Apparatus and method for interacting with multiple quantum states

Abstract

A method for interacting with a plurality of quantum states over respective time intervals includes providing a fiber-coupled optical mode from at least one optical fiber interface that controls optical coupling to and / or from an optical fiber, at least a portion of the optical fiber extending outside an interior of a housing including the at least one optical fiber interface; providing a quantum state from each quantum state emitting element (QSEE) housed in or inside the housing; providing a preferential direction from each of multiple portions of one or more directional structures for an associated element-coupled optical mode that controls optical coupling to and / or from different respective subsets of one or more QSEEs; and scanning a scanning structure housed in or inside the housing to vary overlap between the fiber-coupled optical mode and different respective ones of the element-coupled optical modes over each time interval.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to and benefit of U.S. Provisional Application No. 63 / 440,477, entitled "Managing Coupling of Optical Modes with Quantum State Emitting Devices," filed January 23, 2023, the entire disclosure of which is incorporated herein by reference.

[0002] STATEMENT REGARDING FEDERALLY FUNDED RESEARCH This invention was made with government support under Contract 22-C-0049 awarded by the National Reconnaissance Office. The government has certain rights in this invention.

[0003] The present disclosure relates to managing the coupling of optical modes into quantum state emitting devices. [Background technology]

[0004] To transmit quantum information from one location to another via qubits, one promising platform is to transmit photon qubits (e.g., optical frequency photons). In some cases, quantum repeaters may be used. While it is possible to transmit quantum information point-to-point without quantum repeaters, there may be advantages to utilizing quantum repeaters in some situations. Summary of the Invention

[0005] In one aspect, generally, an apparatus for interacting with a plurality of quantum states over respective successive time intervals in a scanning period includes: a housing including at least one optical fiber interface configured to provide a fiber-coupled optical mode that controls light coupling to and / or from an optical fiber, at least a portion of the optical fiber extending outside an interior of the housing; an array of a plurality of quantum state emitting elements configured to be contained in or within the housing, each quantum state emitting element configured to provide a quantum state; one or more directional structures, portions of the one or more directional structures each configured to provide a preferential direction for an associated element-coupled optical mode that controls light coupling to and / or from a different respective subset of one or more quantum state emitting elements of the quantum state emitting elements; and a scanning structure configured to be contained in or within the housing, the scanning structure configured to vary overlap between the fiber-coupled optical mode and a different respective one of the element-coupled optical modes over each time interval of the successive time intervals.

[0006] Aspects may include one or more of the following features.

[0007] The quantum state emitter includes a plurality of quantum state storage elements, each configured to receive a quantum state that is temporarily stored and emitted after a period of time.

[0008] The certain period of time is longer than the scanning period.

[0009] The quantum state emitting device includes a lattice mismatch.

[0010] Lattice mismatch involves emitting point defects in the crystal lattice material.

[0011] The crystal lattice material includes diamond, silicon, or silicon carbide.

[0012] The optical fiber includes a single-mode optical fiber.

[0013] The housing includes a chamber configured to enclose an array of quantum state emitting devices, one or more directing structures, and a scanning structure.

[0014] The chamber includes a cryochamber configured to maintain a temperature therein below about 10 Kelvin.

[0015] The one or more directional structures include an array of pillars formed from a material that includes quantum state emitting devices.

[0016] The one or more directional structures include a hemispherical optically transmissive material configured to cover multiple of the one or more subsets of the quantum state emitting devices.

[0017] The scanning structure includes a first scanning element configured to move the position of the fiber optic interface along a first axis.

[0018] The scanning structure includes a second scanning element configured to vary, along a second axis substantially perpendicular to the first axis, an overlap between the fiber coupled optical mode and each different one of the element coupled optical modes associated with the directional structure.

[0019] In another aspect, a method for interacting with a plurality of quantum states over respective successive time intervals in a scanning period generally includes: providing a fiber-coupled optical mode from at least one optical fiber interface that controls optical coupling to and / or from an optical fiber, at least a portion of the optical fiber extending outside an interior of a housing including the at least one optical fiber interface; providing a quantum state from each quantum state emitting element in an array of quantum state emitting elements contained in or inside the housing; providing a preferential direction from each of a plurality of portions of one or more directional structures for an associated element-coupled optical mode that controls optical coupling to and / or from different respective subsets of one or more quantum state emitting elements of the quantum state emitting elements; and scanning a scanning structure contained in or inside the housing to vary overlap between the fiber-coupled optical mode and different respective ones of the element-coupled optical modes over each time interval of the successive time intervals.

[0020] In another aspect, in general, an apparatus for interacting with a plurality of quantum states includes a housing including at least one optical interface providing at least 100 fiber-coupled optical modes, wherein at least a portion of the optical interface extends outside an interior of the housing; an array of quantum state emitting elements configured to be contained in or within the housing, each quantum state emitting element configured to provide a quantum state; and one or more directional structures, portions of the one or more directional structures each configured to provide a preferential direction for an associated element-coupled optical mode that controls optical coupling to and / or from a different respective subset of one or more quantum state emitting elements of the quantum state emitting elements.

[0021] Aspects may include one or more of the following features.

[0022] At least one optical interface includes one or more multimode optical fibers.

[0023] The at least one optical interface includes at least one array of a plurality of optical cores.

[0024] The apparatus further includes a scanning structure configured to vary an overlap between at least one fiber-coupled optical mode of the at least 100 fiber-coupled optical modes and at least one element-coupled optical mode of the plurality of element-coupled optical modes over each successive time interval in the scanning period.

[0025] The scanning structure is disposed outside the housing.

[0026] The scanning structure includes one or more spatial light modulators configured to control light coupling to and / or from the at least one optical interface.

[0027] At least one spatial light modulator of the one or more spatial light modulators is disposed outside the housing.

[0028] The scanning structure is configured to vary an overlap between at least one fiber-coupled optical mode of the at least 100 fiber-coupled optical modes and at least one element-coupled optical mode of the plurality of element-coupled optical modes by interfering two or more fiber-coupled optical modes of the at least 100 fiber-coupled optical modes.

[0029] The scanning structure is configured to vary overlap between at least one fiber-coupled optical mode of the at least 100 fiber-coupled optical modes and at least one element-coupled optical mode of the plurality of element-coupled optical modes by selectively exciting one or more optical modes of said optical modes of the at least 100 fiber-coupled optical modes.

[0030] In another aspect, generally, a method for interacting with a plurality of quantum states includes providing at least 100 fiber-coupled optical modes from at least one optical fiber interface, where at least a portion of the optical fiber extends outside an interior of a housing including the at least one optical fiber interface; providing a quantum state from each quantum state emitting element in an array of quantum state emitting elements contained in or inside the housing; and providing a preferential direction from each of a plurality of portions of one or more directional structures for an associated element-coupled optical mode controlling optical coupling to and / or from different respective subsets of one or more quantum state emitting elements of the quantum state emitting elements.

[0031] Aspects may have one or more of the following advantages.

[0032] Quantum repeaters can create entanglement (in some cases on-demand, as a resource) across multiple nodes in a network structure such as the Internet. In some configurations, quantum repeaters can be used to connect any node to any other node in a future "quantum internet" or "quantum-enabled internet." Quantum repeaters with quantum memory also allow for the relaxation of timing constraints, as quantum memory can maintain entanglement between nodes for a finite lifetime and then connect a series of nodes on-demand as needed.

[0033] One advantage of quantum repeaters is that, like classical repeaters, they extend the range over which entanglement can be generated. The fundamental rate loss limit is linearly proportional to the transmission rate; that is, as loss increases, the achievable rate decreases. The use of quantum repeaters allows us to exceed that limit, potentially making quantum-secure communication (e.g., used for secure key transmission), quantum sensing, or distributed quantum computing applications accessible (e.g., at one limit, rates can be changed from values ​​below 1 qubit / s to values ​​above 1000 qubits / s).

[0034] Some of the techniques described herein facilitate the goal of creating high-rate, high-fidelity quantum repeaters with quantum memory for generating entanglement (in some cases on-demand) between remote nodes. In some implementations, the quantum repeaters with quantum memory are relatively low-loss (e.g., to enable high rates), relatively high-fidelity (e.g., reducing decoherence / noise processes), relatively high-rate (e.g., using high scanning speeds, in terms of number of qubits / entanglement generation per second), and have relatively long coherence times.

[0035] Furthermore, some of the techniques described herein can be used to provide a scanning confocal optical interface that is modular and easy to implement in cryogenic environments operating at temperatures below about 10K.

[0036] Other features and advantages will become apparent from the following description, and from the drawings and claims. [Brief explanation of the drawings]

[0037] The present disclosure is best understood from the following detailed description when read in conjunction with the accompanying drawings. It is emphasized that, according to common practice, the various features of the drawings are not to scale. Conversely, the dimensions of the various features are arbitrarily increased or reduced for clarity.

[0038] [Figure 1] FIG. 1 is a schematic diagram of an example quantum state scanning system. [Figure 2] 1 is a schematic diagram of an example satellite uplink scheme. [Figure 3] FIG. 1 is a schematic diagram of an example quantum repeater node. [Figure 4] FIG. 1 is a schematic diagram of an example quantum repeater node. [Figure 5] 1 is a schematic diagram of an example satellite uplink scheme. [Figure 6A] FIG. 1 is a schematic diagram of an example integrated assembly. [Figure 6B] FIG. 1 is a schematic diagram of an example integrated assembly. [Figure 6C] FIG. 1 is a schematic diagram of an example integrated assembly. [Figure 7] FIG. 1 is a schematic diagram of an example array of diamond pillars containing color centers. [Figure 8] FIG. 1 is a schematic diagram of an example quantum repeater node. [Figure 9A] FIG. 1 is a schematic diagram of an example patterned diamond. [Figure 9B] Schematic diagram of a cross section of an example diamond-SIL configuration. [Figure 9C] FIG. 1 is a schematic diagram of an example patterned diamond. [Figure 9D] Schematic diagram of a cross section of an example diamond-SIL configuration. [Figure 9E] FIG. 1 is a schematic diagram of an example patterned diamond. [Figure 9F] Schematic diagram of a cross section of an example diamond-SIL configuration. [Figure 10A-I] 10A to 10I are schematic illustrations of the manufacturing steps for fabricating pillars in diamond. [Figure 11A] FIG. 1 is a schematic diagram of an example entanglement system. [Figure 11B] FIG. 1 is a schematic diagram of an example entanglement system. [Figure 12A] FIG. 1 is a schematic diagram of an example optical scanner. [Figure 12B]FIG. 1 is a schematic diagram of an example optical scanner. [Figure 13A] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 13B] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 13C] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 13D] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 13E] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 13F] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 14A] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 14B] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 14C] FIG. 1 is a schematic diagram of an example scanning procedure. [Figure 15A] FIG. 1 is a schematic diagram of an example scanning structure. [Figure 15B] FIG. 1 is a schematic diagram of an example interface between diamond and fabricated CMOS. [Figure 15C] FIG. 1 is a schematic diagram of an example interface between diamond, fabricated CMOS, and a fiber waveguide core array. [Figure 15D] FIG. 1 is a schematic diagram of an example fiber waveguide core array. [Figure 16A] 1 is a schematic diagram of a portion of an example scanning structure. [Figure 16B] 1 is a schematic diagram of a portion of an example scanning structure. [Figure 16C] 1A-1D are schematic diagrams of various example diamond structures. [Figure 17A] 1 is a schematic diagram of a portion of an example scanning structure. [Figure 17B] 1 is a schematic diagram of a portion of an example scanning structure. DETAILED DESCRIPTION OF THE INVENTION

[0039] In some implementations, a quantum repeater can be configured to include an interface to at least one optical fiber (allowing the signal to be transmitted to another quantum or classical device, e.g., another fiber-coupled device, or a free-space coupled device such as a satellite). Optical signals coupled via the optical fiber interface can interact with atomic memory elements within the quantum repeater's chamber (e.g., cryostat), potentially representing a large number of quantum channels NCH, each containing Nq atomic memories. In particular, for diamond color center spin memories such as SiV or SnV, the quantum repeater can accommodate millions of spatially and spectrally resolved color centers, N CH *N q ~N qubits The device may be configured to interface with

[0040] In some other quantum repeater designs that do not include a fiber optic interface to the cryostat, one approach is to combine a large scanning optical microscope operating outside the cryostat and send an optical signal through an imaging window to interact with an atomic memory sample on a cold finger within the cryostat (e.g., a He cryostat). However, this approach can present certain challenges (e.g., large, expensive, and / or complex design). For example, its large size can lead to misalignments at the wavelength scale (or diffraction limit) due to even small differences in the thermal expansion coefficients of materials, which may require additional motion control, which in turn may further complicate the system.

