Exploring Emerging Applications for Photonic Integrated Circuits
Photonic integrated circuits (PICs) have emerged as key
component-level enablers of next generation datacom/telecom systems that
require smaller, more efficient transceivers and switches. These same
qualities are opening doors across widely varied applications ranging
from vehicle autonomy to quantum computing and beyond.
By: Ana Gonzalez, R&D Manager at EPIC
Miniaturization, higher performance, vibration immunity, reduced footprint and low heat generation are some of the clear benefits of adopting photonic integrated circuit (PIC) technologies for developing new photonic products. PIC technology meets the requirements for the development of new, exciting applications such as point-of-care devices, miniaturized LiDAR, quantum computing, structural monitoring and wearables for healthcare. In this article, we explore some of these emerging applications together with the companies at different levels of the PIC supply chain that can provide the technology required for the design, fabrication, packaging and scaling up volumes for PIC-based modules.
Although PICs have been used extensively in datacom and telecom, there are other application fields in which this technology is attracting increasing attention. An example is the automotive market, as a result of increasing demand for safety and advanced driver-assistance as well as autonomous driving systems, some companies such as Omnitron Sensors, Fastree 3D, Mouro Labs, Beamagine and LuxC are pushing the development of new LiDAR systems. LiDAR based on PIC technologies is potentially cheaper, lighter, more compact and more reliable—PIC based LiDARs have no moving parts.
To meet the increasing demand for automotive LiDAR systems, a growing number of companies have moved into the development of PIC-based LIDAR systems and components. Lumentum is one such company; for short-range LiDAR (10-50 m) and in-cabin monitoring, Lumentum provides high-power 940 nm VCSEL array illuminators. For long-range (200 m), they offer 1550 nm narrow-linewidth DBR diode lasers for long-range frequency-modulated continuous-wave (FMCW) coherent LiDAR. Both of these devices are designed to be time-of-flight (ToF) light sources for flash systems and are really intended to provide more power.
To reach much higher power levels, longer distance, higher resolution and the performance required by automotive LiDAR, Lumentum have developed two innovative solutions: multi-junction addressable VCSEL arrays and bottom emitting devices.
Multi-junction VCSEL arrays.
By increasing the number of junctions, the number of photons emitted can be also increased with the same type of overall wall plug efficiency. This allows the low currents, typically used in two or three junctions, to produce higher levels of power and faster applications.
Bottom emitting devices: These allow fabrication of devices with no external optics and the ability to modify the beam profile. Basically, an external lens is put on the back of the device and the chip is flipped upside down and then mounted top down onto the sub-mount. This allows integration of multi-junction adjustable arrays and optics to create new and novel gallium arsenide patterns, which will be required for the next generation of detector devices (see Figure 1).
Figure 1. Bottom Emitting Integrated Optics Configurations (courtesy by Lumentum)
There are other companies providing VSCEL arrays for LIDAR, such as Array photonics, Bandwidth10 and Astrum. Different types of lasers can be also employed for LIDAR, such as the ones provided by companies including Lumibird, Bright solutions, Eblana photonics, BKtel and RIO.
Another company working on a cost-effective solution for improving vehicle perception is SCANTINEL PHOTONICS, a spin off from Carl Zeiss AG, that uses coherent Frequency-Modulated Continuous Wave (FMCW) ranging to achieve longer distance by using a 1550 nm integrated swept source with a narrow bandwidth and high linearity (see Figure 2). This enables the system to generate reliable, high range three-dimensional images of the environment required for autonomous navigation.
Figure 2. Scantinel’s 1550 nm FMCW LiDAR approach (courtesy by SCANTINEL PHOTONICS)
Coherent ranging allows the photonic integration of the lasers, detectors, and a large part of the optical components on a silicon wafer platform, which eliminates error-prone assembly and calibration steps. The core concept of the system is its scalability to high pixel rates through efficient multiplexing, which is achieved by using a high parallelization of different FMCW channels. Using PICs to create an optical enhanced array (OEA), enables high integration of the system, low power, and solid-state scanning up to a range of 300 meters as well as the option to scale up to high volumes at very competitive price points.
SCANTINEL collaborates with partners such as imec, a leading European research centre in the field of silicon photonics with whom they are working on solid state beam steering, and PHIX, a packaging company for large-volume photonics manufacturing.
Fabricating proof of concept chips or taking proven designs to initial prototype stages is a costly process, so a number of European foundries offer mupliproject wafer (MPF) services including:◊
Silicon Nitride (SiN): CNM, imec, LioniX and LIGENTEC
Indium Phosphide (InP): Smart Photonics, Fraunhofer HHI, 3-5 labs
Silicon Photonics (SiP): VTT, imec, ihp and Cornerstone
Multi project wafer foundries fabricate different chips in the same wafer to reduce costs in the prototyping phase. Another option is the JePPIX Pilot Line that was launched in 2019. JePPIX aims to provide companies with direct access to state-of-the-art manufacturing InP chips from proof of concept to industrial prototyping and pre-production. Applications include fibre-optic communication, biomedical devices, next generation mobile and portable devices, astrobiology, and quantum computing. Foundries also offer Process Design Kits (PDKs) for circuit simulation and mask design to aid in the process of turning concepts into chips. The PDKs can be implemented in different software packages offered through Synopsys, VPI Photonics, Luceda, Lumerical, Nazca and PhotonDesign.
