The ability to simulate physical systems is essential for an understanding of their physical properties and advancing related technologies. However, simulating certain complex systems at scale, including many-body spin models, is often a significant challenge for conventional computing approaches. Quantum simulators, employing controlled quantum systems that mimic the dynamics of a target system, can offer a more efficient approach. This work demonstrates the use of photonic circuits for quantum simulation of disordered spin systems, where quantum properties such as superposition and interference can be leveraged for a more efficient simulation. We fabricate and characterize a low-loss silicon-on-insulator chip with a buried aluminum mirror and a reconfigurable interferometer, which is employed in conjunction with two input photons to map the dynamics of a spin Hamiltonian with four-body interactions. Our results demonstrate several characteristics associated with the target system, such as spin-degenerate ground states and phase transitions, confirming the device’s functionality as a quantum simulator of the spin system.
The ability to simulate physical systems is essential for an understanding of their physical properties and advancing related technologies. However, simulating certain complex systems at scale, including many-body spin models, is often a significant challenge for conventional computing approaches. Quantum simulators, employing controlled quantum systems that mimic the dynamics of a target system, can offer a more efficient approach. This work demonstrates the use of photonic circuits for quantum simulation of disordered spin systems, where quantum properties such as superposition and interference can be leveraged for a more efficient simulation. We fabricate and characterize a low-loss silicon-on-insulator chip with a buried aluminum mirror and a reconfigurable interferometer, which is employed in conjunction with two input photons to map the dynamics of a spin Hamiltonian with four-body interactions. Our results demonstrate several characteristics associated with the target system, such as spin-degenerate ground states and phase transitions, confirming the device’s functionality as a quantum simulator of the spin system.
Silicon-on-insulator (SOI) wafers have been developed for electronic integrated circuits, but are also one of the most promising platforms in photonic integrated circuits due to its low loss, high refractive index, ability to fast phase modulation, and well-developed fabrication technologies. With advantages to handle light with a small volume and high modulation speed, silicon photonics has been intensively investigated as the next generation platform of signal processing, interconnection, and communication devices. Recently, silicon photonics has been attracted great interests in quantum optics communities as optical quantum systems for linear optical quantum computing are achievable using silicon photonic components including beam splitters, phase modulators, and filters and the small footprint of a silicon photonic quantum system has unprecedented performances of high stability and scalability. Here we show our recent results on the fabrication of silicon photonic elements for quantum applications. We fabricated low-loss optical waveguides, efficient fiber-to-waveguide grating couplers, resonator-based optical filters, and tunable beam splitters and tested their performances. A micro-sized heater is attached on a Mach-Zehnder interferometer, showing stable operation as a slow phase modulator. In addition, phase modulators using plasma dispersion effects are under investigation. Finally, the quantum characteristics of photon pairs generated by spontaneous four-wave mixing in silicon waveguides are investigated. These results will be an essential cornerstone for the realization of quantum optical computing and quantum communication devices.
This thesis focuses on optical quantum information processing in two parts. The first part focuses on photonic hardware. Large-scale photonic experiments require the generation of many indistinguishable single photons simultaneously. Traditionally, such photons are generated by spontaneous parametric down-conversion (SPDC). However, the suitability of such sources to generate many photons. We have identified the maximum limits of how many photons SPDC sources can be made. A second requirement of large-scale quantum photonics is having a network with many optical modes while maintaining low optical loss. Such networks are currently lacking. Therefore we introduce the possibility to use a rough surface scatterer for a quantum advantage experiment. Although such a network is far from universal, it has many modes and low loss. Another approach is to use integrated quantum photonics, which is fully reconfigurable and is a promising way of scaling up to large systems in an industrial setting. We have characterized a programmable low-loss 12-mode silicon nitride (SiN). In the second part of the thesis, various proof-of-principle quantum experiments were performed on a 12 mode SiN integrated photonic chip. Firstly, it was shown that these networks can be used to simulate wave propagation through disordered networks. This unconventional approach allows for full experimental access and control to the physical process, which is essential for such experiments. Secondly, we introduced a semi-device independent witness protocol that can in-situ verify the indistinguishability of photons on the chip, which is an essential tool to have when the system size increases and can no longer be efficiently simulated classically. The last experiment is a quantum simulation where we show how a pure quantum state locally thermalizes in a closed system.
