Yole is forecasting that the VCSEL market will grow from $794 million in 2021 to $1.74 billion in 2026.
Since their invention in the 1960s, lasers have found application in virtually every sector, including industrial, scientific, consumer, medical, and defense. Laser technologies have proliferated along with the traditional and emerging applications that have adopted them. A sampling of the available technologies includes diode, fiber, diode-pumped solid-state, CO2, and excimer lasers, and laser applications include material processing, optical communications, automobile front illumination, medical surgery, and 3D sensing, to name a few.
Semiconductor lasers are quantum generators based on an active gain medium of single-crystal semiconductor material. Optical amplification is created via stimulated emission at the transition between quantum energy levels at a high concentration of free charge carriers in the conduction gap.
EE Times Europe reached out to analysts at Yole Développement for insight into the main semiconductor laser technologies being deployed for 3D sensing: edge-emitting lasers (EELs) and vertical-cavity surface-emitting lasers (VCSELs). Highlights of our exchanges with Yole’s Martin Vallo, technology and market analyst, Solid-State Lighting, and Pierrick Boulay, senior technology and market analyst, Solid-State Lighting, follow.
The EEL is a well-established technology that has been in use for decades. Light is emitted at the semiconductor chip’s edges, which act as cavity mirrors, in a waveguide structure parallel to the semiconductor surface. A considerable amount of gain and high output power may be produced, with a relatively lengthy active area of hundreds of micrometers to a few millimeters. Electrically pumped EELs are small and cost-effective laser-emission sources that are suitable in a variety of applications.
Unlike edge-emitting lasers, VCSELs produce light perpendicular to the semiconductor surface. Two distributed Bragg mirrors with alternating layers of high- and low-refractive–index material, with thicknesses of a quarter of the laser wavelength, form the vertical cavity. Electrically pumped quantum wells or quantum dots in the active area between the monolithically produced semiconductor mirrors deliver gain, resulting in single-longitudinal–mode operation.
Because oxide apertures confine both the current and the optical field, the VCSEL can operate in single-transverse mode, making it a small and efficient source of laser emission with high beam quality.
EE Times Europe: Why are VCSELs so well-suited for 3D sensing?
Yole: Various infrared light sources can be used for sensing: LEDs, edge emitters, and VCSELs. LEDs are mature, cheap components and are easy to manufacture. Most of the time, they are used for 2D sensing, as in a driver-monitoring system. On the other hand, edge emitters and VCSELs are perfect light sources for 3D sensing, and the choice of one source or another will depend mainly on the output power needed for the application.
VCSELs are particularly well-suited for 3D sensing in smartphones due to their compact size, ease of manufacture, and ability to use pulse speeds on the order of a nanosecond, which is needed for time-of-flight applications. As a result, Yole is forecasting that the VCSEL market will grow from US$794 million in 2021 to US$1,742 million in 2026.
EE Times Europe : What are the challenges facing VCSELs?
Yole: The main challenges are related to manufacturing, output power, and the wavelengths that could be targeted.
The manufacturing process for VCSELs is quite complex, and each step from the epi wafer to the packaging is critical.
Epitaxial growth is crucial, as it will determine the final manufacturing yield.
The best VCSEL manufacturers can achieve yields of 65% to 70%, which is relatively poor compared with yields for some other components. The output power of VCSELs has been limited, but recently, the use of multi-junction VCSELs that embed multiple active regions has shown an efficiency improvement of 30%.
VCSELs are mainly manufactured on gallium arsenide, [which yields devices that] emit at wavelengths from 850 nm to 940 nm. For longer wavelengths, in the range of 1.3 or 1.5 µm, indium phosphide material has to be used, but manufacturing is much more complex for InP-based VSCELs than for GaAs. Visible VCSELs based on gallium nitride are also emerging, pushed by players like Sony. [The GaN-based VCSELs] could be used in displays for augmented-reality applications.
EE Times Europe: What are the challenges in terms of design, material, and packaging for different applications?
Yole: Semiconductor lasers, both EELs and VCSELs, benefit from their small size, light weight, high reliability, and easy modulation. They have become increasingly popular for new applications in recent years. Generally, we see challenges driven by changes to [enable] lower cost, longer lifetime, higher power, better efficiency, and [higher] reliability. That activity, of course, affects the choice of material and diode type as well as the package. However, the challenges associated with high-power laser diodes [HPLDs] differ from those for low-power lasers for sensing and optical communication.
With the advancement of HPLD technology, packaging remains one of the bottlenecks that affect output power, [high] brightness, [narrow] spectrum, and others. Too much of an increase in the output power could cause undesirable effects, such as catastrophic optical mirror damage or thermal rollover, when more heat is generated than dissipated by the laser device. New, sophisticated packaging approaches have been developed to improve thermal management for different designs, such as single emitters, bars, and stacks.
With an increasing number of emitters, the optical design becomes more complex, which negatively impacts the cost of the micro-optics and the module’s footprint. The other challenge of HPLD is how to focus the light more easily onto a small spot. A laser source with higher brightness can deliver greater power density at the targeted surface, but achieving high brightness through optical design, with beam-combining and optical-coupling processes, is challenging. The beam-shaping components need the ability to work with the high power necessary to handle the electrical power that can pass through the system, without causing damage to the components.