[0041] Some of the techniques described herein enable compact and modular scanners that can be mounted inside a cryostat and perform confocal scanning across a quantum memory on a substrate (e.g., a diamond surface). The module containing the scanner and quantum memory substrate can be designed to be relatively small to reduce motion control overhead. The modules can be small enough to allow multiple modules to be installed in a cryogenic housing where available space may be limited.

[0042] Some implementations of such cryogenically compatible scanning quantum memory systems (or "cryogenically integrated scanners") include the following four subsystems, which provide a platform for integration and automated microscopy of arrays of quantum state emission elements (QSEEs). The distinction between these four subsystems is for clarity, but in some implementations, different subsystems may be combined, modified, or eliminated: (1) an array of quantum state emission elements (e.g., color centers in diamond), which acts as a qubit source and / or quantum memory. For example, the quantum state emission element can be any defect in diamond, such as a color center (e.g., NV, SiV, SnV); and (2) a scanning structure (or simply "scanner"), which scans over the array of quantum state emission elements. This is also known as spatial multiplexing. The scanner may also include a module for single spin qubit manipulation (e.g., a microwave controller). (3) A housing (e.g., a rigid but open platform or a cryogenic chamber, such as an airtight, closed chamber containing a cryogenic system to maintain a low internal temperature) that houses the array at low temperatures, in some cases to improve performance (e.g., to block phonon dephasing). In some implementations, a scanner is located inside the housing. (4) A light source (e.g., a laser) that optically excites multiple quantum state emitting devices, preferably at a frequency that is "on-resonance." In some implementations, the light source may include a laser and a frequency conversion module for spectral multiplexing. When the array is implemented using color centers in diamond, the resonant frequency may be different for each color center due to internal strain in the diamond. Therefore, the laser may be frequency-shifted, possibly by several GHz, when addressing each color center. In some examples, the system may address multiple color centers within a diffraction-limited spot through spectral multiplexing.

[0043] In some examples, there may be frequency conversion associated with downconverting photons to telecom wavelengths (e.g., telecom bands may include one or more of the O, E, S, C, L, and U bands). Nitrogen-vacancy (NV, 636 nm) and silicon-vacancy (SiV, 737 nm) emission photons from on-resonance excitation may be downconverted to more suitable telecom wavelengths (e.g., 1260-1675 nm). One benefit from such conversion is that it allows cryogenically integrated scanners to interface with existing telecom fiber infrastructure.

[0044] At a high level, cryogenically integrated scanners can be configured to function as high-rate, high-fidelity quantum repeaters and memories, with indistinguishable photons at output, which may be necessary to eliminate path information for entanglement distribution. Figures of merit or metrics for a subsystem containing an array of quantum state emitters implemented in diamond color centers are photon collection efficiency, temperature / phonon dephasing, color center density (per area), frequency uniformity (due to the presence or absence of strain), scalability (of the process), and reproducible addressability.

[0045] Color centers in diamond can be created by blanket implantation of nitrogen, silicon, or other Group IV elements followed by annealing. A single color center is then located using an optical microscope, and experiments proceed using the single color center. In other examples, a single color center can be located, and then a cavity can be created around it via lithographic patterning. In these single-color-center scenarios, factors such as density, frequency uniformity, and scalability are usually not important because only a single color center is used in the experiment.

[0046] Because the cryogenically integrated scanner we describe addresses many color centers that act as memory qubits that store entanglement across multiple repeaters (possibly worldwide), the above metrics, which are often inconsequential for a single color center, are factors we consider in various implementations. In one embodiment of a cryogenically integrated scanner, photon collection efficiency can be made much higher through the inclusion of (1) nanopillars, (2) paraboloids etched into diamond, or (3) a metal layer deposited on the backside of thinned diamond to act as a mirror for constructive interference. Essentially, the designed structure can be used to steer optical modes in preferential directions for maximum photon collection efficiency.

[0047] In some examples of cryogenically integrated scanners, we can intentionally design nanostructures (e.g., pillars) at a density that can be addressed by an optical system (confocal or otherwise). The density of addressable color centers can be much higher than that found in diamonds where blanket implantation is used, as it may be necessary to find addressable color centers. In contrast, we can obtain addressable color centers by designing and addressing nanostructures that contain one or more color centers. Furthermore, frequency uniformity can be an important factor due to the requirement for indistinguishability. Improved frequency uniformity can reduce (or eliminate) the burden of having to frequency shift over several GHz to achieve indistinguishability. One advantage of lithographic patterning is the ability to etch fiducial marks into the chip, which provides valuable location information about which color center the scanner is addressing as the chip is scanned. Such information can be used in conjunction with a lookup table for the frequency conversion required to return the resonant frequencies of each individual color center to a common wavelength for indistinguishability purposes.

[0048] In other examples, a solid immersion lens (SIL) may be used to increase photon collection efficiency. In such examples, if there are no nanopillars or similar structures etched into the diamond (perhaps to maintain low internal distortion characteristics for frequency uniformity), etched fiducial marks may be used at the edges or corners of the diamond, which only locally affect distortion far from the addressable color center being used.

[0049] One possible goal of a scanner is to address as many color centers as possible per unit time. Scanner addressing is a form of spatial multiplexing. The laser subsystem may include a frequency converter for spectral multiplexing.

[0050] The following are figures of merit or metrics for the scanner subsystem: total addressable spots (e.g., diffraction-limited spot size, field of view, number of diamond tips), scanning speed, photon collection efficiency in the fiber, and consistency of operation (including repeatable localization (e.g., fiducial marks and convolution into ground truth images)).

[0051] Other scanners used in non-fiber-coupled quantum memory systems are often located outside the cryogenic system and use free-space coupling of light to a device (e.g., a diamond) inside the cryogenic system. Such free-space coupled scanners have potential limitations due to vibrations (e.g., from the pumping action of the cryocooler) and alignment, necessitating the use of two nanopositioner stages, if possible. In such applications, one cryogenically compatible nanopositioner stage may be used inside the cryogenic system to hold the diamond, while a second nanopositioner stage may be located outside the cryogenic system.

[0052] Furthermore, free-space coupling scanners may require a cryogenic system with a window through which light is coupled. Correspondingly, systems utilizing free-space coupling scanners may suffer from infrared (IR) light leakage through the window, potentially limiting the minimum achievable temperature compared to cryogenic systems without such windows. Lower temperatures may be particularly useful because they correspond to a lower probability of phonon dephasing, which translates into longer qubit memory times. Furthermore, blackbody radiation through the window may also adversely affect qubit memory times. Free-space coupling may also require a cryogenically compatible imaging objective or an objective with a sufficiently large working distance outside the window. The former can be expensive and options may be limited, while the latter, while possible, causes a reduction in numerical aperture (NA) and therefore photon collection efficiency. The use of an objective lens may limit field of view (FOV) options; many high-magnification objectives have FOVs of 100 μm × 100 μm. From a scalability perspective, free-space coupling systems may contain only one or two windows, potentially limiting the number of addressable diamonds in the system.

[0053] Many applications, even terrestrial-satellite-terrestrial communications, may require coupling to fiber at some location. Performing such fiber coupling in conjunction with free-space coupling can be challenging because it may require maintaining two systems, one at cryogenic temperature and one at room temperature (e.g., aligning the diamond inside the cryogenic system and then aligning the free-space coupled beam with the fiber outside the cryogenic system with a U-bench setup (or similar)). Finally, free-space coupling can also be susceptible to drift resulting from ambient temperature changes outside the cryogenic system.

[0054] We now describe how some of the above factors can be addressed in some implementations of cryogenically integrated scanners.

[0055] By constructing a small, compact, and integrated scanner inside the cryogenic system, the dimensional change is very small, which can reduce the drift of the laser pointing to the diamond. Furthermore, such an assembly can be designed to be a rigid body with a very high resonant frequency, which reduces the effects of lower frequency vibrations from the environment (e.g., cryogenic system, room). Therefore, the cryogenic integrated scanner can reduce problems arising from alignment and vibration.

[0056] In contrast to free-space coupling, cryogenically integrated scanners can be configured to operate without windows because they include an optical fiber interface that transmits light between the inside and outside of the cryogenic system. This eliminates the problem of IR leakage through windows, potentially enabling lower cryogenic temperatures and reducing system costs. Because cryogenically integrated scanners do not require optical coupling through windows, multiple scanners can be included in a single cryostat, thereby increasing the number of addressable color centers. For example, a single cryogenically integrated scanner can include 10 or 20 fiber links, which is difficult to achieve with free-space coupling systems.

[0057] Furthermore, cryogenic integrated scanners enable a single (cryogenic) temperature system for reading out information from qubits (e.g., color centers in diamond) and coupling them into fibers.

[0058] The design of a cryogenically integrated scanner can also take into account tolerances in the manufacturing process. If the fiber has a mode field diameter of about 5 μm, then roughly speaking, all tolerances in the system will reduce to that number, to a first-order approximation. Typical machining standards can achieve an accuracy of ±1 mil (~25 μm), but this can cause problems from an optical alignment perspective.

[0059] In some implementations, active alignment during assembly may be used (e.g., active compensators, such as piezoelectric elements on lenses, used to compensate for shifts or tolerances by actively fine-tuning the position and angle of optical elements). Thus, instead of one scanning element (e.g., a MEMS mirror), a second scanning element may be used. In such use, one element scans across the sample, the other can compensate for tip and tilt errors, and because the focal plane is actually curved, it has the added benefit of expanding the field of view by correcting for some field curvature. Other implementations may use a microelectromechanical system (MEMS) scanning element and a deformable mirror for wavefront correction. All of these options not only help to mitigate errors arising from tolerances, but also provide flatter fiber coupling efficiency across the entire FOV.

[0060] Cryogenic integrated scanners can also be configured to consider the compatibility of various elements at cryogenic temperatures. For most glass, UV-cured epoxy, and polymer materials, there may be minimal information about the refractive index or coefficient of thermal expansion (CTE). Therefore, careful design and testing is used to determine these material properties in the context of a cryogenic integrated scanner.

