SignificanceGaAs-based high-power edge-emitting semiconductor lasers are the indispensable pump source fiber, solid-state and alkali-vapor laser systems. Over the last two decades, fiber-laser—and high-power laser systems more broadly—have vaulted from the hundred-watt and kilowatt regimes into the tens-of-kilowatts domain owing to steady gains in semiconductor pumping source output power and breakthroughs in the fabrication of supporting components. Currently, the quest for hundreds-of-kilowatt optical powers in high-energy-laser systems and for greater cost effectiveness, reliability, and efficiency in industry pushes these diodes to their limits in terms of power, wall-plug efficiency, brightness, and spectral control.The HighEnergy Laser Scaling Initiative (HELSI) targets 500 kW laser modules and aims to achieve the megawatt class in 2025?2030 while significantly reducing the system volume. Satisfying these demands pump diodes that (i) surpass 70% wallplug efficiency; (ii) remain stable at heatsink temperatures above 50 ℃ to ease thermal management; (iii) maintain precise spectral overlap with the gain fiber’s absorption spectrum across a >50 ℃ range; (iv) lock their wavelength within milliseconds for “instanton” operation; and (v) deliver higher brightness to increase coupling efficiency and suppress stimulated Raman scattering.Meanwhile, emerging manufacturing scenarios accelerate diodelaser innovation. Precision processing such as batterytab welding require the laser source delivery of high power and near diffraction limited beam quality, in addition to narrow linewidths to curb transversemode instability. The abundance of handheld welding tools propels broadarea chips toward higher power levels, thus enabling simpler and more economical system layouts. Current fiberlaser cutters exceed 100 kW, thus imposing stringent demands on diode power density and brightness to mitigate nonlinear effects. Finally, pumping requirements for alkalimetal vapor and ultrafast solidstate lasers demand narrowband, highpower emission at specific wavelengths, and suppressing thermally induced wavelength drifts under high power loading.ProgressTo satisfy increasingly stringent, multidimensional performance demands, breakthroughs in GaAs-based high-power edge-emitting semiconductor lasers have been realized over the past five to ten years. The industrialization of epitaxy and fabrication, in addition to asymmetric-waveguide design, multijunction cascading, and thermally optimized packaging, has pushed continuous-wave outputs of 9xx-nm broad-area single-junction emitters beyond 45 W, while broad-area multijunction chips have achieved continuous-wave power breakthrough at hundred-watt level. Mode-selective architectures curb slow-axis divergence, thus enabling 1 kW of near-diffraction-limited power from 200 μm fibers. On the spectral front, on-chip grating wavelength locking chips has achieved comparable performance to Fabry?Pérot devices in terms of wall-plug efficiency yet offer far superior integrability compared with external volume-Bragg-grating schemes. Continued progress will hinge on scaling outputs against thermal or nonlinear roll-overs, enhancing beam quality while preserving power and efficiency, and expanding on-chip wavelength-locking range without deteriorating electro-optical performance. Satisfying these intertwined goals relies on systematic innovation in semiconductor chip fabrication technologies, chip architecture design, and fiber-coupling techniques.Conclusions and ProspectsGaAs-based high-power edge-emitting semiconductor lasers have advanced considerably in the past decade. Currently, fully industrial 6-inch epitaxial and wafer-level production lines underpin devices whose performance comfortably satisfies, and often exceeds, the demands of modern laser systems. Continuous-wave outputs from 9-xx nm broad-area single-junction emitters surpasses 45 W, while double-junction chips exceed 130 W. The peak electro-optical efficiency in the 7-xx nm has achieved breakthrough peak efficiency exceeding 70%. Meanwhile, the lateral brightness has reached 4.0 W·mm-1·mrad-1 at 45 W, and tapered chips deliver near-diffraction-limited beams (M2<1.2 at 10 W). On-chip grating lockers provide robust spectral control, and single-emitter based modules have transcended the 1 kW barrier through 200 μm fibers. Meanwhile, spectral beam-combining architectures has achieved >53% efficiency at these power levels. From the perspective of current technology trends, GaAs-based semiconductor lasers are advancing toward higher efficiency, superior beam quality, greater environmental robustness, and larger-scale manufacturability.

SignificanceStimulated Raman scattering (SRS) serves as a pivotal nonlinear optical process that effectively overcomes the spectral limitations of conventional gain media, enabling versatile laser emission at specialty wavelengths. This phenomenon finds critical applications across optical communications, quantum photonics, gas sensing, atomic physics, and defense technologies. Currently, three dominant platforms—crystalline materials, optical fibers, and on-chip waveguides—have emerged for Raman laser implementation. While their substantially different optical confinement scales lead to distinct output characteristics, the underlying Raman laser physics maintains fundamental universality. This review systematically examines technological advancements in Raman lasers across these platforms, with particular emphasis on their progress in power/energy scaling, spectral control, wavelength flexibility, and temporal characteristics. The comprehensive analysis provides researchers with macroscopic insights into the state-of-the-art development of mainstream Raman lasers, while offering valuable perspectives for future performance enhancement and cross-platform technological convergence.ProgressAs one of the earliest-developed laser architectures, crystalline Raman lasers face inherent power-scaling limitations due to thermal effects. Conventional gain media such as KGW and YVO4 have achieved maximum continuous-wave (CW) outputs of only 11 W. Nevertheless, their well-established crystal growth techniques enable large-aperture devices capable of generating single-pulse energies approaching the joule level—a critical advantage for high-energy applications. The emergence of synthetic diamond has revolutionized CW power performance, overcoming the intrinsic power limitations of traditional Raman crystals. However, the current maximum diamond substrate dimensions fundamentally restrict energy storage capacity, presenting a key challenge for pulsed energy scaling. Tables 2 and 3 systematically compare the spectral performance of crystalline Raman lasers, highlighting the narrow-linewidth operation and wavelength extension. Based on the authors’ statistical findings, the highest power, the narrowest linewidth, and the broadest spectral coverage of laser output among crystal Raman lasers have all been achieved in the diamond.Raman fiber lasers (RFLs) have emerged as the most prevalent type of Raman lasers due to their high stability, excellent thermal dissipation characteristics, and robust performance. Table 4 summarizes the current research status of narrow-linewidth RFLs, where power scaling remains technically challenging primarily limited by stimulated Brillouin scattering (SBS). Recent advances in laser technologies, including cladding pumping, random lasing, and hybrid-amplification schemes, have progressively enhanced the power capacity of RFLs, with the output power recently surpassing the 10 kW. The progress in power and brightness scaling is systematically reviewed in Fig. 10 and Table 5. The operational wavelength of RFLs can be further extended to 1.8 μm or mid-infrared bands through cascaded Raman conversion or utilization of fluoride or chalcogenide fibers. However, their output power in these spectral regions still lags significantly behind that achieved in the near-infrared regime. In Fig. 12 and Table 8, we provide a comprehensive overview of the technical approaches and recent developments in generating ultrafast laser outputs from RFL systems.In recent years, the rapid advancement of on-chip Raman lasers has been facilitated by the increasing maturity of planar waveguide fabrication techniques. Table 9 summarizes the reported physical properties of on-chip Raman gain materials, while Figs. 15?19 present the structural designs and output characteristics of such lasers based on crystalline silicon, aluminum nitride, chalcogenide glass, diamond, and other materials. These results demonstrate their capability for single Raman line emission, Raman?Kerr optical frequency comb generation, and wavelength tuning.Conclusions and ProspectsBulk crystals, with their high Raman gain coefficients and large interaction volumes, hold significant advantages for high-energy pulsed Raman laser generation. However, their limited thermal conductivity often leads to pronounced thermal effects under CW operation. Recent advances in synthetic diamond fabrication have partially mitigated this limitation, enabling CW diamond Raman lasers to surpass 100 W output power, though this still falls far short of the performance achievable with fiber-based systems. Optical fibers, benefiting from an exceptional surface-to-volume ratio, exhibit outstanding thermal management capabilities. Continuous improvements in fiber design, associated components, and laser diode (LD) manufacturing have propelled Raman fiber lasers to power levels exceeding 10 kW. Nevertheless, their performance in single-frequency operation remains constrained by stimulated Brillouin scattering (SBS), with output power yet to break the 150 W barrier in CW regime. Additionally, progress in wavelength extension and mid-infrared Raman laser generation lags behind power scaling efforts. The maturation of planar waveguide fabrication techniques has spurred growing interest in on-chip Raman lasers, which offer advantages such as low power consumption, compact footprint, and ease of integration. The tightly confined optical modes in microresonators significantly enhance nonlinear effects compared to fibers, enabling low-threshold Raman lasing. However, strong pump excitation can also trigger Kerr optical frequency comb generation, imposing practical limits on further power scaling. As Raman laser technologies across these platforms continue to evolve, multidimensional control over spectral, temporal, and power characteristics will become increasingly accessible. This progress promises to further expand Raman laser applications in advanced manufacturing, scientific research, and defense technologies.

SignificanceLaser is widely used in modern communications, medical applications, and industrial manufacturing due to its high brightness, excellent monochromaticity, superior directionality and strong coherence. All-solid-state lasers exhibit high energy conversion efficiency, excellent beam quality, low thermal effects, compact structure and diverse functionalities, demonstrating significant application potential in advanced manufacturing, lithography, medical treatment and scientific research. As a core part of frequency conversion in solid-state lasers, nonlinear optical (NLO) crystals enable the extension of laser applications into the deep ultraviolet (DUV) region. The performance of NLO crystals largely determines the development of laser technology. Therefore, it is of great practical significance to develop high-performance DUV NLO crystals.A series of representative crystals, such as β-BaB2O4 (β-BBO), LiB3O5 (LBO), CsB3O5 (CBO), CsLiB6O10 (CLBO) and KBe2BO3F2 (KBBF), have been developed for commercial applications in the UV/DUV regions. These crystals exhibit superior NLO performance, including a wide transmission range, high second-harmonic generation (SHG) coefficients, and excellent optical uniformity. Chinese scientists have made significant contributions to the field of DUV NLO crystals. Among them, KBBF crystals are the only NLO crystals that can directly generate DUV lasers through sixth-harmonic generation of a Nd∶YAG laser. However, their strong layered growth habit makes it difficult to obtain large-sized crystals, limiting their application scope to some extent. With the rapid development of modern DUV laser technology, it has become a hot topic of research to explore new high-performance DUV NLO crystals to meet the diverse and customized needs of DUV laser technology.DUV NLO crystals must simultaneously meet the following performance criteria(1) A wide bandgap, corresponding to a wide transparency range and high transmittance (UV cutoff edge <200 nm, bandgap >6.2 eV); (2) Large SHG response, high NLO coefficients (second-order nonlinear susceptibility dij≥0.39 pm/V, at least comparable to KDP’s d36); (3) Moderate birefringence (Δn≈0.05?0.10) and small refractive index dispersion. There are mutual influences and constraints between the above three parameters. The number of crystals capable of simultaneously balancing the three critical factors (SHG response, UV cutoff edge, and birefringence) remains extremely limited. Additionally, it is important that the crystals can grow into large single crystals and be easily processed into usable shapes and sizes. It remains a formidable challenge to explore DUV NLO crystals with application potential.The structure-property relationships of excellent NLO crystals such as β-BBO, LBO, CBO, CLBO, and KBBF have been studied. The results show that boron-oxide anions possess a rich variety of types, diverse structures and excellent microscopic optical properties (such as strong polarizability anisotropy, high hyperpolarizability and wide HOMO-LUMO gap). These properties play a key role in determining the outstanding performance of crystals. The flexible combination of [BO3] planar triangles and [BO4] tetrahedra can form multi-level structures ranging from isolated units to three-dimensional networks. This provides a structural basis for modulating the band gaps and birefringence of crystals. In addition, partial replacement of O atoms in [BO4] units with F atoms can form novel distorted [BOxF4-x] (x=1?3) tetrahedral units. Distorted [BOxF4-x] (x=1?3) tetrahedral units combine with the original [BO3] and [BO4] units to form a new fluorooxoborate system. The new borate derivatives are formed by combining boron-oxygen units with phosphorus-oxygen units. This review primarily focuses on systematically classifying anhydrous inorganic mixed anion borate crystals with DUV NLO properties obtained by experiments in recent years. These compounds not only significantly enrich the structural diversity of inorganic crystals but also greatly expand the synthesis strategies and systems for DUV NLO crystals. Additionally, they provide valuable references for the design and synthesis of DUV NLO crystals with enhanced performance.ProgressThis paper first describes beryllium-containing fluorine-boron mixed anion crystals (Table 1). A representative crystal is KBBF, which exhibits a short UV cutoff edge (147 nm), strong SHG response (1.26×KDP), and moderate birefringence (0.077@1064 nm). To address the toxicity of beryllium, a strategy was proposed to replace Be with Al, Sc, and Y. This prompted the synthesis of fluorine-boron mixed anion crystals containing Al, Sc, and Y (Table 2). To explore fluorooxoborate mixed anion crystals with superior performance, a series of fluorooxoborates represented by NH?B?O?F were synthesized (Table 3). NH4B4O6F exhibits a short UV cutoff edge (155 nm), high SHG response (3×KDP), and large birefringence (0.117@1064 nm). Finally, the role of boron-phosphorus mixed anion NLO crystals (Table 4) and machine learning in advancing research on DUV NLO crystals are discussed.Conclusions and ProspectsThis review systematically summarizes the latest research progress in the DUV NLO field of anhydrous inorganic mixed anion borate crystals. By strategically introducing fluorine and phosphorus-oxygen groups, novel fluorine-boron and boron-phosphorus mixed anion systems have been successfully constructed. These systems have achieved synergistic optimization of key performance parameters such as bandgap, SHG response, and birefringence. The deep integration of theoretical calculations and experimental studies not only reveals the fundamental structure-property relationship between the microscopic anion group arrangement and macroscopic optical properties but also significantly advances crystal design and synthesis methods. In the future, artificial intelligence methods such as machine learning can be used to design and synthesize crystals with enhanced performance.

SignificanceMid-wave infrared (MIR, 3?5 μm) lasers with high power and large energy output have attracted significant attention due to their broad application potential in optoelectronic countermeasures, infrared medicine, and laser communication. As the key component of solid-state MIR laser systems, ZnGeP2 (ZGP) crystals must be large and exhibit low absorption loss to enable efficient nonlinear frequency conversion, thereby achieving hundred-watt-level high power and hundred-millijoule-level large energy output. This review focuses on the development and application of large-size ZGP crystals for MIR laser generation. It systematically summarizes recent advances in ZGP crystal growth, high-power and high-energy MIR laser output, and the miniaturization of MIR lasers. The paper also highlights our group's progress in growing large-size ZGP crystals and fabricating large-aperture optical devices. Finally, current technological challenges are analyzed and future research directions for advancing this field are proposed.ProgressAccording to recent reports from leading domestic and international research institutions (Fig. 1, Fig. 2), the growth size of ZGP single crystals has generally surpassed the Φ50 mm scale. However, due to limitations in post-processing techniques and challenges in maintaining optical homogeneity, the practical device aperture is currently constrained to a maximum of approximately 25 mm×25 mm, with device lengths ranging from 30 to 50 mm (Fig. 8). The advancement of large-size crystal growth technology not only provides a critical foundation for fabricating large-aperture nonlinear optical devices but also enables multiple devices to be sliced from a single ingot, thereby significantly reducing unit production costs and enhancing overall manufacturing efficiency. Despite this progress, large-size ZGP crystals and devices continue to face technical challenges, including the control of compositional and optical uniformity, the suppression of structural defects, and the enhancement of laser-induced damage thresholds. To further improve crystal quality and device performance, ongoing research efforts are focused on optimizing growth parameters, employing elemental doping strategies, and developing advanced post-processing techniques.In the development of high-power MIR laser systems, ZGP-based optical parametric oscillators and amplifiers (OPO/OPA) have emerged as core technological approaches. When combined with Ho∶YAG lasers as efficient pump sources, this configuration has successfully achieved MIR laser outputs exceeding 100 W, thereby significantly advancing the upper power limits of MIR laser technology, as detailed in Table 2. To further enhance output power and optical-to-optical conversion efficiency, research has focused on the integration of multi-stage master oscillator power amplifier (MOPA) architectures, the use of low-absorption ZGP crystals, and the implementation of compact resonator cavity designs. These advancements aim to meet the growing demand for high-power, high-efficiency mid-wave infrared (MWIR) laser systems in critical application areas such as infrared countermeasures, remote sensing, and space-based optical communications.High-energy MIR lasers are primarily realized through OPO and OPA techniques, which utilize nonlinear frequency conversion in ZGP crystals pumped by high-energy 2.1 μm laser sources, as illustrated in Table 3. To enhance pump energy and beam quality while mitigating thermal effects within the nonlinear crystal, a high-energy Ho∶YAG MOPA system is employed, incorporating multi-stage amplification and precise spectral control. Through doping engineering and optimization of crystal growth parameters, the absorption coefficient of ZGP in the pump band is significantly reduced, effectively suppressing thermal lensing and increasing the energy handling capacity of the crystal. These advancements collectively support the development of joule-level, single-pulse MIR laser systems operating at repetition frequency exceeding 1 kHz—systems that are crucial for next-generation applications in directed-energy laser weapons, deep-space exploration, and other frontier scientific and defense technologies.Pulsed thulium (Tm)-doped lasers, commonly configured with linear resonant cavities, are widely used as direct pump sources for ZGP OPO. These systems offer key advantages, including compact size, ease of miniaturization, low lasing threshold, and high peak power density. With ongoing advancements in ZGP crystal growth and post-processing technologies, the absorption coefficient at 1.907 μm has been successfully reduced to below 0.05 cm-1, enabling efficient pumping in the 1.9 μm band. This wavelength coincides with the gain peak of Tm-doped fiber lasers, which is particularly beneficial for suppressing nonlinear optical effects and reducing amplified spontaneous emission (ASE). Moreover, the reduced absorption at this pump wavelength mitigates thermal accumulation within the ZGP crystal, thereby improving both nonlinear frequency conversion efficiency and the output beam quality of mid-wave infrared lasers. Based on these considerations, future development of miniaturized ZGP-based MIR laser systems should focus on three key directions: increasing the power of pulsed Tm-doped fiber lasers, optimizing OPO cavity geometries, and further minimizing the near-infrared absorption of ZGP crystals. These efforts are expected to significantly enhance pump efficiency and support the advancement of high-performance, compact mid-infrared laser sources.Conclusions and ProspectsAll-solid-state lasers based on ZGP crystals for optical frequency conversion have emerged as a leading technology for generating MIR laser sources, owing to their advantages in compactness, lightweight design, high efficiency, high output power, and continuous wavelength tunability. Recent advancements in high-average-power, high-repetition-frequency Ho∶YAG lasers have driven considerable progress in ZGP OPO and OPA systems, enabling mid-infrared laser outputs exceeding 100 W and expanding their potential in applications such as infrared countermeasures, remote sensing, and space-based optical communications. Future research will focus on four key areas: (1) scaling power and energy through multi-stage amplification in Ho∶YAG MOPA-driven ZGP cascaded systems, which have already demonstrated >100 W and >100 mJ single-pulse output; (2) enhancing conversion efficiency beyond the current ~70% limit, which is constrained by phase-matching and pump characteristics, through the development of low-absorption ZGP crystals; (3) achieving miniaturization and integration by employing sub-2 μm pump sources, compact cavity architectures, and optical MOPA systems to reduce system size while maintaining performance; and (4) improving beam quality through image-rotating non-planar ring cavity OPOs, with further enhancements in tunability and power anticipated through advanced resonator designs or MOPA integration. In summary, while the performance of ZGP-based solid-state lasers is presently limited by the output characterization of pump source and the absorption and damage thresholds of ZGP crystals, future advancements in pump engineering and crystal optimization are expected to enable mid-infrared systems with hundreds of watts of output power, joule-level pulse energy, conversion efficiencies exceeding 70%, high beam quality, and compact form factors, thereby meeting the stringent requirements of next-generation optoelectronic applications.

