array. Optics are configured to focus the beams into the ion trap such that each beam in the second array is incident on a corresponding ion in the first array.
[0009] In the disclosed embodiments, the second strength is at least one hundred times greater than any of the first strengths.
[0010] In some embodiments, the second beam is configured to act as an optical tweezer to confine the ions incident upon it, while the ions incident upon the first beam serve as components of the computation segment. In the disclosed embodiments, a second array of beams is configured such that a subset of the second beam is incident upon a corresponding set of ions, whereby the corresponding set is further confined by the optical tweezers.
[0011] Additionally or alternatively, the computation segment includes a single-qubit gate or a multi-qubit gate, each multi-qubit gate including a corresponding group of two or more ions incident to the first beam within the computation segment. In some embodiments, the first and second beams are arranged to define multiple single-qubit gates or multi-qubit gates to which the first beam is incident, wherein the multi-qubit gates are separated from each other by confined ions incident to the second beam. In the disclosed embodiments, the radiation source is configured to redefine the computation segment by switching the corresponding position of the second beam, and single-qubit gate operations or multi-qubit gate operations are applied by the first beam within the redefined computation segment.
[0012] In some of these embodiments, switching the corresponding position of the second beam includes continuing to irradiate the first set of ions with the first set of the second beam while simultaneously initiating irradiation of the second set of ions with the second set of the second beam. In one embodiment, the radiation source is configured to perform a mid-circuit ancilla measurement by applying a second beam from the second set to at least one ion in the first set. Additionally or alternatively, the radiation source is configured to perform cooling and initialization operations on at least one ion in that set. In another embodiment, the radiation source is configured to adiabatically initiate irradiation of the second set of ions with the second beam and terminate irradiation of ions in the first set.
[0013] In the disclosed embodiments, the device includes a third array of detectors configured to sense emitted radiation by detecting the state of qubits, wherein the emitted radiation indicates the corresponding state of the qubits, and wherein optics are configured to direct the emitted radiation from the ion trap onto the detectors.
[0014] In some embodiments, the radiation source includes at least one laser and a splitter, the at least one laser being configured to output coherent radiation and the splitter being configured to split the coherent radiation into multiple beams. In one embodiment, the splitter includes a diffractive optical element (DOE). Additionally or alternatively, the splitter includes a spatial light modulator (SLM). In the disclosed embodiments, the SLM is configured to be divided into at least a first segment and a second segment, wherein each segment is configured to generate different arrangements of corresponding positions of the first and second beams within a second array, and wherein the radiation source includes an actuator configured to switch the coherent radiation incident on the SLM between the first and second segments.
[0015] Additionally or alternatively, the splitter includes a multi-beam acousto-optic deflector (AOD). In some embodiments, the device includes a controller configured to adjust the AOD to correct alignment errors between the beam and ions in the trap. In one embodiment, the device includes a third array of detectors configured to sense radiation emitted by the ions, wherein the controller is configured to detect alignment errors in response to the sensed radiation.
[0016] In some embodiments, the radiation source includes at least one multi-channel acousto-optic modulator (mcAOM) configured to modulate the amplitude of at least one first beam. In one embodiment, the at least one mcAOM is configured to apply different corresponding frequency shifts to different first beams in the first beam, optionally together with additional global or local beams, to drive corresponding Raman transitions of ions incident on the first beam. Additionally or alternatively, the at least one mcAOM includes at least a first mcAOM and a second mcAOM configured to modulate different corresponding groups of beams in a second array.
[0017] Additionally or alternatively, at least one mcAOM is configured to modulate the respective amplitudes of the first and second beams. In the disclosed embodiments, at least one mcAOM is configured to attenuate the first beam such that the second intensity is at least ten times greater than the first intensity.
[0018] In some embodiments, the first beam has a corresponding first frequency within a selected frequency range, and the second beam has a corresponding second frequency outside the selected frequency range. In the disclosed embodiments, the radiation source includes a dichroic beam combiner that aligns the first and second beams to form a second array.
[0019] According to embodiments of the present invention, a method for quantum computing is also provided, the method comprising confining a first array of ions at corresponding positions along an array axis in an ion trap. A second array of beams generating coherent radiation, the beams comprising a first beam and a second beam, the first beam having corresponding first intensities and frequencies selected to excite selected internal and kinematic transitions of the ions, the second beam having a second intensity at least ten times greater than any of the first intensities. The beams are focused into the ion trap such that each beam in the second array is incident on a corresponding ion in the first array. A first quantum computing operation is performed using a first configuration of the first and second beams in the first array. A second quantum computing operation is performed using a second configuration of the first and second beams, wherein corresponding positions of at least some of the first and second beams are switched within the second array.
[0020] According to an embodiment of the invention, a quantum register is further provided, comprising sets of ions at corresponding positions along an array axis within a first array in an ion trap. The ions are configured to perform quantum operations under the control of a radiation source emitting a second array of beams of coherent radiation, the beams comprising a first beam and a second beam, the first beam having a corresponding first intensity and a frequency selected to excite selected internal transitions of the ions, the second beam having a second intensity at least ten times greater than any of the first intensities, and the radiation source switching the corresponding positions of the first and second beams within the second array, and optics focusing the beams into the ion trap such that each beam in the second array is incident on a corresponding ion in the first array.
[0021] Typically, the first beam of driving ions is used to perform single-qubit or multi-qubit operations.