[0061] Regarding fiber connections, standard bayonet-fit connectors (often used at room temperature) may not be suitable for connecting FC connectors. Such connectors are often made from materials with relatively large CTEs. Therefore, in some implementations of cryogenically integrated scanners, custom housings with low CTE materials (e.g., Invar) can be designed and used. Such custom housings can be particularly useful when collimating fibers where the axial distance to the collimating aspheric lens is important. In other implementations, repeatable SMA connectors (e.g., bottomed with connector threads) can be used for guaranteed axial alignment. In such implementations, active alignment of the collimating lens can also be used by placing some UV-cured epoxy in place.

[0062] Cryogenically integrated scanners may be configured to take thermal cycling considerations into account. As the cryogenic system is turned on and off, epoxies can crack or other materials can bond, causing undesired movement of optical elements, resulting in defocusing, aberrations, and generally degraded optical performance (e.g., photon collection efficiency, spot size, which affects the number of addressable spots), leading to inconsistent performance over the lifetime of the quantum repeater. Since the cryostat at each node is theoretically capable of permanent indefinite operation, thermal cycling is not necessarily an operational issue. In practice, quantum repeaters, even during production, may experience at least 10 to 100 thermal cycles over their lifetime for testing, maintenance, etc., if not perhaps thousands of cycles.

[0063] Cryogenic integrated scanners can be designed to operate at both cryogenic and room temperatures with some capability (for testing before lowering the temperature or for extrapolation purposes). Many tests, except for linewidth limit tests (e.g., g2), can be performed at room temperature. To allow the cryogenic integrated scanner to operate at two different temperatures, the system may be thermally insulated by considering the CTE, material data of the metals, the UV-cured epoxy used, etc. Thermal insulation can also involve the use of various materials (e.g., Invar, Molybdenum, Ti-64) to balance the CTE to minimize drift (e.g., allow the axial position to stay at the focus).

[0064] In some implementations, cryogenically integrated scanners contain multiple optical elements inside the cryogenic chamber. The use of small elements can be advantageous because it is easier to control CTE and material changes with temperature (thermalization) over smaller dimensional lengths. Smaller elements also allow for a more rigid assembly, correspondingly pushing the first modal frequency response to higher frequencies and therefore reducing vibration issues. However, it can be beneficial to build a system with a high NA (which means a low f / #, or large beam diameter and / or small focal length) for photon collection efficiency. While a short focal length may be desirable, it can also make assembly difficult (e.g., actively aligning a 2 mm hemispherical lens or a ¼ inch optic with micron accuracy within a cylindrical cell can be difficult). One solution could be to increase the beam diameter, but this could require a larger mirror, which could slow down the speed of the scan because mirror velocity decreases as a square root of size.

[0065] Power dissipation in the components of a cryogenically integrated scanner can also be a design consideration. For example, dilution refrigerators have finite cooling capacities at different temperatures (e.g., many refrigerators have cooling capacities on the order of a few hundred microwatts at 100 mK). Such limitations can mean that the scanner should have low power dissipation to maintain low temperatures. Some implementations of cryogenically integrated scanners use electrostatically actuated MEMS mirrors, which are capacitively driven and therefore can have minimal power dissipation. Because the scanning element is driven resonantly and the Q factor is inversely proportional to temperature, energy dissipation can decrease rapidly at low temperatures.

[0066] In solid-state MEMS devices, where the mechanical components are actuated via charge carriers, the intrinsic carrier concentration can decrease exponentially at low temperatures (e.g., silicon-based MEMS devices). At cryogenic temperatures, the MEMS device may cease to function because most of the charge carriers have been drained away. Some implementations of cryogenically integrated scanners circumvent this problem by coating most of the device (not just the mirror portion) with metal, ensuring that there are enough charge carriers to actuate the mirror.

[0067] Because the optical fiber may pass through different temperature conditions, the cryogenically integrated scanner can perform polarization maintenance across temperature gradients.

[0068] Cryogenically integrated scanners may perform spatial and spectral multiplexing. Multiple color centers may be addressed per diffraction-limited spot (e.g., through the use of narrow linewidth lasers and differential distortion between color centers, which leads to differences in resonant frequency greater than the narrow linewidth). Cryogenically integrated scanners may also multiplex by using multiple laser beams within a single assembly for speed (e.g., as in LiDAR systems with rotating heads and 16, 32, or 64 "lines").

[0069] The following is a non-exhaustive list of scanning elements that may be used in a cryogenically integrated scanner: spatial light modulators (SLMs), digital micromirror devices (similar to DMD-DLP displays), galvos, deformable mirrors, or fiber endoscopic type designs with the ability to fully fold by 180 degrees.

[0070] Cryogenic integrated scanners may use monostatic or bistatic designs. In a monostatic design, optical pumping is provided through a first path or fiber, and optics or other beam-guiding structures (e.g., nanopillars) are designed to preferentially guide the light emitted from the diamond toward the first fiber. The light can be coupled into the first fiber and separated from the first path (used for optical pumping) by using a circulator or fiber splitter. In a bistatic design, optical pumping is provided through a first path or fiber, and optics or other beam-guiding structures (e.g., nanopillars) are designed to preferentially guide the light emitted from the diamond in another direction, and the light is coupled into a second path or fiber. The first fiber can be used as the pumping path, connecting the on-resonance laser to the color center, while the second fiber can be used as the emission path, connecting the color center's emitted photons to the uplink (quantum) channel. In a monostatic design, excitation photons can be frequency shifted upon excitation and emission by the same electro-optic modulator (EOM), depending on the frequency shifting method used. In a bistatic design, a second EOM is used for the second fiber, or the second fiber may be fed back to the same EOM as the first fiber (e.g., by using a time-gated switch to accept light from the laser or emission side). While a bistatic design can eliminate splitter losses at the input and output, a monostatic design can use a splitter because emitted photons from the color center do not return to the laser but are instead diverted to the uplink channel. For both monostatic and bistatic designs, the emitted light is coupled into the uplink channel, which may be the quantum channel. The light can be sent to a beam splitter to interfere with other photons from another node, photons arriving from a space satellite source, or somewhere else in the world (e.g., another quantum repeater node).

[0071] The cryogenically integrated scanner may use a calibration procedure to generate a lookup table of frequencies at predetermined locations (e.g., cryogenically scanning across a field of view and testing each site (pillar and / or diffraction-limited spot) for fluorescence). Each site can be found by exciting each diffraction-limited spot or pillar with an off-resonance laser having a higher energy (e.g., 532 nm) than the color center wavelength (e.g., 737 nm). Sites that fluoresce indicate the presence of one or more color centers at that location. Photoluminescence excitation measurements can then be performed by scanning a low-linewidth resonant laser around the resonant wavelength region (e.g., 737 nm), and all resonant lines can be located (e.g., one per color center). Spectral lines due to other transitions can be discarded. The end result is that for each location, the lookup table contains one or more resonant frequencies for that color center. If the color center cannot be found, or if the two lines are too close (e.g., within the linewidth of the laser), or if there is some problem (e.g., low photon count rate from that spot or pillar), the cryogenically integrated system may be designed not to address that spot during operation. SiV may have a linewidth of about 150-200 MHz. For on-resonance excitation, the cryogenically integrated scanner may use a laser with a linewidth of about 100 kHz.

[0072] The calibration described above describes how to find the resonant frequency at each location (pillar or diffraction-limited spot). Each spot is mapped to a specific x,y position corresponding to a specific drive signal sent to the scanning element (e.g., MEMS). Additional calibration procedures can be performed initially or from time to time as needed as things move out of and back into alignment due to thermal effects, vibration, or otherwise, to generate a coordinate system. This map-generating calibration routine can take advantage of the fact that lithographically patterned fiducial marks can have very accurate positions relative to addressable color centers, since they have errors on the order of the mechanical tolerances (<100 nm) of the e-beam lithography writing tool.

[0073] The diamond tip may have more than two fiducial marks (since a plane is defined by three points). Specifically, there may be four fiducial marks at all four corners of the field of view, rather than at all four corners of the diamond tip. To perform the map generation routine, the scanning area is expanded. In normal operation, this scanning area is sufficient to cover all addressable color centers. In this routine, the scanning area is larger than during normal operation to ensure that the scanning element scans over the fiducial marks. After scanning the larger area, the scanning element has the associated drive signals (to the scanning element), or equivalently, the x,y positions of the fiducial marks, and therefore the positions of all addressable color centers (because the relative distances between the fiducial marks are known). The fiducial marks are pillars or other nanostructures encoded with a unique pattern (e.g., a QR code). For example, to image the pillars, an on-resonance laser scans the peaks and valleys of the "QR code" to acquire a topographical surface map.

[0074] Cryogenically integrated scanners can suffer from reduced performance due to an increased diffraction-limited spot size across the FOV, especially for high-speed / high-NA systems. Part of the problem is due to a non-flat focal plane. While fiber collection efficiency can also be reduced, focusing on the aforementioned spot size issue, one possible correction is to position multiple diamond chiplets at a slight angle to each other to create a slightly curved focal plane. Such a design can be particularly beneficial when building scanning systems with FOV performance across multiple diamond chiplets.

[0075] The cryogenically integrated scanner may also include a microwave stripline to control one or more of the color center's electrons, the color center's nuclear spin, or the color center's neighboring carbon nuclear spin. For example, the microwave stripline is a copper wire approximately 50 μm long connected to an RF generator (external to the "housing") that can transmit pi / 2, pi, or other pulses to manipulate the spin state (rotation on the Bloch sphere). This microwave stripline can simultaneously control the spins of all color centers. Many other control methods exist that do not involve a single microwave line (e.g., focused acoustic waves, other types of focused electromagnetic waves). A CMOS chip aligned, preferably bonded, to the backside of the diamond tip can be used to transmit focused microwave signals to each pillar / color center. Additionally, a solid-state chip with an acoustic emitter (such as a micromechanical device) can be used to focus acoustic waves on the pillars or color centers.

[0076] Cryogenic integrated scanners can be used to deliver entanglement (e.g., single and dual time bins, frequency bins) between two quantum memories (e.g., two color centers). Single time bin entanglement between two quantum memories can be generated using the following procedure: (1) apply a pi / 2 microwave pulse to each quantum memory to generate a coherent, substantially equal superposition of two quantum states; (2) excite the respective color center in each quantum memory using an on-resonance laser (depending on the spin, the color center electron is either excited and emits a photon or not); (3) interfere the two photon channels (e.g., with a beam splitter) to erase either path information; and (4) measure the output of the beam splitter (e.g., with a photodetector). For a single time bin, if either two clicks or zero clicks are measured, the attempt is considered a failure and discarded. If one click is measured, the two quantum memories are entangled.

[0077] Dual-time-bin entanglement between two quantum memories can be generated using the following procedure: (1) applying a microwave pi / 2 pulse to the color center; (2) laser excitation followed by photon emission; (3) applying a microwave pi pulse to flip the qubit state; (4) laser excitation again followed by photon emission; (5) interfering the two photon channels (e.g., with a beam splitter) to erase either path information; and (6) measuring the beam splitter output (e.g., with a photodetector). In contrast to single-time-bin-based entanglement, the criterion for successful entanglement using dual-time-bin entanglement is a single click in both time bins. This can be advantageous compared to the single-time-bin version, since in the single-time-bin version, a no-click scenario can register as a one-click scenario due to dark counting in the detector. Such an event is less likely to occur with the dual-rail time base. An alternative entanglement basis can use the polarization of light in combination with polarization-sensitive transitions of qubits, where the entanglement criterion is related to the detected polarization state of the detected light (e.g., clockwise or counterclockwise circular polarization).