SignificanceIn the context of global digital transformation, silicon photonics has achieved significant advancements in optical communications, interconnects, and computing, owing to its CMOS compatibility, superior integration capability, and cost advantages, demonstrating substantial market potential. The development of integrated lasers on silicon photonics platforms has been constrained by silicon’s inherently low luminescence efficiency as an indirect bandgap material. Heterogeneous integration technology, which combines high-efficiency luminescent materials with established silicon platforms, has emerged as the most promising solution to overcome laser performance limitations. Bonding technology presents notable advantages, enabling the integration of high-quality epitaxial materials onto large-scale silicon photonic wafers while eliminating complex alignment requirements. Molecular bonding enables the formation of robust covalent interfacial bonds at relatively low annealing temperatures, making it particularly appropriate for silicon photonics/III?V material heterogeneous integration. This technique has demonstrated technological maturity through successful implementation in commercial optical module products. Silicon heterogeneous integrated lasers have developed through various technological pathways, maintaining significant research interest. Substantial progress has been achieved in critical aspects including on-chip architecture design, material preparation, and integration methodology. This paper presents a comprehensive review of current developments in bonding-based silicon heterogeneous integrated lasers, innovative research, and recent progress in domestic research initiatives.ProgressFirst, we discuss the implementation scheme of silicon heterogeneous integrated single longitudinal mode lasers and the research progress of three types of lasers, including distributed feedback (DFB) lasers, Bragg grating reflection (DBR) lasers, and broadband tunable laser with cascaded micro-ring resonators. By incorporating both direct phase fluctuations and indirect carrier-density-fluctuation-induced quantum noise from spontaneous emission and using the modified Schawlow?Towns?Henry equation, an explicit analytical relationship between the laser linewidth and the physical parameters of practical laser cavity designs are established. The implementation of high-Q optical cavities enables substantial linewidth reduction in semiconductor lasers. This approach is particularly advantageous for silicon heterogeneous integrated lasers, where the optical mode distribution both the undoped silicon waveguide and III?V gain region synergistically reduces optical losses. Silicon photonics offers unique advantages for implementing low-loss, large-scale optical structures, including broadband tunable reflectors and high-Q grating cavities, which are critical for achieving laser linewidth compression. The DFB laser represents a fundamental photonic device architecture that utilizes a spatially periodic Bragg grating structure integrated throughout the active gain medium to achieve simultaneous optical feedback and spectral mode selection. This configuration exhibits distinctive operational advantages of single-mode operation, spectral stability, and high coherence. In heterogeneous integrated DFB lasers, the Bragg gratings are typically patterned either directly on or adjacent to the silicon rib waveguides. Control of the optical mode distribution through waveguide geometry optimization to realize minimize optical loss while maintaining sufficient modal gain enables high-power narrow-linewidth operation. In the DBR lasers, the gratings are located at both ends of the active optical gain region. Thus, long grating can be designed to realize narrow linewidth emission. The mismatch in thermo-optic coefficients between the active and passive regions induces mode hopping during operation, necessitating real-time feedback control of the output optical power for stable performance. The incorporation of cascaded silicon micro-ring resonators as laser cavities enables wavelength-tunable operation. The design of cascaded micro rings also involves a trade-off between insertion loss and filtering bandwidth. Second, we introduce the quantum dot (QD) heterogeneous integrated laser that exhibits superior performance characteristics owing to its low linewidth enhancement factor. Compared to conventional quantum well lasers, the QD lasers experience narrower linewidth, lower threshold current, enhanced temperature stability, defect tolerance, and inherent feedback insensitivity. Third, we analyze narrow-linewidth heterogeneously integrated lasers incorporating ultra-low loss silicon nitride waveguides and the challenges in the customized fabrication process. Then, we examine emerging research frontiers in silicon heterogeneous integrated laser that are attracting significant academic attention. Recent advances in heteroepitaxial III?V/Si lasers are examined, focusing on both vertical and lateral epitaxial integration approaches. The micro-transfer printing (μTP) fabrication methodology and resulting photonic device characteristics are examined. The Ge/Si laser employing strain engineering and n-type doping are discussed. Finally, we introduce the research progress of domestic universities and research institutions in the field of heterogeneous integrated lasers, including low-temperature bonding processes, heteroepitaxial growth techniques, and Ge/Si lasers.Conclusions and ProspectsThrough sustained research and development efforts, silicon heterogeneous integrated laser technology has evolved from fundamental research to commercial implementation, establishing a comprehensive technological framework encompassing novel material integration, advanced micro-nanofabrication, and photonic device engineering. This article presents a thorough review of current developments in heterogeneous integrated light source technology. Heterogeneous integration technology is experiencing a transformative phase of development. The silicon photonics industry is transitioning from 8-inch to 12-inch wafer platforms, driven by enhanced process capabilities, 3D integration compatibility, and economic advantages. The integration of multiple optoelectronic materials on a common substrate to achieve multifunctional photonic chips represents a significant development trend. While silicon photonics maintains dominance in optical modules for various applications, heterogeneous integration is overcoming critical performance limitations, enabling expansion into emerging fields including optical computing, lidar, and integrated optical gyroscopes. Heterogeneous integration technology is anticipated to become the fundamental enabling technology for future photonic integrated circuits, providing essential technical support for significant applications of next-generation silicon photonics technology.

SignificanceIntegrated photonics technology represents a rapidly advancing field that attracts substantial attention for its capacity to transcend the dimensional and performance constraints of traditional integrated circuits through photon-based information transmission. Lithium niobate (LN) emerges as a particularly promising integrated photonics platform, attributed to its remarkable optical characteristics, including broad transparency range, high refractive index, and pronounced electro-optic and acousto-optic effects, along with its compatibility with semiconductor micro-nano fabrication processes. As light sources fundamentally determine the performance and capabilities of integrated photonic systems, research into LN-based on-chip lasers holds critical importance. Recent years have witnessed significant enhancement in LN waveguide, microcavity, microring, and resonant structure performance through Smart-cut technology and improved mechanical etching processes. These developments have facilitated the evolution of rare-earth-doped LN microcavity lasers, with ongoing performance optimization efforts. Furthermore, substantial progress has been achieved in heterogeneously integrated LN on-chip laser research. This article synthesizes recent developments in LN on-chip lasers to establish a foundation for future advancements.ProgressThis paper presents the fabrication flow of LN wafers using Smart-cut technology (Fig. 1). It then introduces rare-earth-doped LN microcavity lasers, beginning with an overview of three different doping techniques (Fig. 2), followed by the fabrication process of rare-earth-doped LN microdisk lasers using photolithography-assisted chemo-mechanical etching (PLACE) (Fig. 3). The research progress on rare-earth-doped LN microdisk lasers is then discussed, including key performance parameters and observed nonlinear phenomena (Fig. 4). Next, the fabrication process of microring lasers using electron-beam lithography combined with helium ion etching is demonstrated (Fig. 5). The research progress on rare-earth-doped LN microring lasers is subsequently reviewed (Fig. 6). This is followed by an introduction to the advancements in single-mode whispering gallery mode microcavity lasers based on the Vernier effect (Fig. 7) and mode control (Fig. 8). In addition to whispering gallery mode microcavity lasers, FP cavity lasers using Sagnac loop reflectors as end mirrors are also discussed (Fig. 9). Furthermore, the progress in electrically pumped rare-earth-doped microcavity lasers is summarized (Fig. 10). Finally, recent advances in on-chip lasers fabricated through the heterogeneous integration of III?V materials with LN are reviewed (Fig.11).Conclusions and ProspectsDespite notable advances, lithium niobate-based on-chip lasers face persistent challenges, including relatively low threshold powers and conversion efficiencies in rare-earth-doped thin-film lasers, alongside integration complexities stemming from uniform doping-induced absorption and refractive index variations. Although heterogeneous integration with III?V materials demonstrates superior performance potential, this approach requires additional refinement in interface loss control, reliability, and chip-level packaging. Future research directions should emphasize advanced local doping techniques, innovative resonator designs, and enhanced wafer-scale integration methods to achieve stable, efficient, and high-performance LN-based on-chip lasers, facilitating expanded applications in optical communications, quantum information, and artificial intelligence technologies.

SignificanceLasers are pivotal tools in numerous cutting-edge technologies, offering unparalleled performance owing to their high brightness, monochromaticity, directionality, and coherence. These attributes make them indispensable in fields such as optical communication, high-resolution displays, high-precision sensing, advanced manufacturing, and biomedical imaging. As the demand for compact, efficient, and tunable laser sources increases, the search for new, high-performance gain media becomes increasingly critical.Metal halide perovskites have recently attracted extensive interest in the laser community because of their remarkable optoelectronic properties. These include high carrier mobility, broad spectral tunability, strong light absorption, high photoluminescence quantum yield, large optical gain, and low lasing thresholds. Moreover, they are compatible with low-cost and scalable solution-processing methods, making them ideal for next-generation integrated photonic devices.Among various perovskite forms, single crystals are prominent owing to their superior crystalline quality, significantly lower trap densities, and enhanced carrier dynamics. These attributes result in reduced nonradiative recombination, longer carrier lifetimes, lower lasing thresholds, and improved stability under operational conditions. Therefore, perovskite single crystals represent a key material platform for advancing laser technology, warranting a comprehensive overview of their recent progress and remaining challenges.ProgressThis review provides a systematic summary of the recent advancements in perovskite single-crystal lasers, covering material properties and gain mechanisms, crystal growth techniques, laser devices, and applications.Through the careful modification of the A, B, and X sites of the ABX? perovskite structure, the bandgap can be tuned to span the visible spectrum, with halide substitution from Cl to Br to I progressively lowering the bandgap. Low-dimensional perovskites also exhibit strong exciton binding energies owing to confinement effects, which further enhance light?matter interactions. The optical gain properties of perovskites are closely associated with carrier recombination mechanisms, including monomolecular, bimolecular, and Auger processes. When the carrier density exceeds ~1018 cm-3, stimulated emission can occur, enabling optical gain. Compared with polycrystalline films, single crystals demonstrate significantly reduced trap-assisted nonradiative recombination and offer an ideal platform for exciton?photon coupling and polariton formation. Their lack of grain boundaries inhibits ion migration and moisture ingress, thus enhancing thermal and environmental stability.Bulk single crystals are predominantly grown via solution-based methods, including temperature cooling crystallization, inverse temperature crystallization, slow evaporation crystallization, and antisolvent vapor-assisted crystallization. While these methods yield high-purity crystals, they are often limited by slow growth rates or environmental sensitivity. For micro- and nanoscale single crystals, advanced methods such as space confinement growth, solution epitaxy, chemical vapor deposition, and surface energy-controlled method enable better control over crystal morphology, thickness, and orientation, which are essential for integration into photonic devices. Furthermore, patterned crystal arrays have been achieved using stamping, surface-functionalized substrate-assisted growth, capillary force-assisted crystallization, photolithography-assisted crystallization, inkjet printing, and laser patterning, paving the way for on-chip integration and scalable device fabrication.Several types of perovskite single-crystal laser devices have been developed. Nanowire lasers, benefiting from their one-dimensional geometry, naturally form Fabry?Pérot cavities with excellent waveguiding and reflectivity at the end facets. Early studies demonstrated wavelength-tunable lasers across the visible spectrum using solution-processed MAPbX3 nanowires, whereas later research achieved color tuning, single-mode operation, and threshold reduction via halide composition control and substrate engineering. Plasmonic nanowire lasers further enhance emission through near-field coupling with surface plasmons, enabling subwavelength light confinement. Perovskite microplates and nanoplates can form whispering gallery mode cavities in various shapes, supporting high-Q, tunable, and even single-mode lasing. Other microcavity designs include distributed feedback (DFB) lasers, vertical-cavity surface-emitting lasers (VCSELs), and plasmonic lasers that exploit strong coupling and feedback mechanisms. Although bulk single crystals are not ideal for traditional lasing owing to reabsorption, they can support random lasing through internal scattering from structural imperfections. In addition, heterojunction-based lasers exhibit significant potential by leveraging types I and II band alignments to reduce thresholds and enhance thermal stability. Examples include CsPbBr3/GaN and CsPbI2Br/WSe2 systems that enable effective carrier confinement and enhanced exciton?photon interactions. Beyond fundamental studies, perovskite single-crystal lasers have demonstrated functional applications in emerging technologies. In display systems, they offer structural color tuning and polarization control through nanostructure engineering. In sensing applications, perovskite nanowires have enabled sensitive, low-cost gas detection platforms. For integrated photonics, they have been employed in polariton lasers and waveguide-based circuits, offering a scalable route toward on-chip coherent light sources.Conclusions and ProspectsDespite these promising developments, several challenges must be addressed to fully realize the potential of perovskite single-crystal lasers. Crystal growth methods must be further optimized for uniformity, reproducibility, and scalability. Control over optical properties such as emission wavelength, cavity geometry, and dimensionality is still limited, particularly for achieving ultra-low threshold and electrically pumped operation. While single crystals exhibit fewer defects than polycrystals, surface defects remain problematic, resulting in undesired nonradiative losses. Moreover, high carrier densities are required for lasing intensify Auger recombination and thermal degradation, hindering continuous-wave and electrically driven operation. To overcome these problems, future research should focus on understanding and mitigating Auger recombination, improving crystal growth techniques for shape and quality control, and exploring stable, lead-free, or 2D perovskite compositions. Efficient thermal management strategies will also be essential for high-duty cycle or continuous lasing. Furthermore, efforts toward patterning and array integration at the wafer scale are key to transitioning perovskite single-crystal lasers from laboratory demonstrations to commercial applications in optoelectronics, communications, and integrated photonic circuits.