[0022] The invention will be more fully understood from the following detailed description of embodiments thereof, taken in conjunction with the accompanying drawings, in which: Brief description of the attached diagram Figure 1 This is a schematic block diagram illustrating a quantum computing system according to an embodiment of the present invention; Figure 2 This is a block diagram schematically illustrating a trapped ion array configured as a quantum bit in a quantum computer according to an embodiment of the present invention; Figure 3A and Figure 3B This is a schematic front view of a laser beam array with two different configurations according to an embodiment of the present invention, which is used to drive a plurality of single-qubit gates and multi-qubit gates in a trapped ion array; Figure 4This is a schematic side view of a multi-beam optical trapping and excitation subsystem used in a quantum computing system according to an embodiment of the present invention; Figure 5 This is a schematic detailed view of a multi-beam generation and modulation module according to an embodiment of the present invention; Figure 6A and Figure 6B These are respectively, according to embodiments of the present invention. Figure 5 Schematic side and top views of the multichannel acousto-optic modulator used in the module; Figure 7 This is a schematic detailed view of a multi-beam generation and modulation module according to another embodiment of the present invention; and Figure 8 This is a schematic side view of a multi-beam optical trapping and excitation subsystem used in a quantum computing system according to an alternative embodiment of the present invention.
[0023] Detailed description Overview Practical applications of quantum computing require large multi-qubit registers. Trapped ions are well-suited for this purpose. For example, ion arrays can be used to form quantum registers with hundreds or even thousands of qubits, preferably arranged at equal intervals and held in traps (e.g., linear RF Paul traps or arrays of multiple interconnected linear traps) along the array axis. Such arrays are commonly referred to as ion crystals.
[0024] However, significant challenges remain in implementing this approach. As the number of ions in the array increases, the heating of ion motion modes caused by electric field noise increases rapidly. Furthermore, as the array size increases, the frequency spacing between adjacent motion modes becomes more compact. Consequently, resolving and driving specific modes for high-fidelity qubit coupling becomes difficult.
[0025] The embodiments of the invention described herein address these problems by dividing the array into multiple groups of adjacent ions in a reconfigurable manner while maintaining connectivity between all ions in the crystal. Each group can then be used as a separate register for a corresponding single-qubit or multi-qubit gate. (Alternatively, some ions can be isolated to serve as ancilla-qubits for intermediate circuit measurements in fault-tolerant codes.) To prevent interactions between ions in different groups, adjacent ion groups are separated from each other by a barrier comprising an optically confined, reconfigurable set of ions. During circuit operation, any ion in the entire crystal can be reconfigured as barrier ions using "optical tweezers," which are high-intensity, focused light fields that powerfully hold the barrier ions.
[0026] To adequately limit the intensity, the laser beam used as optical tweezers is typically at least ten times stronger than the beam used to excite ions in a multi-qubit gate, and can be at least one hundred times stronger or greater. Thus, barrier ions isolate adjacent ion groups in the array from each other by suppressing vibrations that propagate from one group to the next. In some embodiments, for more efficient separation between adjacent gates, optical tweezers are applied to pairs of adjacent ions, or even to groups of three or more adjacent ions.
[0027] To enable flexible and efficient programming of quantum computers using such segmented ion arrays, it is desirable that the size and position of the computational segments be variable, allowing different kinds of gate operations to be performed in parallel, sequentially, or both. For this purpose, in embodiments of the invention, the positions of the tweezers bundle and the excitation bundle can be changed such that ions that were part of a group used in one computational segment of a phase of computation become barrier ions in the next phase, and other ions that were barrier ions in one phase become components of a computational segment in another phase. For example, a series of computational schemes for realizing such an exchange between logic ions and barrier ions are described in U.S. Provisional Patent Application 63 / 490,007, filed March 14, 2023, the disclosure of which is incorporated herein by reference.
[0028] In the embodiments of the invention described herein, a radiation source (including a single laser or multiple lasers) of an array of coherent radiation beams is used to address and segment ions in the ion trap. Optical components focus the beams into the ion trap. In the embodiments described below, each beam in the beam array is incident on a corresponding ion in the ion array; however, alternatively, some or all beams may be incident on corresponding groups of two or more adjacent ions. The beam array includes an excitation beam and a tweezers beam, the excitation beam having an amplitude selected to excite selected internal and kinetic transitions of the ions, the amplitude including amplitude, frequency, and phase, and the tweezers beam having an intensity at least ten times greater than the intensity of any of the excitation beams. The radiation source includes optical components that allow the respective positions of some or all of the excitation beams and tweezers beams to be interchanged within the beam array, thereby enabling dynamic changes to the configuration of the ion groups constituting the computational segment throughout the ion array.
[0029] In this embodiment, in addition to controlling the beam intensity to generate the excitation beam and tweezers beam (and the excitation beam for operation on auxiliary qubits to realize intermediate circuit measurements), the radiation source also includes a modulator that modulates the corresponding amplitude of the excitation beam. (As previously mentioned, the term "amplitude" refers to complex amplitude, which includes the amplitude, frequency, and phase of the beam under discussion.) In some embodiments, the modulator includes at least one multichannel acousto-optic modulator (mcAOM) that applies different corresponding frequencies, amplitudes, or phases to some or all of the excitation beams to drive corresponding internal transitions of the ions to which the beam is incident. In the example embodiments described below, the excitation beam drives Raman transitions of the ions, i.e., transitions between internal atomic energy levels (with or without vibrational sidebands) of the ions' electronic transitions; therefore, the excitation beam is referred to as a Raman beam. Alternatively, some excitation beams may be applied to cause optical deflection of internal transitions to enhance optical control for realizing intermediate circuit measurement operations within a register. Additionally, the principles of the invention can be applied to drive other internal transitions and sidebands of qubit arrays.
[0030] In some embodiments, the mcAOM also modulates the tweezer bundle, for example, at a frequency relative to the excitation beam. As described above, the mcAOM can impose a significant attenuation on the excitation beam to produce the desired intensity difference between the tweezer bundle and the excitation beam.
[0031] In one embodiment, a spatial light modulator (SLM) can be used to split a laser beam while modulating the beam amplitude to produce a high-intensity tweezers beam and a low-intensity excitation beam.
[0032] In some embodiments, alignment errors between the beam and the ions in the trap can be corrected as follows: an array of detectors can be used to detect misalignment, which senses radiation emitted by the ions in the array, as further described below. The alignment error can then be corrected using an SLM or an acousto-optic deflector (AOD).
