Arnan Mitchell

Optical tweezers have found widespread application in microrheology, where fluctuations in particle position can be used to infer the viscosity of the medium around the trapped object. In addition to translational studies, rotational optical tweezers are also used to determine viscosity from the balance between applied torque and Stokes drag from the surrounding medium. Here we show that rotational optical tweezers need not require spherical birefringent particles for measuring viscosity. In our case, we harness torque due to orbital angular momentum to spin a microfabricated spiral phase plate and use this for microrheology. Using heavy water as a test medium, we obtain a viscosity of 1.25 ± 0.05 cP, with a fitted central value within 0.4% of the reference value.

Microwave photonic (MWP) filters are key devices for microwave systems because of their low loss, high bandwidth, and electromagnetic interference immunity. Sharp roll-off is important for accurate spectral filtering and effective suppression of interference. However, electronic MWP filters struggle to simultaneously achieve sharp transitions as well as reconfigurability. In this paper, we demonstrate reconfigurable MWP filters with very sharp roll-off using a transversal filter method based on a Kerr optical microcomb. Four different types of single-band MWP filters with ultrahigh roll-off rates up to ~32.6 dB/GHz and a minimum 20-dB shape factor of ~1.15 are achieved. In addition, by programming tap coefficients, we demonstrate band-pass filters with tunable centre frequencies verying from 5 GHz to 15 GHz with a range of filter responses. The reconfigurable MWP filters with sharp roll-off offer a powerful, versatile, and stable solution for applications needing high spectral selectivity, such as high-resolution radar imaging, next-generation wireless networks, and advanced biomedical sensing.

Dilated convolution is pivotal for dense prediction tasks such as semantic segmentation, as it expands the receptive field to capture multi‐scale contextual information without increasing parameter complexity. However, its electronic implementation often faces bottlenecks due to non‐contiguous memory access and high computational load. Here, we report a Reconfigurable Optical Dilated Convolution Accelerator (RODCA) based on a wavelength‐dispersion co‐tuning mechanism. By dynamically controlling optical frequency comb spacing and dispersion, RODCA enables reconfigurable dilation rates while improving spectral efficiency. The system achieves an equivalent computing throughput of 13.858 Tera operations per second (TOPS). We experimentally validate the RODCA using Laplacian kernels and semantic segmentation tasks, where it achieves a pixel accuracy of 88.46% and a mean Intersection over Union (mIoU) of 76.08%, closely matching in‐silico results. This work demonstrates RODCA's capability for efficient multi‐scale feature extraction, highlighting its potential for next‐generation photonic computing in complex AI applications.

Mode-division multiplexing is a common technology in photonic integrated circuits, which can combine various optical mode signals into a single channel, such as a multimode waveguide, to improve the communication system’s capacity. The optical mode (de)multiplexer plays a critical role in mode-division multiplexing by converting multiple single-mode signals from different channels into corresponding multimode signals, respectively, and then combining them into the same physical channel and vice versa. Here, we propose an ultra-broadband optical mode (de)multiplexer based on modal-field redistribution assisted by subwavelength grating (SWG) structures in lithium niobate on insulator. As a proof of concept, we design and fabricate a three-channel mode (de)multiplexer. The simulation results show an operating bandwidth of 800 nm (from 1.2 to 2.0 μm) with an insertion loss of less than 1 dB, and the experimental results show that the insertion loss and crosstalk of all devices are less than 1.0 dB and −15.9 dB in the wavelength range of 1500–1600 nm, which is basically consistent with the simulation results. This work holds significant promise for the future implementation of high-performance photonic integrated circuits (PICs) on the LNOI platform.

Electro-optic modulators (EOMs) are pivotal for optical communication systems, converting electrical signals to the optical domain. To further maximize spectral efficiency and boost transmission capacity, endowing EOMs with dual-band operation capability has become imperative. Here, we demonstrate a dual-band silicon nitride-loaded thin-film lithium niobate modulator that supports energy-efficient operation in both the O- and C-band using a unified photonic-electronic architecture. The device integrates dual-band grating couplers and bidirectional dual-band multiplexers with capacitively loaded traveling-wave electrodes engineered for high-speed modulation. By tailoring the waveguide geometry to minimize inter-band group-index disparity, the modulator achieves simultaneous velocity matching across widely separated wavelengths. The proposed EOM achieves PAM-8 transmission at a raw line rate of 600 Gbps before FEC overhead at both 1310 and 1550 nm, with corresponding power consumption of 0.82 and 0.89 fJ/bit, respectively. Our results establish a scalable route toward multi-band thin-film lithium niobate photonic interconnects for future high-capacity optical communication systems.