SignificanceQuantum cascade lasers (QCLs) represent a groundbreaking advancement in semiconductor photonics, leveraging heterostructure band engineering to generate coherent light in the mid- to far-infrared spectral range (3?20 μm). This spectral region, often termed the “molecular fingerprint region,” is of paramount importance due to its association with the vibration-rotation energy transitions of gas molecules and biomolecules. Since their first experimental demonstration in 1994, QCLs have rapidly evolved into one of the cornerstone technologies in infrared photonics, driven by their exceptional performance characteristics. Their key attributes, including broad wavelength tunability, high output power, compact size, and room-temperature operation, collectively position QCLs as indispensable tools for a wide range of applications.In gas sensing, QCLs enable the precise detection of trace gases at parts-per-billion (ppb) or even parts-per-trillion (ppt) levels, leveraging their high spectral resolution and sensitivity. Their ability to operate in the fingerprint region allows for unambiguous identification of specific molecules, making them ideal tools for monitoring pollutants, greenhouse gases, and hazardous substances in industrial and environmental contexts. In medical diagnostics, QCLs facilitate non-invasive detection of biomarkers, such as nitric oxide for asthma or acetone for diabetes, offering real-time, label-free analysis with minimal patient discomfort. These applications underscore the transformative potential of QCLs in addressing global challenges related to health, environment, and industry.However, despite their remarkable capabilities, conventional QCLs have historically faced a significant limitation: high power consumption. This issue has constrained their deployment in portable devices, battery-powered systems, and field-deployable sensors, where energy efficiency is a critical requirement. For instance, in remote environmental monitoring or mobile healthcare diagnostics, the need for low-power-consumption operation is paramount to ensure prolonged functionality without frequent recharging or replacement of power sources. Addressing this challenge is essential not only for expanding the utility of QCLs beyond laboratory settings but also for facilitating their adoption in practical, real-world applications.ProgressIn recent years, significant progress has been made in optimizing the low-power-consumption operation of QCLs through multi-dimensional engineering strategies. These innovations can be roughly divided into two areas: facet engineering and waveguide architecture optimization, both of which have played crucial roles in reducing power consumption while maintaining performance.In the design and fabrication of infrared QCLs, reducing facet-induced mode loss is a vital technical aspect for lowering device power consumption. Utilizing high-reflectivity facet coatings can substantially decrease mirror loss and threshold current density. This effectively reduces the laser’s threshold power consumption and enhances the device’s output optical power and wall-plug efficiency, significantly improving the laser’s overall performance. Another approach to increasing cavity facet reflectivity is adopting a distributed Bragg reflector mirror structure. Additionally, introducing sub-wavelength metallic apertures at the facet can modify the optical field phase. This suppresses diffraction loss while improving modal reflectivity and transmissivity. Compared to that of devices without sub-wavelength metallic apertures, the threshold power consumption is reduced by 25%.Waveguide structure design is another key technical approach to reducing device power consumption. The conventional waveguide structure design of QCLs is often limited by high threshold current density and high optical mode loss, making it difficult to meet the demands of low-power-consumption application scenarios. In recent years, researchers have made significant progress in suppressing non-radiative losses, enhancing optical field confinement, and improving thermal conduction by optimizing the geometric parameters, material systems, and fabrication technologies of waveguides. Among these, buried heterostructure quantum cascade lasers (BH-QCLs) have a significant impact on reducing device power consumption. Firstly, this fabrication process doesn’t require creating an electrical injection window narrower than the ridge, enabling the fabrication of ultra-narrow ridges. The preparation of ultra-narrow ridges helps reduce the active region area, thereby achieving low-power-consumption QCLs. Secondly, the BH structure, by embedding the active region in semi-insulating material, effectively suppresses lateral current spreading, reducing unnecessary losses and thus lowering the device’s total power consumption. Moreover, the BH structure can significantly enhance the device’s heat dissipation capability and reduce thermal resistance. This characteristic alleviates the problem of increased threshold current caused by heat accumulation, thereby reducing power consumption. Research on ridge waveguide quantum cascade lasers (RW-QCLs) is relatively limited. This mainly stems from the following reasons: Firstly, compared to BH-QCLs with waveguide structures fabricated using wet etching, RW-QCLs have greater sidewall roughness, leading to higher scattering losses and increased device threshold power. Secondly, the electrode and insulator layers on the waveguide sidewalls of RW-QCLs will cause waveguide mode absorption, introducing additional optical losses. Lastly, as the active region of BH-QCLs is buried in high thermal conductivity material, compared to RW-QCLs, BH-QCLs can more effectively reduce device thermal resistance, resulting in a lower active region temperature. To avoid epitaxy regrowth, researchers have introduced lateral gratings on the ridge sidewalls as an alternative to the BH structure, achieving low-power-consumption QCLs comparable to the BH-QCLs.Conclusions and ProspectsThis paper reviews the recent developments of low-power-consumption infrared QCLs, emphasising the different schemes for reducing device power consumption by optimising mirror and waveguide losses. Comparing reported device configurations reveals a general strategy: shrink the active region while reducing device losses. Most current research focuses on buried heterojunction structures, which enhance heat dissipation and lower scattering/absorption losses from waveguide sidewalls. However, the epitaxial regrowth of semi-insulating materials increases process complexity and device costs. Traditional ridge waveguides, though avoiding complex epitaxy regrowth, introduce significant waveguide losses.Currently, although the power consumption of infrared quantum cascade lasers has been reduced to the hundred-milliwatt level, their output power is still low and cannot meet the requirements of practical applications. Finding the optimal balance between device power consumption and output power will be the main direction for future development. Novel active region designs and waveguide structures may provide new pathways to address these challenges.Firstly, the experimental work introduced in this paper focuses on reducing the current in the active region, while there has been little study on the relatively high bias voltage required for the operation of the active region. Secondly, whispering-gallery modes can eliminate the impact of mirror losses, but this also leads to lower output power. How to increase their output power will be an interesting direction for future research. Lastly, the introduction of artificial intelligence is bound to shorten the design and optimization cycle of devices. It can efficiently search for the optimal solution in a vast parameter space and may provide entirely new design configurations and ideas for the field. It is foreseeable that high-efficiency, low-power-consumption infrared quantum cascade lasers will bring about tremendous changes in fields such as the low-altitude economy, green research, intelligent industry, and national defense security, and will drive an epoch-making leap in infrared photonic technology.

ObjectiveThe 3?5 μm mid-infrared band coincides with the atmospheric transmission window and contains absorption spectral lines of various gas molecules. It is also the response band of various detectors in photoelectric countermeasures. Mid-infrared band lasers are widely used in fields such as imaging, medicine, aviation, military and other fields. Quartz fiber lasers are currently the fiber lasers with the highest output power and the most mature development. However, due to the high phonon energy of the matrix glass, it has always been difficult for quartz matrix fiber lasers to achieve laser output in the mid-infrared band. Fluoride-based glass has the advantages of low phonon energy, wide transmission window and high solubility of rare earth ions. Therefore, rare earth-doped fluoride fibers have become the mainstream gain medium for the research of wide-band, especially mid-infrared band fiber lasers. Fiber Bragg gratings, as key components of mid-infrared lasers, have the advantages of anti-electromagnetic interference, small size, light weight, suitability for harsh environments, high sensitivity, strong reliability, and low cost. This paper mainly provides a systematic overview of four mainstream fabrication processes, research status and applications of fiber gratings in mid-infrared fiber lasers.Peogress In 2024, T. Boilard and colleagues from Laval University in Canada achieved a 1.7 W 3.92 μm laser output in a commercial InF?-based all-fiber laser system heavily doped with Ho3?. As shown in Figure 12, they used an 888 nm laser diode as the pump source. By constructing an all-fiber system with high-reflectivity fiber Bragg gratings (HR-FBGs) and optimizing the fiber structure to enhance the pump efficiency, they obtained a 1.7 W 3921.5 nm laser output at a pump power of 120 W in a 1.7 m long double-clad indium fluoride-based fiber. HR-FBGs with 98% and 80% reflectivity were inscribed at both ends of the gain fiber. The overall slope efficiency was 9.2%. This experiment also demonstrated that even under conditions of high rare-earth ion doping concentration and significant heat generation from non-radiative transitions, all-fiber laser devices still exhibit high stability.Conclusions and ProspectsAs a key type of fiber grating, fiber Bragg grating (FBG) features small additional loss, small size, and compatibility with optical fibers and other optical devices, making it widely used in optical communication and optical sensing. As a device that can be coupled with optical fibers and compatible with optical devices, FBG can also be widely applied in fiber lasers. In particular, its inherent optical fiber characteristics enable it to integrate with fiber lasers, making the laser structure more compact, loss lower, anti-interference ability stronger, and stability higher. Due to its excellent filtering and reflecting properties, FBG is widely used in optical sensing, fiber lasers, and optical communication systems. However, compared with silica glass, fluoride glass has a more complex ionic composition. During femtosecond laser processing, ion migration occurs in the femtosecond laser irradiation area due to heat accumulation and diffusion. In addition, the non-uniformity of the refractive index modulation (RIM) region in fiber gratings may cause significant broadband loss due to Mie scattering. Therefore, the migration of main gain ions may affect the fiber gain of fluoride fiber gratings in fiber lasers. Scattering in FBG not only reduces its efficiency but also causes energy loss. Thus, how to reduce FBG loss remains an urgent problem to be solved in the future. Furthermore, exploring low-loss splicing techniques for heterogeneous fibers (e.g., splicing AlF?-based glass fibers with InF?-based glass fibers) and fabricating high-reflectivity FBGs using femtosecond laser writing technology will further help to achieve all-fiber mid-infrared fiber lasers, reduce the influence of environmental factors and cavity mirror coupling on laser output, and extend the laser’s working time.

SignificanceSince the invention of the first laser in 1960, lasers have demonstrated an irreplaceable and significant role in many key fields such as communication, medicine, manufacturing, and scientific research. Thanks to their excellent monochromaticity, directionality, and coherence, they have greatly advanced human societal development. As a quantum device that converts input energy into high-coherence light output, a laser is mainly composed of three core elements: the pump source, the gain medium, and the resonant cavity. Among them, the pump source provides energy input for the device; the resonant cavity provides optical feedback and waveform limitation for photons produced by stimulated emission; the gain medium is the core component of the laser, whose function is to convert external pumping energy into highly coherent laser output through the stimulated emission process. As a key medium for optical amplification, the gain medium has physical and optical properties that directly affect the key performance indicators of lasers, including output power, beam quality, energy conversion efficiency, and long-term operational stability. Therefore, in-depth research on gain media is not only a core topic in laser physics, but also a key to promoting the wide application of related technologies in fields such as communication, industry, medical care, and scientific research.With the increasingly urgent demand for devices to develop towards processability, miniaturization, integration, and flexibility, laser technology is also facing technical problems of breaking through the constraints of traditional architectures. Due to the physical limitations of the resonant cavity length and heat dissipation design, traditional lasers face challenges in size reduction and flexible adaptation, which have become key technical obstacles restricting the development of on-chip photonic integration and flexible optoelectronic devices. Against this background, the development of new gain media for miniaturized lasers has become a research hotspot. The rise of solution-processable gain media provides a breakthrough solution for the miniaturization and flexibility of lasers. Compared with traditional gain materials, solution-processable gain media can be directly film-formed through low-cost solution processes such as spin-coating, inkjet printing, or roll-to-roll printing. Their preparation process does not require high temperature, high pressure, or complex lithography technology, which not only greatly simplifies the production process but also significantly reduces the manufacturing cost. Meanwhile, the solution method endows the materials with excellent flexible compatibility, enabling them to adhere to curved substrates or irregular surfaces, laying the foundation for the application of miniaturized lasers in emerging fields such as flexible optoelectronics and implantable medical devices.Carbon dots (CDs) are zero-dimensional nanomaterials with a size less than 10 nm, consisting of a carbon skeleton at the core and a surface shell layer rich in functional groups. As a new type of carbon-based nanomaterials, they feature simple preparation methods, low cost, and wide availability of raw materials—including natural biomass (e.g., leaves and fruit peels)—and can be synthesized through various approaches such as hydrothermal method, microwave method, and laser ablation. Carbon dots not only exhibit low toxicity and excellent biocompatibility, but also have significant advantages in optical properties—including wavelength tunability, photoluminescence quantum efficiency, and photostability. In the process of promoting the sustainable development of miniaturized lasers, the safety and environmental friendliness of the materials are particularly critical. Carbon dots have become ideal gain materials for the development of miniaturized lasers due to their high chemical stability, low toxicity, and good solution processability. The carbon dot laser reported in 2012 triggered research into the use of carbon dots in low-toxicity lasers. Subsequently, researchers significantly improved their performance by optimizing the resonant cavity design and surface modification strategies, laying a solid foundation for their practical application.ProgressThis review comprehensively summarizes the research advancement of carbon dot lasers during the miniaturization process, systematically outlining the challenges and solutions in their development. Firstly, it presents the advantages of carbon dots in solution-processable miniaturized lasers and reviews the current research status of carbon dot lasers (Figs.1 and 2). Subsequently, it discusses the gain mechanisms of carbon dots and analyzes the potential influencing factors on the performance of carbon dot lasers (Fig.3). Next, it summarizes comprehensive optimization strategies ranging from the selection of gain media to the design of resonant cavities (Figs.4?6). Finally, it elaborates on the current application progress of carbon dot lasers (Fig.7) and anticipates their future far-reaching application development as well as the challenges they encounter. The current development objective in this domain centers on preparing carbon dot materials with low threshold, high optical gain, and excellent stability to fulfill the requirements of continuous-wave laser and electrically pumped laser systems. Although the exploration of the gain mechanism of carbon dots has achieved initial progress, the systematic summaries of the mechanism and optimized theoretical models remain insufficient. It is notable that carbon dots retain a distinctive competitiveness in the field of functional materials by virtue of their convenient synthesis process, outstanding biocompatibility, and significantly reduced environmental toxicity.Conclusions and ProspectsCarbon dot lasers have demonstrated unique competitiveness in the field of green optoelectronic devices due to their environmental friendliness and solution processability. With the in-depth research on the mechanism of carbon dot lasers and the greatly expanded application fields, it is expected to promote the leapfrog development of lasers towards low environmental load and sustainable manufacturing.

SignificanceNonlinear optical (NLO) crystals are key materials for expanding the applicable wavelengths of all-solid-state lasers. The development of deep ultraviolet (DUV, wavelength ≤200 nm) NLO crystals has been a highly focused research direction in the fields of optoelectronic materials and laser technology. Based on DUV NLO crystals, DUV solid-state lasers developed through cascaded frequency doubling technology offer advantages such as high photon energy, narrow linewidth, good coherence, tunability, high spectral resolution, and compact structure. These properties make them highly valuable in applications such as laser spectroscopy, condensed matter physics, and chemical reaction dynamics. Currently, practical DUV NLO crystals are still in great shortage. The KBe?BO?F? (KBBF) crystal has long been regarded as the only material that can be practically applied for DUV laser frequency doubling output, highlighting the significant challenges faced in the development of this field. KBBF crystal exhibits a typical honeycomb-like layered structure. In the past decade, significant progress has been made in exploring DUV NLO crystals based on the layered structure. This article introduces a series of representative novel DUV NLO crystals, emphasizing the critical role of the layered structure design strategy in the exploration of DUV NLO crystals, and discusses the challenges and opportunities in further developing this type of crystals.Progress An excellent DUV NLO crystal must achieve a balance among the following three core properties1) a short UV cutoff edge, preferably below 170 nm, to meet the basic transmission requirement for 177.3 nm laser output; 2) a relatively large second-harmonic generation effect, at least comparable to that of KH?PO? (KDP, d36=0.39 pm/V @ 1064 nm); 3) sufficient birefringence (0.07?0.1) to enable phase-matching in the DUV region. In addition, it should possess good crystal growth habit, stable physicochemical properties, and a high laser damage threshold. In the past decade, significant progress has been made in the exploration of DUV NLO crystals based on layered structures. This article takes the layered structure as the core feature and reviews representative crystals discovered in the past ten years that have the potential for direct frequency doubling output of DUV lasers. By classifying the groups in the two-dimensional functional layers into primary functional groups and auxiliary connecting groups, the discussion is expanded from two aspects: intralayer assembly and interlayer connectivity. The crystals are evaluated based on four key properties: second harmonic generation effect, UV cutoff edge, birefringence, and phase-matching cutoff wavelength (λPM). The article begins with a description of the classical KBBF and Sr?Be?B?O? (SBBO) crystal structures and their properties, and then covers the exploration achievements of novel DUV NLO crystals in seven areas: 1) Be-based borates; 2) Al-based borates containing B3O6 groups; 3) rare earth borates containing B3O6 groups; 4) fluorooxoborates; 5) borophosphate complex salts; 6) hydroxyborates; 7) other compound systems, including carbonates, fluorphosphates, organic small molecule crystals, and theoretically predicted structures. Most of the newly reported compounds in the work feature layered structures and exhibit excellent comprehensive properties, showing potential for second harmonic generation of DUV lasers below 200 nm.Conclusions and ProspectsChina has made outstanding achievements in the design and performance optimization of DUV NLO crystals. Scientists in this field have effectively regulated the balance of the second harmonic generation effect, UV cutoff edge, and birefringence through the rational assembly of functional groups, advancing the realization of phase-matching in the DUV region. The layered structure design approach emphasized in this article has played a key guiding role in the exploration of DUV NLO crystals, both deepening and expanding the theory of anionic groups, as well as providing clear directions for future material design. Layered structures have played a crucial role in fully exploiting the potential of excellent functional groups. Therefore, developing new functional groups and layered structures is of great significance in further exploration of novel DUV NLO crystals.The article points out that the development of DUV NLO crystals still faces several challenges1) For high-performance DUV NLO crystals, it is necessary to break through the large crystal growth technology. 2) The difficulty in extending the phase-matching cutoff wavelength of many crystals into the DUV region is mainly due to dispersion issues. Systematically studying the structure-performance relationship of refractive index dispersion to effectively regulate the smaller refractive index dispersion in the DUV region is an important breakthrough. 3) By integrating artificial intelligence (AI) technology with the structure-performance relationship of DUV crystals and first-principles calculations, the research efficiency of DUV NLO crystals can be enhanced. Additionally, the newly proposed additional periodic phase (APP) technology for the phase-matching condition provides new opportunities for crystals that previously struggled to achieve phase-matching in the DUV region.