Luceda Photonics, based in Belgium, provides software and services for an integrated approach, enabling photonics IC engineers to enjoy the same first-time-right design experience as electronic IC designers. To this end, they have their IPKISS integrated photonics design platform based on Python. This platform is a scripting environment that covers the complete photonic IC design flow up to measurement feedback for true component characterization and validation. The components rely on a single, centrally defined model for a smooth transition between the different design stages such as layout, physical and circuit simulation. This makes the design flow robust, reduces design errors and saves considerable design time.
The IPKISS platform is modular and can be extended to integrate with EDA design flows via their IPKISS.eda module. This module can be plugged into the IPKISS platform to allow parametric cells to be exported to EDA tools, thereby enabling PIC designers to enjoy the benefits of a professional EDA environment and the ability to exercise good control over the details of complex components. It affords automation and control across all levels from the component to the circuits - a feature particularly attractive for both LiDAR and quantum computing applications.
Another company working in the field of simulation software is VPI Photonics. They have participated in a number of industry-leading research projects, the most recent of which is PlasmoniAC, an EU-funded project that aims to develop a radically new circuit-technology for neuromorphic computing based on plasmons. The idea is to create high-speed neuromorphic chips that are low-cost, energy-efficient, and compact with the aim of strengthening the competitiveness of the European photonics industry to play a greater role in the global neuromorphic and deep learning market. The project basically leverages the energy and size efficiency of plasmonic circuits and applies them to neuromorphic computing architecture. VPI’s contribution is to develop an add-on model library for plasmonic devices for use in conjunction with their photonic integrated circuit models to allow their customers to build and simulate neuromorphic circuits as well as evaluate their performance.
PICs are envisioned as a key technology in the near-term deployment of metropolitan quantum-key-distribution-based secure systems that leverage the fact that entangled photons can be individually generated, modulated and routed on the PIC. Companies such as QuSide Technologies and Quix develop PIC based devices for random number generation. Another application is the development of quantum computing being pursued by companies such as XANADU, a quantum photonics computer company with a mission to build quantum computers that are useful and available to people everywhere. They currently have three cloud machines available, which are based on silicon nitride PIC systems that generate gaussian process sampling for near-term applications. The advantage of silicon nitride is that the machines can operate primarily at room temperature and the system easily integrates into existing telecommunication infrastructure.
Personal health and agri-food applications are under development at OnePlanet Research Center, which was established in 2019 to initiate fundamental and applied research into applications to improve health and access to healthy and sustainable food. The centre focus on three main areas: sensing, including non–invasive, electrochemical, imaging techniques; digitizing and data analysis; and various applications as diverse as wearables, nitrogen sensor boxes and AI models. As shown in Figure 3, the photonics sensor technology is based on silicon nitride and is being developed by imec.
Health applications include devices to detect early markers in urine samples that could point to potential health issues and to provide insights into changes that occur in a certain timeframe; wearables for measuring mental well-being and stress; and digital techniques to measure how individuals absorb food and to detect problems in the gastrointestinal system.
Agri-food applications aim to use sensor and digital technology to make food production and processing more sustainable. Specifically, to enable farmers to monitor every tree, plant and animal individually so they can respond quickly and precisely at the right place and time. The aim is save time and use resources more effectively and efficiently while protecting the environment as much as possible. In this way, it may be possible to grow crops in places where the soil is less fertile, which has in turn led to food shortages.
Related to healthcare and environmental applications, the MIRPHAB Pilot Line was launched to scale up products based on miniaturized Mid-IR spectroscopic sensors, including PICs acting as the spectrometer by using arrayed waveguide gratings (AWG). Mid-IR light in the 3- 12 μm wavelength band interacts strongly with molecular vibrations that present unique adsorption spectrums that give superior detection capabilities and unambiguous detection of chemicals in gas and liquids, enabling high sensitivity and real-time detection, qualities that present interesting applications for the development of wearables, breath analysers, point-of-care applications and detection of chemicals for industry.
Optics11 offers high-end optical sensing systems for both industrial and life science applications. Their main focus is on developing high-end tuneable laser-based fiber Bragg grating (FBG) integrators for optical sensing systems that incorporate a high-end optical acoustic emission system called Optima, which can sample up to megahertz rates. The FBG integrators have high accuracy and high precision. The Optics11 FAZ 14 series integrators can sample up to kilohertz rates. The technology also features a broad portfolio of FP and FBG based to sensors to measure strain, acceleration, temperature, and pressure. The company has experience in a wide range of applications and is currently engaged in research for the next generation of PIC based interrogators for low-cost, high volume applications, while maintaining high performance.
Optics11’s technology is used in real-time structural health monitoring applications, for example, in bridges, caves and wind turbines to detect displacements and the need for repairs. Other industrial applications include acoustic emission for detecting partial discharge in high voltage applications and also road traffic monitoring, which involves installing fibre FBG arrays under the road surface to monitor the flow, speed and wight of traffic.