Microring modulators are the most fundamental and crucial building blocks for next-generation big data applications, including optical interconnect, photonic computing, and optical wireless communication. However, achieving exceptionally fast modulation rates for microring modulator remains a formidable challenge due to the photon lifetime limitations and lumped electrode architectures. In this work, we propose and experimentally demonstrate an ultra-high-speed microring modulator on a silicon nitride-loaded lithium niobate on insulator platform. By integrating low-loss Euler-bend waveguides with fully symmetrical capacitively loaded traveling-wave electrodes reinforced by a dummy waveguide, the device achieves a record-breaking modulation rate of 517 Gbps (200 GBaud PAM6). This exceptional performance is accompanied by an electro-optic bandwidth exceeding 110 GHz and an ultra-low energy consumption of merely 1.5 fJ/bit. To demonstrate the performance and application prospects of this modulator, we successfully verify its capability to accelerate feature extraction at a modulation rate of 40 GBaud in photonic computing and support breakthrough wireless multi-target communication with a single-channel capacity of 360 Gbps (180 GBaud PAM4), highlights its transformative potential across a broad spectrum of next-generation photonic systems.

Synthetic dimensions provide a powerful tool that uses comparatively simple structures to probe high‐dimensional topological physics, in which edge states emerging at lattice boundaries are of great importance. However, the demonstration of lattice boundaries in synthetic dimensions is relatively nascent. In this work, we realize an integrated coupled ring system in a thin‐film lithium niobate photonic platform that enables the simulation of one‐dimensional frequency crystal lattice with sharp boundaries, attaining suppression for two coupling terms with a single auxiliary cavity. Their effect on tight‐binding lattice dynamics was verified by acquiring discretized band structures of an site lattice. The ability to create robust frequency‐space boundaries is a key step toward the realization of topological systems that harness bulk‐edge correspondence as well as optical information processing in a photonic chip.

Multiplexed optical sensing using integrated photonic platforms demands scalable interrogation architectures that do not compound hardware complexity with increasing numbers of sensing channels. Here, we demonstrate simultaneous readout of two cascaded silicon nitride (Si₃N₄) racetrack microring resonators (MRR) with different free spectral ranges (FSR), using a single electro-optic optical frequency comb (EO-OFC) generated from one continuous-wave laser source. By tuning the radio-frequency (RF) modulation frequency to match the spectral separation between selected resonances of the two rings, individual comb lines are aligned to each resonator, enabling independent and simultaneous resonance tracking without wide-range wavelength scanning. A programmable waveshaper demultiplexes the carrier and first-order sideband channels to dedicated photodetectors, allowing real-time monitoring of both rings in parallel. The system was validated by bulk refractive-index sensing with glycerol solutions at five concentrations (0.039–0.6% v/v). Both rings achieved sensitivities in the range of 26.8–38.5 nm/RIU and limits of detection (LoD) on the order of 10-5 RIU (2.15–2.96 × 10-5 RIU), with OFC-multiplexed performance matching single-wavelength baseline measurements on the same chip. The carrier and sideband channel assignments were interchangeable, confirming architectural flexibility. These results establish EO-OFC interrogation as a viable, scalable route to multiplexed microring resonator sensing, with direct implications for parallel biosensing, environmental monitoring, and food safety applications.

Physically unclonable functions (PUFs) have emerged as a promising hardware security primitive, and recent years have seen growing interest in realizing PUFs on photonic integrated circuits (PICs). We classify existing PIC-based PUFs into electrically configured and optically encoded designs. Notably, security properties remain insufficiently analyzed, with mathematical unclonability often assumed without supporting argument. To illustrate this issue, we introduce a modeling attack against the subclass of optically encoded PUFs that operate in a linear optical regime. We show that, for a representative simulated PUF construction, observing as few as 200 challenge-response pairs allows learning the challenge-response mapping. Under canonical modeling assumptions, linear optically encoded PUFs are therefore mathematically clonable, calling for a careful assessment of their suitability for authentication protocols that rely on mathematical unclonability.