SignificanceLasers based on stimulated Raman scattering (SRS) are critical means for achieving high-brightness, narrow linewidth, low noise and special wavelength and frequency laser outputs, which play an irreplaceable role in fields such as space exploration, high-energy physics, LiDAR, and optoelectronic countermeasures. Particularly in scenarios requiring precise wavelength control or ultrahigh beam quality, Raman lasers have emerged as a key solution due to their unique spectral manipulation capabilities and superior coherence properties.Due to its superior photothermal properties, diamond exhibits remarkable advantages in nonlinear laser systems. Firstly, diamond possesses the highest thermal conductivity among all natural materials (~2000 W/(m·K) at room temperature), providing excellent thermal management capabilities during high-power laser operation. The superior heat dissipation performance of diamond allows it to withstand pump power densities 1?2 orders of magnitude higher than other Raman-active media, effectively mitigating thermal lensing and optical damage. Secondly, its ultra-broad optical transmission range (from ultraviolet to microwave regime) makes it an ideal medium for full-spectrum laser applications. Most importantly, due to its high atomic density, strong bonding properties and highly symmetric face-centered cubic lattice, diamond demonstrates outstanding nonlinear optical responses. The Raman gain coefficient of diamond reaches approximately 10 cm/GW (@1064 nm), which can maintain high Raman gain while achieving high beam quality output. Moreover, the exceptional chemical stability of diamond makes it highly suitable for extreme environment laser systems. These combined advantages enable diamond-based Raman lasers to serve as next generation high-power, high-brightness laser sources, offering efficient wavelength conversion, outstanding thermal stability and long-term operational reliability.In recent years, the advancement of diamond microwave plasma chemical vapor deposition (MPCVD) preparation technology has promoted the rapid development of diamond lasers. At present, Raman lasers with diamond as the gain medium exhibit excellent performance under high-power continuous and pulsed pumping in the ultraviolet to infrared wavelength range. The development of laser technology relies on high-quality diamond materials as the foundation, and high optical quality is the prerequisite for diamond to achieve laser applications. For Raman laser applications, the optical uniformity of diamond crystals should be better than 1×10-5. However, defects in the crystal can cause local stress and deteriorate the optical uniformity of the crystal. Therefore, reducing the defect density of diamond crystals and improving their optical performance are urgent issues that need to be addressed in diamond lasers at present. It is necessary to summarize the existing optical grade single crystal diamond preparation technology to prospect the future development of this field.ProgressThis article introduces the types and origins of diamond defects and systematically reviews the research progress of diamond defect suppression technology and diamond Raman laser technology in recent years. We summarize the mainstream crystal defect suppression schemes from three stages: pre-growth, growth process and post-growth. Firstly, perform pre-treatment to remove substrate dislocations before growth, including employing high-pressure and high-temperature (HPHT) seed crystals with lower defect density, utilizing diamond ultra-precision machining technology to obtain surfaces with lower roughness and applying hydrogen-oxygen plasma or metal heating to etch dislocations, thereby reducing the impact of substrate dislocations or subsurface damage caused by polishing. Friel et al. achieved diamond epitaxial layers with dislocation density below 400 cm-2 and stress birefringence Δn better than 1×10-5 using Ar/Cl2 inductively coupled plasma etching method (as shown in Figure 3). During the growth process, promoting lateral growth by preparing masks or directly patterning the substrate can effectively suppress the epitaxial growth of dislocations. Professor Zhu Jiaqi’s team from Harbin Institute of Technology employed photolithography technology to fabricate patterned iron array etching pits on the surface of CVD (chemical vapor deposition) single crystal diamond. They observed that the average defect density in the lateral growth region was reduced to approximately one-third of that in the vertical growth region. In addition to lateral epitaxial growth, doping with elements of large atomic radii can also effectively pin dislocations in diamond. Table 1 summarizes the dislocation densities and full width at half maximum of Raman shift before and after diamond growth on various substrates and under different conditions. Moreover, low-pressure and high-temperature (LPHT) or HPHT annealing after growth has been shown to significantly enhance the optical quality of diamond. Hang Yin’s team from the Shanghai Institute of Optics and Fine Mechanics systematically investigated the effects of annealing temperature, annealing time and cooling rate on CVD diamond during HPHT annealing. Furthermore, annealing can be integrated with substrate pretreatment and growth process optimization to improve crystal quality.At present, research on diamond Raman lasers is constantly breaking through, covering various aspects such as high-power output, brightness optimization, multi-wavelength control, single longitudinal mode operation, beam synthesis and ultra-short pulse generation. The continuous wave output power of diamond Raman laser has reached 100 W, the peak pulse power has attained the level of megawatts and the output beam quality approaches the diffraction limit.Conclusions and ProspectsMaterials quality directly determines the performance of lasers. Diamond possesses excellent photothermal properties and nonlinear characteristics, which play a vital role in Raman laser applications. Currently, diamond crystals still face challenges of limited size and poor optical uniformity. Substrate pre-treatment, lateral epitaxial growth or element doping during the growth process and annealing treatment after growth can all effectively improve the optical properties of diamond. Integrating the three treatment stages with crystal dislocation dynamics calculation analysis can elucidate the dislocation mechanisms, achieve precise dislocation control and fundamentally address optical uniformity issues. Beyond process optimization, surface treatment is also critical. Advancing high-precision diamond polishing techniques to improve crystal surface quality and optimizing optical coatings to minimize reflection losses are essential for further performance enhancement of diamond Raman lasers. With the continuous innovation and improvement of preparation and surface treatment technologies, diamond lasers will provide novel light source solutions for important scientific research and defense applications including biological detection, LiDAR, optoelectronic countermeasures and high-energy physics.

SignificanceMetasurfaces—quasi-two-dimensional planar materials composed of artificially designed subwavelength structural units—have garnered significant attention for their capabilities in optical and electromagnetic wave manipulation. By precisely engineering nanostructures, metasurfaces enable flexible control over light properties (amplitude, phase, polarization, and wavefront distribution) at subwavelength scales, thereby overcoming the diffraction limits of conventional optical components. In laser technology, traditional beam control methods rely on bulky optical elements, which increase system complexity and spatial footprint while impeding device miniaturization. In contrast, metasurfaces offer ultra-thin profiles, lightweight characteristics, and high integration density, enabling multifunctional optical control through structural design. These advantages highlight their immense potential for advanced laser beam manipulation.The introduction of metasurfaces provides a promising new approach for laser beam manipulation. Their subwavelength thickness enables extreme miniaturization of beam control devices, facilitating direct integration of beam shaping functionalities on the laser chip. For example, integrating metasurfaces with semiconductor lasers enables beam collimation, focusing, and mode control without additional optical components, substantially reducing system complexity and footprint. In vertical-cavity surface-emitting lasers (VCSELs), whose emission direction is perpendicular to the substrate, metasurfaces are directly fabricated on the VCSEL substrate for monolithic integration. However, for edge-emitting lasers, monolithic integration requires waveguide structures to couple the beam prior to metasurface manipulation, addressing their limited emission facet size. Furthermore, metasurfaces offer unparalleled design freedom, enabling capabilities unattainable with conventional optics—such as high-order vortex beam generation, multi-focal shaping, and dynamically tunable wavefront modulation—thus expanding the functional potential of laser systems.From a fabrication perspective, metasurfaces leverage standard semiconductor planar processes, showing exceptional compatibility with laser chip production. This compatibility facilitates monolithic integration, enabling beam manipulation at the chip level, which enhances system mechanical stability and integration density. Such integration not only streamlines packaging and assembly but also improves reliability and environmental adaptability, rendering metasurface-laser systems particularly suited for size/weight-constrained applications.The application of metasurfaces in laser beam manipulation has been demonstrated across diverse laser systems, including semiconductor lasers, external-cavity semiconductor lasers, and solid-state lasers. In semiconductor lasers, metasurfaces enable critical functions such as single-mode stabilization, on-chip beam shaping, and structured light generation. For external-cavity lasers, they act as high-efficiency feedback elements and mode selectors, facilitating stable output of high-purity, high-power orbital angular momentum (OAM) beams. In solid-state lasers, metasurfaces break the cavity symmetry to precisely manipulate OAM beams. These multifaceted applications not only underscore the functional versatility of metasurfaces but also offer innovative pathways for optimizing performance and expanding the functionality of laser systems. Consequently, a systematic review of recent progress is essential to guide future developments in this field.ProgressRecent years have witnessed the extensive utilization of metasurfaces for optical field manipulation across diverse laser types, including semiconductor lasers, external-cavity semiconductor lasers, and solid-state lasers. Numerous research teams have explored metasurface-based optical control in semiconductor lasers. A collaborative team from Beijing University of Technology and the Institute of Semiconductors, Chinese Academy of Sciences, demonstrated monolithic integration of metasurfaces on 980 nm VCSELs, enabling on-chip beam collimation, deflection, non-diffracting Bessel beam generation (Figs.1?2), and vortex beams (Figs.3?4). In the work reported by Prof. Capasso's team, a metal grating structure was integrated on the cavity facet of the QCL laser to achieve beam collimation and control the laser beam polarization (Fig.9). The team of Professor Luo Yi, Professor Yu Siyuan, and Professor Cai Xinlun proposed a method to monolithically integrate the DFB with a micro-ring-based optical vortex emitter through the waveguide structure (Fig.10), which can realize the vertical emission of the OAM beam. Wen et al. and Dong et al. implemented metasurfaces fabricated on polymer support layers above VCSELs to achieve distinct functionalities: Wen et al. demonstrated circularly polarized beam emission (see Fig.14(a)); Dong et al. realized a vortex microlaser generating tunable topological charges (see Fig.14(c)). Seghilani et al. developed a compact external-cavity surface-emitting semiconductor laser incorporating an intra-cavity all-dielectric metasurface. This system generated high-power, highly coherent Laguerre?Gauss modes with programmable OAM charges (Fig. 17). Metasurfaces have also proven instrumental in solid-state lasers. Prefessor Forbes’ team reported an intracavity metasurface that directly converted linearly polarized light into customizable high-purity OAM states via geometric phase manipulation (Fig.20). R. Chriki et al. demonstrated twisted-beam generation with arbitrary topological charges using geometric phase metasurfaces (Fig.21).Conclusions and ProspectsThis review synthesizes advances in metasurface-enhanced laser systems. Precise optical-field manipulation has improved laser performance, enabling breakthroughs in multi-beam output, polarization control, and structured light generation. With future progress in materials, fabrication, and intelligent algorithms, metasurface-based lasers will play pivotal roles in communications, quantum technologies, biomedicine, and manufacturing—transitioning from labs to industrial deployment in fields like lidar and precision machining.

SignificanceWith the rapid evolution of silicon photonics in applications such as optical communications, sensing, and computing, the demand for efficient, compact, and reliable on-chip light sources continues to grow. Compared to quantum well (QW) gain media, self-assembled semiconductor quantum dots (QDs) offer intrinsically low threshold current densities, superior thermal stability, and ultra-small linewidth enhancement factors. These characteristics make QD lasers particularly well suited for compact, energy-efficient, and temperature-robust photonic integrated circuits (PICs), especially when they are heterogeneously integrated on silicon or other low-index passive platforms.ProgressIn recent years, considerable advances have been made in QD lasers and heterogeneous integration technology. Gaussian-shaped QD gain spectra allow sub-milliampere thresholds and continuous-wave operation up to 220 ℃ due to inhomogeneous broadening, and p-type doping further enhances the high-temperature performance. A near-zero linewidth enhancement factor —an order of magnitude lower than those of typical QWs—contributes to narrower linewidths and improved tolerance to feedback, enabling isolator-free operation.Various types of heterogeneously integrated QD lasers have been demonstrateddistributed feedback (DFB) lasers achieve linewidths down to 26 kHz, side-mode suppression ratios (SMSR) exceeding 60 dB, and a 3 dB modulation bandwidth of 13 GHz [Figs. 6(a)‒(c)]; tunable lasers based on ring resonators using the Vernier effect provide tuning ranges up to 47 nm while maintaining SMSRs over 50 dB [Figs. 6(d)‒(f)]; comb lasers show the ability to generate flat optical frequency combs and exhibit low-noise radio-frequency spectra [Figs. 6(g) and (h)], supporting high-speed multi-channel data transmission with a low bit-error-ratio.Beyond silicon, wafer-scale integration with low-index materials such as silicon nitride (SiN), tantalum pentoxide (Ta2O5, tantala), and lithium niobate (LiNbO3) has been demonstrated through multi-layer bonding, intermediate coupler, and narrow n-contact taper coupling (Fig. 7), combining QD gain with ultralow-loss or strong electro-optic waveguides.Conclusions and ProspectsHeterogeneous integration of QD lasers combines the advantages of quantum dot gain media with the scalability of silicon photonics. Progress in this area has led to the demonstration of devices with ultralow threshold currents, high-temperature operation capabilities, sub-megahertz linewidths, broad wavelength tunability, and high-speed modulation performance.Despite these advancements, several critical challenges remain. Managing thermal expansion mismatch at the bonded interfaces remains a key issue affecting device reliability during post-bonding processing. Furthermore, the heterogeneous integration of quantum dot lasers with low-refractive-index passive platforms—such as silicon nitride, silicon carbide (SiC), and lithium niobate, requires further investigation and optimization. Additionally, leveraging mature silicon photonics platforms to achieve large-scale high-density optoelectronic integration without compromising yield continues to be a central goal. Overcoming these obstacles is essential to advancing the scalability and performance of heterogeneously integrated quantum dot laser systems.With continued advances in material growth, bonding techniques, and integration design, heterogeneously integrated quantum dot lasers are poised to play a key role in integrated silicon photonic systems. These developments will accelerate the transition of on-chip light sources from laboratory research to real-world deployment, enabling transformative progress in large scale data interconnects, energy-efficient optical computing, and precision sensing.

SignificanceUltrashort pulse lasers, characterized by pulse durations of less than 1 ps and peak power densities exceeding 109 W/cm2, are indispensable in various domains, including attosecond science, laser fusion, ultrafast photonics, precision machining, and biomedicine, due to their ultrafast temporal characteristics. Nonetheless, accurately measuring their fundamental parameters, such as pulse duration and phase, presents considerable challenges. Traditional electronic measurement techniques are limited by their picosecond-level time resolution, rendering them inadequate for assessing sub-picosecond pulses. Nonlinear detection methods, which utilize optical autocorrelation to translate time-domain information into spatial or spectral domains via frequency doubling or summation effects in materials, have emerged as pivotal in overcoming these measurement limitations. The efficacy of nonlinear optical materials, which serve as the fundamental media for this technology, is critical, as their properties including phase matching capability, response bandwidth, and signal strength directly influence detection accuracy, sensitivity, and applicability. Consequently, the advancement of high-performance bulk optical materials is of paramount importance for fostering progress in ultrafast science and its industrial applications.Progress This paper provides an overview of the nonlinear optical response characteristics of two kinds of prevalent bulk optical materials utilized in the detection of ultrashort laser pulsescrystalline and glass materials. It also highlights recent advancements in the domain of ultrashort laser pulse detection. In the initial section concerning nonlinear crystal materials, it is noted that crystals have emerged as the predominant choice for early ultrashort laser pulse detection, attributed to their superior nonlinear coefficients and the well-established phase-matching techniques. This category is primarily divided into two types: single-domain crystals and multi-domain crystals. Single-domain crystals facilitate efficient frequency doubling via birefringence phase matching, however, they necessitate precise mechanical angle adjustments. O’Shea et al. demonstrated the use of a 1 mm thick BaB2O4 (BBO) crystal in conjunction with angular scanning to achieve synchronous time-domain and phase measurements, achieving an accuracy comparable to that of 100 μm thick KH2PO4 (KDP), albeit with limited anti-vibration capabilities. Multi-domain crystals encompass periodically polarized crystals, which enhance conversion efficiency through quasi-phase matching (QPM) but exhibit a relatively narrow bandwidth. Miao et al. introduced a design for an aperiodic domain structure that significantly broadens the bandwidth by a factor of 100, enabling high-precision pulse measurements at ultra-low energy levels ranging from 124 aJ to 9.5 fJ, with sensitivity that is eight orders of magnitude greater than those of single-domain crystals. Additionally, random multi-domain crystals, such as strontium barium niobate (SBN), possess one-dimensional ordered and two-dimensional disordered needle-like ferroelectric domain structures, which provide an infinite array of inverted lattice vectors to mitigate phase mismatch. Fischer and colleagues leveraged the SBN transverse second harmonic generation (TSHG) effect to achieve real-time measurements of individual pulses (ranging from 30 fs to 200 fs) without mechanical scanning, covering a dynamic range from 30 fs to 1 ps and spanning wavelengths from 800 nm to 2200 nm. The subsequent section addresses nonlinear glass materials. Despite their isotropic nature, second-order nonlinearity can be induced through the disruption of local symmetry. In recent years, these materials have garnered significant attention due to their broad response characteristics. The microcrystalline glass composite strategy primarily involves the formation of randomly oriented microcrystals through a crystal-glass composite, facilitating wide-band and wide-angle responses via random quasi-phase matching (RQPM). Researchers in this field have manipulated the nonlinear response of materials through two principal approaches to enable the measurement of ultrashort laser pulses. The first approach involves entropy engineering to control grain size. Feng et al. successfully synthesized high-crystallinity nanocrystalline composite glass (NIG) by adjusting the entropy value of niobium-silicate glass, resulting in a grain size of less than 10 nm in the high-entropy system. This study marked the first observation of a lateral frequency doubling signal under a weak scattering background, achieving pulse measurement within the 870?1300 nm range (with an error margin of less than 10 fs), while also eliminating the need for mechanical alignment. The second enhancement pertains to the mixed base effect, as demonstrated by Lin et al., who incorporated large-radius alkali metal ions (K?, Rb?, Cs?) to mitigate lattice defects in LiNbO3. This modification resulted in a doubling of the frequency doubling intensity and a blue shift of the absorption edge to 356 nm. Utilizing the NGC-4Cs sample, the measurement bandwidth was extended to 780?820 nm for the first time, thereby encompassing the operational range of the titanium-sapphire laser. Furthermore, Yan et al. employed the Na? mixing effect to facilitate real-time monitoring of pulse distortion phenomena, such as splitting and forward tilt, achieving a sensitivity of 600 fs delay and a tilt angle of 10°.Conclusions and ProspectsCurrently, both nonlinear crystals and glass ceramics possess distinct advantages in the domain of ultrashort pulse detection. Nonlinear crystals are characterized by their high precision, however, their manufacturing processes are intricate and their operational bandwidth is constrained. Conversely, microcrystalline glass offers a broad bandwidth response and does not necessitate mechanical alignment, although there is a need for enhancement in signal strength. Future research endeavors should concentrate on several key areas. Firstly, optimization of material design is essential. This involves the precise control of internal grain size, distribution, and defect concentration within microcrystalline glass to balance scattering loss with nonlinear enhancement effectively. Additionally, the exploration of novel natural multi-domain crystals or artificial metamaterials is crucial for expanding the phase-matching bandwidth. Secondly, the integration and innovation of technology should be prioritized. This includes the implementation of lateral frequency doubling to develop a real-time single-pulse detection system, as well as the incorporation of machine learning algorithms to enhance inversion efficiency. Furthermore, promoting the utilization of microcrystalline glass in integrated photonic chips will facilitate the advancement of miniaturized and multifunctional detection platforms. In summary, the ongoing innovation in nonlinear optical materials is anticipated to propel ultrashort pulse detection towards higher precision, integration, and multifunctionality, thereby presenting new opportunities for ultrafast science and its applications.

