Open Access
Add to BibSonomy
Abstract
Narrow linewidth lasers have a wide range of applications in the fields of coherent optical communications, atomic clocks, and measurement. Lithium niobate material possesses excellent electro-optic and thermo-optic properties, making it an ideal photonic integration platform for a new generation. The light source is a crucial element in large-scale photonic integration. Therefore, it is essential to develop integrated narrow linewidth lasers based on low-loss LNOI. This study is based on the multimode race-track type add-drop microring resonator with multimode interferometric coupler (MMRA-MRR) of the DFB laser self-injection-locked, to achieve the narrowing of linewidth to the laser. The microring external cavity was used to narrow the linewidth of the laser to 2.5 kHz. The output power of the laser is 3.18 mW, and the side-mode suppression ratio is 60 dB. This paper presents an integrated low-noise, narrow-linewidth laser based on thin-film lithium niobate material for the communication band. This is significant for achieving all-optical device on-chip integration of lithium niobate material in the future. It has great potential for use in high-speed coherent optical communication.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Lithium niobate material has a wide spectral transparency window (400 nm∼5 µm), excellent electro-optical, thermo-optical properties, high refractive index contrast. Based on these advantages, optoelectronic devices such as lithium niobate on insulator (LNOI) electro-optical modulators [1–5], filters [6,7], passive resonant cavities [8,9], and optical frequency combs [10,11] have been extensively researched. In order to realize the monolithic all-optical integration of the same material, the light source based on lithium niobate material is an indispensable part. Light sources based on lithium niobate materials have been reported. In 2022, Loncar's group first demonstrated a thin-film lithium niobate integrated electrically pumped laser [12], and in the same year, Ya Cheng's group realized a single frequency laser with a linewidth of 322 Hz using LN microdisk [13], and the following year they realized a narrow linewidth light source based on LNOI at a wavelength of 980 nm [14], making research on narrow linewidth laser sources in the 1550 nm communication band necessary.
Solid-state and fiber lasers exhibit excellent narrow linewidth performance, but their large size limits their application in communication systems. Semiconductor narrow linewidth lasers are crucial in researching narrow linewidth lasers due to their small size and ease of use. Currently, the main research direction for achieving miniaturization advantages in narrow linewidth is through heterogeneous integrated external cavity with gain or laser chip. Narrow linewidth lasers have been researched using various materials for external cavities, including Si [15], SiO2 [16,17], Si3N4 [18–20], and LNOI [13,21,22]. Among these materials, LNOI combines their advantages and has a very low waveguide transmission loss, which effectively prolongs the photon lifetime and results in a better linewidth narrowing effect. Furthermore, the narrow linewidth laser used in LNOI heterogeneous integration can be easily integrated monolithically with other LNOI-based electro-optical devices. This expands the range of applications for the narrow linewidth light source.
This paper presents a multimode race-track type add-drop microring resonator with a multimode interferometric coupler (MMRA-MRR) structure based on LNOI material. The MMRA-MRR structure effectively increases the injection feedback from the outer cavity to the laser, while the multimode waveguide reduces transmission loss due to sidewall roughness during the microring's transmission process. The use of multimode waveguide necessitates a larger coupling area than the traditional circular microring can provide for the required coupling efficiency. Therefore, a race-track type microring design was implemented to increase the coupling area. This design ensures a high Q factor of the microring and enables high feedback optical injection into commercial distributed feedback (DFB) lasers. The lasers’ phase noise is significantly reduced due to the self-injection locking effect, resulting in a narrowed output linewidth of 2.5 kHz and an output power of 2 mW. Figure 1 shows the schematic of the hybrid integrated narrow linewidth laser. The successful demonstration of LNOI low-noise and narrow linewidth lasers has great potential for on-chip all-optical integration and high-speed coherent optical communication systems.
Fig. 1. Schematic diagram of narrow linewidth laser hybrid integration with MMRA-MRR external cavity chip and commercial DFB laser. A phase shifter is used to control the phase of the feedback light to ensure that it is in an optimal self-injected locking state. Thermal electrodes on the microring are used to tune the resonant wavelength of the microring.