[0078] Frequency shifting to generate indistinguishable photons for entanglement purposes can be achieved in several ways. In one example, sideband modulation by an amplitude modulator (e.g., at an RF frequency) can shift the input beam so that it is on-resonant with the color center. Emitted photons can be weighted and inversely frequency-shifted with a probability determined by the amount of power in that band. In another example, serrodyne modulation can be performed by an EOM / phase modulator. An increasing portion of a voltage ramp (e.g., linear in rate) is applied to the EOM to generate a phase shift proportional to the voltage. The EOM generates a shift in the frequency of the light it is modulating, which shift is equal to the time derivative of the accumulated phase. The duration of the increasing portion of the ramp is made equal to the characteristic lifetime of the color center being addressed, after which a decreasing portion of the voltage ramp is applied to achieve an equal but opposite phase shift and corresponding frequency shift. This ramp can be used to shift the beam into resonance with the color center and then return the photons emitted from the color center to the original laser frequency used as input to the EOM. The applied voltage waveform is a sawtooth waveform. Furthermore, each addressable color center can be synchronized with a waveform generator, so that there can be a unique voltage ramp at each color center, each with its own resonant frequency due to the local distortion field.

[0079] FIG. 1 illustrates an example quantum state scanning system 100 configured to interact with multiple quantum states over respective successive time intervals during a scan. Light enters a housing 102 (e.g., providing a cryogenic environment) via a fiber optic interface 104 optically coupled to an optical fiber 105. In some examples, at least a portion of the fiber optic interface 104 extends outside the interior of the housing 102. An array 106 of QSEEs (e.g., color centers in a diamond, silicon, or silicon carbide sample) is configured to be contained in or inside the housing 102, with each QSEE configured to provide a quantum state. The array 106 of QSEEs is coupled to the fiber optic interface 104 at least in part through the use of one or more directional structures 108. Portions of the one or more directional structures 108 are each configured to provide a preferential direction for a respective element-coupled light mode. The mode controls light coupling to and / or from different respective subsets of one or more QSEEs in the array 106 of QSEEs. The one or more directional structures 108 may include an array of pillars formed in a material containing the QSEEs, or a hemispherical optically transmissive material configured to cover multiple QSEEs in the subsets of one or more QSEEs. The fiber optic interface 104 provides the fiber-coupled optical mode to a scanning structure 110 configured to be housed on or within the housing 102. The scanning structure 110 can vary the overlap between the fiber-coupled optical mode and different respective ones of the element-coupled optical modes over each of successive time intervals.

[0080] Referring again to FIG. 1 , the scanning structure 110 may include various features for varying the propagation direction and / or mode size of the coupled optical mode. In some examples, the scanning structure 110 may include a first scanning element (not shown) configured to translate the optical fiber interface 104 along a first axis. The scanning structure 110 may also include a second scanning element (not shown) configured to vary the overlap between the fiber-coupled optical mode and different ones of the element-coupled optical modes associated with the directional structure 108 along a second axis substantially perpendicular to the first axis. The second scanning element may include a reflective surface configured to rotate about an axis substantially parallel to the first axis. Alternatively, in some implementations, an electromechanical steering mechanism may be used to move and / or rotate the optical fiber about multiple axes relative to the element-coupled optical modes. Light emitted from the multiple QSEEs may be transmitted outside the housing 102 via the optical fiber 105 and delivered to the optical device 112. The optical device 112 (e.g., an electro-optic modulator) may be connected to (or integrated with) control equipment (e.g., to shift the frequency of light to generate substantially indistinguishable photons for entanglement generation). The optical device 112 may also be connected to a satellite or otherwise transmit optical modes from the optical fiber 105 to another location.

[0081] 2 illustrates an example satellite uplink scheme 200. A first quantum repeater node 202A and a second quantum repeater node 202B (e.g., located at multiple users (Alice and Bob) respectively) transmit photons from a first array of QSEEs 203A and a second array of QSEEs 203B to a satellite 204. At the satellite 204, photons from the first quantum repeater node 202A may interfere with photons from the second quantum repeater node 202B (e.g., by using a beam splitter) and then be detected (e.g., by using a photodiode), thereby performing one or more entanglement operations that may result in entanglement between pairs involving QSEEs from both the first array of QSEEs 203A and the second array of QSEEs 203B. Such entanglement operations may be performed such that photons transmitted to the satellite 204 can entangle with quantum memories stored in each QSEE.

[0082] FIG. 3 illustrates an example quantum repeater node 300 that may be used in an uplink scheme (e.g., to transmit photons to a satellite, as shown in FIG. 2). An array of QSEEs 302 (e.g., SiV or NV color centers in a diamond sample used as quantum memory) is configured as a memory qubit within a housing 304 (e.g., a dilution refrigerator or cryostat). The array of QSEEs 302 may be positioned in the focal plane of a microscope 306 (e.g., a confocal optical microscope), which may be located partially or entirely outside the housing 304. The microscope 306 may optically access the quantum memory 302, for example, through a window (not shown) in the housing 304. An optical scanner 308 (e.g., a galvanometer mirror, a MEMS mirror, or a piezoelectric scanning mirror) may scan the optical mode transmitted through the microscope 306 across the various optical modes of spatially distinct QSEEs or clusters of two or more QSEEs, thereby multiplexing the optical signals from the various QSEE locations onto a single optical channel 309. The single optical channel 309 is optically coupled to an EOM 310. The EOM 310 may be configured to provide a frequency shift between the zero-phonon line (ZPL) frequency of the QSEE and a carrier frequency of at least one of: (1) light transmitted through a beam splitter 312 and generated by a laser 314; or (2) output channels 316, which may be optically coupled to a satellite (not shown) or an optical fiber (not shown). The frequency shift provided by the EOM 310 shifts the frequency of photons emitted by the QSEE to a single frequency, providing indistinguishability in the output channels 316, which may then be used for entanglement generation. In some examples, the laser 314 may be tuned to the carrier frequency of the output channels 316.

[0083] FIG. 4 illustrates an example quantum repeater node 400 that may be used in an uplink configuration. An array of QSEEs 402 is disposed in the focal plane of a 2D optical scanner 403, which is disposed in a housing 404 (e.g., a dilution refrigerator). In this example, the 2D optical scanner 403 multiplexes optical signals from various spatially distinct QSEEs or clusters of QSEEs into a single channel 406. The single channel 406 is optically coupled (e.g., by a fiber interface) to an EOM 410, which can provide a frequency shift to light propagating within the single channel 406. A beam splitter 412 is disposed between a laser 414 and the EOM 410, and an output channel 416 is optically coupled to the single channel 406 and the laser 414 via the beam splitter 412. In some examples, the beam splitter 412 and the laser 414 may be co-integrated on a photonic integrated circuit. Furthermore, the array of QSEEs 402 and the 2D optical scanner 403 may be assembled together in an integrated assembly.

[0084] FIG. 5 illustrates an example satellite uplink system 500. A first array of QSEEs 502A and a second array of QSEEs 502B are disposed within a first housing 504A and a second housing 504B (e.g., dilution refrigerators), respectively. One or more QSEEs from the first array of QSEEs 502A are remotely entangled with one or more QSEEs from the second array of QSEEs 502B. In some examples, photonic nanostructures in the form of diamond nanopillars may be etched into diamond, with each nanopillar containing a color center or cluster of color centers configured for use as a QSEE. In such examples, the nanopillars can direct quantum emission from each QSEE into a well-defined optical mode along the nanopillar's waveguide axis. The first array of QSEEs 502 are optically coupled to a frequency conversion element 506, which shifts optical signals provided by a terrestrial network 508. The optical signals of single photons emitted by the QSEEs in the first array 502A and the second array 502B are optically coupled to a first EOM 510A and a second EOM 510B, respectively, configured to provide a frequency shift to the optical signals for photon indistinguishability and entanglement. Free-space optics can be used to couple the single photons emitted by the QSEEs to the satellite 512, where photons from the first EOM 510A are made to interfere with photons from the second EOM 510B, thereby constituting an entanglement operation. The terrestrial network 508 may also be in communication with other terrestrial networks.

[0085] FIG. 6A shows an example integrated assembly 600A (e.g., for performing 2D optical scanning). An optical fiber 602 outputs light 604 that is collimated by a collimating lens 606A (e.g., a lens or stack of lenses). After collimation, the light 604 is directed to a scanning mirror 608 (e.g., a 2D MEMS mirror). The scanning mirror 608 scans the light 604 across a lens 610 (e.g., HyperSIL). A diamond 612 containing a color center is disposed (e.g., via bonding or clamping) on ​​the back side of the lens 610. The back side may be coincident with the focal plane of the lens 610. The diamond 612 may be heat-sinked to the cold finger of a cryostat (not shown), which may be less than 4 Kelvin, 1 Kelvin, or less than 100 milliKelvin, depending on the application and experimental constraints. As the scanning mirror 608 scans the angle of the beam formed from the light 604, the focal spot scans across the diamond 612, thereby providing spatial multiplexing of single-photon signals from the color centers of the diamond 612 into the optical fiber 602. The laser excitation (e.g., from light provided by a laser optically coupled to the optical fiber 602) and quantum emission from the color centers of the diamond 612 are directed along the same optical path in the forward and reverse directions, respectively. While the light entering and exiting the scanning mirror 608 is depicted as being approximately perpendicular in this example (i.e., at an angle of incidence of approximately 45 degrees relative to the normal to the plane of the mirror), other angles of incidence are possible, and lower angles of incidence, in particular, may be advantageous because they more fully utilize the surface area of ​​the scanning mirror 608. For example, with the scanning mirror 608 configured with an angle of incidence of 22.5 degrees, the scanning mirror 608 presents a larger cross-sectional area to the light 604 than with an angle of incidence of 45 degrees. Another consideration is the placement of the scanning mirror 608 relative to the lens 610. In some examples, scanning mirror 608 may be positioned a focal length away from lens 610. At this distance, as scanning mirror 608 scans the optical mode of light 604 through a range of angles, lens 610 refracts light 604 into a cone that is incident normally on the surface of diamond 612.If the scan mirror 608 is not the focal length away from the lens 610, the cone may tilt as the scan mirror 608 scans through a range of angles, which may be non-ideal.

[0086] FIG. 6B shows an example integrated assembly 600B. Laser light from an optical fiber 602 (e.g., 0.2 numerical aperture) is collimated by a collimating lens 606B, and the collimated light is redirected by a scanning mirror 608 to a first objective lens 614A and a second objective lens 614B, which collectively focus the collimated light onto the color center of a diamond 612. The first objective lens 614A (e.g., a 1 mm diameter spherical-convex / flat lens) partially focuses the light, and the second objective lens 614B (e.g., a HyperSIL spherical-flat lens) focuses the light just outside the planar back surface of the second objective lens 614B, where the diamond 612 is bonded, clamped, or otherwise attached. Thus, the color center of the diamond 612 can be located at the focal point formed by the two objective lenses. The single photon emissions from the color centers occur at the focal points of the two objective lenses, are collected by the two objective lenses, and are returned to the optical fiber 602 along the same path.