[0033] The embodiments described herein use certain types and arrangements of active and passive optical components to generate beam arrays input to ion traps. Alternatively, those skilled in the art will understand upon reading this specification that other components and configurations may be used to generate beam arrays (including excitation beams and tweezer beams) and to switch the position of beams within the array, and that these components and configurations are considered to be within the scope of this invention. Furthermore, the types of beam arrays described herein can be applied not only to arrays of trapped ions but also to other types of quantum computing media, such as arrays of neutral atoms.
[0034] System Description Figure 1 This is a block diagram schematically illustrating a quantum computing system 20 according to an embodiment of the present invention. System 20 is presented as a non-limiting example of an application environment that can utilize arrays of excitation beams and tweezers beams.
[0035] In this example embodiment, atomic source 22 injects a stream of neutral atoms, such as calcium atoms, into a high-vacuum chamber 26. Radiation source 28 directs several radiation beams into the vacuum chamber 26, including a beam tuned to ionize the atoms injected by source 22. (In this example, as described above, it is assumed that system 20 is based on electronic transitions, and that radiation source 28 comprises a laser emitting a coherent radiation beam; however, incoherent beams could alternatively be used, for example, to perform ionization detection and tweezers beams.) The resulting atomic ions are trapped in an ion trap 24 (e.g., a Paul trap), which confines the ions using RF and DC fields along a designated axis within the vacuum chamber 26. Magnetic fields may also be applied to the ion trap 24 to separate the different spin components of the ions' electronic states into Zeeman levels.
[0036] An electronic qubit control and computation processor 32 drives a radiation source 28 to direct an additional beam of light toward the trapped ions in order to perform quantum computation operations and then read out the results. Typically, the results are read out by tuning a laser beam to the absorption line of a qubit state and then measuring the resulting fluorescence emission using an optical detector 30. Each step in this example is defined by appropriate program code, multiple gates are used in parallel, and may include a redefined set of ions constituting a register on which the gates operate at each step. Alternatively, the set of ions can be defined and applied in a quantum simulation.
[0037] Figure 2 This is a block diagram schematically illustrating an array of trapped ions 40 of qubits configured in a quantum computer (e.g., system 20) according to an embodiment of the present invention. Radiation source 28 ( Figure 1 Several different laser beam inputs are provided to the ion trap 24 for different purposes. An ionizing laser 42 ionizes atoms output from the atomic source 22 to produce ions 40, which are held in the trap. An additional cooling laser 44 and a repumping laser 53 cool the ions to their electronic and motion ground states by pumping appropriate state transitions of the ions while detuning the laser frequency to produce mechanisms such as Doppler cooling, sideband cooling, polarization gradient cooling, cooling via electrically induced transparency (EIT), and / or other cooling methods known in the art. The repumping laser 53 is used to facilitate the return of ions trapped in metastable states to states where they can interact with the main laser system.
[0038] Cooled ions 40 are held in a linear array by an electromagnetic field along axis 38 within trap 24. The Coulomb repulsion and trapping field between ions 40 determine the equilibrium distance between them and the phonon frequencies of the normal vibrational modes (both transverse and longitudinal) of the ion 40's motion within the array. These normal vibrational modes generate vibrational sidebands at optical transition frequencies between the states of ion 40, which are used for quantum computing.
[0039] To define and manipulate multi-qubit gates (including two, three, or more qubits) and single-qubit gates, radiation source 28 ( Figure 1 The system includes an excitation source 46 comprising one or more lasers that direct radiation beams at multiple different frequencies to strike ions 40. These beams can be generated entirely by the same laser with appropriate amplitude, frequency, and phase modulation, or by multiple different lasers. Among these beams are a Raman beam 48 and a tweezers beam 50. The tweezers beam 50 is tuned to use a high-intensity optical field to confine selected ions 40, thus defining a multi-qubit register among the confined ions. The Raman beam 48 (also called the excitation beam) is tuned to coherently excite and manipulate the internal and motional transitions of the individual ions 40 within these multi-qubit registers, thereby driving gate operations for performing quantum computing. Furthermore, a global Raman beam 49 propagates at an angle relative to the Raman beam 48 to perform multi-qubit gate operations. For example, the global Raman beam can propagate in reverse relative to the individual Raman beams 48 to provide maximum photon momentum transfer to the ions in the multi-qubit gate.
[0040] Finally, the readout beam 52 and the shelving beam 51 are tuned to selectively excite internal transitions in ion 40 for the purpose of reading out gate states. The shelving beam 51 selectively drives the ion to a long-lived metastable state. When the ion is shelved in this metastable state, it will not emit photons in response to the readout beam 52 during the readout process. The readout beam 52 causes ion 40 to fluoresce, the intensity of which depends on the corresponding gate state. The resulting fluorescence emission is measured using an optical detector 30, and the calculation results are controlled by the qubit and computed by a computational processor 32. Figure 1 (Received). As part of this scheme, the operation sequence of the optical pulses can also be used for optical shifting, shelving, storage, and protection of data qubits from errors caused by photon scattering during the measurement sequence.
[0041] To operate multi-qubit gates in embodiments of the invention, various coherent manipulation techniques can be used, such as quadrupole optical transitions for photonic qubits or Raman beam transitions for hyperfine or Zeeman qubits. In this example, the Raman beam 48 is excited at a set of frequencies. 0± m Each group of ions was coherently irradiated with 40 excitation frequencies corresponding to selected internal transitions of the ions. Centered on 0. As described below, beam 48 is modulated, for example, by a suitable mcAOM, to coherently include the internal transition frequencies. Multiple sidebands of 0 m The frequency components in the qubit. For example, the beam 48 output by a laser operating at approximately 400 nm can be modulated by mcAOM to drive the ions at the frequency of the sideband of the S1 / 2 Zeeman splitting transition of calcium 40. The internal qubit level frequency ω0 is determined by the externally applied magnetic field B.
