Periodic Reporting for period 1 - QLASS (Quantum Glass-based Photonic Integrated Circuits)
Periodo di rendicontazione: 2024-01-01 al 2025-06-30
The second quantum revolution aims beyond the understanding of basic science to its application in novel technologies. In this framework, the development of integrated quantum photonic processors (QPICs) holds the potential to make powerful quantum computation widely accessible. Particularly, photonics provides unique capabilities for quantum communications, sensing, simulation, and computing.
Compared to photonic circuits intended for classical information processing, quantum operation poses new challenges. Above all, photon losses must be minimized in quantum photonics, since information is irreversibly destroyed with each lost photon. Similarly, today’s quantum computers aren’t very good at handling mistakes, unlike regular computers, which can easily fix small errors in data. Near-term quantum computers are often classified as noisy intermediate-scale quantum (NISQ) devices. However, while numerous algorithms have been devised for a large, error-corrected quantum computers, practical applications for NISQ computers are lacking, leaving a chasm to be filled to promote a virtuous cycle of investment and continued development essential for a thriving quantum computing space. Fortunately, intermediate-scale quantum photonic devices offer the prospect of achieving quantum advantage over classical computers for specific tasks. While classical computers are fundamentally inefficient at handling complex quantum mechanical systems – such as the molecular quantum dynamics – engineering controllable quantum devices optimized to efficiently characterize quantum systems shows great promise. Although many different platforms for QPICs are being developed, none has shown the versatility of femtosecond laser writing (FLW) in realizing diverse quantum systems. For these reasons, QLASS aims at leveraging the capabilities of FLW to fabricate 3D photonic circuits with unprecedented number of modes within glass. Compared to other QPIC techniques, using glass yields extremely small interface losses – ideal for modular architectures connecting multiple chips – with speed, affordability, and optimization for end-user goals. Incorporating high-performance single-photon sources, superconducting nanowire single-photon detector, and electronics enabling reconfigurable state manipulation via control of an exceptionally large number of cryogenic-detector channels, we will create an end-to-end quantum photonics platform. Of particular interest is implementing Variational Quantum Algorithms (VQAs), which have emerged as leading candidates for near-term advantage, for which our platform is ideally suited. We will develop software for end users to translate VQAs into FLW circuits, with error mitigation methods to enhance QPIC performance.
Compared to photonic circuits intended for classical information processing, quantum operation poses new challenges. Above all, photon losses must be minimized in quantum photonics, since information is irreversibly destroyed with each lost photon. Similarly, today’s quantum computers aren’t very good at handling mistakes, unlike regular computers, which can easily fix small errors in data. Near-term quantum computers are often classified as noisy intermediate-scale quantum (NISQ) devices. However, while numerous algorithms have been devised for a large, error-corrected quantum computers, practical applications for NISQ computers are lacking, leaving a chasm to be filled to promote a virtuous cycle of investment and continued development essential for a thriving quantum computing space. Fortunately, intermediate-scale quantum photonic devices offer the prospect of achieving quantum advantage over classical computers for specific tasks. While classical computers are fundamentally inefficient at handling complex quantum mechanical systems – such as the molecular quantum dynamics – engineering controllable quantum devices optimized to efficiently characterize quantum systems shows great promise. Although many different platforms for QPICs are being developed, none has shown the versatility of femtosecond laser writing (FLW) in realizing diverse quantum systems. For these reasons, QLASS aims at leveraging the capabilities of FLW to fabricate 3D photonic circuits with unprecedented number of modes within glass. Compared to other QPIC techniques, using glass yields extremely small interface losses – ideal for modular architectures connecting multiple chips – with speed, affordability, and optimization for end-user goals. Incorporating high-performance single-photon sources, superconducting nanowire single-photon detector, and electronics enabling reconfigurable state manipulation via control of an exceptionally large number of cryogenic-detector channels, we will create an end-to-end quantum photonics platform. Of particular interest is implementing Variational Quantum Algorithms (VQAs), which have emerged as leading candidates for near-term advantage, for which our platform is ideally suited. We will develop software for end users to translate VQAs into FLW circuits, with error mitigation methods to enhance QPIC performance.
QLASS activities involve three main tasks: (i) generation, (ii) manipulation, and (iii) detection of photons.
(i) we developed an active 8-channel demultiplexer to be interfaced with the quantum emitter working at a wavelength of 928 nm. The system relies on the effective integration of different active elements ranging from a quantum emitter, a quantum switch and two demultiplexers.
(ii) we fabricated a 3D QPIC with an impressive number of modes (128) and 40 thermal shifters on two distinct depths leveraging FLW technology.