[0087] FIG. 6C illustrates an example integrated assembly 600C. In this example, multiple metering standards 616 represent different sections of a differential metering stack including a first material 618, a second material 620, and a third material 622. The differential metering stack can be used to maintain optical alignment, for example, by reducing changes in the focus of light 604 on diamond 612 when integrated assembly 600C is cryogenically cooled. In some examples, first material 618 is Invar (coefficient of thermal expansion (CTE) of 1.26E-6 / 1K), second material 620 is Ti64 (CTE of 8.8E-6 / 1K), and third material 622 is Mo (CTE of 5E-6 / 1K). In general, the three materials, their positions, and their lengths can be selected to result in integrated assembly 600C that is substantially thermally insulating. In some examples, the thermal insulation design can be designed so that after a temperature change in the integrated assembly 600C, the change in the optical path length of the light 604 between the focal point and the lens is substantially similar to the change in the focal point of the lens. Light 604 emitted from the optical fiber 602 (e.g., numerical aperture 0.12) is collimated by the collimating lens 606C, and the collimated light passes through a window 607 and is redirected by a scanning mirror 608 to a first objective lens 614C and a second objective lens 614D (e.g., a solid immersion lens (SIL)). This focuses the collectively collimated light onto the color center of the diamond 612.

[0088] FIG. 7 shows an example array 700 of diamond nanopillars, including nanopillars 702 fabricated from diamond containing color centers. The nanopillars 702 can be fabricated using standard nanolithography and plasma etching techniques. The nanopillars 702 can have a variety of shapes, including cylindrical, tapered, conical, parabolic, or other shapes, with the distal tip of each nanopillar 702, where the color center of each nanopillar 702 is located, having an approximate diameter of approximately 200-400 nm. The nanopillars 702 can have a numerical aperture of approximately 0.4 in some examples. The center-to-center pitch 704 between the nanopillars 702 can be approximately 1-3 μm, resulting in approximately 1,000 nanopillars 702 in an area of ​​diamond that is approximately 30 μm by 30 μm, or approximately 1 million nanopillars 702 in an area of ​​diamond that is approximately 1 mm by 1 mm. The nanopillars 702 may be arranged in a regular array, such as a square lattice, hexagonal lattice, or other regular pattern, or in an irregular pattern (i.e., a non-repeating pattern). The location and size of the nanopillars 702 may be specified during a nanolithography step. In some examples, the size and spacing of the nanopillars 702 may be selected to form a foveated pattern to compensate for spot size degradation at the edge of the field of view of the scanning system. In some examples, the nanopillars 702 may be positioned to coincide with the locations of previously characterized color centers in the diamond.

[0089] 8 shows an example quantum repeater node 800 that may be used in an uplink configuration. Light emitted by a laser 802 is split (e.g., with a 90 / 10 splitting ratio) by a fiber splitter 804. In some examples, 90% of the light is sent to an uplink channel 806, and 10% is sent, perhaps indirectly, by fiber and collimation elements, to an EOM 808 that is optically coupled to a scanning mirror 810 disposed within a housing 811. The scanning mirror 810 is optically coupled to objective optics 812, which transmits the light to a nanopillar array 814. A magnetic field coil 816 and a microwave antenna 818 are disposed near the nanopillar array 814. A computer 820 is configured to control the laser 802, an AWG 816 (arbitrary waveform generator) that controls the EOM 808, a mirror controller 818 that controls the scanning mirror 810, and a magnet and microwave controller 820 that controls the magnetic field coil 816 and microwave antenna 818. The MEMs scanner is controlled by a MEMs controller located external to the dilution refrigerator and connected to it via an electrical cable. The amount of frequency shift imparted to the laser light by the EOM 808 is controlled by the AWG 816. The AWG 816, mirror controller 818, and magnet and microwave controller 820 are all synchronized by a common clock to ensure that the QSEEs across the nanopillar array 814 can be addressed by control signals (e.g., microwave and optical signals) having desired, pre-characterized frequencies, and that single photons emitted by color centers via the nanopillars of the nanopillar array 814 are all spectrally multiplexed onto a common carrier frequency to ensure photon indistinguishability. Fiber splitter 804 reflects and transmits a portion of the incident light from laser 802 , directing indistinguishable single photon emissions into uplink channel 806 .

[0090] FIG. 9A shows an example patterned diamond 900A from a top view. An array of pillars 902A, a recessed region 903, and a ledge 904 have been patterned into the diamond (e.g., by lithography). A first dashed line 906A indicates a cross section (i.e., a plane) formed out of the plane of the paper. The pillars may be formed by removing an annulus of diamond material around each pillar. The diamond material outside the annulus remains and is the same height as the pillar, with both heights equal to the height of the original, unpatterned diamond. In this example, the annulus around each pillar overlaps, so that the diamond material outside the pillar array 902A is completely removed except to form a ledge 904, which may be formed using an electron beam of a size equal to or smaller than the area of ​​the recessed region 903. The array of pillars 902A and ledge 904 may be used as a support structure for a SIL (solid immersion lens) (not shown). As a result, the pressure applied to bring the pillar into contact with the SIL is distributed over a larger area.

[0091] FIG. 9B shows a cross section of an example diamond-SIL configuration 900B through the plane indicated by the first dashed line 906A in FIG. 9A. A SIL 908 is attached to at least one of the array of pillars 902A or the ledge 904. The SIL 908 may be used to reduce light reflection from the end of each pillar of the array of pillars 902A. This reflection could otherwise be significant if the pillars terminated in air due to refractive index mismatch. To reduce this reflection, the end of each pillar may be located within 100 nm of the flat surface of the SIL 908 or may be in direct contact with the SIL 908, depending on other design constraints. In some examples, the SIL 908 is a hemispherical lens pre-loaded into the final processed (i.e., patterned) diamond.

[0092] FIG. 9C shows an example patterned diamond 900C from a top view. An array of pillars 902C, ledges 904, and annular rings 905 have been patterned into the diamond (e.g., by lithography). A second dashed line 906C indicates a cross section (i.e., a plane) formed out of the plane of the paper. The pillars may be formed by removing an annulus of diamond material around each pillar. The diamond material outside the annulus remains and is the same height as the pillars, with both heights equal to the height of the original, unpatterned diamond. In this example, the radius of each annulus is smaller than the spacing between the pillars, resulting in annular rings 905 distributed throughout the array of pillars 902C. The array of pillars 902C, ledges 904, and annular rings 905 may be used as support structures for a SIL (not shown). As a result, pressure applied to bring the pillars into contact with the SIL is distributed over a larger area.

[0093] Figure 9D shows a cross section of an example diamond-SIL configuration 900D through the plane indicated by the second dashed line 906C in Figure 9C. A SIL 908 is attached to at least one pillar, ledge 904, or one or more annular rings 905 in an array of pillars 902C. The ledges 904, annular rings 905, and an array of pillars 902C are made from the same material (e.g., diamond) but are shown in different colors and outlines for added clarity.

[0094] FIG. 9E shows an example patterned diamond 900E from a top view. An array of pillars 902E and recessed regions 903 have been patterned (e.g., by lithographic techniques) into the diamond. A third dashed line 906E indicates a cross section (i.e., a plane) formed out of the plane of the paper. The array of pillars 902E may be used as a support structure for a SIL (not shown). In this example, the supporting diamond structure that is not part of the array of pillars 902E is completely or partially removed. This is done by patterning an etch mask only where the pillars are to be formed and etching all of the diamond around that area in a single step, or by etching the annulus around each pillar in one etch step and etching away all or part of the support material in a subsequent step. Thus, pressure applied to the pillars to bring them into contact with the SIL 908 may be distributed only on the pillars and not on the support material. Such an arrangement may have the advantage that in the case of an uneven diamond surface, it reduces the possibility that the support material (which has been removed in this example) will introduce an undesirable space (i.e., gap) between the end of the pillar and the flat surface of the SIL908.

[0095] Figure 9F shows a cross section of an example diamond-SIL configuration 900F through the plane indicated by the third dashed line 906E in Figure 9E. A SIL 908 is attached to at least one pillar of an array of pillars 902E.

[0096] Referring again to Figures 9A, 9B, 9C, 9D, 9E, and 9F, the patterned diamond may be square, with widths and lengths given by 2 mm x 2 mm or 4.5 mm x 4.5 mm, in some examples. The thickness of the patterned diamond may be less than 0.5 mm or 20 μm. However, the thickness of the diamond may be selected according to the desired alignment with the SIL to optimize properties such as mechanical robustness, substrate transparency, optical interface with the SIL or multi-waveguide array, or substrate flexibility. A flexible substrate may be achieved by reducing the thickness of the diamond until the thickness of the substrate supporting the pillars is less than 2 μm. A flexible substrate may be advantageous for an inverted pillar arrangement. The array of pillars may be 0.5 mm x 0.5 mm in size, with each pillar having a diameter of approximately 150 to 500 nm and a height of 500 nm to 2 μm, and the array of pillars having a pitch of 1 to 5 μm. In general, the extent of the pillar array may be selected to be larger than the field of view of a confocal microscope, or in the case of a multi-waveguide array, larger than the diameter of the multi-waveguide array. Depending on the desired configuration for the SIL, various pillar shapes may be used. The pillars may have a cylindrical shape or an inwardly tapered shape, so that light is directed away from the bulk diamond substrate and toward the SIL. In the case of an inverted diamond substrate in which the pillars are located on the distal side of the SIL or multi-waveguide array, a pillar shape may be selected in which the pillars taper to a larger diameter as they proceed from the distal end toward the bulk diamond substrate. Parabolic shapes may also be used.

[0097] The pillars may be arranged in a regular array, either with square lattice spacing, hexagonal lattice spacing, or other regular arrays, or they may be arranged irregularly. An irregular arrangement may be advantageous for aligning the pillars with the positions of color centers, whose lateral positions may be random. A narrower inter-pillar pitch may be advantageous because it may result in a higher density of QSEEs within the array. However, the inter-pillar pitch may be designed to be large enough to leave enough air between the pillars at the distal ends so that the pillars function primarily as independent waveguides, thereby preventing significant leakage of the mode of one pillar into the mode of an adjacent pillar.

[0098] 10A, 10B, 10C, 10D, 10E, 10F, 10G, 10H, and 10I show exemplary fabrication steps for processing pillars in diamond.

[0099] FIG. 10A shows a diamond layer 1002 that can be used as a starting material for a nanofabrication process and can be prepared, for example, by performing dry etching to relax the surface strain layer, blanket implantation with Si, and annealing at up to 1100° C.

[0100] FIG. 10B shows a SiN layer 1004 deposited (e.g., by PECVD) on the diamond layer 1002, an e-beam resist layer 1006 (e.g., ZEP) spun on the SiN layer 1004, and a charge dissipation layer 1008 (e.g., ESpacer) spun on the e-beam resist layer 1006.

[0101] FIG. 10C shows the SiN layer 1004 on the diamond layer 1002, the e-beam resist layer 1006 on the SiN layer 1004, and the charge dissipation layer 1008 on the e-beam resist layer 1006 after patterning the region 1110 (e.g., via e-beam lithography).

[0102] FIG. 10D shows the SiN layer 1004 on the diamond layer 1002 and the e-beam resist layer 1006 on the SiN layer 1004 after patterning the patterned region 1110 and removing the charge dissipation layer (e.g., the charge dissipation layer of FIG. 10C).