The spectral broadening of laser arrays is a result of non-uniform emitting wavelengths from individual emitters. The spectral width is one of the key specifications of laser bars; it is crucial to control the spectral performance to improve production yield and thereby reduce costs and gain competitiveness. Achieving temperature and stress uniformity across the laser bar, thus eliminating thermal and stress effects, remains a challenge.
The increasing performance of low-power laser assemblies leads to many power-hungry solutions. This is an unsustainable trend, particularly for datacom interconnects and computing applications, as well as for LiDAR used in future autonomous electric vehicles. Therefore, we see a high demand for dense integration of photonics with electronics. Integrated optoelectronics using III-V compound semiconductor technology on low-cost silicon substrates has shown exciting possibilities. One key driver for using silicon for photonics is the advantages of scaling up to 300-mm wafer CMOS processes; this drives the cost down significantly — up to 60%. A further advantage is in the low-loss silicon waveguides that connect various photonic elements, including lasers, modulators, photodetectors, and other passive devices.
Improvements in the yield of silicon photonics-based devices coupled with the application-driven requirements enable the integration of a larger number of devices on a single silicon photonic integrated circuit [PIC]. We expect that heterogeneous integration [direct wafer bonding] will be preferred in the coming years.
EE Times Europe: How will lasers contribute to the enhancement of AI and ML applications?
Yole: Artificial-intelligence and machine-learning applications primarily use silicon chips in deep neural networks. AI chips [also called AI accelerators] are specially designed accelerators for applications based on artificial neural networks [ANNs]. Most commercial ANN applications are deep-learning applications. The main system trends in data centers are disaggregation, scale-up, heterogeneous acceleration, and the convergence of high-performance computing [HPC] and the cloud. The goal is to bring technologies to the market that can contribute to building more efficient systems for workloads in HPC and AI training/inference. As HPC moves into the cloud, the systems will require a low-latency and high-bandwidth communication fabric for AI accelerators. Integrated lasers will serve for optical interconnection of HPC and AI/ML workloads. This solution is very new and is still in R&D.
EE Times Europe: Do you think the laser materials-processing market is a potential area of growth?
Yole: The laser-based materials-processing market is a traditional laser market. Laser materials processing has been used for over four decades and plays a vital role in modern manufacturing and the economy. While laser cutting, welding, marking, and drilling processes have reached maturity and wide industrial acceptance, new developments in recent years in additive manufacturing and micro/nanofabrication have enabled new capabilities that lasers can bring to the manufacturing industry. Laser-based systems are increasingly gaining share within the materials-processing market because of the greater precision, processing speeds, and flexibility this technology enables. EELs for kilowatt material processing are projected to grow from US$409 million in 2020 to US$468 million in 2026 at a compound annual growth rate for the period of 2%. This moderate growth will continue to be driven by new applications as well as the automated-manufacturing trend.
EE Times Europe: What future innovations do you see for edge-emitting lasers?
Yole: It is important to emphasize that edge emitters make up a core technology of multiple laser systems targeting many end markets. Generally, edge-emitter technology has been driven by multiple laser applications’ demands on intrinsic laser diode parameters for different end markets.
For example, the traffic in data centers is growing exponentially. Therefore, for optical communication, it is essential to increase the bandwidth and speed of the optical modules at all levels of DC infrastructure while simultaneously decreasing the footprint and power consumption. In terms of discrete laser diodes, innovations stem from the DFB [distributed-feedback] laser design by enhancing the multi-quantum well of EA [electro-absorption] modulators for high-speed operations and reducing the inductance of the wires by bonding short wires for high bandwidth. Future innovations will likely result in the integration of PICs, also known as silicon photonics. Integrating complex electro-optical circuits onto a single silicon chip enables the creation of new form factors while increasing the bandwidth and distance, significantly improving power efficiency and density in ways that simply could never be addressed with discrete devices.
EE Times Europe: What are the trends in high-power laser applications?
Yole: Currently, we observe more dynamics in additive manufacturing, semiconductor manufacturing, and medical applications, especially in dermatology and surgery.
The additive manufacturing industry is going through extraordinary times. 3D printing is enabling digital supply chains and establishing itself as a viable technology for on-demand manufacturing. In consumer electronics, laser micromachining is widely used for high-volume semiconductor manufacturing to achieve the desired precision, quality, throughput, and cost per machined part. For devices such as smartphones, tablets, and wearables, laser micromachining is used in the fabrication of semiconductor chips and their packaging, electronic components, and the circuitry connecting these components in the form of PCBs.
The manufacture of high-resolution touchscreen displays also relies on many laser-micromachining processes. The camera windows, the sensors, and even the external enclosure utilize laser micromachining to achieve the required machined results.
Lasers for medical applications reflect recent research on the laser’s principles, technologies, and applications in diagnostics, therapy, and surgery. A wide variety of lasers serve medical applications, depending on such factors as the optical wavelength, output power, and pulse format required. In many cases, the laser wavelength is chosen such that certain specific substances, such as pigments in tattoos or caries in teeth, will absorb light more strongly than the surrounding tissue so that they can be more precisely targeted.
This article was originally published on EE Times.
Maurizio Di Paolo Emilio holds a Ph.D. in Physics and is a telecommunication engineer and journalist. He has worked on various international projects in the field of gravitational wave research. He collaborates with research institutions to design data acquisition and control systems for space applications. He is the author of several books published by Springer, as well as numerous scientific and technical publications on electronics design.