[0042] Raman beams 48 and 49 illuminate each ion 40 with selected frequency components at optimal amplitude within a gating time T to drive the multi-qubit gate from its initial state to an entangled target state. In one embodiment of the invention, individual modulation of Raman beam 48 allows each ion 40 to be driven with its own amplitude vector, which is typically different from the amplitudes applied to the other ions. Other embodiments may also use a global Raman beam or a semi-global Raman beam (within each segment) to operate the desired entangled gate within each register segment, or use a global beam that achieves local operation through individually operated optical shifts.
[0043] After the computation cycle is completed, the shelving beam 51 and the readout beam 52 are directed toward the ion 40 to read the state of the qubit register defined by the tweezers beam 50. The readout beam 52 is tuned to the absorption line of one of the states of the ions in the register. As described above, the absorption of laser radiation by the ions in the appropriate state results in fluorescence, which is detected by the optical detector 30 (…). Figure 1 The detector 30 measures the intensity of fluorescence emission to detect the final state of the gate.
[0044] Processor 32 typically includes a general-purpose computer with appropriate interfaces to other components of system 20. Processor 32 is software-driven to perform the functions and calculations described herein. This software may be stored on a tangible, non-transitory computer-readable medium, such as optical, magnetic, or electronic storage media.
[0045] Array of Raman beams and tweezers beams Figure 3A and Figure 3B This is a schematic front view of an array 60 of laser beams 62 in two different configurations according to an embodiment of the present invention, which is used to drive several multi-qubit gates in an array of trapped ions 40. The beams 62 are numbered sequentially from left to right in the figure. Optics in system 20 (as shown in the figure below) focus each beam 62 onto a corresponding ion 40 in trap 24. Although for simplicity... Figure 3A / Figure 3B Only twenty-five beams are shown in the diagram, but in practice, radiation source 28 can output a much larger number of beams simultaneously.
[0046] Beam 62 comprises a Raman beam with a low intensity 64 and a tweezers beam with a higher intensity 66. As previously mentioned, intensity 66 is typically at least ten times greater than intensity 64, and can be at least one hundred times greater or even more than one thousand times greater. In the example embodiment, each Raman beam delivers 50 μW of optical power to the vicinity of the corresponding ion in trap 24, while each tweezers beam delivers 300 mW of optical power. As further described below, embodiments of the invention utilize mcAOM and one or more beam-splitting components to achieve this large dynamic range while still maintaining precise control over all beams.
[0047] A multi-qubit register 68 is defined by ions excited by Raman beams of low intensity 64 in each group, which can be used for both single-qubit and multi-qubit gate operations. Register 68 includes groups of qubits that can be manipulated together to perform complex quantum computations. Conversely, beams 62 of high intensity 66 define potential barriers 70 between registers 68. These barriers 70 are used to limit the heating rate within the ionic crystal, reduce spectral congestion due to vibrational modes in the crystal, and isolate mechanical vibrations within the registers from each other, thereby ensuring that computations performed in one register do not interfere with computations in adjacent registers. Figure 3A and Figure 3B In the configuration shown, one or more tweezer bundles with a high intensity 66 are incident on the corresponding single ion, ion pair, or ion group, thereby further confining the ions locally by optical tweezers. The barrier 70 can be created by one or more tweezed ions as shown in these figures. Alternatively, the barrier can be created by a small number of tweezed ions plus one or more untweezed ions between some of the tweezed ions, which correspondingly reduces overall mechanical movement crosstalk mediated by the effective Coulomb force between adjacent segments due to the expanded separation.
[0048] Different registers 68 can each include the same number of adjacent ion qubits, or they can include any number of ions, including registers containing only one or two qubits. Figure 3A and Figure 3B In the example shown, the tweezers bundle is used to define the size of the multi-qubit registers. Raman beams are arranged to define multiple single-qubit and multi-qubit gates within these registers (with a low-intensity beam of 64 incident on them), and to mitigate residual computational crosstalk between the registers, which are separated from each other by trapped ions incident on a high-intensity beam of 66. Figure 3A and Figure 3BDuring the transition between configurations, the size and position of register 68 are redefined whenever needed by swapping some or all of the positions of the low-intensity beams and high-intensity beams in array 60.
[0049] For example, they can be executed sequentially. Figure 3A and Figure 3B The configuration shift between the two allows maintaining the initial positions of the tweezer bundles and applying a new set of tweezer bundles in addition to the existing bundles. In this new configuration, the old and new positions of the tweezer bundles are applied to create a new (second) bundle array. The new tweezer bundles create another confined set of ions within each register 68 to enable further quantum operations, such as single-operation intermediate circuit-assisted measurements performed within the second array. Furthermore, cooling operations can be performed by the redefined first bundle and / or other cooling bundles within the second array.
[0050] In this example, another step in the sequence of switching the corresponding positions of the second bundle is to switch the initial corresponding positions of the second bundle as follows: Figure 3B The new beam positions are shown to redefine the configuration of the multi-qubit ion segments. This transition can be performed in a controlled manner, such as by gradually withdrawing the first set of tweezers and gradually introducing the second set of tweezers during an adiabatic transition, or through faster operation techniques.
[0051] Optical subsystem Figure 4 System 20 according to an embodiment of the present invention ( Figure 1 A schematic side view of the multi-beam optical trapping and excitation subsystem 71 used in the ion trap 24. Subsystem 71 forms beams 62 and transmits these beams from excitation source 46 to ion trap 24 (e.g., Figure 3A / Figure 3B (As shown).
[0052] The optical subsystem 71 includes a splitter 72 that divides the coherent radiation output from one or more lasers in the excitation source 46 into multiple beams. The splitter 72 may include, for example, a diffractive optical element (DOE), a spatial light modulator (SLM), or a multi-beam deflector, such as an acousto-optic deflector (AOD) or a micromirror array, or may include multiple single-channel deflectors. The splitter 72 divides the coherent radiation into multiple beams at different angles. Depending on the type of splitter, these beams may have equal (or substantially equal) intensities, or they may have substantially different intensities depending on whether they are used as Raman beams or tweezer beams during a given time interval.