SignificanceWith the rapid development of emerging applications such as artificial intelligence (AI), autonomous driving, mobile internet, and high-speed computing, the information transmission capacity and processing speed required by human society are increasing exponentially. As a result, ever-growing requirements are placed on improving the transmission capacity of optical fiber communication systems. Traditional methods of increasing the transmission capacity of the communication system, such as boosting the transmission rate of a single channel or reducing the channel spacing, have already approached the theoretical limits defined by Shannon. Further attempts to improve transmission using the above approaches lead to the effects like four-wave mixing and increased crosstalk between channels, which negatively impact system stability. Therefore, expanding the gain bandwidth of current optical fiber communication systems has become an urgent problem to be solved.However, traditional rare-earth-doped optical fibers are limited by the inherent bandgap widths of rare-earth ions, making it difficult to meet the growing demand for data transmission rates and expanded bandwidth. Rare-earth-doped optical fibers can typically only achieve amplification in the C-band (1530?1565 nm) and part of the S-band (1460?1530 nm) or L-band (1565?1625 nm). In addition, rare-earth-ion-doped fiber lasers and amplifiers also result in a large portion of the low-loss area of quartz optical fibers not being effectively utilized. Although erbium-doped fiber amplifiers (EDFAs) are widely used in communication networks due to their high gain and low noise, their amplification range is still limited to the C+L band (1530?1620 nm), leaving a significant portion of the low-loss window of quartz fibers (1100?1700 nm) underutilized. Thus, developing ultra-wideband gain media capable of bridging the gaps in traditional rare-earth-doped fibers is essential for enhancing the transmission bandwidth of optical fiber communication systems.In the past few years, various techniques have been developed to fabricate bismuth-doped glass fibers, mainly including the modified chemical vapor deposition (MCVD) approach, rod-in-tube approach, molten-core approach, and phase separation approach. Especially, these studies have focused on the luminescence properties of bismuth-doped glasses and optical fibers with various compositions, which exhibit excellent broadband emission due to the ability of bismuth ions to adopt multiple valence states in various matrices. Currently, broadband optical amplification covering the range of 1100?1800 nm has been successfully achieved using different matrix materials such as aluminosilicate (1150?1300 nm), phosphosilicate (1300?1450 nm), silicate (1400?1500 nm), and high-germanium silicate glasses (1600?1800 nm). Hence, bismuth-doped glass fibers have been demonstrated to effectively compensate for the emission deficiencies of rare-earth-doped fibers in the near-infrared wavelength range, emerging as a research hotspot for novel gain media to expand communication capacity.ProgressWe elaborate on the research progress of bismuth-doped glass fibers in the entire 1100?1800 nm band. First, the luminescence mechanism of bismuth-doped glasses and their broadband luminescence characteristics are introduced. On this basis, the fabrication methods of bismuth-doped glass optical fibers are explained. Besides, the latest research progress, device performance, and unique advantages of using bismuth-doped optical fibers to build various types of optical fiber amplifiers and lasers are summarized. Finally, we extensively analyze the current challenges and issues in this field and provide an outlook on future development directions.Conclusions and ProspectsBismuth-doped glass optical fibers with outstanding broadband characteristics and long-lived near-infrared emission hold great promise for the development of high-efficiency, broadband, and tunable optical communication devices, including amplifiers and lasers. However, significant scientific and technical problems still need to be further explored to promote the development of the new laser gain medium in both academic and engineering aspects. For example, optimizing the fabrication method of bismuth-doped glass optical fibers to solve the contradiction between high bismuth active center concentration and low unsaturated loss. Additionally, to broaden the near-infrared luminescence spectra of bismuth-doped glass optical fibers, the co-doping mechanism of bismuth and other rare-earth ions should be explored. If these challenges can be overcome in the near future, the broadband amplification and luminescence properties of bismuth-doped fiber amplifiers and lasers will be significantly enhanced, offering substantial economic and societal benefits.

SignificanceMetal halide perovskites (MHPs) exhibit high absorption, tunable bandgap, and high carrier mobility, making them promising for next-generation photovoltaics, light-emitting diodes (LEDs), and coherent light sources. However, the toxicity of lead with persistent environmental effects severely limits commercial viability and sustainability. Recently lead-free metal halide perovskites (LFMHPs) have attracted growing attention as environmentally friendly alternatives with compelling opto-electronic characteristics. In addition to ecological benefits, LFMHPs exhibit diverse structural and photophysical properties, which provide new approaches in coherent light generation. Their compatibility with solution-based synthesis, soft lattice characteristics, and defect tolerance extend applications for miniaturized, tunable, and flexible photonic devices, particularly in the field of semiconductor lasers. With the growth of global demand for lead-free compatible technology, LFMHPs stand as a promising platform to meet the dual goals of performance and sustainability in photonic applications.Progress The attempt to develop LFMHPs has mainly followed two substitution strategiesisovalent cation substitution (e.g., Sn2+, Ge2+) and heterovalent cation substitution (e.g., Ag+/Bi3+ or Cu+/Sb3+ combinations). These approaches derive a series of perovskite structures, ranging from traditional 3D ABX3 lattices to low-dimensional and double perovskite configurations such as A2B(I)B(III)X6 and A3B(III)2X9 (Fig.3). These diverse structural motifs not only influence band structure and stability, but also govern exciton dynamics and emission mechanisms. Among isovalent systems, Sn-based perovskites (e.g., ASnX3) show potentiality due to their direct bandgaps, tunable PL spectrum from visible to near-infrared regions, and high carrier mobility (Fig.4). Their lasing performance has been demonstrated in various cavities, for instance, (PEA)2SnI4 and CsSnBr3 single crystals have exhibited low-threshold lasing and high Q-factor. To mitigate issues such as Sn2+ oxidation and lattice degradation, recent studies have introduced molecular additives and low-dimensional structures that enhance photostability and emission performance. The substitution of Pb2? ions with heterovalent cations introduces disparities in electronic configuration and oxidation state, which profoundly affect the crystal structure, bandgap, and electronic state distribution of the host lattice. In such systems, excitons are strongly localized by both lattice defects and the presence of heterovalent dopants, often leading to the formation of self-trapped excitons (STEs). STEs arise when excitons become localized due to strong exciton-phonon coupling, resulting in emission energies significantly lower than the material’s bandgap. The localization typically manifests as a pronounced Stokes shift, increased exciton binding energy, and extended photoluminescence lifetimes. These characteristics can be effectively investigated through temperature-dependent photoluminescence spectroscopy and transient absorption measurements. Materials such as Cs2AgNaInCl6 and Cs3Cu2X5 exhibit broad emission spectra, large Stokes shifts, and high PLQYs (photoluminescence quantum yields) up to 80% (Fig.5).The photophysical characteristics of LFMHPs render them suitable as gain media for various laser configurations.A wide variety of laser architectures such as whispering gallery mode (WGM), Fabry?Pérot (FP), distributed feedback (DFB), vertical cavity surface emitting lasers (VCSELs), and random lasers (RL) have been realized using different LFMHP compositions and nanostructures. In 2023, Lin and colleagues developed Cs3Cu2I5 thin films via dual-source co-evaporation and integrated them within a vertical cavity formed by two distributed Bragg reflectors (DBRs), achieving lasing at 440 nm under femtosecond excitation with a threshold of 1.12 μJ/cm2. By incorporating an additional gain layer and DBR, dual-wavelength lasing at 458.8 nm and 505.6 nm was realized with a threshold of 5.62 μJ/cm2. Concurrently, Petrozza's group patterned dielectric gratings using electron-beam lithography and reactive ion etching, followed by spin-coating a ~110 nm-thick PEA2SnI4 perovskite layer. At 77 K, the resulting DFB laser demonstrated a narrow linewidth (0.9 nm) and low threshold energy density (19 μJ/cm2), highlighting the potential of LFMHPs in high-coherence light generation. Alternative cavity strategies include bio-inspired platforms: in 2016, Sum and co-workers employed natural photonic crystal structures derived from butterfly wings as optical resonators. Using CsSnI3 films with 20% SnF2 additive to suppress Sn2+ oxidation, they achieved single-mode near-infrared lasing with a threshold energy density of ~15 μJ/cm2, demonstrating compatibility with non-planar and flexible substrates. Further extending this approach, Lee’s team synthesized CsSnI3 quantum dots and embedded them in a cholesteric liquid crystal (CLC) resonator to construct a tunable DFB laser. The device exhibited ultralow threshold energy (0.15 μJ per pulse), narrow linewidth (0.20 nm), and wavelength tunability from 582 to 606 nm. Impressively, it retained ~87% of its initial efficiency after six months of storage under ambient temperature and high humidity, underscoring the environmental resilience of optimized LFMHP-based laser architectures (Table 1).Conclusions & Prospects LFMHPs have progressed from eco-friendly alternatives to a distinct class of photonic semiconductors, characterized by exceptional compositional versatility and tunable optoelectronic properties. Their demonstrated compatibility with multiple lasing architectures combined with low thresholds, wide spectral tunability, and defect-tolerant nature, position them as strong candidates for future coherent light sources. Nonetheless, achieving practical deployment demands further progress in enhancing environmental and operational stability, mitigating defect-induced nonradiative losses, and deepening the understanding of exciton?lattice coupling. Advances in material design, surface passivation, microcavity integration, and theoretical modeling will be essential to overcome current limitations. LFMHP-based lasers hold strong prospects for enabling high-performance and sustainable photonic technologies.

SignificanceUpconversion (UC) laser has the characteristics of anti-Stokes displacement, monochromatism, and high stability. Hence, rare earth (RE) ion-doped upconversion micro/nano laser has shown potential applications in various fields, such as biomedicine, holographic projection, visible light communication, data storage, and new-generation display technologies, resulting in extensive attention in recent years. The optical feedback from photon scattering of the porous upconversion nanoparticles clusters has been reported to produce upconversion random lasers. Light bouncing back and forth between two reflective surfaces or internal surfaces has been utilized to achieve modulated upconversion lasing emission. In addition, plasmonic cavities with enhanced electromagnetic fields can amplify the upconversion process within the sub-diffraction-limiated volumes and produce highly efficient upconversion lasers. In this review, the recent advances in RE ions-doped upconversion materials for random, Fabry?Perot (F?P)/whispering gallery mode (WGM) cavity-, and plasmonic cavity-modulated upconversion micro/nano lasers are overviewed. The main factors affecting the output of upconversion micro/nano lasers are summarized. Current challenges and future directions of the upconversion micro/nano lasers are also discussed.ProgressFirst, upconversion nanoparticles (UCNPs) can be employed as cavities for feedback and resonance based on their own scattering effect to produce random lasing emission. For example, a type of high-temperature-operated compact self-cooling laser has been demonstrated using Ba2LaF7∶Yb3+, Er3+ nanocrystals (NCs)-embedded glass-ceramics. Additionally, by using core-shell UCNPs as the gain medium, a highly efficient single-segment white random laser and tunable random lasing emission from 309 to 363 nm have also been achieved.Unlike random lasers induced by the scattering effect, F?P cavities and whispering gallery modes (WGMs) are commonly employed as micro-resonator geometries for upconversion lasing emission owing to their high quality factors. F?P microcavities, constructed with two or more parallel mirrors, allow light to bounce back and forth between the reflective surfaces. Resonance occurs when the optical path length equals an integer multiple of the light wavelength. Based on this principle, upconversion lasing emission has been achieved by designing an F?P cavity—consisting of a quartz tube sandwiched between a distributed Bragg reflector and an Al mirror—with a NaYF4∶Yb3+, Er3+@NaYF? core-shell NC solution as the gain medium. However, the relatively long cavity length makes it challenging to obtain a finely structured lasing spectrum.WGM microcavities can confine light in a narrow ring along the equatorial surface of the cavity via total internal reflection. WGM cavities, which have various shapes such as bottles, spheres, toroids, and rings, have been further explored and demonstrated as one of the optimal cavities for microlasers with low thresholds and narrow linewidths, owing to their high quality factors and small mode volumes. A bottle-like WGM microcavity enables upconverted blue, green, red, and deep ultraviolet lasing emission by coating a drop of silica resin containing UCNPs onto an optical fiber. To reduce the size of upconversion microlasers, a single hexagonal NaYF4∶Yb3+, Tm3+, Er3+ microrod was utilized to achieve multicolor upconversion lasing emission, supported by total internal reflection between its six flat surfaces. After depositing efficient energy-looping NCs onto the surface of a polystyrene microsphere, low-threshold upconversion lasing emission pumped by continuous wave was achieved. Furthermore, WGM microcavities fabricated from RE ion-doped glass ceramics exhibited upconversion lasing emission with extremely low thresholds, reaching the microwatt level.One strategy to lower the threshold of lanthanide upconverting lasers is to enhance the upconversion efficiency of the gain medium. Plasmonic structures with locally enhanced electromagnetic fields can amplify the upconversion process within sub-diffraction-limited volumes. As typical plasmonic materials, silver and gold nanoparticles with tailorable resonance modes matching upconversion excitation or emission wavelengths have been shown to enhance upconversion efficiency. Accordingly, plasmonic arrays consisting of silver nanopillars were fabricated to provide a high-quality single-mode lattice plasmon with a sharp resonance peak. After depositing NaYF4∶20%Yb3+, 20%Er3+@NaYF4 NPs onto the surface of the plasmonic nanoarray to form a microcavity, continuous-wave upconversion lasing with an ultralow threshold of 70 W/cm2 was achieved.A typical laser device consists of three crucial componentsa gain medium, a pumping source, and an optical cavity. These are also the main factors affecting the output of upconversion lasers. To achieve lasing emission, the gain medium must satisfy the requirement of population inversion, meaning that the population in the excited states should exceed that in the ground state. Furthermore, to realize net gain after each feedback cycle, the number of emitted photons in the gain medium must exceed the losses induced by scattering or re-absorption. Increasing the pumping power leads to successive amplified spontaneous emission and stimulated emission (lasing action). For the optical cavity, it must have a high quality factor, i.e., the optical loss should be low.Conclusions and ProspectsThis review summarizes recent progress in RE ion-doped upconversion micro/nano lasers. The excellent frequency conversion properties of upconversion nanoparticles provide numerous opportunities for upconversion lasing, such as the upconversion random lasing, WGM/F?P cavity-modulated upconversion lasing, and plasmonic cavity-based upconversion lasing. Moreover, the emission wavelength of upconversion microlasers has been expanded from the deep ultraviolet to the near-infrared region. Low-threshold continuous-wave-pumped upconversion lasing has been achieved through the design of microcavities with WGM and lattice plasmon modes, showing potential applications in solid-state holographic display, underwater monitoring, high-speed information transmission, bioimaging and tracking, and water purification. However, challenges remain. For instance: (1) Further exploration is needed to improve upconversion efficiency, thereby reducing the lasing threshold; (2) It is necessary to enhance the thermal stability of the gain medium to avoid the influence of thermal effects on laser performance under high pumping power; (3) Smart cavities with high-Q-factor WGMs, F?P cavities, and plasmonic enhancement should be further developed for various applications.

SignificanceGallium nitride based ultraviolet laser diodes (GaN-based UV LDs) have great application potential in fields such as biomedicine and optical communications due to their wide direct band gap, high stability, and wavelength tunability (Fig. 1). This paper reviews the development history and current situation of GaN-based UV LDs, and briefly introduces the structure of ultraviolet A (UVA) LDs (Fig. 2). Meanwhile, this paper analyzes the technical bottlenecks from the aspects of material growth, process preparation, packaging, and reliability. One of the important challenges of GaN-based lasers is to obtain high quality epitaxial materials, mainly from aspects such as solving the lattice mismatch between AlGaN and the substrate, the defect density of epitaxial materials, and the doping efficiency in p-AlGaN. In terms of device manufacturing processes, key processes such as etching, passivation, and ohmic contact also have significant influences on the photoelectric characteristics of GaN-based UV LDs. Finally, developing new packaging materials that are resistant to ultraviolet rays and have high thermal conductivity, optimizing the heat sink structures, and improving the quality of optical coatings on the cavity surface are the keys to enhancing the lifespan and power of devices. To break through the current technological bottlenecks, the future development of GaN-based UV LDs will revolve around multiple dimensions such as material optimization, new structure design, and system integration, to promote the development of GaN-based UV LDs towards high power, long life, and high reliability.ProgressIn this paper, we review recent progress in GaN-based UV LDs, addressing critical challenges in materials, device fabrication, and reliability. Firstly, for material development, stress management in high-Al-content AlGaN remains a key focus, typically addressed through lateral epitaxy or patterned GaN templates (Fig. 3). Defect reduction is another major hurdle, and our studies on carbon impurities in AlGaN reveal that increasing carbon concentrations raises resistivity in Mg-doped GaN due to compensation effects, a phenomenon mitigated by optimizing the ratio of the growth source flow rate of group V elements to that of group III elements (Fig. 4). The p-type doping efficiency can be enhanced through growth condition adjustments or superlattice structures (Fig. 5). Device etching, passivation, and ohmic contact processes significantly impact laser performance. The high contact resistance of p-AlGaN layers often requires complex metal stacks like Ni/Au or Pd/Ru combined with high-temperature annealing to achieve stable low-resistance. Thirdly, reliability presents one unique challenge under UV operation. High-energy photons induce photoaging, while elevated current densities accelerate degradation. Our work demonstrates that UV-excited interactions between ionized air (producing OH?) and ambient dust form SiO? deposits on cavity (Fig. 6). Concurrently, current stress exacerbates material defects, underscoring the importance of robust optical resonator design and packaging strategies for device longevity.Conclusions and ProspectsTo overcome current technological bottlenecks, the future development of GaN-based UV LDs will focus on material optimization, novel structural designs, and system integration. In device structural innovation, vertical-cavity surface-emitting lasers (VCSELs) emerge as a key direction for UV lasers due to their low threshold current, narrow beam divergence, and ease of 2D integration. High-performance UV VCSELs can be achieved by integrating high-reflectivity AlGaN-based or dielectric distributed Bragg reflectors (DBRs) while optimizing the optical confinement in the active region. Meanwhile, distributed feedback (DFB) lasers enable single-mode and narrow-linewidth emission through grating structures, meeting the demands of high-precision spectroscopy and optical communications. Additionally, emerging approaches such as micro-cavity lasers (e.g., photonic crystal lasers) and topological insulator lasers may offer new pathways to enhance UV laser performance. For short-wavelength expansion, the realization of deep-UV (UVC) lasers depends on breakthroughs in high-Al-content AlGaN or AlN materials. Techniques such as AlN template technology, nanostructured substrates, and strain engineering can improve the epitaxial quality of these wide-bandgap semiconductors.