2. Principle and simulation
The external cavity of a passive chip external cavity laser (ECL) injects light into the laser chip through an optical chip circuit. Any slight perturbation of the feedback light during laser operation can disrupt its function, resulting in various states [23]. When the intensity of the feedback light reaches a certain value, the gain of the laser's other wavelengths collapses to the wavelength of the feedback light. This causes the feedback wavelength of the light to have an advantage in gain and mode competition, resulting in the laser's self-injection lock (SIL). As a result, the frequency noise of the laser's outgoing light is significantly reduced, achieving linewidth narrowing of the laser.
To provide a more intuitive numerical representation of the compressed narrow linewidth laser, we used the chirp reduction theory proposed by Kazarinov and Henry [24] and Tran et al [25] to simulate the ECL in this work. To make this theoretical model more relevant to our design, we split the DFB laser chip into a gain chip with a reflector for the DFB. This makes the entire system equivalent to a composite cavity laser. As shown in Fig. 2(a). The feedback effect of the external cavity in this system is equated to a reflection ${r_{eff}}(\omega )$, $\; {r_{eff}}(\omega )$ is a function of all the variables of the external cavity, including the coupling loss $|{{t_{coupling}}^2} |$, the passive waveguide loss $|{{t_{passive}}^2} |$, the transfer function T of the microring, and the phase change due to the passive waveguide.
Fig. 2. (a) The composite cavity can be equated to a single-cavity laser with equivalent reflectivity, and reff denotes the equivalent reflectivity of the external cavity. (b) Results of calculating the F, A, and B parameters.
The 3 dB loss in the coupling region corresponds to $|{{t_{coupling}}^2} |= 0.5$, the passive waveguide transmission loss factor $|{{t_{passive}}^2} |$= $exp({ - 2\alpha {L_p}} )$, $\alpha $ is the transmission loss factor of the passive waveguide, and Lp is the length of the passive waveguide. $\varphi $ is the phase change generated by the transmission of the passive waveguide in the outer cavity. ${T_{drop}}$ is the transmission function at the drop end of add-drop microring.
$\phi $ is the phase shift in microring, a is the amplitude attenuation coefficient in the ring and $a = \textrm{exp}({ - \alpha {L_R}} )$, LR is the circumference of the microring.
According to [23], then the following expression can be used and the line width after narrowing can be expressed as:
F2 is the narrowing factor of the external cavity on the linewidth of the laser, $\mathrm{\Delta }{\upsilon _0}$ is the linewidth of the original laser, ${\tau _0}$ is the photon lifetime in the active region of the DFB laser ${\tau _0} = \frac{{2{n_g}{L_a}}}{c}$, and ${\alpha _H}$ is the linewidth enhancement factor. The specific parameters are shown in Table 1.
Table 1. Compound cavity laser simulation parameters
3. External cavity MRR chip design and measurements
Firstly, a race-track type add-drop microring resonator was prepared. In the transmission region of the microring, a 5 µm wide waveguide was designed. The transmission loss of the waveguide is primarily caused by the contact between the optical mode field and the waveguide's sidewall. The rough sidewall results in light transmission loss. The design of the wide waveguide well avoids the contact between the optical mode field and the sidewalls, which effectively reduces the transmission loss of light. However, the use of wide waveguides can result in a decrease in coupling efficiency between adjacent waveguides due to the increased distance between the optical mode fields. As a result, an additional design is required to increase the coupling coefficient in the coupling region. The use of a race-track type microring in the design increases the length of the coupling region, resulting in better performance in the critical coupling state. In Ref. [26], an all-pass type microring with a coupling region length of 380 µm achieved a high-quality transmission spectrum with a rejection ratio of over 10 dB. When designing the microring, we made sure to set the coupling zone length to be larger than 380 µm to account for the difference in critical coupling conditions between the add-drop type microring and the all-pass type microring. Additionally, we set the gap value between the straight waveguide and the microring to 0.8 µm.
The microring underwent spectral testing using a broad-spectrum light source (ASE) as the optical input. The testing was performed on a six-axis high-precision coupling stage. The input light was coupled to the LNOI chip using a lensed fiber. Spot size converters (SSC) were designed at both the input and output ends of the chip to match the optical mode fields of the waveguide and the fiber, reducing the coupling loss. The microring's transmission spectrum was observed using a high-precision (0.04 pm) optical spectrum analyzer (OSA, APEX AP2060A) coupled to the output end via a lensed fiber. The spectra are presented in Fig. 3(d). The microring's FSR is 55 GHz, and its Q factor reaches 4.9 × 105 while the intrinsic Q factor of the microring is Qi = 2 × 106.