[0053] Following the splitter 72, the lens 74 guides the beam array onto a multichannel acousto-optic modulator (mcAOM) 76, which modulates the amplitude, frequency, and phase of each beam. Specifically, the mcAOM 76 applies different corresponding frequencies and phase shifts to different beams to drive corresponding Raman transitions in the ions incident on the first beam. The lens 74 can advantageously be configured as a Fourier transform lens, having the splitter 72 at its front focal plane and the mcAOM 76 at its rear focal plane. Thus, the lens 74 produces a spatial Fourier transform of the beams, where the deflection angle of each beam output from the splitter 72 is converted into a lateral position on the mcAOM 76, in the form of an array of equally spaced collimated beams.
[0054] The size of each collimated beam is scaled so that downstream optics will produce a focused beam on each ion, smaller than the distance between the ions. The mcAOM electrode design makes its acoustic column wider than each laser beam in its specific channel. This design allows for adjustment of the laser beam position within the acoustic mcAOM column to optimize laser / ion interactions.
[0055] Furthermore, transitions between optical tweezer configurations can be further controlled by the mcAOM. For example, when dividing a laser beam using a DOE, the intensity of the active tweezers beam is gradually withdrawn, while a new tweezers beam is adiabatically introduced. In all cases, the mcAOM enables both fast and slow control without delays or flickering caused by reloading new patterns into the mcAOM.
[0056] exist Figure 4 In the illustrated embodiment, the same laser in excitation source 46 generates both the Raman beam and the tweezers beam, and mcAOM 76 modulates the amplitudes of both the Raman beam and the tweezers beam. (In other embodiments, as described below, different lasers may be used to generate the Raman beam and the tweezers beam.) When splitter 72 includes a DOE or multi-beam deflector, the beams split by this splitter can all have high intensity, sufficient to allow any one of the beams 62 to be incident on the ion 40 with high intensity 66 (e.g., Figure 3A / Figure 3B (As shown). Therefore, the mcAOM 76 is controlled to apply substantial attenuation of ten times or even one hundred times or more to the high-intensity beam from the splitter 72 to be used as a Raman beam. The mcAOM 76 can also apply additional attenuation to the Raman beam even when the splitter generates Raman beams and tweezers beams with different intensities.
[0057] Alternatively, when the splitter 72 includes a multi-beam acousto-optic deflector (AOD), the controller 32 ( Figure 1A drive signal can be generated to adjust the corresponding intensity of each beam output from the AOD. This method allows for rapid switching between different configurations of tweezers and Raman beams, and the beams output from the AOD will have different frequency shifts depending on the beam angle. The beam intensity can be controlled using a beam detector and a feedback loop (not shown). Additionally or alternatively, the drive signal applied to the mcAOM 76 can be adjusted to compensate for intensity and frequency deviations of the beams output from the AOD.
[0058] In other embodiments where splitter 72 includes an SLM, the SLM can be driven to modulate the intensity of the beam such that the beam intended to be used as a tweezers beam has a higher intensity in each computation cycle than the beam intended to be used as a Raman beam. For example, the SLM may include a liquid crystal on silicon (LCoS) device, such as an LCoS device manufactured by HOLOEYE Photonics AG (Berlin, Germany). Controller 32 ( Figure 1 The SLM can be driven to generate a diffraction pattern that splits the incident laser beam into multiple beams with a desired intensity distribution. The SLM can also be driven to correct aberrations in the optics of subsystem 71.
[0059] Commercially available SLM devices (such as the aforementioned HOLOEYE device) have frame rates of 100 Hz or lower, meaning that switching between different configurations of the Raman beam and tweezers beam will typically take at least 10 ms. In one embodiment, in order to switch between different configurations of the Raman beam and tweezers beam (e.g., in...) Figure 3A and Figure 3B To enable faster switching between configurations, the SLM is divided into at least two segments. Each segment of the SLM is driven to generate different arrangements of the tweezers beam and the Raman beam at corresponding positions within array 60. Fast actuators (such as rotating mirrors or AODs) switch the coherent radiation beam output by excitation source 46 between the first and second segments of the SLM. Although the switching response of the SLM itself is slow, this arrangement allows for rapid movement of the beam configuration, thereby enabling the efficient execution of a series of quantum operations with high fidelity. When using rotating mirrors, switching will take a few milliseconds, and when using AODs, switching will take even less time.
[0060] When different segments of the SLM are used to generate different beam patterns, the incident angle of the beam on the mcAOM 76 will change. This change may lead to a mismatch between the incident angle and the Bragg angle of the acoustic grating formed in the mcAOM (e.g., ...). Figure 6A (As shown). To mitigate this problem, the driving frequency applied to the mcAOM can be adjusted to match different incident angles.
[0061] In addition to the amplitude modulation function mentioned above, the mcAOM 76 can also apply a frequency shift to the tweezers beam relative to the Raman beam. vice versaThis allows the frequency range of the tweezers beam to be outside the frequency range of the Raman beam. This frequency separation is useful in reducing crosstalk effects between beams in the mcAOM. For this purpose, a detuning of approximately 80 MHz between the tweezers beam and the Raman beam should be sufficient. For example, the channel of the mcAOM 76 used to modulate the tweezers beam can be driven at a frequency of approximately 260 MHz, while the channel used to modulate the Raman beam can be driven at approximately 160 MHz. Although the latter frequency may result in lower diffraction efficiency in the mcAOM, the efficiency loss is negligible given the much lower intensity of the Raman beam relative to the tweezers beam.
[0062] Following the mcAOM 76, the telescope 78 directs the modulated beam to the ion trap 24. The telescope 78 also allows for adjustment of the beam spacing to precisely match the spacing between the ions 40 in the trap. Additionally or alternatively, the drive signal applied to the mcAOM 76 can be adjusted to compensate for deviations in the shape and focus of the beam 62 on the respective ions 40.
