LIDAR (Light Detection and Ranging) is emerging as a necessity for fully automated self-driving automotive applications. In order to sample the far field with sufficient resolution for this application the system must incorporate many optical elements, leading to challenges for manufacturability and size. Due to the density of optical components required, LIDAR is well suited for photonic integration in order to achieve miniaturization and scalable manufacturability. This talk will give an overview of LIDAR, the components required for a chip-scale solution, and silicon photonics progress with respect to this goal.
Automotive self-driving capability is highly developed using camera vision but can benefit from the addition of independent sensor platforms for redundancy and further reduction of MTBF (mean time between failure). Frequency Modulated Continuous Wave (FMCW) Light Detection and Ranging (LIDAR) is particularly suited to this application by virtue of its capability to directly detect velocity as well as range with high resolution. However, such a system requires the dense integration of multiple optical components including lasers, amplifiers, phase and amplitude control low-noise photodiodes, mode converters, and optical waveguides. These must further be integrated in a compact form factor that can be manufactured in high volume. Silicon photonics using the Intel Hybrid Silicon platform can enable such optical integration on a silicon chip in a scalable high-volume manufacturing process, thus enabling chip-scale solid-state LIDAR.
With silicon photonics going fabless, large-scale silicon photonic integrated circuits (PICs) have recently become a reality. Many of these PICs feature system reconfigurability to benefit from the cost-effective mass manufacture of a universal platform. However, reconfigurable silicon PICs relying on the weak, volatile thermo-optic or electro-optic effect of silicon usually suffer from a large footprint and energy consumption. Recently, phase-change materials have shown great promise for energy-efficient, ultra-compact and ultra-fast non-volatile integrated photonic applications. Here, by integrating phase-change materials, Ge2Sb2Te5 (GST) with silicon microring resonators, we demonstrate a non-volatile, programmable, energy-efficient, and compact platform over the telecommunication range. By measuring and fitting the output spectra of the microrings covered with various lengths of GST in the amorphous and crystalline states, we characterize the strong broadband attenuation (~7.3 dB/μm) and optical phase (~0.70 nm/μm) modulation effects of the platform. By adjusting the energy and number of free-space laser pulses applied to the GST, we perform reversible and quasi-continuous tuning of the GST state, and the subsequent tuning of the attenuation and resonance of the microring resonators enabled by the thermo-optically-induced phase changes. Designed to achieve near critical coupling of the microring resonators when the GST is in the amorphous state, a non-volatile 1×1 optical switch with high extinction ratio as large as 33 dB is demonstrated. Our research constitutes the first step towards future large-scale programmable silicon PICs. With appropriate design, a broadband low-loss 2×2 optical switch could be electrically controlled which would be the building block for a future non-volatile routing network and optical FPGA.
J. J. Zheng, A. Khanolkar, P. P. Xu, S. Deshmukh, J. Myers, J. Frantz, E. Pop, J. Hendrickson, J. Doylend, N. Boechler, and A. Majumdar, "GST-on-silicon hybrid nanophotonic integrated circuits: a non-volatile quasi-continuously reprogrammable platform," Opt. Mater. Express 8(6), 1551-1561 (2018).
Free-space beam steering using optical phased arrays is a promising method for implementing free-space communication links and Light Detection and Ranging (LIDAR) without the sensitivity to inertial forces and long latencies which characterize moving parts. Implementing this approach on a silicon-based photonic integrated circuit adds the additional advantage of working with highly developed CMOS processing techniques. In this work we discuss our progress in the development of a fully integrated 32 channel PIC with a widely tunable diode laser, a waveguide phased array, an array of fast phase modulators, an array of hybrid III-V/silicon amplifiers, surface gratings, and a graded index lens (GRIN) feeding an array of photodiodes for feedback control. The PIC has been designed to provide beam steering across a 15°x5° field of view with 0.6°x0.6° beam width and background peaks suppressed 15 dB relative to the main lobe within the field of view for arbitrarily chosen beam directions. Fabrication follows the hybrid silicon process developed at UCSB with modifications to incorporate silicon diodes and a GRIN lens.