ObjectiveMid-infrared (3‒4 μm) lasers are critical for applications such as free-space communication, infrared countermeasures, and gas sensing, demanding concurrent high power and beam quality. Interband cascade lasers (ICLs) based on antimonide superlattices combine high gain with low power consumption, offering significant advantages in this spectral region. However, conventional ICLs suffer from low vertical thermal conductivity due to their 2000+ layer superlattice structure, limiting active stages (<10) and forcing narrow ridge widths (<6 μm) to maintain fundamental-mode operation. This restricts the output power to tens of milliwatts, insufficient for long-range applications. Widening the ridge to boost the power inevitably excites higher-order transverse modes, degrading the beam quality. This work addresses this trade-off by integrating multimode interference (MMI) couplers into ICLs. The MMI structure selectively suppresses odd-order modes via self-imaging effects, enabling wide-ridge designs without compromising beam quality, representing a crucial advancement for power-intensive mid-infrared applications.MethodsWe design and fabricate Fabry-Pérot (FP) and MMI-ICL devices with three ridge widths of 8, 11, and 14 μm (Fig. 1). The symmetric 1×1 MMI structure with a width of 36 μm and length of 1356 μm features the input/output waveguides centering on the MMI region and connected via 34° tapers to minimize reflection losses. The MMI length (LMMI) is calculated using the self-imaging principle, targeting the first-order position for TE0/TE2 modes enhancement and TE1 mode suppression. COMSOL Multiphysics simulations are performed to analyze the modal losses as a function of ridge width for TE0, TE1, and TE2 modes (Fig. 2).A 7-stage ICL epitaxial structure is grown on n-GaSb via molecular beam epitaxy (MBE). Ridge waveguides and MMI couplers are patterned by contact lithography and wet etching with a H3PO4-based solution. The devices feature SiO2 passivation, Ti/Pt/Au top contacts, and Ge/Au/Ni/Au back contacts. Cleaved 3-mm-long cavities are mounted on copper heat sinks. The beam profiles (Fig. 4) are measured at 3× threshold current using a Pyrocam IV beam analyzer. Power-current-voltage curves (Fig. 6) are recorded under continuous wave (CW) operation.Results and DiscussionsFor FP lasers, increasing the ridge width from 8 μm to 14 μm results in a progressive degradation of beam quality as shown in Figs. 4(a)‒(c): the 8-μm device exhibits a single-lobe TE0 profile, the 11-μm device shows a dual-lobe TE1-dominated pattern, and the 14-μm device displays a triple-lobe TE0/TE1/TE2 mixed mode. In contrast, the output remains a single-lobe TE0-dominant profile across all ridge widths [Figs. 4(d)‒(f)] for the MMI-ICL samples, demonstrating the MMI effectiveness in suppressing higher-order modes.In addition, the MMI-ICLs achieve near-diffraction-limited performance as shown in Fig. 5. The MMI-ICL with a 11-μm ridge shows a divergence angle of 26°, corresponding to a beam quality factor (M2) of 1.23 (1.1× diffraction limit), compared to 38° for the FP counterpart. For the 14-μm device, the MMI-ICL reduces divergence angle from 41° to 18° (M2=1.57), achieving near-diffraction-limited performance. Weak side lobes observed in the 14-μm MMI-ICL [Fig. 3(b)] are attributed to the minor TE2 contributions, consistent with the simulated 96% transmission of this mode through the MMI section.The MMI-ICLs demonstrate an enhanced optical power without increasing threshold performance (Fig. 6). The enlarged active area in the MMI section boosts the output power across all ridge widths compared to those of the FP lasers. This improvement highlights the MMI ability to address the core trade-off between beam quality and power in wide-ridge semiconductor lasers.ConclusionsThis work demonstrates a breakthrough in high-power and high-beam-quality ICLs via integrated MMI couplers. MMI-ICLs with 14-μm ridge widths maintain a single-lobe output, outperforming the conventional FP designs and existing sidewall-grating technologies. The 11-μm MMI-ICL achieves the M2 factor of 1.23 (26° divergence angle), representing a significant advancement toward diffraction-limited performance. The MMI-ICL platform resolves the critical power-beam quality trade-off, enabling next-generation mid-infrared sources for long-range sensing and communication. Future work will optimize MMI designs and explore phase-locked arrays to further enhance output power while maintaining beam coherence.

ObjectiveThe optoelectronic devices based on type-II quantum wells (QWs) and superlattices (SLs) have important applications in the field of mid-infrared (mid-IR) technology. Type-II QWs and SLs based on III-V semiconductors are a type of artificial microstructure materials designed through band engineering. It can be formed by alternating growth of InAs, GaSb, AlSb, and their multicomponent alloys to form periodic structures. The type-II QWs and SLs can be grown using the molecular beam epitaxy (MBE) technique, which produces a uniform material over a large area. The corresponding bandgap covers the infrared spectrum of 2.7‒30 μm, offering low Auger recombination efficiency and high quantum efficiency. The optoelectronic devices based on type-II QWs and SLs, such as interband cascade lasers, quantum cascade lasers, infrared detectors, and solar cells, have important applications in military and civilian fields, including infrared detection, industrial detection, gas sensing, healthcare. However, the type-II QWs and SLs used in mid-infrared lasers and detector devices usually contain thousands of heterointerfaces that are susceptible to element mixing and diffusion. It can lead to lattice distortion and stress generation. Therefore, achieving high-quality epitaxial growth in type-II QWs and SLs faces significant challenges.MethodsType-II QWs and SLs are epitaxially grown by a solid-source MBE system, which is equipped with valved arsenic and antimony crackers. The crystalline surface quality of the as-grown layer is monitored in-situ by reflection high energy electron diffraction (RHEED) in real time. Interband cascade laser(ICL) devices are fabricated from the two wafers by wet etching and contact photolithography. They are mounted epi-side up on copper heat sinks with uncoated facets for laser testing. The Fourier transform infrared spectrometer is used for recording the lasing spectra. In order to reduce the interfacial mixing of type-II QWs and SLs materials and balance the strain, it is particularly important to calibrate the growth rate and lattice matching. Here, the interfacial strain is controlled by interface engineering in the MBE growth, resulting in perfectly lattice-matched type-II QWs and SLs. Together with the rigorous calibration on the growth rates and doping concentrations, high-quality ICL structures composed of a large number of type-II QWs and SLs grown on InAs (001) substrates are successfully demonstrated.Results and DiscussionsFirst, the crystal lattice matching, growth rate and doping concentration of type-II QWs and SLs are calibrated. The lattice matching of type-II QWs and SLs is calibrated by inserting an AlAs interface. As shown in Fig. 1, the cascade superlattice in the interband cascade laser achieves good lattice matching with the InAs substrate. The growth rates of InAs, GaSb, and AlSb are calibrated using RHEED oscillation, and the carrier doping concentration is calibrated using the Hall effect. Then, we apply the growth and calibrated parameters to the interband cascade laser structure. As shown in Fig. 3, the interband cascade laser structure is fully strained to the InAs substrate, reducing the defects generated during film relaxation. The as-grown surface defects density is as low as 102/cm2. The atomic force microscope reveals a series of clear, regular atomic terraces. These atomic terraces are about 0.33 nm high, which corresponds to one monolayer InAs, and the root mean square roughness in the 3 µm×3 µm scanning area is only 0.13 nm. To see the heterointerfaces more clearly, cross-sectional scanning transmission electron microscope (STEM) shows the atomic-level flatness of type-II QWs and SLs interfaces (including InAs/AlSb and InAs/GaInSb). Finally, the high quality interband cascade laser materials are fabricated into laser devices. Figure 4 shows that we obtain a series of high performance interband cascade lasers. At the wavelengths of 4 μm and 8 μm, the highest operating temperature and the low threshold current density are achieved in our works.ConclusionsIn this work, the interface engineering is used to precisely control interface strain, resulting in the type-II QWs and SLs with perfect lattice matching. By using molecular beam epitaxy technology, high-quality interband cascade laser structures are epitaxially grown on the InAs (001) substrates, with growth parameters strictly calibrated such as growth temperature, growth rate, and doping concentration. The surface measured by atomic force microscope yields a root-mean-square roughness of 130 pm on an ICL structure with thickness more than 8 μm. Meanwhile, the defect density under optical microscope is in the level as low as 102 cm-2. The cross-sectional STEM image clearly shows the sharp hetero-interfaces of type-II SLs. Ultimately, breakthroughs on the device performance are demonstrated for InAs-based ICLs. For example, the shortest lasing wavelength of 4.02 μm in pulsed operation with the lowest threshold current density of 232 A/cm2 is achieved at 300 K. And the highest operating temperature of 275 K is also achieved in an ICL lasing in the long wavelength infrared region (8.22 μm). These results demonstrate that improving the materials quality of type-II superlattices can enhance the performance of infrared optoelectronic devices.

ObjectiveDual-wavelength laser sources have attracted considerable attention for various applications in optical communications, interferometry, microwave generation, and terahertz wave generation. Square microcavity lasers with directly connected output waveguides, as compact single-cavity devices, have been demonstrated to simultaneously emit the 0th and 1st order transverse modes. Deformed square microlasers with circular sides are designed to enlarge the mode spacing. However, the mode spacing exhibits high sensitivity to circular-side deformation, imposing stringent requirements on fabrication precision. In this paper, to achieve widely tunable dual-mode spacing in microcavity lasers and relax the fabrication tolerance, deformation modulation in square microlasers is proposed and demonstrated.MethodsA deformed square microcavity laser with two opposite circular vertices and a waveguide connected to an undeformed vertex is proposed and demonstrated to realize dual-mode lasing (Fig. 1). The characteristics of the passive-cavity transverse-electric (TE) modes are investigated by the two-dimensional finite element method with the commercial software. The devices are fabricated using projection lithography and inductively coupled plasma etching techniques. Microlasers are mounted on a thermoelectric cooler, and the output signals are coupled into a tapered fiber from the cleavage facet of the output waveguide and are measured by an optical spectrum analyzer with a resolution of 0.02 nm. To demonstrate the potential applications, the autocorrelation curves of the dual-mode lasers with different mode spacings are measured by an autocorrelator, confirming their potential application in sub-THz wave generation. Additionally, a microlaser with dual-mode spacing of 0.82 nm is used as a pump source in an optical frequency comb generation system.Results and DiscussionsSimulation results (Fig. 2) reveal that circular vertices can enlarge the dual-mode wavelength interval between the 0th and 1st order transverse modes. The deformed square microcavities with side length of a=30 μm (circular vertex radius of δ=0.4?20 μm), a=26 μm (δ=0.4?16 μm), and a=20 μm (δ=0.4?12 μm) can theoretically achieve the dual-mode wavelength spacings of 0.22?1.76 nm, 0.38?1.81 nm, and 0.73?2.71 nm, respectively. The simulated magnetic field distributions of the 0th and 1st order transverse modes (Fig. 3) illustrate that the 1st order mode is more effectively regulated by the circular vertices than the 0th mode, thus adjustable dual-mode intervals can be realized by changing the circular vertex radius. Experimental results (Fig. 6) show that the device with a=30 μm exhibits excellent dual-mode lasing characteristics. Adjustable dual-mode intervals from 0.46 nm to 1.61 nm are realized when the circular vertex radius increases from 6 μm to 18 μm. The microlasers exhibit pure dual-mode spectra, and the side mode suppression ratios exceed 20 dB. The deformed microlasers with a=30 μm and δ=12, 17 μm are selected for autocorrelation curve measurements (Fig. 7), and the corresponding autocorrelation curves exhibit approximately sinusoidal waveforms with periods of 9.30 ps and 6.07 ps, corresponding to dual-mode frequency spacings of 108 GHz and 165 GHz, respectively, showing their potential application in sub-THz wave generation. Additionally, dual-mode microlasers are used to generate optical frequency combs based on the dual-pump scheme (Fig. 8), and a 0.82-nm-spaced comb with 90 comb lines is generated successfully.ConclusionsDeformed square microcavity lasers with two circular vertices are designed and fabricated, emitting dual modes with a widely tunable mode spacing. The circular vertex deformation is employed to modulate the wavelengths and Q factors of the 0th and the 1st order transverse modes. As the deformation increases, the wavelength of the 1st order transverse mode blueshifts faster than that of the 0th order transverse mode, leading to an enlarged dual-mode spacing. For the microcavity laser with a side length of 30 μm and an output waveguide width of 1.5 μm, as the circular vertex radius increases from 6 μm to 18 μm, the dual-mode spacing expands from 0.46 nm to 1.61 nm. Additionally, the circular vertex deformation disrupts the mode field near the vertices, thereby suppressing higher-order transverse modes. These deformed microcavity lasers feature pure dual-mode lasing spectra with side-mode suppression ratios exceeding 20 dB. In a single deformed microcavity laser, the dual-mode spacing hardly varies with the current. It is expected that the dual-mode spacing in a single microcavity laser can be adjusted via current injection control enabled by a patterned electrical injection window. This compact deformed square microcavity laser exhibits excellent process reproducibility and dual-mode lasing stability, which has potential applications in sub-THz wave generation and optical frequency comb generation based on a dual-pump scheme.

ObjectiveBlue-green lasers hold broad application prospects in fields such as underwater laser communication, laser display, and high-density optical storage. In particular, the 486 nm laser corresponds to the Fraunhofer dark line (H-β line) in the light-transmitting window of seawater, which can effectively suppress the influence of solar background light noise. With continuous breakthroughs in semiconductor laser technology, it has become a new choice for blue-green light sources. To achieve blue-green emissions, the first approach is to directly realize diode lasers based on the gallium nitride gain system. However, restricted by device structure, conventional edge-emitting and surface-emitting semiconductor lasers have encountered bottlenecks in achieving high powers and small divergence angles. By comparison, a semiconductor disk laser (SDL) inherits both the flexible external cavity structure of solid-state lasers and the flexible bandgap design of semiconductor lasers, which is expected to become a new solution for high-brightness blue-green lasers. However, previous researches have mainly focused on the improvement of continuous-wave (CW) power, with relatively few reports on the high-energy pulse characteristics required for underwater laser communication.MethodsFigure 1 presents the designed chip structure. The InGaP and Al0.3Ga0.7As layers serve as an etch-stop layer and a window layer, respectively. A five-periods 8-nm-thick In0.17Ga0.83As/GaAs double quantum well structure is employed to provide optical gain at the target wavelength, and Fig. 2 shows the simulated gain spectrum. A tensile-strained GaAs0.94P0.06 layer is grown after each quantum well (QW) period for stress compensation. A 26-pair AlAs/GaAs distributed Bragg reflector centering at 980 nm is then integrated. The designed chip is then grown on a 2° off-cut GaAs substrate via metal-organic chemical vapor deposition (MOCVD). To optimize heat dissipation, a high-thermal-conductivity diamond or high-purity copper is employed as a transition heat sink. The semiconductor chip and the diamond heat sink are pre-metalized with a titanium-platinum-gold (Ti/Pt/Au) alloy before soldering. The substrate is subsequently removed using wet etching, resulting in a gain chip thickness of approximately 5 μm.Results and DiscussionsWith the heat sink temperature maintained at 288 K, the 972 nm gain chip is evaluated in a V-shaped resonant cavity, and an output power of 15 W is attained (Fig. 5). The beam quality factors Mx2 and My2 in the horizontal and vertical directions at the maximum output power are approximately 1.91 and 1.39, respectively. The calculated brightness reaches 595.4 MW/(cm2·sr). Following the evaluation of the gain chip, a blue-green emission with a CW power of 4.27 W and a slope efficiency of 18.8% is realized by intracavity frequency doubling (Fig. 6). The blue-green emission exhibits a center wavelength of approximately 486 nm with a spectral width of ~1 nm, positioning it near the Fraunhofer absorption line in solar radiation. The Mx2 and My2 are approximately 4.07 and 2.77, respectively, yielding a calculated brightness of 160 MW/(cm2·sr). By controlling the pump, the operating characteristics under quasi-continuous waves are further investigated. At a duty cycle of approximately 8%, a peak power of 14.9 W, corresponding to a pulse energy of 1.32 mJ, is further attained (Fig. 7).ConclusionsBlue-green lasers hold broad application prospects in fields such as underwater laser communication, laser display, and high-density optical storage. Here, we design a high-performance semiconductor disk laser chip based on an InGaAs multi-quantum well-active region structure. In a V- shaped resonant cavity, a continuous-wave power of approximately 15 W at 972 nm is achieved, with a calculated brightness of 595.4 MW/(cm2·sr). By further integrating an intracavity nonlinear frequency conversion unit, the semiconductor disk laser can emit a 4.27 W, 486 nm blue-green emission and the brightness reaches 160 MW/(cm2·sr). Additionally, a pump modulation mechanism is employed, and a 486 nm blue-green laser with a microsecond pulse duration and a kilohertz repetition rate is realized. The peak power and single-pulse energy reaches 14.9 W and 1.32 mJ, respectively.