[0063] Following the telescope 78, these beams pass through a dichroic beam splitter 80 and are then focused into the ion trap 24 by the objective optics 82. The objective optics 82 reduces the spot size of the beam 62 incident on the ions 40 to near the diffraction limit, approximately 1 µm or less for an objective optics with NA=0.5, and reduces the spacing between the beams to the interval between the ions 40 along the axis 38, typically a few micrometers.
[0064] To read out the intermediate and final states of register 68, excitation source 46 and mcAOM 76 are adjusted to direct the readout beam toward the qubit of interest in the register. Absorption of photons in the readout beam causes ions 40 to emit fluorescent radiation indicating the corresponding state of the quantum gate. Objective optics 82 collects the fluorescent radiation emitted by ions 40 in ion trap 24 and directs the emitted radiation toward dichroic beamsplitter 80, which reflects the emitted radiation onto detector array 84. Detector array 84 measures the intensity of the resulting fluorescence emission and reads the result to controller 32.
[0065] In some embodiments, the controller 32 uses signals output from the detector array 84 to detect alignment errors between the beam 62 and the ion 40. For example, if the fluorescence emission level received from a given ion is lower than the fluorescence emission level of other ions excited in a similar manner, it can be concluded that the beam incident on that ion is misaligned. These alignment errors may include positional deviations of the beam 62 relative to the corresponding ion 40, as well as defocus and other aberrations. When the splitter 72 includes an SLM or AOD, the drive signal applied to the splitter can be adjusted to correct for alignment errors, as well as certain errors in beam shape and focusing.
[0066] Additionally or alternatively, the controller 32 may modify the drive signal applied to the mcAOM 76 in order to correct errors caused by defocus and other aberrations.
[0067] Other specific excitation and measurement techniques can be applied to detect alignment errors. For example, the Rabi frequency of the transitions of ion 40 induced by a Raman beam can be measured, and ions exhibiting low Rabi frequencies can be identified as misaligned. As another example, the optical offset generated by the spatial light field in the narrowband quadrupole transitions of each ion in the array can be measured by scanning the frequency of the laser light that excites the transition. Since the optical offset is proportional to the intensity of the beam incident on each item, it can be used as an indicator of misalignment.
[0068] Multi-channel acousto-optic modulator Figure 5 This is a schematic detailed view of the multi-beam generation and modulation module 87 in the optical subsystem 71 according to an embodiment of the present invention. An excitation source 46 outputs a source beam 88, which is then split into multiple input beams 90 by a splitter 72. A Fourier transform lens 74 guides the beams 90 into an acousto-optic crystal 86 in a mcAOM 76. As the input beams 90 pass through the acousto-optic crystal 86, they are diffracted by the acoustic waves in the crystal, causing a change in the direction of the diffracted beams. These diffracted beams form an array of output beams 92. This diffraction process modulates the output beams 92 such that each beam has its own intensity, frequency, and phase.
[0069] Figure 6A and Figure 6B These are schematic side and top views of the mcAOM 76 according to an embodiment of the present invention. An array of piezoelectric transducers 94 is fixed along a surface of an acousto-optic crystal 86, which is transverse to the direction of the beam 90. An electrical drive signal applied to the piezoelectric transducers 94 causes the transducers to generate acoustic waves 95, which propagate through the crystal 86 with amplitude, frequency, and phase determined by the drive signal. The acoustic waves 95 produce periodic changes in the refractive index of the acousto-optic crystal 86, thereby effectively forming a dynamic diffraction grating. When the input beams 90 pass through the acousto-optic crystal 86, they are diffracted by this dynamic diffraction grating, thereby generating an array of modulated output beams 92. Each output beam 92 in the array has different characteristics, such as intensity, phase, and frequency, which are determined by the modulation applied by the acousto-optic crystal 86.
[0070] The acoustic wave 95 is terminated by an absorber 96 on the opposite side of the acousto-optic crystal 86. Additionally, any residual energy in the input beam 90 that directly passes through the mcAOM is blocked to prevent it from impacting the ions.
[0071] Optical subsystem using multiple acousto-optic modulators Figure 7This is a schematic detailed view of a multi-beam generation and modulation module 99 according to another embodiment of the present invention. Module 99 can be incorporated into optical subsystem 91. This module differs from the previous embodiment in that it includes two mcAOMs 76a and 76b, which modulate different corresponding groups of input beams 90. Because mcAOMs have limited resolution, each mcAOM can only be used to modulate a certain number of beams. Using two mcAOMs 76a and 76b, the number of modulated output beams 92 generated by module 99 can be doubled. Alternatively, even more mcAOMs (with appropriate splitting and combining optics) can be used to generate even more modulated output beams.
[0072] Splitter 97 divides the input beam 90 between mcAOM 76a and mcAOM 76b, which are similar in design and operation to mcAOM 76 as described above. The two mcAOMs 76a and 76b operate in parallel, each modulating a subset of the input beam 90 to generate a corresponding array of output beam 92. After the modulation process, the two arrays of output beam 92 are recombine into a single array by beam combiner 98. This single array of output beam 92 is then directed toward ion trap 24 for application to ions 40.
[0073] Separation of tweezers beam source and Raman beam source Figure 8 This is a schematic side view of a multi-beam optical trapping and excitation subsystem 100 according to an alternative embodiment of the present invention. Subsystem 100 can replace subsystem 71 ( Figure 4 This is used in quantum computing system 20. Subsystem 100 differs from subsystem 71 in that, in subsystem 100, the Raman beam 48 and the tweezers beam 50 are generated by separate, independent laser sources. This separation facilitates more efficient use of laser power because the Raman beam 48 has a much lower intensity than the tweezers beam 50, and the tweezers beam for each ion does not need to be coherently matched with the other tweezers beam or Raman beam, as described above. A splitter 72 and / or an mcAOM 76 can be driven to correct any alignment errors between the Raman beam and the tweezers beam to form a single, uniformly spaced beam array for input to the ion trap 24.