Free-space beam steering using optical phase arrays are desirable as a means of implementing Light Detection and
Ranging (LIDAR) and free-space communication links without the need for moving parts, thus alleviating vulnerabilities
due to vibrations and inertial forces. Implementing such an approach in silicon photonic integrated circuits is
particularly desirable in order to take advantage of established CMOS processing techniques while reducing both device
size and packaging complexity.
In this work we demonstrate a free-space diode laser together with beam steering implemented on-chip in a silicon
photonic circuit. A waveguide phased array, surface gratings, a hybrid III-V/silicon laser and an array of hybrid III/V
silicon amplifiers were fabricated on-chip in order to achieve a fully integrated steerable free-space optical source with
no external optical inputs, thus eliminating the need for fiber coupling altogether. The chip was fabricated using a
modified version of the hybrid silicon process developed at UCSB, with modifications in order to incorporate diodes
within the waveguide layer as well as within the III-V gain layer. Beam steering across a 12° field of view with ±0.3° accuracy and 1.8°x0.6° beam width was achieved, with background peaks suppressed 7 dB relative to the main lobe within the field of view for arbitrarily chosen beam directions.
In this paper we outline recent results which combine defect mediated Photo-Detectors (PDs) in a Ring Resonator (RR)
structure. By exploiting the multiple-pass of the optical signal through the detector, we are able to significantly decrease
the size of the detector structure while maintaining good responsivity (typically 0.1 A/W). In such a geometry the
detector bandwidth is not capacitance limited, while the leakage current is reduced toward 1 nA. We also show that these
PDs may be used in the drop port of a RR to monitor the propagating signal. These devices have applicability in
multiplexing and potential for integration with high speed modulation functionality.
Recently, low threshold Raman silicon lasers based on ring resonator architecture have been demonstrated. One of the
key elements of the laser cavity is the directional coupler that couples both pump and signal light in and out of the ring
resonator from the bus waveguide. The coupling coefficients are crucial for achieving desired laser performance. In this
paper, we report design, fabrication, and characterization of tunable silicon ring resonators for Raman laser and amplifier
applications. By employing a tunable coupler, the coupling coefficients for both pump and signal wavelength can be
tailored to their optimal values after the fabrication, which significantly increases the processing tolerance and improves
the device performance.
We report simulation results for a directional coupler between silicon waveguides in different layers of a three-dimensional (3D) optical circuit. The coupling length is 1.4 mm. The device is manufacturable using standard CMOS technology provided individual waveguide layers can be vertically stacked. In simulations of coupling efficiency the design exhibits negligible loss with respect to translational and rotational misalignments of up to 0.5 μm. Efficiency degradation is less than 5% for etch depth and waveguide width variations of 0.4 μm, and less than 1 dB over the range of standard lithographic tolerances for variations from layer to layer in feature width, depth, and alignment.
This paper describes work investigating the impact of lattice defects on the attenuation of optical signals at wavelengths
around 1550nm in silicon rib waveguides. Using Fourier transform infrared spectroscopy it is shown that high energy proton irradiation of silicon induces excess optical absorption peaked at a wavelength of 1800nm, but extending below 1600nm. This absorption is related to the introduction of silicon divacancy defects. It is further demonstrated that silicon divacancy concentration is accurately determined for a range of proton doses using positron annihilation spectroscopy and successfully predicted using an analytical expression proposed previously. Low loss rib waveguides
were fabricated in silicon-on-insulator substrates. These waveguides were subsequently implanted with silicon ions at an energy of 2.8MeV through photolithographically defined mask windows of various lengths. The additional optical loss as a result of the defects introduced by the implantation process was accurately determined. For a dose of 2.5x1014cm-2, the loss is greater than 500dBcm-1. Finally, it is shown that excess absorption can be predicted using the same analytical expression for the determination of vacancy concentration, thus providing a straightforward method for the design of integrated, on-chip optical absorbers in silicon photonic circuits.