Fig. 3. (a) Fabricated add-drop microring with a coupling region length of 390 µm photographed through a microscope. (b) Localized scanning electron microscope (SEM) magnification of the coupling region. (c) Transmission spectrum of the microring (APEX AP2060A), FSR = 55 GHz. (d) Magnification of the transmission spectra at the resonance wavelength of λ=1550.364 nm,3 dB bandwidth is 3.18 pm and Q = 4.9 × 105.
4. LNOI narrow linewidth laser and measurements
Based on the MMRA-MRR external cavity chip, the linewidth narrowing of a commercial laser chip is realized. The structure schematic of the narrow linewidth laser is shown in Fig. 1. A multimode interferometric coupler (MMI) is used to connect the input end and the drop end of the add-drop microring together, so that the two-way feedback of light is realized. This design effectively increases the intensity of the feedback light, which can make up for the insufficiently high Q of the microring. The light intensity feedback R = 0.4 was calculated by the matrix equation. Figure 4 illustrates the schematic transmission of light in the MMRA-MRR structure of the outer cavity. The SIL state of the laser is dependent on the resonant wavelength of the external cavity and the phase. If the laser wavelength does not match the resonant wavelength, achieving the SIL state becomes difficult [15]. Maintaining the stability of the SIL state requires careful consideration of the phase condition [27,28]. Therefore, the outer cavity has two sets of thermal electrodes designed to adjust the wavelength and phase of the feedback light, ensuring the laser is in a stable SIL state.
Fig. 4. The external cavity chip enables the feedback light (green arrow) to be injected into the DFB laser chip, and the laser self-injects into the locked compression narrow linewidth.
The laser chip and external cavity chip of the narrow linewidth laser are coupled face-to-face. The laser chip is affixed to a thermoelectric cooler (TEC) which stabilizes the output wavelength by controlling the laser's working temperature. The voltage on the thermal electrode of the MMRA-MRR external cavity chip is adjusted to align the resonance wavelength of the microring with the output wavelength of the laser. Simultaneously, the phase shifter of the MMRA-MRR external cavity chip is adjusted to ensure that the feedback light satisfies the phase matching condition. A narrow linewidth laser spectrum is tested using an OSA, as depicted in Fig. 5(e). The laser achieves a side-mode suppression ratio (SMSR) of 60 dB. Additionally, the output power of the narrow linewidth laser is 2 mW when the injection current of the laser chip is set at 200 mA and the TEC temperature is controlled at 24 °C. By increasing the injection current of the laser chip further, the output power can reach 3.18 mW at an injection current of 260 mA.
Fig. 5. Physical and test schematic diagram of LNOI narrow linewidth laser and test result diagram, (a) laser chip placed on the TEC and face-to-face coupled with the MMRA-MMR external cavity chip, (b) noise PSD linewidth measured with three beams of light beat frequency. FRM: Faraday rotating mirror, DAC: digital analog converter, CIR: circulator, ISO: optical isolator, OC: optical fiber coupler, PD: photodetector. (c) Frequency noise of a free-running DFB versus a DFB laser with a MMRA-MMR external cavity chip, the frequency noise of the LNOI-based hybrid integration narrow linewidth laser is reduced by a factor of 169, (d) the phase noise of the hybrid integration narrow linewidth laser is significantly suppressed. (e) Emission spectrogram of the LNOI-based narrow linewidth laser with SMSR = 60 dB.
The narrow linewidth laser testing method typically involves using the delayed self-heterodyne method. This method utilizes the principle of Michelson interference, where the first laser being tested is split into two and sent through different fiber arms with varying delays to eliminate coherence. The resulting beat frequency converts laser phase or frequency fluctuations into changes in light intensity, allowing for the measurement of phase noise, linewidth, and other performance parameters. To eliminate the coherence of the two laser signals, a delayed fiber, also known as the delayed self-heterodyne interference (DSHI) technique, must be used. Longer delay fibers are required to measure very narrow linewidths using this method, which increases the experimental difficulty and reduces the accuracy of the test. Various methods have been proposed to make the measurement of ultra-narrow linewidth more accurate. The equipment used for testing single-frequency laser noise can divide the laser into three beams with equal phase differences. The data of the differential phase change over time is obtained using the beat frequency technique. Fourier transform and spectrum analysis are then carried out to obtain the power spectral density of the frequency noise and phase noise. This allows for an estimation of the equivalent linewidth of the laser's power spectral density (PSD) [29].