[0103] Figure 10E shows the SiN layer 1004 on the diamond layer 1002 and the e-beam resist layer 1006 on the SiN layer 1004 after removing the patterned areas that result in the disk pattern (e.g., patterned areas 1110 in Figure 10D).

[0104] FIG. 10F shows the SiN layer 1004 on the diamond layer 1002 and the e-beam resist layer 1006 on the SiN layer 1004 after the pattern of the removed e-beam resist layer 1006 has been transferred (e.g., by etching) to the SiN layer 1004.

[0105] FIG. 10G shows the SiN layer 1004 on the diamond layer 1002 after stripping the e-beam resist layer (eg, e-beam resist layer 1006 in FIG. 10F).

[0106] FIG. 10H shows the SiN layer 1004 on top of the diamond layer 1002 after reactive ion etching with O 2 plasma has been performed to transfer the pattern of the SiN layer 1004 to the diamond layer 1002 and form diamond pillars 1012.

[0107] FIG. 10I shows the diamond layer 1002 and diamond pillars 1012 formed after stripping away the SiN layer (eg, SiN layer 1004 in FIG. 10H).

[0108] FIG. 11A shows an example entanglement system 1100A that provides entanglement across a network of quantum repeaters, including “P” nodes and “Q” nodes. In some examples, the number of quantum repeaters exceeds 10,000. A diamond 1102 includes an array of patterned pillars 1103, each of which contains one or more spectrally distinct color centers that can be used as QSEEs. The pillars 1103 can be used to direct the emission of the color centers (i.e., QSEEs) into well-defined output modes. The diamond 1102 can be maintained at cryogenic temperatures within a housing 1104. A pair of controllable mirrors 1108 can be used to scan the spatial overlap between the carrier mode (e.g., the carrier mode of a pump laser 1112 passing through a first polarizing beam splitter 1110A) and the output modes of the various QSEEs. The pump laser 1112 is optically coupled to a frequency conversion module 1113 (e.g., an EOM) by a mirror 1114. Within a given pillar 1103, the laser light within the pillar 1103 can be modulated (e.g., in frequency) to spectrally select and address a single color center, whose emission is then counter-modulated by the same amount back to the carrier frequency of the light output by the pump laser 1112 and sent to the frequency conversion module 1113. The emission from the QSEE in one of the pillars 1103 is optically coupled to a "Q" fiber 1118A, which is optically coupled to a second polarizing beam splitter 1110B. Here, this emission is superimposed with the emission from a second remote repeater node, labeled "P." The second remote repeater node is optically coupled to the "P" fiber 1118B, and the output port of the second polarizing beam splitter 1110B is monitored by a single photon detector 1122.

[0109] FIG. 11B shows an example entanglement system 1100B and an example pulse timing sequence 1130 (e.g., derived from an excitation laser 1112). The pulse timing sequence 1130 includes approximately 50 resonant excitation pulses per color center, with each pillar 1103 in the diamond 1102 including approximately 10 color centers. Each pillar 1103 may be addressed within 1 microsecond, in some examples. Within that 1 microsecond, there may be 10 clock cycles, each consisting of 50 resonant excitation pulses. In some examples, for each clock cycle, the laser may be modulated to a different frequency (e.g., using an electro-optic modulator controlled by an AWG (arbitrary waveform generator)) to address each of the spectrally distinct color centers. Over the next microsecond, the excitation / collection mode may be spatially scanned to overlap with the output mode of another pillar 1103. In this example, the excitation laser 1112 may be modulated to a frequency of ω l , which passes through the first EOM 1124A and the first AWG 1126A. This allows for varying the frequency and pulse shape (e.g., frequency ω l Two pulses of frequency ω m and ω n A controllable mirror 1108 allows the light from the pump laser 1112 to be scanned on the pillar 1103 along a scan path 1128. A second EOM 1124B and a second AWG 1126B convert the pulses at frequency ω m and ω n Two pulses of ω l may be used to return the frequency

[0110] FIG. 12A shows an example optical scanner 1200A including a fiber 1202 and a scannable fiber interface 1204. The fiber 1202 (e.g., a single-core or multi-core fiber) is inserted through the wall of a housing 1206 and attached to the scannable fiber interface 1204 via a piezoelectric clamp 1208. The fiber 1202 has a length R and a diameter D after the piezoelectric clamp 1208. The end of the fiber 1202 may be shaped to provide a lens effect that focuses light emitted by the fiber 1202 to a focal region at a distance from the end of the fiber 1202. The piezoelectric clamp 1208 translates with an applied electrical signal to excite one or more of multiple bending modes 1209A of the fiber 1202. This causes the multiple optical modes of the fiber to be scanned across a diamond 1210, which is fixed relative to a leaf spring 1212 and a holding structure 1214. The holding structure may be integrally formed from the wall of the housing 1206. In some examples, the fiber 1202 may be azimuthally asymmetric. As a result, it has two bending modes with different frequencies in different directions. By electrically driving the piezoelectric clamp 1208 at one frequency, only one of the bending modes 1209A is driven, resulting in motion along a well-defined axis. Thus, the resonant frequency of the fiber bending mode 1209A sets a limit on the scan rate along the axis of the bending mode 1209A. Scanning along an axis perpendicular to the bending mode 1209A can be achieved by exciting an orthogonal bending mode of the fiber 1202 or by translating the fiber 1202 relative to the diamond 1210 via the piezoelectric clamp 1208. The scannable fiber interface 1204 may include a threaded adjustment screw (not shown) to enable translation of the fiber 1202 along its propagation axis, which is perpendicular to the plane of the diamond 1210. Thus, the adjustment screw allows alignment of the focal spot of the fiber 1202 to the diamond 1210. The adjustment screws and housing 1206 may be made of a material that has similar thermal expansion / contraction properties as the fiber 1202. As a result, the focal spot may be aligned at room temperature and the system may remain aligned after being cooled to cryogenic temperatures.The fiber 1202 may be driven at or near one or more resonant frequencies of its bending modes 1209A, which moves the end of the fiber 1202 either translationally (as shown in FIG. 12A ) or rotationally (as shown in FIG. 12B ), enabling scanning of the overlap of the fiber optical modes with various optical modes associated with pillars in the diamond 1210. Translation can be achieved by exciting the fundamental bending mode 1209A of the fiber 1202, while rotation of the fiber end can be achieved by exciting higher-order bending modes such that nodal points of the fiber motion are located near the tip of the fiber 1202.

[0111] 12B shows an example optical scanner 1200B. A piezoelectric clamp 1208 can excite one or more rotational modes 1209B of the fiber 1202. The fiber 1202 can be weighted toward its end to shift the position of the nodal point 1216 toward the weight. If higher-order bending modes are driven such that the tip of the fiber 1202 moves primarily by rotation, a system of relay optics 1218, such as a pair of lenses in a 4f configuration and a focusing lens, can be used to convert the rotational motion of the end of the fiber 1202, and thus the light 1220 emitted by the fiber 1202, into translational motion of the light 1220 across the diamond 1210 and to focus the light 1220 onto the diamond 1210. The SIL 1222 may be used to reduce total internal reflection (e.g., in a hemispherical configuration) or may be used in combination with the relay optics 1218 as an additional focusing element (e.g., in a Weierstrass configuration).

[0112] Referring again to Figures 12A and 12B, diamond 1210 may be structured into a nanophotonic structure. In this case, the numerical aperture of the end of fiber 1202 is preferably matched to the numerical aperture of the optical mode of the nanophotonic structure. For example, if the nanophotonic structures are cylindrical, tapered conical, or parabolic pillars, they may have a numerical aperture of less than 0.7, less than 0.4, or less than 0.15. Thus, the end of the fiber shaped to function as a lens element can be designed to have a similar numerical aperture so that the optical mode of the nanophotonic structure is well matched to one or more modes of fiber 1202. In some examples, fiber 1202 can be driven in one direction by resonantly driving bending mode 1209A and in another direction by translating fiber 1202 along its propagation axis.

[0113] 13A, 13B, 13C, 13D, 13E, and 13F show various example scanning configurations.

[0114] 13A shows an example scanning configuration 1300A. A fiber 1302 enters a housing 1304 and terminates near a diamond 1306 held in a sample holder 1308, so that both the fiber 1302 and the diamond 1306 are in the focal plane of a lens 1310. The lens 1310 is positioned a focal length away from the plane of the diamond 1306, so that light emitted from the fiber 1302 and light emitted from QSEEs (e.g., color centers), represented as dots in the diamond 1306, are collimated by the lens 1310 and incident on a scanning mirror 1312 (e.g., a MEMs mirror). The scanning mirror 1312 may be positioned one focal length from the lens 1310 and on the opposite side of the lens 1310 from the diamond 1306 and the fiber 1302. Thus, movement of the scanning mirror 1312 along two axes translates the optical fiber modes across the diamond 1306, the overlap of which is scanned across the various QSEEs. The scanning mirror 1312 need not be positioned exactly one focal length from the lens 1310, but such placement can be advantageous in terms of translating the motion of the scanning mirror 1312 into translation of the fiber focal spot across many different optical modes on the diamond 1306, and in terms of translating the optical modes of the fiber across the diamond 1306 without rotating them about an axis orthogonal to the plane of the diamond 1306. In this example, a first beam 1314A is emitted by the fiber 1302, passes through the lens 1310, and is reflected by the scanning mirror 1312, thereby generating a second beam 1314B that is redirected through the lens 1310 and incident on the diamond 1306. One or more QSEEs in diamond 1306 re-emit the light as a second beam 1314B, which then propagates back into fiber 1302 in the reverse order.

[0115] FIG. 13B shows an example scanning structure 1300B. In this example, a hyperSIL 1316 is used as the focusing element. As in FIG. 13A, the fiber 1302 terminates in the same plane as the QSEE within the diamond 1306, which is located at the focal plane of the hyperSIL 1316, so that both the fiber mode and the emission from the QSEE are collimated by the hyperSIL 1316. The scanning mirror 1312 may be located one focal length away from the hyperSIL 1316. In general, the focal length within the (high-index) hyperSIL 1316 is typically longer than the focal length outside the hyperSIL 1316 in its low-index vacuum environment. In both the configurations of FIGS. 22A and 22B, the termination of the fiber 1302 can be positioned close to the QSEE so that the angle of incidence of the fiber mode on the scanning mirror 1312 deviates minimally from normal incidence. This increases the cross-sectional area of ​​the scanning mirror 1312 relative to the fiber mode. This can be advantageous as it allows for an increased number of addressable modes given the scan rate and mirror size of the scanning mirror 1312.

[0116] 13C shows an example scanning configuration 1300C. In this example, a fiber lens 1318 is used in combination with a hyperSIL 1316 to scan light from a fiber 1302 across a diamond 1306. The fiber 1302 and fiber lens 1318 may be positioned near the diamond 1306 and the hyperSIL 1316 to reduce the angle of incidence on the scanning mirror. The fiber interface terminates in the focal plane of the fiber collimating lens, such that the fiber mode is collimated by the fiber collimating lens and the collimated mode is incident on the scanning mirror 1312. The diamond 1306 may be positioned in the focal plane of the hyperSIL 1316. As a result, emission from the QSEE in the diamond 1306 is collimated by the hyperSIL 1316 and the collimated emission is incident on the scanning mirror 1312. In this configuration, the fiber lens 1318 and the hyperSIL 1316 can be selected to optimize the overlap of the fiber mode with that of the QSEE. In particular, the fiber mode and the mode of the QSEE do not need to have the same numerical aperture, as they can be aligned by appropriate selection of the focal lengths of the hyperSIL 1316 and the fiber lens 1318. The scanning mirror 1312 can be positioned at a distance of one focal length of the hyperSIL 1316 from a point on the surface of the hyperSIL 1316 to increase the number of addressable modes on the diamond 1306.