[0074] The tweezers bundle 50 can be wavelength-shifted relative to the Raman beam 48. Therefore, the two beam sets can be combined using a dichroic beam combiner 102, which aligns the tweezers bundle and the Raman beam to form an array 60 for input to the ion trap 24.
[0075] Alternatively or additionally, the Raman beam 48 may have a polarization orthogonal to the tweezer beam 50. In this case, the dichroic beam combiner 102 may be replaced by a polarization beam splitter.
[0076] The embodiments described above are exemplified by way of example, and the invention is not limited to what is specifically shown and described above. Rather, the scope of the invention includes combinations and sub-combinations of the various features described above, as well as variations and modifications of these features that would occur to those skilled in the art upon reading the foregoing description and that are not disclosed in the prior art.
Claims
1. A device for quantum computing, comprising: An ion trap is configured to hold a first array of ions at corresponding positions along the array axis; A radiation source, configured as a second array of beams emitting coherent radiation, the beams comprising a first beam and a second beam, the first beam having a corresponding first intensity and a frequency selected to excite selected internal transitions of the ions, the second beam having a second intensity at least ten times greater than any of the first intensities, and the radiation source being configured to switch the corresponding positions of the first beam and the second beam within the second array. and An optical device configured to focus the beam into the ion trap such that each beam in the second array is incident on a corresponding ion in the first array.
2. The apparatus according to claim 1, wherein, The second strength is at least one hundred times greater than any of the first strengths in the first strength.
3. The apparatus according to claim 1, wherein, The second beam is configured to act as optical tweezers to confine the ions incident upon the second beam, while the ions incident upon the first beam serve as components of the computation segment.
4. The apparatus according to claim 3, wherein, The second array of the beams is configured such that a subset of the second beams is incident on a corresponding set of ions, whereby the corresponding set is further confined by the optical tweezers.
5. The apparatus according to claim 3, wherein, The computation segment includes a single-qubit gate or a multi-qubit gate, each multi-qubit gate including a corresponding group of two or more ions incident on the first beam within the computation segment.
6. The apparatus according to claim 5, wherein, The first beam and the second beam are arranged to define a plurality of single-qubit gates or multi-qubit gates to which the first beam is incident, wherein the multi-qubit gates are separated from each other by confined ions to which the second beam is incident.
7. The apparatus according to claim 6, wherein, The radiation source is configured to redefine the computation segment by switching the corresponding position of the second beam, and within the redefined computation segment, the first beam applies a single-qubit gate operation or a multi-qubit gate operation.
8. The apparatus according to claim 7, wherein, Switching the corresponding position of the second beam includes: continuing to irradiate the first set of ions with the first group of the second beam, while simultaneously irradiating the second set of ions with the second group of the second beam.
9. The apparatus according to claim 8, wherein, The radiation source is configured to perform intermediate circuit-assisted measurements by applying at least one of the second beams in the second group to at least one ion in the first group.
10. The apparatus according to claim 9, wherein, The radiation source is configured to perform a cooling operation and initialization on at least one ion in the first set of ions after the intermediate circuit-assisted measurement is completed.
11. The apparatus according to claim 8, wherein, The radiation source is configured to adiabatically initiate irradiation of the second set of ions and terminate the second beam irradiation of ions in the first set.
12. The apparatus of claim 3, further comprising a third array of detectors configured to sense radiation emitted by qubits, wherein, The emitted radiation indicates the corresponding quantum state of the qubit, and the optical device is configured to direct the emitted radiation from the ion trap to the detector.
13. The apparatus according to any one of claims 1-12, wherein, The radiation source includes at least one laser and a splitter, the at least one laser being configured to output coherent radiation and the splitter being configured to divide the coherent radiation into multiple beams.
14. The apparatus according to claim 13, wherein, The splitter includes a diffractive optical element (DOE).
15. The apparatus according to claim 13, wherein, The splitter includes a spatial light modulator (SLM).
16. The apparatus according to claim 15, wherein, The SLM is configured to be divided into at least a first segment and a second segment, wherein each segment is configured to generate different arrangements of corresponding positions of the first beam and the second beam within the second array, and wherein the radiation source includes an actuator configured to switch between the first segment and the second segment coherent radiation incident on the SLM.
17. The apparatus according to claim 13, wherein, The splitter includes a multi-beam acousto-optic deflector (AOD).
18. The apparatus of claim 17, further comprising a controller configured to adjust the AOD to correct alignment errors between the beam and the ions in the trap.
19. The apparatus of claim 18, further comprising a third array of detectors configured to sense radiation emitted by the ions, wherein, The controller is configured to detect the alignment error in response to sensed radiation.
20. The apparatus according to any one of claims 1-12, wherein, The radiation source includes at least one multi-channel acousto-optic modulator (mcAOM), which is configured to modulate the corresponding amplitude of at least the first beam.
21. The apparatus according to claim 20, wherein, The at least one mcAOM is configured to apply different corresponding frequency shifts to different first beams in the first beam in order to drive corresponding Raman transitions of the ions incident on the first beam.
22. The apparatus according to claim 20, wherein, The at least one mcAOM includes at least a first mcAOM and a second mcAOM, the first mcAOM and the second mcAOM being configured to modulate different corresponding groups of beams in the second array.
23. The apparatus according to claim 20, wherein, The at least one mcAOM is configured to modulate the corresponding amplitudes of the first beam and the second beam.
24. The apparatus according to claim 23, wherein, The at least one mcAOM is configured to attenuate the first beam such that the second intensity is at least ten times greater than the first intensity.
25. The apparatus according to any one of claims 1-12, wherein, The first beam has a corresponding first frequency within a selected frequency range, and the second beam has a corresponding second frequency outside the selected frequency range.
26. The apparatus according to claim 25, wherein, The radiation source includes a dichroic beam combiner that aligns the first beam and the second beam to form the second array.