As shown in Fig. 5(b), the principle involves calculating the differential phase noise PSD directly from the DAC sampling output laser signal differential phase information. The laser signal frequency noise PSD can then be calculated using the following formula.
The spectral width of the laser field can be calculated based on the frequency noise PSD of the laser signal. However, the solution process is complex. To simplify the analysis process, the laser spectrum is divided into two parts. In the low-frequency part, the laser noise is inversely proportional to the sampling frequency. Therefore, the noise level is initially high but gradually reduces with an increase in the sampling frequency. In the high-frequency part, the laser noise is mainly white noise, which is independent of the sampling frequency and tends to level. Theoretically, the low-frequency part of the spectrum under 1/f noise cannot be solved analytically. However, the spectral line function under white noise in the high-frequency part follows a Lorentz function. If the white noise level in the high-frequency part is ${S_{\nu 0}}(f )$, the Lorentzian linewidth of the laser is $\mathrm{\Delta }{\upsilon _{Lorentzian}} = \pi {S_{\nu 0}}(f )$ [17].
The experimentally measured frequency noise of the directly packaged commercial DFB laser chip and the narrow linewidth laser hybrid integration with MMRA-MRR are shown in Fig. 5(c). The narrow linewidth laser hybrid integration with MMRA-MRR suppresses the high-frequency frequency noise of the common laser by 22.3 dB, and successfully narrows the linewidth of the laser to 2.5 kHz.
5. Discussion
Traditionally, self-injection locking of lasers using all-pass microring is achieved by utilising the reverse light scattered by Rayleigh in the microring. However, this method requires a high Q factor of the microring. However, this method requires a high Q factor of the microring, making it even more challenging to achieve for the LNOI material, which is harder to etch. To compensate for the lack of Q factor, this paper proposes a MMRA-MRR structure that is easier to implement in the LNOI material system. As a result, a narrow linewidth laser based on LNOI hybrid integration has been successfully demonstrated. The electro-optical modulator based on LNOI has been rapidly and maturely developed due to the excellent electro-optical properties of the LNOI material. Additionally, the realized LNOI hybrid integration narrow linewidth laser can be directly monolithic with the electro-optical modulator of the same material to realize the monolithic narrow linewidth transmitter front-end. Si3N4 system narrow linewidth lasers, when integrated with other devices, generate coupling losses that reduce the overall link performance. This work has potential for practical application of monolithic device integration in the future.
The experimental test results show minor deviations from the simulated structure in Fig. 2(b). These deviations are believed to be caused by uncontrollable factors during device fabrication and the coupling test process. For example, inconsistencies in the fabrication process may cause the gap value of the two coupling zones of the add-drop type microring to deviate from the design. The mismatch of the temporal mode field between the end-face mode-spot converter and the laser chip coupling causes insertion loss, which greatly reduces the emitted power of the laser. This, in turn, affects the linewidth of the LNOI hybrid integration narrow linewidth laser. Since the accuracy of the drive current source (Thorlabs ITC4001) used is not high enough, there will be drive current noise, and this noise has a non-negligible effect on the noise of narrow linewidth lasers [30,31].The final output optical power is approximately 10 dB lower than the output power of the separately packaged DFB laser (injection current 260 mA, temperature 25°C) due to losses in the coupling between chip-to-chip and chip-to-fiber. And the coupling loss between the DFB chip and the LNOI chip is about 6 dB.
This paper demonstrates the excellent results achieved by the hybrid integrated narrow linewidth laser based on LNOI external cavity chip. In comparison to previous work, this study has successfully achieved narrow linewidth lasers on the LNOI platform with excellent results in several metrics. This achievement is a significant contribution to the realization of all-optical device integration on the LNOI platform. Table 2 compares the performance metrics of the narrow linewidth lasers achieved on the lithium niobate platform in recent years. Most of the previous studies were based on the external cavity of silicon nitride [18–20], silicon dioxide [16,17], etc. Although excellent linewidth performance has been achieved, coupling between chips of different materials will bring huge loss, which does not meet the development needs of integrated microwave photonics in the long run. Compared with Ref. [13], the microring resonant cavities demonstrated in this study are superior to microdisk resonators in terms of preparation process and integration, and are more suitable for the development of integrated miniaturized narrow linewidth lasers. Moreover, this paper innovatively proposes the LNOI-based MMRA-MRR external cavity structure, and this design to some extent compensates the problem of insufficiently high Q factor of the microring caused by the difficulty of lithium niobate etching, and provides a strong optical feedback to the laser chip to compress the linewidth of the laser.