[0117] 13D shows an example scanning configuration 1300D. A first beam 1314A is emitted from fiber 1302, reflected off concave mirror 1320, reflected off scan mirror 1312, reflected back off concave mirror 1320, and directed toward the QSEE at diamond 1306. The emission from the QSEE can follow the same path in the reverse direction. Concave mirror 1320 may be configured with a fixed curvature and a fixed distance from the three elements (fiber 1302, scan mirror 1312, and diamond 1306) to bring the fiber mode (e.g., first beam 1314A) to a focal point at scan mirror 1312 and back to a focal point at diamond 1306. Alternatively, concave mirror 1320 may be configured with a fixed curvature and a fixed distance from the three elements such that the fiber mode is collimated by concave mirror 1320, then incident on scan mirror 1312 as a collimated beam, and subsequently focused to diamond 1306 by a second reflection from concave mirror 1320. Fiber 1302, scan mirror 1312, and diamond (e.g., QSEE) may be arranged coplanar with one another, but may also be arranged in a non-coplanar configuration. In some examples, two reflections from concave mirror 1320 result in the fiber mode being focused to diamond 1306. This is either by initially collimating the light on the first reflection and subsequently refocusing it on the second reflection, or by creating an intermediate focus between the first and second reflections (e.g., at the position of scan mirror 1312).

[0118] FIG. 13E shows an example scanning structure 1300E. The fiber 1302 terminates at the same plane as the QSEE of the diamond 1306, and a hemispherical SIL (solid immersion lens) 1321 is in contact with the diamond 1306. In this example, the fiber 1302 may terminate as a bare fiber (i.e., not in a fiber interface) disposed on the flat surface of the SIL 1321. The fiber 1302, diamond 1306, and SIL 1321 are housed on a first threaded plug 1324A inserted into a tubular housing 1326. A lens 1310 is fixed to the tubular housing 1326 at a distance of one focal length from the plane of the fiber 1302 and diamond 1306, and the distance can be controlled by rotating the first threaded plug 1324A. At the other end of the tubular housing 1326 is a scanning mirror 1312 attached to a second threaded plug 1324B. In some examples, the distance between the scan mirror 1312 and the lens 1310 is one focal length of the lens 1310. However, this distance may not be as important as the distance between the lens 1310 and the fiber 1302 and diamond 1306. The second threaded plug 1324B includes an electrical feedthrough 1328 for controlling the scan mirror 1312. In this configuration, the lens 1310 collimates the fiber-coupled mode and the QSEE-coupled mode, and the scan mirror 1312 scans the collimated fiber-coupled mode, which overlaps with the various QSEE-coupled modes. The SIL 1321 can provide a high-index medium for the QSEE-coupled mode to propagate, so that the numerical aperture of the QSEE-coupled mode is better matched to the numerical aperture of the fiber 1302, and the mode can be well-matched by the lens 1310. While the numerical aperture of a typical single-mode fiber is 0.12, the numerical aperture of the QSEE coupled mode can be 0.3-0.4 for various tapered pillar nanostructures terminating directly into air. For tapered nanostructures terminating into SIL1321, the numerical aperture can be reduced to 0.1-0.15, which better matches the fiber numerical aperture.

[0119] FIG. 13F shows an example scanning structure 1300F. Fiber 1302 terminates in the same plane as the diamond 1306 QSEE. A scannable concave mirror 1330 can be scanned either rotationally or translationally (e.g., by piezoelectric actuation). A first beam 1314A emitted from fiber 1302 is reflected once by scannable concave mirror 1330, resulting in a second beam 1314B that is brought to a focal point in the plane of the diamond 1306 QSEE. The second beam 1314B can be swept across various QSEEs by movement of scannable concave mirror 1330.

[0120] 14A, 14B, and 14C illustrate various aspects of an example scanning structure having multiple optical modes. In some examples, simultaneous scanning of multiple QSEE coupled optical modes can be performed in parallel.

[0121] FIG. 14A shows an example scanning structure 1400A. A fiber 1402 can transmit and emit multiple different spatial modes that exit the fiber 1402 as an array of multiple parallel light beams 1404. Each of the multiple parallel light beams 1404 has an intensity profile that diverges as it propagates away from the end of the fiber 1402. A first lens 1406A collimates the diverging light beams 1404 and simultaneously redirects their axes to converge on a scanning mirror 1408. The light beams 1404 are reflected by the scanning mirror 1408 and pass through a second lens 1406B. This causes the various collimated light beams 1404 to focus onto an array of QSEEs 1410 at various different lateral focal positions of a diamond 1412. Thus, the light beams 1404 corresponding to the different fiber modes are imaged to an array of focal spots corresponding to the positions of the QSEEs 1410. In some examples, the second lens 1406B may instead be a pair of relay and objective lenses (e.g., as shown in FIG. 14B ), or other combinations of lenses or other optical elements. Rotation of the scanning mirror 1408 may cause the entire array of focal spots to translate laterally across an array of QSEEs 1410. Emission from the QSEEs is directed toward the second lens 1406B and coupled into corresponding fiber modes. This allows laser excitation and photon collection from the QSEEs to occur in parallel.

[0122] 14B shows an example scanning structure 1400B. In this example, a mode-matching element 1414 operates in the Fourier plane with respect to the output of the fiber 1402. The output of the fiber 1402 contains multiple independent spatial modes that are collimated by a first lens 1406A. The multiple independent spatial modes are directed to a mode-matching element 1414 (e.g., a spatial light modulator) that includes an array of independently controlled phase-adjusting elements (not shown), after which the reflected light is collected by an imaging system including a first relay lens 1416A and a second relay lens 1416B and directed to a scanning mirror 1408 and then to an objective lens 1418. The multiple phase-adjusting elements may be individually adjusted to match the modes of the fiber 1402 to a pattern of spatially separated focal spots in the plane of the QSEE 1410 that may be scanned across the QSEE 1410 by the scanning mirror 1408.

[0123] 14C shows an example scanning procedure 1400C that includes a scan path 1420 for each mode of a multimode fiber. The scan paths 1420 are associated with respective light beams 1404 that may be scanned in parallel over a QSEE 1410 in a diamond 1412 (e.g., using scanning structures 1400A or 1400B of FIG. 14A or FIG. 14B, respectively).

[0124] In general, a fiber waveguide core array can be used to transmit light between one or more optical fibers and one or more diamonds containing multiple QSEEs. The fiber waveguide core array may be positioned at its distal end in close proximity to a printed circuit board (PCB) or CMOS chip having current-carrying traces and electrical connections for providing electromagnetic fields to manipulate the state of the QSEEs. The PCB or CMOS may be configured with holes or trenches. This allows the distal end of the fiber waveguide core array to be flush with the surface of the PCB or CMOS chip, and the diamond sample may be attached to the distal end of the fiber waveguide core array inserted into the hole or trench in the PCB or CMOS chip. Such an arrangement may have the advantage of placing the QSEE in the plane of the electrical elements of the PCB or CMOS chip, where the electromagnetic field may be maximized or where the electromagnetic field may exhibit advantageous properties such as maximum or minimum gradient.

[0125] FIG. 15A shows an example scanning structure 1500A. In this example, a single-mode fiber 1502 is optically coupled to a mode coupler 1504 (e.g., a spatial light modulator). The mode coupler 1504 can convert the single mode of the single-mode fiber 1502 into two or more modes in a fiber interface 1505, such as a fiber waveguide core array containing array elements (not shown). The fiber interface 1505 enters a housing 1508 (e.g., maintained at a temperature below 4 Kelvin) and is optically coupled to a diamond 1510. The fiber interface 1505 can support many modes that are significantly isolated from each other but may be weakly coupled to a few other modes. For example, the fiber interface 1505 can be a fiber waveguide core array with 100 or more modes, with each mode weakly interacting with fewer than 10 neighboring modes and minimal or no interaction with the remaining other modes. In another example, the fiber interface 1505 may be a fiber waveguide core array with as many as 100,000 or more modes, with each mode interacting with fewer than 10 neighboring modes and minimal or no interaction with the remaining other modes. A diamond 1510 containing multiple QSEEs may be placed directly at the distal end of the fiber interface, such that pump light from one mode of the fiber interface 1505 overlaps with at least one QSEE-coupled optical mode. Emission from the QSEE-coupled optical mode may be transported back to the proximal end of the fiber interface 1505 and remain primarily in the same fiber mode, with weak leakage to adjacent optical modes. The composite mode may then be reconfigured by the mode coupler 1504 to have significant overlap with a single Gaussian spatial mode. Light from multiple QSEEs that may overlap the same fiber interface mode may be distinguished by the spectral frequency of their emission (e.g., if the QSEEs are non-uniformly spectrally distributed).

[0126] FIG. 15B shows an example interface 1500B between a diamond 1510 and a fabricated CMOS 1512. The diamond 1510 includes pillars 1514 in which QSEEs 1516 (e.g., color centers) reside. RF control circuitry 1518 applies control signals (e.g., RF or microwave) to the QSEEs 1516, allowing them to manipulate their respective quantum states. The pillars 1514 are inserted into CMOS wells 1520 so that the RF control circuitry 1518 can more easily interact with the QSEEs 1516. In some examples, each CMOS well 1520 may have multiple pillars 1514 and / or multiple QSEEs 1516 inserted therein. The diamond 1510 may include multiple such pillars, and the fabricated CMOS 1512 may include multiple such CMOS wells.

[0127] FIG. 15C shows an example interface 1500C between diamond 1510, fabricated CMOS 1512, and a fiber waveguide core array 1506 including array elements 1507.

[0128] FIG. 15D shows an example fiber waveguide core array 1506 including array elements 1507.

[0129] In some examples, a diamond sample may be nanostructured and positioned relative to the distal end of a fiber waveguide core array (e.g., as shown in Figures 16A and 16B, where the fiber waveguide core array replaces the SIL). To multiplex the various modes of a fiber waveguide core array into a single fiber input / output mode, a scanning confocal setup may be used, with the proximal end of the fiber waveguide core array positioned in the focal plane of the confocal setup, such that scanning the confocal spot across the focal plane (e.g., via scanning a MEMs mirror, galvanometer mirror, or any other scanning arrangement) results in overlap of the modes of the single input / output fiber with the individual modes of the fiber waveguide core array. An array of QSEEs may be randomly positioned relative to the fiber waveguide core array. In that case, the elements should be packed more densely than the array elements (i.e., cores) to ensure at least one element per core. Alternatively, an array of QSEEs may be aligned with each core of the fiber waveguide core array. To achieve the latter, the arrangement of the cores may first be imaged, and then an array of QSEEs with the same arrangement may be fabricated (eg, using standard nanofabrication methods).