27. The apparatus according to any one of claims 1-12, wherein, The first beam has a first polarization, and the second beam has a second polarization orthogonal to the first polarization, wherein the radiation source includes a polarization beam combiner that aligns the first beam and the second beam to form the second array.
28. A method for quantum computing, comprising: The first array of ions is trapped in the ion trap at corresponding positions along the array axis; A second array of beams that generate coherent radiation, the beams comprising a first beam and a second beam, the first beam having a corresponding first intensity and having frequencies selected to excite selected internal and kinematic transitions of the ions, the second beam having a second intensity that is at least ten times greater than any of the first intensities of the first intensity; The beam is focused into the ion trap such that each beam in the second array is incident on a corresponding ion in the first array; A first quantum computing operation is performed using a first configuration of the first beam and the second beam in the first array; as well as A second quantum computing operation is performed using a second configuration of the first and second beams, in which corresponding positions of at least some of the first and second beams are switched within the second array.
29. The method according to claim 28, wherein, The second strength is at least one hundred times greater than any of the first strengths in the first strength.
30. The method according to claim 28, wherein, The second beam is configured to act as optical tweezers to further confine the ions incident upon the second beam, while the ions incident upon the first beam serve as components of the computation segment.
31. The method according to claim 30, wherein, The computing segment is driven to perform single-qubit gate operations or multi-qubit gate operations, each multi-qubit gate operation being performed on a corresponding group of two or more ions within the computing segment to which the first beam is incident.
32. The method according to claim 31, wherein, Performing the second quantum computing operation includes redefining the computing segment by switching the corresponding positions of the first and second beams within the second array, and applying a single-qubit gate operation or a multi-qubit gate operation by the first beam within the redefined computing segment.
33. The method according to claim 32, wherein, Switching the corresponding position of the second beam includes: continuing to irradiate the first set of ions with the first group of the second beam, while simultaneously irradiating the second set of ions with the second group of the second beam.
34. The method of claim 33, further comprising applying at least one second beam from the second beam in the second group to at least one ion in the first group to perform an intermediate circuit-assisted measurement.
35. The method of claim 34, further comprising, after completing the intermediate circuit-assisted measurement, applying at least one beam to cool and initialize at least one ion in the first set of ions.
36. The method according to claim 33, wherein, Switching the corresponding position includes adiabatically initiating irradiation of the second set of ions and terminating irradiation of the first set of ions by the second beam.
37. The method of claim 30, wherein, Focusing the beam into the ion trap includes applying optics to focus the beam onto the ions, and wherein the method includes sensing the emitted radiation by directing radiation emitted by the qubit through the optics onto a detector array, wherein the emitted radiation indicates the corresponding state of the qubit.
38. The method according to claim 28, wherein, The second array that generates the beams includes dividing the coherent radiation output by at least one laser into multiple beams.
39. The method according to claim 38, wherein, Dividing the coherent radiation includes applying a diffractive optical element (DOE) to the coherent radiation output by the at least one laser.
40. The method according to claim 38, wherein, Dividing the coherent radiation includes applying a spatial light modulator (SLM) to the coherent radiation output by the at least one laser.
41. The method according to claim 40, wherein, Applying the SLM includes dividing the SLM into at least a first segment and a second segment, wherein each segment is configured to generate different arrangements of corresponding positions of the first beam and the second beam within the second array, and switching the coherent radiation incident on the SLM between the first segment and the second segment.
42. The method according to claim 38, wherein, The division of the coherent radiation includes applying a multi-beam acousto-optic deflector (AOD) to the coherent radiation output by at least one laser.
43. The method according to claim 42, wherein, Modulating the corresponding amplitude includes controlling the AOD to correct alignment errors between the beam and the ions in the trap.
44. The method of claim 43, further comprising sensing radiation emitted by the ions, wherein, Controlling the AOD includes detecting the alignment error in response to sensed radiation.
45. The method according to any one of claims 28-44, wherein, The second array for generating the beam includes modulating the corresponding amplitude of at least the first beam using at least one multi-channel acousto-optic modulator (mcAOM).
46. The method according to claim 45, wherein, Modulating the corresponding amplitude includes applying different corresponding frequency shifts to different first beams in the first beam in order to drive corresponding Raman transitions of the ions incident on the first beam.
47. The method according to claim 45, wherein, Modulating the corresponding amplitude includes applying at least a first mcAOM and a second mcAOM to modulate different corresponding groups of beams in the second array.
48. The method of claim 45, further comprising using the at least one mcAOM to modulate the corresponding amplitude of the second beam.
49. The method according to claim 48, wherein, Modulating the corresponding amplitude includes attenuating the first beam such that the second intensity is at least ten times greater than the first intensity.
50. The method according to any one of claims 28-44, wherein, The first beam has a corresponding first frequency within a selected frequency range, and the second beam has a corresponding second frequency outside the selected frequency range.
51. The method of claim 50, further comprising using a dichroic beam combiner to align the first beam and the second beam to form the second array.
52. The method according to claim 28, wherein, The first beam has a first polarization, and the second beam has a second polarization orthogonal to the first polarization, wherein the method includes aligning the first beam and the second beam using a polarization beam combiner to form the second array.
53. A quantum register comprising collections of ions at corresponding positions along an array axis within a first array in an ion trap. in, Ions are configured to perform quantum operations under the control of a radiation source that emits a second array of beams of coherent radiation, the beams comprising a first beam and a second beam, the first beam having a corresponding first intensity and a frequency selected to excite selected internal transitions of the ions, the second beam having a second intensity at least ten times greater than any of the first intensities, and the radiation source being configured to switch the corresponding positions of the first and second beams within the second array, and optics focusing the beams into the ion trap such that each beam in the second array is incident on a corresponding ion in the first array.
54. The quantum register according to claim 53, wherein, The first beam drives the ions to perform single-qubit or multi-qubit operations.