Table 2. This work is compared with previous LNOI-based narrow linewidth lasers
6. Conclusion
In conclusion, this work demonstrates that the LNOI hybrid integration narrow linewidth laser achieves a 2.5 kHz linewidth with a SMSR as high as 60 dB in the 1550 nm communication band. Additionally, the laser output power can reach up to 3.18 mW. The laser is based on the design of the MMRA-MRR external cavity to realize the SIL of the laser. The aim is to achieve a high-power narrow linewidth laser in the order of hundreds of Hz by optimizing the structural design and reducing the coupling loss. This will have significant application value in realizing LNOI monolithic integrated transmitter front-end and high-speed coherent optical communication.
Funding
National Key Research and Development Program of China (2021YFB2800402, 2022YFB2803200).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. M. Xu, Mingbo He, Hongguang Zhang, et al., “High-performance coherent optical modulators based on thin-film lithium niobate platform,” Nat. Commun. 11, 1–7 (2020). [CrossRef]
2. C. Wang, Mian Zhang, Xi Chen, et al., “Integrated lithium niobate electro-optic modulators operating at CMOS-compatible voltages,” Nature 562(7725), 101–104 (2018). [CrossRef]
3. M. He, Mengyue Xu, Yuxuan Ren, et al., “High-performance hybrid silicon and lithium niobate Mach–Zehnder modulators for 100 Gbit s -1 and beyond,” Nat. Photonics 13(5), 359–364 (2019). [CrossRef]
4. P. Kharel, C. Reimer, M Luke, et al., “Breaking voltage-bandwidth limits in integrated lithium niobate modulators using micro-structured electrodes: erratum,” Optica 8(9), 1218 (2021). [CrossRef]
5. K. Qu, Shanghong Zhao, Xuan Li, et al., “Ultra-flat and broadband optical frequency comb generator via a single mach-Zehnder modulator,” IEEE Photonics Technol. Lett. 29(2), 255–258 (2017). [CrossRef]
6. Y. Ding, Shiqi Tao, Xuanhao Wang, et al., “Thermo-optic tunable optical filters with GHz-bandwidth and flat-top passband on thin film lithium niobate platform,” Opt. Express 30(12), 22135 (2022). [CrossRef]
7. M. Han, J. Li, J. Wei, et al., “Ultra-Wideband Tunable Microwave Photonic Filter Based on Thin Film Lithium Niobate,” Photonics 10(10), 1080 (2023). [CrossRef]
8. M. Zhang, C. Wang, R. Cheng, et al., “Monolithic ultra-high-Q lithium niobate microring resonator,” Optica 4(12), 1536–1537 (2017). [CrossRef]
9. R. Gao, Ni Yao, Jianglin Guan, et al., “Lithium niobate microring with ultra-high Q factor above 108,” Chin. Opt. Lett. 20(1), 011902 (2022). [CrossRef]
10. M. Zhang, Brandon Buscaino, Cheng Wang, et al., “Broadband electro-optic frequency comb generation in a lithium niobate microring resonator,” Nature 568(7752), 373–377 (2019). [CrossRef]
11. C. Wang, M. Zhang, M. Yu, et al., “Monolithic lithium niobate photonic circuits for Kerr frequency comb generation and modulation,” Nat. Commun. 10(1), 1–6 (2019). [CrossRef]
12. A. Shams-Ansari, Dylan Renaud, Rebecca Cheng, et al., “Electrically pumped high power laser transmitter integrated on thin-film lithium niobate,” Optica 9(4), 408–411 (2022). [CrossRef]
13. J. Lin, Farajollahi Saeed, Fang Zhiwei, et al., “Electro-optic tuning of a single-frequency ultranarrow linewidth microdisk laser,” Adv. Photonics 4(03), 1–9 (2022). [CrossRef]
14. T. Huang, Yu Ma, Zhiwei Fang, et al., “Wavelength-Tunable Narrow-Linewidth Laser Diode Based on Self-Injection Locking with a High-Q Lithium Niobate Microring Resonator,” Nanomaterials 13(1), 1 (2022). [CrossRef]