ObjectiveThe quantum cascade laser (QCL), an infrared semiconductor laser based on unipolar carrier transitions, operates via electron inter-subband transitions within a quantum cascade structure. Owing to its unique emission coverage across the mid- to far-infrared spectrum, this technology demonstrates broad application potential in trace gas detection, free-space optical communication, and infrared countermeasures. Molecular beam epitaxy (MBE) and metal organic chemical vapor deposition (MOCVD) are two primary growth techniques for QCL fabrication. MBE has long been the preferred method for QCL growth due to its superior precision in parameter control and interfacial abruptness. In 2020, Wang et al. used MBE to fabricate a mid-wave QCL device that achieved a room-temperature continuous-wave output power of 5.6 W, a wall-plug efficiency of 22%, and a lasing wavelength of 4.9 μm, representing state-of-the-art performance. In contrast, MOCVD, despite having slightly lower precision in growth parameter control and interfacial abruptness, offers advantages in cost-effectiveness and growth efficiency, making it a promising technique for large-scale QCL production. In 2018, Botez et al. at the University of Wisconsin employed MOCVD to develop a mid-wave QCL device with a room-temperature continuous-wave output power of 2.6 W, a wall-plug efficiency of 12%, and a lasing wavelength of 5 μm. Although MOCVD-based QCLs currently underperform MBE-based ones, they have significant room for improvement and are expected to narrow the gap with MBE through optimized material growth and device design in the future, promoting widespread QCL applications. Moreover, several research groups have enhanced optical power by laser beam combining to overcome single chip power limitations.MethodsWe successfully grew the full QCL epitaxial structure via MOCVD. Devices of various specifications were then fabricated from the epitaxial wafer. Systematic investigations were conducted on how ridge width and cavity length affect device performance. For a device with a 10 mm cavity length and 12 μm ridge width, the peak output power attained 6.3 W. In addition, spatial beam combining technology was utilized to integrate QCL chips (6 μm ridge width, 5 mm cavity length) into QCL modules M7 (7 chips) and M19 (19 chips). The current?voltage?power (I?V?P) and spectral characteristics of these modules were rigorously evaluated. For M7, further temperature-dependent and lifetime tests were also executed.Results and DiscussionsAt 288 K, the device with a 5 mm cavity length and 6 μm ridge width showed a threshold current of 0.28 A, a threshold current density of 0.93 kA/cm2, a maximum optical power of 2.2 W at 1.1 A, and maximum wall-plug efficiency of 16.4% at 0.75 A (Fig. 7). When the cavity length increased to 7.5 mm and 10 mm, the threshold current rose to 0.39 A and 0.60 A. The maximum optical power correspondingly increased to 2.9 W and 4.0 W, with the maximum wall-plug efficiency slightly decreasing to 15.5% and 15.3%. With a fixed 10 mm cavity length, varying the ridge width from 6 μm to 12 μm caused the threshold current, maximum optical power, and maximum wall-plug efficiency to change accordingly (Fig. 8). At 293 K and 0.8 A, the M7 module delivered 9.3 W of optical power with 12.2% wall-plug efficiency. The inset lasing spectrum revealed a peak wavelength of 4.6 μm. Despite decreasing with rising temperature, at 347 K, the M7 module still maintained over 3 W of optical power and over 5% wall-plug efficiency (Fig. 9). All three M7 modules exhibited an optical power exceeding 9.0 W, with no optical power degradation during the lifetime test, and their failure-free times were 2045, 963, and 761 h (Fig. 10). Operating at 1.0 A, the M19 module achieved the highest reported optical power of 22.4 W for a mid-infrared QCL module. However, its wall-plug efficiency decreased compared to the M7 module, likely due to thermal effects (Fig. 11).ConclusionsWe fabricated QCL devices and modules with a lasing wavelength of 4.6 μm by using MOCVD. The device with a 6 μm ridge width and 10 mm cavity length had a room temperature continuous power of 4.0 W and a maximum wall-plug efficiency of 15.3%. Another device with a 12 μm ridge width and 10 mm cavity length achieved a room temperature continuous power of 6.3 W and a maximum wall-plug efficiency of 15.3%, setting a world-record output power for single ridge devices. Using chips with a 6 μm ridge width and 5 mm cavity length, we made M7 modules (7 chips) and M19 modules (19 chips). At 0.8 A, the M7 module had a continuous output power of 9.3 W and a wall-plug efficiency of 12.2%. The M19 module, at 1.0 A, achieved a continuous output power of 22.4 W and a wall-plug efficiency of 7.8%. The M19 module’s output power represents the highest level yet reported.

ObjectiveOrthogonally dual-polarization lasers have significant applications in fields such as interferometry, precision measurement, and terahertz wave generation. They have become a research hotspot in solid-state laser technology. An increasing number of researches is focusing on generating a stable dual-polarization laser. The Yb∶Ca3Gd2(BO3)4 (Yb∶CGB) crystal, which has an orthorhombic crystal structure, exhibits anisotropy in both absorption and emission spectra. Due to the comparable polarized emission cross-sections of Yb∶CGB crystal along a, b and c axes, it is highly convenient to achieve a dual-polarization laser using this crystal. Additionally, the polarization state of the laser can also be switched by adjusting the pump power or using output couplers with different transmissivity values. The polarization properties of the lasers based on Yb∶CGB crystals are crucial for the generation of an orthogonal dual-polarization laser and hold great research significance.MethodsTo investigate the anisotropy in continuous-wave (CW) laser action of Yb∶CGB crystals, samples are prepared by cutting along the a-, b-, and c-crystallographic axes. The dimensions of samples are 3 mm×3 mm×6 mm, with the end faces (3 mm×3 mm) polished but not coated with antireflection layers. A compact plane-concave resonator is employed to study the laser performances of the Yb∶CGB crystal. A plane mirror, serving as the input mirror, is coated for high transmittance (>98%) at 808?980 nm and high reflectance (>99.9%) at 1020?1200 nm. A concave mirror with a radius of curvature of 25 mm is used as the output coupler, whose transmittance can be chosen from T=0.5% to T=20%. The length of the cavity is approximately 23 mm. To remove the heat generated during the laser operation, the crystal sample is wrapped by indium foil and mounted in a water-cooled copper holder maintained at 12 ℃ and placed close to the input mirror in the cavity. The pump source is a fiber-coupled diode laser emitting unpolarized radiation at 975 nm (bandwidth of less than 0.5 nm). The fiber core diameter and the numerical aperture are 105 μm and 0.22, respectively. Using a fiber output focusing lens, the focused pump beam spot is reimaged into the crystal. The laser output power is measured using a dynamometer, while the spectrum is recorded with a spectrometer. A Glan prism is employed to distinguish lasers with different polarization states.Results and DiscussionsThe continuous wave laser performances of a-cut, b-cut, and c-cut Yb∶CGB crystals are systematically studied. The most efficient laser action is achieved under the output coupling of T=5%. Under this condition, the maximum output powers are 7.07, 6.20, and 6.61 W for the a-cut, b-cut and c-cut crystals, respectively. The laser spectra for different output couplers are also measured and shown in Fig. 1(b), Fig. 3(b) and Fig. 5(b). It is clear that the laser wavelength shifts towards a shorter wavelength as the transmittance of the output coupler increases.With a Glan prism, the polarization states of the laser are carefully investigated. For the a-cut Yb∶CGB crystal, with an output coupler (T=0.5%), the laser exhibits an orthogonal dual-polarization state over the entire pump power range. Using the output couplers with T=3% and T=5%, the laser maintains a single-polarization state (E∥c) when Pin is less than 10.54 W and 13.5 W, respectively. Once the incident pump power exceeds these values, the laser with an E∥b polarization state starts to oscillate. The changes in output power for different polarization states as incident pump power increases are shown in Fig. 2. When the output coupling transmittance is relatively large, such as T=10%, 15%, and 20%, the output laser maintains its linear polarization state of E∥c over the entire pump power range. Similar observations regarding the changes in laser polarization states are made for the b-cut and c-cut crystals, with the specific details shown in Fig. 4 and Fig. 6.ConclusionsEfficient continuous-wave laser operations are demonstrated with a-, b-, and c-cut Yb∶CGB crystal samples, producing the maximum output power of 7.1 W, and the optical-to-optical efficiency is 24% with respect to the incident pump power. Through changing the output coupler utilized, the polarization state of the laser can be successfully modulated. What’s more, with an output coupler (T=0.5%), an orthogonal dual-polarization laser is achieved over the entire pump range. All the results indicate that the Yb∶CGB crystal is suitable for generating orthogonal dual-polarization lasers, which can be used in precision metrology and precision metrology.

ObjectiveMid-infrared (3?5 μm) lasers play a pivotal role in gas sensing and free-space communication due to their alignment with molecular absorption lines and low atmospheric scattering. However, conventional second-order distributed feedback (DFB) surface-emitting lasers suffer from poor far-field symmetry, characterized by a dual-lobe emission pattern, which severely compromises the beam quality. This work addresses this critical challenge by proposing a hybrid second- and fourth-order DFB grating design for interband cascade lasers (ICLs). The innovation lies in the fundamental optimization of the optical mode distribution to achieve single-lobe far-field emission, thereby enhancing beam collimation and radiation efficiency. This advancement is essential for practical applications requiring a high beam quality, such as in portable gas sensors and long-range optical communication systems.MethodsThe ICL structure is epitaxially grown on an n-type GaSb substrate using molecular beam epitaxy. A hybrid grating combining the second-order (period of 0.95 μm) and fourth-order components is fabricated via electron-beam lithography and dry etching techniques (Fig. 4). COMSOL Multiphysics simulations are conducted to analyze the electric field distributions and the coupling coefficients under symmetric and antisymmetric modes (Figs. 1 and 3). Key structural parameters, such as a grating duty cycle of 0.3, are optimized to balance coupling strength and fabrication feasibility. The devices are processed into 4.5-μm-wide ridge waveguides, with Si3N4/SiO2 passivation and Ti/Pt/Au metallization. Performance metrics, including spectral purity, power-current-voltage (P-I-V) characteristics, and far-field patterns, are systematically evaluated under continuous-wave (CW) operation at 25 °C.Results and DiscussionsThe hybrid grating ICL demonstrates stable single-mode lasing with a side-mode suppression ratio (SMSR) of 30 dB, a significant improvement over the 20 dB SMSR of conventional second-order DFB devices (Fig. 5). The enhanced spectral purity stems from the higher coupling coefficient of hybrid grating, which suppresses mode hopping.The hybrid grating enables the surface-emitting power comparable to the edge-emitting power, achieving a surface radiation efficiency exceeding 50% (Fig. 6). In contrast, the uniform second-order gratings exhibit only a 25% efficiency. Despite a slight threshold current increase due to additional waveguide losses, the hybrid design maintains a competitive output power.One of the most significant achievements of this work is the elimination of the dual-lobe far-field pattern characteristic of conventional devices. The hybrid grating ICL produces a symmetric single-lobe profile with a divergence angle of 0.7° (Fig. 7), representing a remarkable 61% reduction compared to the 1.8° divergence angle of the second-order DFB lasers. Simulations confirm that the hybrid grating redistributes the electric field maxima, reducing antisymmetric mode losses and enabling single-lobe emission (Fig. 3).ConclusionsThis work demonstrates a notable advancement in mid-infrared laser technology by integrating hybrid second- and fourth-order DFB gratings into ICLs. The successful achievement of single-lobe far-field emission with a low divergence angle of 0.7°, enhanced spectral purity (30 dB SMSR), and a high surface emission radiation efficiency (>50%) paves the way for high-beam-quality mid-infrared sources. The hybrid grating design overcomes the inherent limitations of traditional DFB lasers, offering a robust solution for applications in gas sensing, lidar, and free-space communication, where precise beam control is essential. Future research will focus on scaling output power while maintaining beam quality, further enhancing the industrial applicability of the proposed technique.

ObjectiveDual-wavelength external cavity quantum cascade lasers (EC-QCLs) hold significant potential in free-space communication, simultaneous detection of multi-component trace gases, and difference-frequency terahertz (THz) generation. However, their high-performance stable output has long been hindered by challenges such as mode competition and power insufficiency. This study aims to address these limitations by optimizing optical configuration and feedback control, thereby achieving a dual-wavelength EC-QCL with high power, broad tunability, and excellent single-mode characteristics. The proposed design provides a robust and reliable light source for advanced applications requiring precise spectral control and multi-wavelength operation.MethodsA double blazed grating Littrow external cavity system is designed in the experiment, and dual-wavelength output is achieved based on a Fabry-Perot (FP) quantum cascade laser (QCL) chip with a central wavelength of 8.6 μm. The QCL adopts the design of a bound-to-continuum active region structure. To improve thermal stability and power efficiency, the device optimizes thermal management through secondary epitaxy of high thermal conductivity InP∶Fe material and a diamond-copper composite heat dissipation system. The optical path of the system is shown in Fig. 1. The light beam is divided into two paths by a beam splitter, which are coupled to two independent gratings (with a groove density of 150 line/mm) respectively, forming a dual feedback channel. A polarizer is innovatively introduced to regulate the intensity of the optical path feedback. Based on Malus law, the reflected light needs to pass through the polarizer twice, and the total feedback light intensity is attenuated to cos4θ. This mechanism dynamically balances the gain competition between the dual-wavelength modes and prevents the disappearance of the dual-wavelength phenomenon caused by mode competition. In the experiment, the axial position of one grating is fixed, and the angle of the other grating is dynamically adjusted to achieve flexible combined output of a fixed wavelength and a continuously tunable wavelength.Results and DiscussionsUnder continuous-wave (CW) operation at 293 K, the FP-QCL achieves a total dual-facet power of 1.89 W (wallplug efficiency of 11.91%) with high thermal stability (T0=140 K, T1=246 K). In pulsed mode (50 kHz repetition rate, 2 μs pulse width), the electroluminescence spectrum shows a 33.5 meV full width at half-maximum , and the far-field profiles (5.5 mm size in slow-axis direction and 3.2 mm size in fast-axis direction) confirm the fundamental transverse mode operation (Fig. 2). A single-grating configuration achieves a tuning range of 8.10?8.97 μm (1234.7?1114.4 cm-1) with >0.8 W output (Fig. 3). Dual-grating operation enables two wavelength combinations: fixed 8.82 μm with tunable 8.33?8.74 μm (Fig. 4) and fixed 8.56 μm with tunable 8.33?8.82 μm (Fig. 5). Both configurations maintain single-mode operation with a total power of >0.7 W, surpassing that of the existing dual-wavelength EC-QCLs. FTIR analysis confirms the spectral independence and the tunable power distribution. The polarization feedback balances net gain, while the broad-gain active region and low-loss cavity enable a high-power performance.ConclusionsAn external cavity quantum cascade laser modulated by a double blazed grating is constructed, which can achieve arbitrary dual-wavelength output within a tunable range. In the experiment, the wavelengths of 8.82 μm and 8.56 μm are fixed, and by rotating another grating, the dual-wavelength outputs of arbitrary combinations of 8.82 μm with tunable 8.33?8.74 μm and 8.56 μm with tunable 8.33?8.82 μm are obtained. The dual-wavelength outputs are independent of each other, and the total power is greater than 0.7 W. Compared with those of QCLs in the same wavelength band, the power performance is significantly improved, and the good single-mode characteristics are maintained. This also confirms that the external cavity regulation technology can realize the dual-wavelength output of a single-tube laser, which is of great significance for improving the accuracy of gas detection and multi-gas detection. Meanwhile, it also has potential application prospects in generating difference-frequency terahertz signals.

ObjectiveLiDAR is a crucial component that is indispensable in fields such as advanced automotive assisted driving, unmanned aerial vehicle (UAV) mapping, and intelligent sensing. With the continuous expansion of its market, higher requirements have been put forward for laser chips. Currently, the LiDAR systems usually adopt discrete multi-junction edge-emitting diode lasers, which face issues such as a large divergence angle, poor beam quality, and a high wavelength temperature drift coefficient. It is difficult to achieve lateral mode control and longitudinal wavelength locking through traditional technologies, resulting in low long-distance resolution and poor signal-to-noise ratio of LiDAR. Therefore, obtaining laser chips with high power, low divergence angle, and high beam quality has become the key.MethodsIn order to achieve high-power and low-divergence-angle diode lasers, this paper uses an asymmetric double-sided Bragg reflection waveguide to expand the near-field distribution of the vertical mode, achieving stable single-mode operation with a large optical mode size. Experimentally, a single-junction diode laser with high power and nearly-circular beam laser output is designed, which improves the vertical divergence angle by reducing the refractive index and the thickness of the defect layer. On this basis, two multi-junction Bragg reflection waveguide laser structures are designed. Multiple active regions share the same waveguide mode, achieving narrow single-beam and dual-beam laser outputs, respectively.Results and DiscussionsThe optoelectronic characteristics of the single-junction lasers are measured under different operation conditions. 70 μm and 140 μm stripe-width lasers show continuous powers of 6.4 W and 8.7 W. The full width at half maximum (FWHM) of the vertical divergence angle is 7.3°, and the vertical divergence angle containing 95% of the power is approximately 15.8°, which is much lower than that of traditional edge-emitting diode lasers. To improve the heat dissipation of the laser, lasers with a large stripe width are fabricated, and the power measured under quasi-continuous operation conditions exceeds 42 W. High power, ultra-low divergence angle and nearly circular beam output is achieved from such a 905 nm Bragg reflection waveguide laser. Two types of 905 nm multi-junction Bragg reflection waveguide lasers are fabricated, which show a pulse peak power exceeding 59 W. The measured vertical far-field distributions display completely different profiles due to different waveguide mechanisms, which show narrow single-beam and twin-beam outputs.ConclusionsIn this study,an asymmetric double-sided Bragg reflection waveguide epitaxial structure is adopted, which utilizes the photonic bandgap effect to confine the optical field. Through modifying the optical defect layer to suppress the local optical waveguide effect, the optical mode size in the vertical direction is significantly extended. The vertical divergence angle is obviously reduced, which is only 15.8° with 95% power content. Such a high-power, nearly-circular laser output is beneficial for simplifying the optical system and reducing application costs. To further enhance the peak power of the laser, this paper develops two types of multi-junction Bragg reflection waveguide lasers with different waveguide mechanisms. Multiple active regions share the same waveguide mode, and the active regions are cascaded through tunnel junctions. By introducing tunnel junctions at the position of the optical field node near the active regions, the optical absorption loss can be reduced. This structure can be integrated with a grating or an advanced waveguide structure to achieve the control of longitudinal and lateral modes, thereby obtaining laser output with stable wavelength and high brightness.