[0130] FIG. 16A shows a portion of an example scanning structure 1600A. Light 1602 enters a first SIL 1604A and is optically coupled through a first diamond 1606A to a parabolic pillar 1608A containing a QSEE 1610. The pillar (e.g., parabolic pillar 1608A) can have a shape such as a tapered cone, a parabolic shape, or other waveguiding shape that redirects quantum state emission into a directional output mode, thus functioning as a waveguiding element. The pillar can be located on the distal side of the diamond (e.g., the first diamond 1606A shown in FIG. 16A) or the proximal side of the diamond (e.g., the second diamond 1606B shown in FIG. 16B). The parabolic pillar 1608A is inserted into a first RF PCB 1612A. In the distal case, emissions from the multiple QSEEs 1610 are directed primarily by the parabolic pillar 1608A towards the first diamond 1606A, and thus towards the imaging system. The proximal surface of the diamond may advantageously be located on the first SIL 1604A or hyperSIL (not shown) so that reflections from the interface to the vacuum environment are minimized. In this case, the thickness of the diamond substrate and the propagation length through the SIL 1604A may be selected so that the parabolic pillar 1608A is located one focal length from the proximal surface (i.e., the curved surface) of the first SIL 1604A, so that their emissions are collimated by the proximal SIL surface. If the first SIL 1604A is fabricated from a material with a similar refractive index to the first diamond 1606A (e.g., diamond or cubic zirconia), it has the advantage that reflections from the diamond-SIL or diamond-hyperSIL interface are minimized, and so is refraction of the optical mode at that interface. Optical modes at the diamond-SIL interface (or diamond-hyperSIL interface) of a waveguide element (e.g., parabolic pillar 1608A) may exist in the diffraction regime due to their confinement by the waveguide element; in this case, the use of the first SIL 1604A (or hyperSIL) helps to further minimize or reduce subsequent diffraction of the mode as it propagates through the first SIL 1604A (or hyperSIL).This has the advantage of reducing the numerical aperture of the modes in the optical system compared to the numerical aperture of the equivalent modes when transferred directly from the nanostructured waveguide element to the vacuum environment at the planar diamond-vacuum interface. The first diamond 1606A may be fixed to the flat surface of the first SIL 1606A or hyperSIL 1604A by a clamping mechanism or by adhesive bonding, or it may simply rest in contact with the surface and be held in place by surface bonding forces. In the latter case, the first diamond 1606A on which the waveguide elements are arrayed may be reduced to a minimum thickness (e.g., less than 5 μm or less than 1 μm) to facilitate the surface bonding forces becoming the dominant force. As a result, the diamond's stiffness is less than the surface bonding forces, and the first diamond 1606A deforms under the surface bonding forces and becomes flush with the flat surface of the first SIL 1604A. The distal location also has the advantage that, for example, electrical connections for carrying radio frequency signals generated by the first RF PCB 1612A to control the quantum state of the QSEE 1610 can be positioned close to the QSEE 1610 (e.g., less than 100 μm, or less than 5 μm).

[0131] FIG. 16B shows a portion of an example scanning structure 1600B. Light 1602 enters a second SIL 1604B, passes through a second diamond 1606B, and is optically coupled to a parabolic pillar 1608A containing a QSEE 1610. A leaf spring 1614 secures the second diamond 1606B relative to the second SIL 1604B. In this proximal case, multiple QSEEs 1610 are positioned proximal to the second diamond 1606B. This has the advantage that the diamond substrate is not in the optical path, so it does not introduce aberrations or scattering losses into the optical system. Furthermore, if a SIL or hyperSIL is used (e.g., the second SIL 1604B), the waveguide element is pressed against the distal SIL / hyperSIL surface, thereby minimizing the gap between the two materials.

[0132] 16C shows various example diamond structures. An unstructured diamond 1620 may include a QSEE 1610, while a structured diamond may have parabolic pillars 1614A, conical pillars 1614B, or cylindrical pillars 1614C. There may or may not be multiple QSEEs 1610 within each pillar.

[0133] FIG. 17A shows a portion of an example scanning structure 1700A. Light 1702 enters a first hyperSIL 1704A, passes through a first diamond 1706A, and is directed toward two pillars 1707, one of which includes a QSEE 1708. In this example, the surface of the first diamond 1706A is curved to partially or fully match the lower curved surface of the first hyperSIL 1704A. In some examples, the back surface of the first hyperSIL 1704A is curved to match the focal plane of the first hyperSIL 1704A, so that the diamond conforms to the hyperSIL surface. The hyperSIL surface may be curved (e.g., as shown in FIG. 17A) or composed of multiple planar segments that approximate a curved surface (e.g., as shown in FIG. 17B) to match the curvature of the focal plane of the hyperSIL or the curvature of the composite imaging system, which may generally not be planar. The diamond substrate (e.g., the portion of the first diamond 1706A not including the pillars 1707) can be made thin enough that the surface bonding forces overcome the rigidity of the diamond substrate and the substrate conforms to the hyperSIL surface, which has the advantage that the various waveguiding elements can be positioned simultaneously along the focal plane of the imaging system.

[0134] 17B shows a portion of an example scanning configuration 1700B. Light 1702 enters a second hyperSIL 1704B and is directed through a second diamond 1706B that includes multiple planar segments that approximate a curved surface.

[0135] While the present disclosure has been described in connection with certain embodiments, it is to be understood that the present disclosure is not limited to the disclosed embodiments, but on the contrary is intended to cover various modifications and equivalent arrangements included within the scope of the appended claims, which scope is to be accorded the broadest interpretation so as to encompass all modifications and equivalent structures permitted under law.

Claims

1. 1. An apparatus for interacting with a plurality of quantum states over respective successive time intervals during a scan period, comprising: a housing including at least one optical fiber interface configured to provide a fiber-coupled optical mode that controls light coupling to and / or from an optical fiber, at least a portion of the optical fiber extending outside an interior of the housing; an array of quantum state emitting devices configured to be contained in or within the housing, each quantum state emitting device configured to provide a quantum state; and one or more directional structures, each of a plurality of portions of the one or more directional structures configured to provide a preferential direction for an associated element-coupled light mode that controls light coupling to and / or from a different respective subset of one or more quantum state emitting devices of the quantum state emitting devices; a scanning structure configured to be contained in or inside the housing, the scanning structure configured to vary an overlap between the fiber coupled optical mode and a different one of the element coupled optical modes over each time interval of the successive time intervals.

2. The apparatus of claim 1 , wherein the plurality of quantum state emitters comprises a plurality of quantum state storage elements each configured to receive a quantum state that is temporarily stored and emitted after a period of time.

3. The apparatus of claim 2 , wherein the fixed time period is longer than the scanning period.

4. The device of claim 1 , wherein the plurality of quantum state emitting devices comprises a plurality of lattice mismatches.

5. The device of claim 4 , wherein the plurality of lattice mismatches comprises a plurality of emitting point defects in a crystal lattice material.

6. The apparatus of claim 5 , wherein the crystal lattice material comprises diamond, silicon, or silicon carbide.

7. The apparatus of claim 1 , wherein the optical fiber comprises a single-mode optical fiber.

8. The apparatus of claim 1 , wherein the housing comprises a chamber configured to enclose the array of quantum state emitting devices, the one or more directional structures, and the scanning structure.

9. 10. The apparatus of claim 8, wherein the chamber comprises a cryogenic chamber configured to maintain a temperature within the cryogenic chamber below about 10 Kelvin.

10. The device of claim 1 , wherein the one or more directional structures comprise an array of pillars formed in a material comprising the quantum state emitting devices.

11. 11. The device of claim 10, wherein the one or more directional structures comprise hemispherical optically transparent material configured to cover more than one of the subset of one or more quantum state emitting devices of the plurality of quantum state emitting devices.

12. The apparatus of claim 1 , wherein the scanning structure includes a first scanning element configured to move the position of the fiber optic interface along a first axis.

13. 13. The apparatus of claim 12, wherein the scanning structure includes a second scanning element configured to vary, along a second axis substantially perpendicular to the first axis, an overlap between the fiber coupled optical mode and a different respective element coupled optical mode of the plurality of element coupled optical modes associated with a directional structure.

14. The apparatus of claim 13 , wherein the second scanning element includes a reflective surface configured to rotate about an axis substantially parallel to the first axis.

15. 1. A method for interacting with a plurality of quantum states over respective successive time intervals in a scan period, comprising: providing a fiber-coupled optical mode from at least one optical fiber interface that controls optical coupling to and / or from an optical fiber, wherein at least a portion of the optical fiber extends outside an interior of a housing that includes the at least one optical fiber interface; providing a quantum state from each quantum state emitting device in an array of quantum state emitting devices contained in or inside the housing; providing a preferential direction for an associated element-coupled light mode from each of a plurality of portions of one or more directional structures that controls light coupling to and / or from different respective subsets of one or more quantum state emitting devices of the plurality of quantum state emitting devices; scanning a scanning structure contained in or inside the housing to vary, over each of the successive time intervals, an overlap between the fiber coupled optical mode and a different one of the element coupled optical modes; A method comprising:

16. 1. An apparatus for interacting with a plurality of quantum states, comprising: a housing including at least one optical interface providing at least 100 fiber-coupled optical modes, at least a portion of the optical interface extending outside an interior of the housing; an array of quantum state emitting devices configured to be contained in or inside the housing, each quantum state emitting device configured to provide a quantum state; and one or more directional structures, portions of the one or more directional structures each configured to provide a preferential direction for an associated element-coupled light mode that controls light coupling to and / or from a different respective subset of one or more quantum state emitting devices of the plurality of quantum state emitting devices.

17. 17. The apparatus of claim 16, wherein the at least one optical interface comprises one or more multimode optical fibers.

18. 17. The apparatus of claim 16, wherein the at least one optical interface comprises at least one array of multiple optical cores.

19. 17. The apparatus of claim 16, further comprising a scanning structure configured to vary an overlap between at least one fiber-coupled optical mode of the at least 100 fiber-coupled optical modes and at least one element-coupled optical mode of the plurality of element-coupled optical modes over each successive time interval in a scan period.

20. The apparatus of claim 19 , wherein the scanning structure is disposed outside the housing.

21. 20. The apparatus of claim 19, wherein the scanning structure comprises one or more spatial light modulators configured to control light coupling to and / or from the at least one optical interface.

22. The device of claim 21 , wherein at least one of the one or more spatial light modulators is disposed outside the housing.

23. 20. The apparatus of claim 19, wherein the scanning structure is configured to vary overlap between at least one fiber-coupled optical mode of the at least 100 fiber-coupled optical modes and at least one element-coupled optical mode of the plurality of element-coupled optical modes by interfering two or more fiber-coupled optical modes of the at least 100 fiber-coupled optical modes.

24. 20. The apparatus of claim 19, wherein the scanning structure is configured to vary overlap between at least one fiber-coupled optical mode of the at least 100 fiber-coupled optical modes and at least one element-coupled optical mode of the plurality of element-coupled optical modes by selectively exciting one or more of the optical modes of the at least 100 fiber-coupled optical modes.

25. 1. A method for interacting with a plurality of quantum states, comprising: providing at least 100 fiber-coupled optical modes from at least one optical fiber interface, wherein at least a portion of an optical fiber extends outside an interior of a housing containing the at least one optical fiber interface; providing a quantum state from each quantum state emitting device in an array of quantum state emitting devices contained in or inside the housing; providing a preferential direction for an associated element-coupled light mode from each of a plurality of portions of one or more directional structures that controls light coupling to and / or from different respective subsets of one or more of the quantum state emitting devices; A method comprising:

Images