15. N. M. Kondratiev, V. E. Lobanov, A. V. Cherenkov, et al., “Self-injection locking of a laser diode to a high-Q WGM microresonator,” Opt. Express 25(23), 28167 (2017). [CrossRef]
16. G. Zhang, Qizhuang Cen, Tengfei Hao, et al., “Self-Injection Locked Silica External Cavity Narrow Linewidth Laser,” J. Lightwave Technol. 41(8), 2474–2483 (2023). [CrossRef]
17. M. A. Tran, D. Huang, and J. E. Bowers, “Tutorial on narrow linewidth tunable semiconductor lasers using Si/III-V heterogeneous integration,” APL Photonics 4(11), 1 (2019). [CrossRef]
18. K. J. Boller, Albert van Rees, Youwen Fan, et al., “Hybrid integrated semiconductor lasers with silicon nitride feedback circuits,” Photonics 7(1), 4 (2019). [CrossRef]
19. C. Xiang, P. A. Morton, and J. E Bowers, “Ultra-narrow linewidth laser based on a semiconductor gain chip and extended Si3N4 Bragg grating,” Opt. Lett. 44(15), 3825 (2019). [CrossRef]
20. J. Li, S. Yang, M. Chen, et al., “Hertz-linewidth hybrid integrated laser based on high-Q Si3N4 microring reflectors,” Proc. SPIE 11891, 118910V (2021). [CrossRef]
21. M. Li, Lin Chang, Lue Wu, et al., “Integrated Pockels laser,” Nat. Commun. 13, 5344 (2022). [CrossRef]
22. V. Snigirev, Annina Riedhauser, Grigory Lihachev, et al., “Ultrafast tunable lasers using lithium niobate integrated photonics,” Nature 615(7952), 411–417 (2023). [CrossRef]
23. L. A. Coldren, “Diode lasers and photonic integrated circuits,” Opt. Eng. 36(2), 616 (1997). [CrossRef]
24. R. F. Kazarinov and C. H Henry, “The Relation of Line Narrowing and Chirp Reduction Resulting from the Coupling of a Semiconductor Laser to a Passive Resonator,” IEEE J. Quantum Electron. 23(9), 1401–1409 (1987). [CrossRef]
25. M. A. Tran, Duanni Huang, Joel Guo, et al., “Ring-Resonator Based Widely-Tunable Narrow-Linewidth Si/InP Integrated Lasers,” IEEE J. Sel. Top. Quantum Electron. 26(2), 1–14 (2020). [CrossRef]
26. C. Wei, J. Li, J Jia, et al., “Ultrahigh-Q lithium niobate microring resonator with multimode waveguide,” Opt. Lett. 48(9), 2465 (2023). [CrossRef]
27. A. Savchenkov, S. Williams, and A Matsko, “On Stiffness of Optical Self-Injection Locking Anatoliy,” Photonics 5(4), 43 (2018). [CrossRef]
28. R. R. Galiev, N. M. Kondratiev, I. A Lobanov, et al., “Optimization of Laser Stabilization via Self-Injection Locking to a Whispering-Gallery-Mode Microresonator,” Phys. Rev. Appl. 14(1), 014036 (2020). [CrossRef]
29. D. Xu, Fei Yang, Dijun Chen, et al., “Laser phase and frequency noise measurement by Michelson interferometer composed of a 3 × 3 optical fiber coupler,” Opt. Express 23(17), 22386 (2015). [CrossRef]
30. W. H. Burkett, B. Lü, and M Xiao, “Influence of injection-current noise on the spectral characteristics of semiconductor lasers,” IEEE J. Quantum Electron. 33(11), 2111–2118 (1997). [CrossRef]
31. G. P. Agrawal and R Roy, “Effect of injection-current fluctuations on the spectral linewidth of semiconductor lasers,” Phys. Rev. A 37(7), 2495–2501 (1988). [CrossRef]
32. J. Zhang, Zihan Li, Johann Riemensberger, et al., “Fundamental charge noise in electro-optic photonic integrated circuits,” arXiv, arXiv:2308.15404, (2023). [CrossRef]
33. J. Ling, Jeremy Staffa, Heming Wang, et al., “Self-Injection Locked Frequency Conversion Laser,” Laser Photonics Rev. 17, 30–34 (2023). [CrossRef]