ObjectiveHigh-power 2?3 μm mid-infrared (MIR) femtosecond lasers have emerged as indispensable tools in advanced scientific and industrial applications, including time-resolved spectroscopy, environmental trace-gas detection, and label-free biomedical imaging, owing to their unique operation in the molecular fingerprint region. Traditionally, such laser sources are generated through optical parametric oscillation (OPO) or difference-frequency generation (DFG) techniques. However, these approaches suffer from several intrinsic limitations, such as relatively low conversion efficiency, complex system architecture, and stringent phase-matching requirements. To achieve higher output power levels, further amplification of the oscillator output becomes necessary, but this introduces a critical challenge: the finite gain bandwidth of the amplifier medium leads to the significant gain-narrowing effect, which not only broadens the pulse duration but also degrades the overall system performance. While spectral pre-shaping techniques can partially address this issue by carefully tailoring the seed spectrum into a saddle-shaped profile prior to amplification, which mitigates gain saturation near the central wavelength, these methods inevitably increase optical complexity and incur substantial energy losses. Therefore, the development of a high-power MIR femtosecond oscillator capable of directly generating a saddle-shaped spectrum represents a significant breakthrough in this field.MethodsIn this work, we demonstrate a Kerr-lens mode-locked oscillator based on polycrystalline Cr∶ZnS through optimizing cavity design and intracavity dispersion control. A 1908 nm fiber laser serves as the pump source, delivering up to 25 W output power with a 6.0 mm beam diameter and a 0.6 mrad divergence angle. The pump beam is focused into the polycrystalline Cr∶ZnS gain medium (dimensions of 2 mm×2 mm×7 mm; Cr2+ doping concentration of 4×10-19 cm-3) using a 100 mm focal-length plano-convex lens. The crystal is wrapped in indium foil and mounted in a copper holder with active temperature stabilization, maintained at 18 ℃ via a semiconductor thermoelectric cooler (TEC) coupled to a water-cooled copper block heat sink. The femtosecond oscillator employs an asymmetric X-folded cavity with a total length of 2.113 m, comprising two curved high reflectors (R1 and R2, radius of 100 mm, high transmittance @1.6?1.9 μm, high reflection @2.0?2.7 μm), a flat-end high reflector (M1, high reflection @2.0?2.7 μm), and a flat output coupler on a CaF? substrate (OC, partial reflectivity of 50%@2.0?2.7 μm). Using standard ABCD matrix analysis, the beam waist radius inside the crystal is estimated at 32.5 μm, while the pump beam waist radius is measured at 30 μm, ensuring optimized mode overlap for soft-aperture Kerr-lens mode-locking. The end mirror (TM1) is mounted on a translation stage, and mode-locked operation is initiated by manually displacing TM1 to induce cavity perturbation.Dispersion compensation is achieved through three chirped mirrors (CM1?CM3, providing -250 fs2 dispersion compensation) and two third-order dispersion mirrors (TM1?TM2, delivering -3000 fs3 dispersion compensation ). This configuration enables precise broadband group delay dispersion (GDD) compensation across the 2.1?2.6 μm spectral range, which is crucial for maintaining stable femtosecond pulse oscillations.Results and DiscussionsAt the incident pump power of 5.85 W, Kerr-lens mode-locking is achieved with an output power of 1.44 W, corresponding to an optical-to-optical conversion efficiency of 24.6%. The mode-locked spectrum spans 2.1?2.6 μm. Pulse characterization is performed via second-harmonic-generation frequency-resolved optical gating (SHG-FROG). The FROG trace displays normalized intensity, while the retrieved pulse profile exhibits a normalized intensity inversion error below 0.004. The reconstructed temporal intensity yields a pulse duration of 46 fs. The retrieved spectral intensity closely matches the experimental spectrum, validating the pulse retrieval fidelity. By combining the pulse repetition rate, average power, output coupler transmittance, and laser beam waist radius, the calculated B-integral within the cavity is 4.97. This substantial nonlinear phase shift indicates strong Kerr nonlinearity, where self-phase modulation (SPM) continuously generates new spectral components. The resulting spectral reshaping suppresses the central intensity while amplifying the sidebands, ultimately forming a saddle-shaped spectrum. The spectrum shows a fundamental frequency signal-to-noise ratio (SNR) exceeding 60 dB at 70.99 MHz. Beam quality factors are measured as Mx2=1.04 and My2=1.12. Power stability monitoring over 1 h reveals an average output of 1.44 W and the root mean square value of the average power instability of 0.27%, confirming robust long-term operation.ConclusionsWe demonstrate a Kerr-lens mode-locked laser based on polycrystalline Cr∶ZnS, achieving stable femtosecond operation with a saddle-shaped spectrum. At 5.85 W pump power, the oscillator delivers a 46 fs pulse with a 211.9 nm bandwidth and a 1.44 W average power. This distinctive spectral profile makes the source particularly valuable for broadband mid-infrared generation and ultrashort pulse amplification technologies.

ObjectiveWith the explosive growth of cloud computing, artificial intelligence, and data centers, there is an increasing demand for compact, low-power, and high-bandwidth optical interconnect devices. The 1060 nm vertical-cavity surface-emitting laser (VCSEL) offers a favorable trade-off between fiber dispersion, attenuation, and low-cost packaging compatibility. However, achieving ultra-high-speed modulation while maintaining thermal stability and manufacturability remains a significant challenge. This study proposed a novel 1060 nm VCSEL design based on a composite double oxide aperture structure, aiming to achieve a modulation bandwidth exceeding 30 GHz, a low threshold current, enhanced single-mode stability, and improved process compatibility and scalability.MethodsThe VCSEL structure was grown on GaAs substrates by metal-organic chemical vapor deposition (MOCVD), incorporating 21 pairs of p-type and 32 pairs of n-type GaAs/AlGaAs distributed Bragg reflectors (DBRs). Graded AlGaAs layers were introduced within the DBRs to reduce heterointerface resistance. The active region consisted of five-period strain-compensated InGaAs/GaAsP multiple quantum wells (MQWs), designed to enhance differential gain and increase the relaxation oscillation frequency. A composite elliptical oxide aperture with a major axis of 3.9 μm and a minor axis of 2.4 μm was formed by sequential wet thermal oxidation of Al0.98Ga0.02As layers. The oxidation process was monitored in real-time using infrared microscopy, enabling precise lateral oxidation control. This dual-aperture structure significantly enhanced both current and optical confinement. Crosslight PICS3D simulations were employed to optimize the overlap between the standing-wave field and MQW position, thereby maximizing gain efficiency. Material gain spectra from 300 K to 360 K were simulated to analyze the peak gain degradation and wavelength redshift behavior. An equivalent circuit model was developed to extract parasitic resistances and capacitances for accurate modeling of dynamic modulation characteristics (Fig. 6).Results and DiscussionsSimulations show strong optical confinement within the oxide aperture. The material gain peak decreases from 4167 cm-1 to 3331 cm-1 as temperature increases from 300 K to 360 K, with a wavelength redshift rate of approximately 0.42 nm/K, indicating excellent thermal stability. The optical mode and gain region overlap exceeds 85%, ensuring efficient photon generation. Electro-optical measurements show a low threshold current of 0.25 mA and a slope efficiency of 0.32 W/A. Under a bias current of 5.5 mA at 25 ℃, the device achieves a small-signal 3 dB modulation bandwidth of 33.5 GHz, with a K-factor of 129.83 ps and a D-factor of 13.77 GHz/(mA)1/2. Parasitic parameters extracted from the equivalent circuit model reveal that the bandwidth is predominantly governed by intrinsic carrier-photon dynamics, with minor external limitations. Notably, the device maintains stable dual-transverse-mode operation across a wide current range from 0.9 mA to 4.95 mA, effectively suppressing spatial hole burning and intermodal competition. Compared with recent reports on 1060 nm VCSELs, the proposed device achieves comparable or even higher bandwidth without the need for complex ion implantation or surface etching, offering a simplified and fabrication-friendly design.ConclusionsThis work demonstrates a high-performance 1060 nm VCSEL integrating a composite elliptical oxide aperture and strain-compensated MQWs, achieving a modulation bandwidth up to 33.5 GHz along with excellent thermal stability and modal control. The design leverages strong current and optical confinement, optimized quantum well positioning, and low parasitic characteristics to realize ultrafast modulation dynamics. The simplified fabrication process ensures high compatibility with generic VCSEL platforms, offering a promising solution for deployment in next-generation data centers and high-performance computing environments.

ObjectiveAmong rare earth ions, Dy3+ has attracted significant attention in optical material research due to its exceptional luminescent properties in the visible spectrum, demonstrating broad application prospects. Dy3+ exhibits two primary emission peaks in the blue and yellow regions, corresponding to the 4F9/2→6H15/2 (~480 nm) and 4F9/2→6H13/2 (~570 nm) electronic transitions respectively, with the yellow emission being particularly prominent. However, the relatively long lifetime of the lower energy level 6H13/2 in Dy3+ yellow emission can lead to population bottlenecking, thereby limiting luminescence efficiency. One effective solution involves introducing other rare earth ions to facilitate energy transfer. Given the proximity between the 7F4 energy level of Tb3+ and the 6H13/2 level of Dy3+, energy transfer can help depopulate the blocked ions at 6H13/2 level of Dy3+, consequently reducing its fluorescence lifetime and enhancing yellow emission. To date, several Dy3+-doped crystals have successfully achieved yellow laser output under commercial blue laser diode pumping. Nevertheless, research on yellow lasers remains exploratory, and the search for novel Dy3+-doped gain media continues to be a crucial direction in this field. CaWO4 crystals, as excellent host materials for rare earth activator ions, possess numerous advantageous characteristics including outstanding physicochemical stability and wide optical bandgap, finding extensive applications in luminescent materials, scintillators, and sensors. However, current studies on CaWO4 optical properties in the visible spectrum remain limited, with existing research on Dy3+-doped tungstates primarily focused on phosphor applications. Considering the rich energy level transitions of Dy3+ and Tb3+ in the visible range and the superior properties of CaWO4, Dy3+,Tb3+ co-doped CaWO4 crystals hold potential for achieving continuous yellow laser output. Based on this rationale, we employed the Czochralski method to grow CaWO4 crystals co-doped with 3.8% (atomic fraction) Dy3+ and 3% Tb3+ (Dy3+, Tb3+∶CaWO4). We systematically analyzed their room-temperature polarized absorption spectra, low-temperature polarized fluorescence spectra, and fluorescence lifetimes. Through Judd?Ofelt theory, relevant spectroscopic parameters were calculated. Furthermore, using a blue semiconductor laser as the pump source, we conducted all-solid-state continuous yellow laser experiments with a simple plano-concave resonator configuration.MethodsDy3+, Tb3+∶CaWO4 crystal was grown using the Czochralski method. High-purity (99.99%) Dy2O3, Tb4O7, Na2CO3, and CaCO3 were weighed according to the stoichiometric ratio Dy0.038Tb0.03Na0.068Ca0.864WO4. The resulting crystal, with dimensions of Φ21 mm×40 mm, is shown in Fig.1. The actual doping concentrations of rare earth ions in the crystal were determined using inductively coupled plasma emission spectroscopy (Agilent ICP-OES 725 ES). The crystal structure was characterized using an X-ray diffractometer (Rigaku Smartlab). For optical performance testing, the crystal was cut into 4 mm×4 mm×2 mm samples and polished. Polarized emission spectra and fluorescence lifetimes were measured at room temperature using a fluorescence spectrometer (Edinburgh FLS-980), while polarized absorption spectra were obtained using a UV-Vis-NIR spectrophotometer (Hitachi UH 4150). The primary vibrational modes and phonon energy characteristics of the crystal were analyzed using a Raman spectrometer (HORIBA LabRAM Odyssey). For the yellow continuous-wave laser experiment, a multi-tube blue LD was used as the pump source with a maximum output power of 10 W. A biconvex focusing lens with a 70 mm focal length was employed to focus the pump light into the crystal. The experimental sample was a cut 3 mm×3 mm×25 mm (a×c×a) Dy3+,Tb3+∶CaWO4 crystal with laser-grade polished 3 mm×3 mm end faces. During the experiments, the crystal was wrapped in thin indium foil and mounted on a custom copper block with circulating water cooling, where the water temperature was controlled by a constant-temperature water bath. The laser cavity adopted a simple plano-concave configuration featuring an input mirror coated for enhanced blue pump light transmission (T>99%) and high yellow band (500?600 nm) reflectivity (R>99%), along with a plano-concave output coupler (50 mm radius of curvature) coated for 1.8% transmittance at 574 nm yellow light. The cavity length was maintained at approximately 50 mm throughout the experiment.Results and DiscussionsInductively coupled plasma-atomic emission spectroscopy (ICP-AES) analysis revealed actual doping concentrations of 3.19% for Dy3+ and 2.72% for Tb3+. The XRD pattern of the Dy3+, Tb3+∶CaWO4 crystal (shown in Fig. 2(a)) indicates that all diffraction peaks align with the standard pattern of pure CaWO4, confirming that the crystal retains the structure of CaWO4. Figure 2(b) illustrates the unit cell structure of CaWO4, which belongs to the scheelite family and exhibits a tetragonal structure. The polarized Raman spectrum (presented in Fig. 3) shows no additional peaks compared to pure CaWO4, indicating that Dy3+ and Tb3+ doping does not alter the vibrational modes or crystal structure of CaWO4. However, the characteristic peaks exhibit slight redshifts. This phenomenon may be attributed to the introduced Dy3+, Tb3+, and Na? ions perturbing the WO42- tetrahedra, resulting in elongated W—O bond lengths and reduced vibrational frequencies. The absorption cross-sections of the crystal at 450 nm were determined to be 0.726×10?21 cm2 for π-polarization and 0.526×10?21 cm2 for σ-polarization (Figs. 4(a) and 4(b)). According to the Judd?Ofelt theory, the fluorescence branching ratio for the ?F9/2→?H13/2 transition is 79.21%, and the radiative lifetime of the ?F9/2 level is 449.5 μs. Emission spectra and fluorescence lifetimes were measured at five distinct temperatures (77 K, 150 K, 200 K, 250 K, and 300 K). As evident from Fig. 5, the emission cross-sections progressively decrease with increasing temperature, primarily attributed to the continuous broadening of the full width at half maximum (FWHM) as temperature increases. Figure 6(a) displays the fluorescence lifetime fitting curve at 300 K as a representative case, while Fig. 6(b) presents the temperature-dependent lifetime relationship. The data reveal a gradual increase in lifetime with increasing temperature. The wavelength of the blue pump source undergoes a red shift with increasing power. In this experiment, the three-tube blue LD had an adjustable power range of 0?10 W, with the output wavelength gradually shifting from 448 nm to 450 nm. At the maximum pump power of 10 W (450 nm wavelength), the crystal absorbed 6.8 W of pump power, achieving a maximum absorption efficiency of 59.43%. After optimizing the laser resonator parameters and temperature control (1?30 ℃), only white fluorescence was observed (Fig. 8) with no laser generation. Potential limitations include: suboptimal doping concentrations, insufficient pump power, and non-ideal output couplers. Future work will employ higher-power pumps or beam-combining/ dual-end-pumped configurations to optimize the system, aiming to achieve laser output.ConclusionsLarge-size high-quality Dy3?, Tb3+∶CaWO4 single crystals were successfully grown using the Czochralski method. The segregation coefficients of Dy3+ and Tb3+ were determined to be 0.8393 and 0.9072, respectively, indicating that CaWO4 is an excellent host matrix for Dy3+ doping. XRD and Raman spectroscopy studies confirmed that Dy3+, Tb3+∶CaWO? crystals maintain the same structure as pure CaWO4. Compared with previously reported Dy3+, Tb3+∶CaWO4 crystals with a lower Tb3+ doping concentration (1.28% in atomic fraction), increasing the Tb3+ concentration to 3.8% in this study led to an enhanced fluorescence branching ratio for the ?F9/2→?H13/2 transition of Dy3+. Meanwhile, XRD and Raman spectra showed negligible changes, suggesting that the crystal structure remains stable. The segregation coefficient of Tb3+ increased from 0.7742 to 0.9072, whereas that of Dy3+ decreased from 0.8974 to 0.8393, likely due to crystal field competition caused by differences in ionic radii. Room-temperature measurements yielded absorption cross-sections of 0.726×10-21 cm2 (π-polarized) and 0.526×10-21 cm2 (σ-polarized) at 450 nm. Temperature-dependent polarized emission spectra and fluorescence lifetime measurements at five temperatures in the range of 77?300 K revealed decreasing emission cross-sections and increasing fluorescence lifetimes with rising temperature. All-solid-state laser experiments using 450 nm blue excitation produced intense white light emission from Dy3+, Tb3+∶CaWO4 crystals, though yellow laser output was not achieved. These results demonstrate the crystal’s potential as a gain medium for yellow lasers and in white LED applications, making it a promising candidate material for all-solid-state lighting devices.