Models of interacting complex systems provide the fundamental statistical physics reference frame for the study and the understanding of associative memories, machine learning, and the dynamics of neural networks. On the other hand, simulating complex multi-synaptic interactions on a classical hardware is computationally demanding due to the super-linear scaling of the system complexity. Photonic quantum technologies provide a promising solution to these limitations by leveraging on their inherent speed and parallel processing ability in order to simulate complex networks. Recently, a connection between multiphoton processes and generalized p-body Hopfield models has been theoretically established. Here, we design and demonstrate an experimental platform that exploits single photons distributed across a set of optical modes, in which controlled arrays of binary phase shifters act as Ising-like neurons. We focus specifically on a fully connected Hopfield Hamiltonian with four-body local interaction terms, realized via two-photon processes. Through quantum simulations on programmable photonic processors, the study identifies three distinct regimes: a memory retrieval phase, a spin-glass memory "black-out" phase, and a paramagnetic phase. Experimental results confirm successful memory retrieval at low storage capacities and temperatures, where the system consistently relaxes to fixed points with high memory overlap, effectively reconstructing the stored patterns. Future research will extend the platform design to investigate networks with local or dilute interactions, while advances in the realization of scalable photonic circuits will enable architectures that encompass very large numbers of interacting spins.
Quantum simulators are controllable quantum sys-tems that can be used to reproduce the states and the dynamics of other quantum systems object of study. The ease to manipulate single photons on a chip at room temperature without deco-herence issues has made integrated photonics one of the most viable technologies for processing quantum information. Many demonstrations have been reported over the past fifteen years, included the simulation of complex and many-body interactions. Combining quantum light sources, reconfigurable circuits and high-efficiency single-photon detectors together on a single chip is an extremely challenging goal, still taking a lot of scientific and technological efforts worldwide. Here, we report emergent results along this direction. On a silicon nitride photonic platform, we separately show a source of photon pairs based on a microring, high-performance wavelength filters based on Mach-Zehnder in-terferometers and silicon single-photon avalanche diodes directly coupled to the photonic waveguides. The demonstration of these fundamental building blocks represents a starting point for the realization of a photonic quantum simulator operated entirely on a single chip working at room temperature.
Recently there has been a worldwide effort to produce scalable qubit implementations for quantum information protocols, including quantum key distribution, quantum metrology and quantum computing---technologies which have the power to revolutionise their respective fields. Silicon photonics is in a prime position to provide the physical, photonic implementations of these technologies due to it's high-yield, high-performance and inherently scalable CMOS manufacturing infrastructure. The on-chip production of more than one uncorrelated photon pair (more than two photons) has remained a technological challenge. Here, we seek to overcome this using careful quantum photonic design, and state-of-the-art Silicon photonics technology. Our four-photon device expands the capabilities of integrated quantum photonics to fully entangled four photons, specifically the four photon graph states. In the graph state formalism, each vertex corresponds to a qubit, and each edge corresponds to a controlled-phase gate (an entangling operation) performed between the qubits it links. It has been shown that these states are very powerful---simple measurement strategies on certain graph states provide universal quantum computing. Interestingly, certain classes of graph states are equivalent under single-qubit transformations. These are the global entanglement classes. Our device uses both a reconfigurable entangling gate and local transformations to fully explore the space of 4 qubit graph states. The device employs four ring resonator sources, on-chip spectral demultiplexing, a reconfigurable entangling gate, and single-qubit transformations for each qubit. Using these capabilities, we can verify our entangled quantum states using entanglement witnesses, local tomography, and by running toy measurement-based quantum computations. These devices will be both the largest and most diverse integrated photonic quantum state generators produced to date. A vital demonstration of this work is high-quality quantum interference between spontaneous four-wave mixing photon-pair sources on the same chip. These ideas and proof-of-concept demonstrations point to large scale quantum technology, built using today's technologies.
Summary form only given. The efficient characterization and validation of the underlying model of a quantum physical system is a central challenge in the development of quantum devices and for our understanding of foundational quantum physics. However, the impossibility to efficiently predict the behaviour of complex quantum models on classical machines makes this challenge to be intractable to classical approaches. Quantum Hamiltonian Learning (QHL) [1, 2] combines the capabilities of quantum information processing and classical machine learning to allow the efficient characterisation of the model of quantum systems. In QHL the behaviour of a quantum Hamiltonian model is efficiently predicted by a quantum simulator, and the predictions are contrasted with the data obtained from the quantum system to infer the system Hamiltonian via Bayesian methods.