(iii) we have been working on waveguide-integrated superconducting nanowire single-photon detectors (WI-SNSPDs). In the first phase of the project a 16 channel detector array has been developed, combining the cryogenic cooling equipment with vacuum pumps and the required electronics for system control and detector biasing and readout.
Overall, the quantum photonic platform developed in QLASS relies on a feedback architecture that enables real-time reconfiguration of photonic circuits to prepare and evolve complex quantum wavefunctions for computation. To this aim, coincidences of photon detection events across the SNSPD array are measured to build correlation histograms. Custom-developed software will interpret these results and optimize the circuit configuration in the next iteration based on the computational objective (e.g. energy minimization in VQAs). Finally, we developed the QLASS software package, an open-source Python library that enables quantum chemistry calculations on photonic hardware. The implementation successfully adapts the VQE algorithm—a hybrid quantum-classical approach for finding ground state energies of molecular systems—to work with photonic quantum computers using dual-rail encoding.
(i) we developed an active 8-channel demultiplexer to be interfaced with the quantum emitter working at a wavelength of 928 nm. The system relies on the effective integration of different active elements ranging from a quantum emitter, a quantum switch and two demultiplexers.
(ii) we fabricated a 3D QPIC with an impressive number of modes (128) and 40 thermal shifters on two distinct depths leveraging FLW technology.
(iii) we have been working on waveguide-integrated superconducting nanowire single-photon detectors (WI-SNSPDs). In the first phase of the project a 16 channel detector array has been developed, combining the cryogenic cooling equipment with vacuum pumps and the required electronics for system control and detector biasing and readout.
Overall, the quantum photonic platform developed in QLASS relies on a feedback architecture that enables real-time reconfiguration of photonic circuits to prepare and evolve complex quantum wavefunctions for computation. To this aim, coincidences of photon detection events across the SNSPD array are measured to build correlation histograms. Custom-developed software will interpret these results and optimize the circuit configuration in the next iteration based on the computational objective (e.g. energy minimization in VQAs). Finally, we developed the QLASS software package, an open-source Python library that enables quantum chemistry calculations on photonic hardware. The implementation successfully adapts the VQE algorithm—a hybrid quantum-classical approach for finding ground state energies of molecular systems—to work with photonic quantum computers using dual-rail encoding.
We are nearing a historic juncture in the history of computing. Fundamentally new kinds of computers that were once only theoretical possibilities are rapidly approaching fruition.
QLASS is addressing challenging aspects of quantum photonic platform development from end-to-end. We developed a feedback architecture were each block is optimized using a specifically selected technology. During the first phase of the project, we developed an active 8-channel demultiplexer to be interfaced with the quantum emitter working at a wavelength of 928 nm to achieve high performance in terms of generation rate, photon purity, and indistinguishability; a 3D quantum photonic processor featuring 128 modes and 40 phase shifters allowing a high degree of reconfigurability; a 16-channel SNSPD detection module allowing high-efficiency detection; an eight-channel integrated time measurement circuit allowing coincidence measurements down to the ns range; a 64-channel modular phase shifter driver enabling the active control of a large number of phase shifters and a software package enabling quantum chemistry calculations on photonic hardware. These blocks will be extended and combined in the second part of the project to develop an integrated quantum photonic platform providing high flexibility and minimal losses. Our technology will contribute to paving the way for compact, high performance, reliable, cost-effective components, that will enable quantum technology to be introduced in the market. The applications are expected to reach a wide variety of industries. Our principal use case is solving problems in the design of lithium-ion batteries to achieve improvements in capacity and efficiency crucial for attaining EU technological and sustainability goals.
QLASS is addressing challenging aspects of quantum photonic platform development from end-to-end. We developed a feedback architecture were each block is optimized using a specifically selected technology. During the first phase of the project, we developed an active 8-channel demultiplexer to be interfaced with the quantum emitter working at a wavelength of 928 nm to achieve high performance in terms of generation rate, photon purity, and indistinguishability; a 3D quantum photonic processor featuring 128 modes and 40 phase shifters allowing a high degree of reconfigurability; a 16-channel SNSPD detection module allowing high-efficiency detection; an eight-channel integrated time measurement circuit allowing coincidence measurements down to the ns range; a 64-channel modular phase shifter driver enabling the active control of a large number of phase shifters and a software package enabling quantum chemistry calculations on photonic hardware. These blocks will be extended and combined in the second part of the project to develop an integrated quantum photonic platform providing high flexibility and minimal losses. Our technology will contribute to paving the way for compact, high performance, reliable, cost-effective components, that will enable quantum technology to be introduced in the market. The applications are expected to reach a wide variety of industries. Our principal use case is solving problems in the design of lithium-ion batteries to achieve improvements in capacity and efficiency crucial for attaining EU technological and sustainability goals.