Open Access
Add to BibSonomy
Abstract
Researchers have developed a uniform grating DFB with an integrated active optical feedback waveguide (UG-AFDFB) to enhance the modulation speed of direct modulation lasers (DMLs) while reducing costs based on identical active layer designations. However, this design has difficulties in obtaining high single mode yield (SMY) and low relative intensity noise (RIN) as a result of the strong optical feedback caused by the integrated active feedback waveguide (AFW) and the random phase of the facet phase. In this paper, a partial corrugated grating DFB with an integrated active optical feedback waveguide (PG-AFDFB) is proposed to address this issue. Comparison of SMY, S21, RIN, modulation eye pattern, and frequency chirp parameters between UG-AFDFB and PG-AFDFB based on time-domain transmission line laser mode reveals that PG-AFDFB with an optimized grating couple parameter κ performs significantly better than UG-AFDFB under the same conditions. Furthermore, the performance of PG-AFDFB is not sensitive to the random phase of the rear facet phase. Even when κ ranges from 6000 /m to 12000 /m, the current in the AFW is between 0 mA and 20 mA, and the length of the AFW ranges from 50 µm to 100 µm; the SMY of PG-AFDFB remains above 80%.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
With the rapid increment of optical communication capacity, particularly in data center and passive optical network systems, the demand for high-speed optical transmitters is increasing. Direct modulation laser (DML) is one of the most attractive solutions for passive optical network (PON) and Ethernet due to its low cost and low power consumption [1–4]. Therefore, it is crucial to further improve the modulation bandwidth of DML and simplify its manufacturing process. There are two main technological approaches to improving the modulation bandwidth of a DML, enhancing carrier-photon resonance (CPR) and introducing photon-photon resonance (PPR). To understand the influence of CPR, the intrinsic response bandwidth of the DML can be expressed using the following formula.
In order to simultaneously consider the bandwidth enhancement effect brought by short cavity length, avoid thermal effects, and reduce cleavage difficulty, a DFB laser with integrated passive feedback waveguide (IFB-DFB) is widely studied [9–13]. By introducing passive feedback waveguides, the PPR effect and detuned loading (DL) effect can also be used to further enhance bandwidth to 50 GHz or even 65 GHz [9,13]. The PPR effect mainly utilizes the resonance between the main mode and side mode of the laser, increasing the 3 dB bandwidth by resisting the rolling down of frequency peak of CPR. To obtain a relatively flat response, a suitable PPR frequency is essential. The PPR frequency is generally adjusted by changing the phase of the optical feedback, which is related to the injection current of the passive waveguide [14]. However, the butt-joint regrowth process makes the fabrication of IFB-DFB complex, especially for quantum well lasers containing aluminum. Furthermore, the single-mode yield (SMY) would also be degraded due to the complex epitaxial growth process, thereby increasing production costs.
An approach of using identical active layer (IAL) has been introduced in the design process of electro-absorption modulated laser (EML) and DML due to its advantages in epitaxial process [8,15–17]. Using this technology, an EML array with a modulation bandwidth of 56 Gb/s has been realized in the L-band [17]. Additionally, an active optical feedback waveguide (AFW) has been proven to be effective in enhancing the modulation bandwidth, with reported modulation bandwidths of 24 GHz [8] and 27 GHz [18] for DMLs, respectively. However, the AFW also has some drawbacks, such as the increase in the threshold current due to the additional absorption loss introduced by the active waveguide. Moreover, the lasing mode stability could be worse with an increase in the injection current of the active guide, which is necessary to obtain a suitable feedback phase. Typically, to achieve a higher output power, the rear and front faces of the laser are coated with high reflection (HR) and anti-reflection (AR) films, respectively. However, in a DFB laser, the random phase of the HR face can negatively impact the SMY and output power, which can be even more significant in a uniform grating DFB laser with an integrated active feedback waveguide (UG-AFDFB).
Therefore, to sample the fabrication of laser and introduce an AFW to enhance modulation speed, meanwhile improving the stability of laser, a partial corrugated grating DFB with integrated active optical feedback waveguide (PG-AFDFB) is demonstrated in this paper. Although some works have used the partial grating to improve the stability of DFB with optical feedback and EML [19–22], the passive feedback waveguide relates to a complex butt-joint epitaxial process regrowth make the cost expensive and low yield. In this paper, the output characteristics including SMY, small signal response, output characteristics, relative intensity noise (RIN), eye pattern, and frequency chirp are simulated and compared with the previous UG-AFDFB. As the simulation results shown, PG-AFDFB with an optimized grating couple parameter κ performs significantly better than UG-AFDFB under the same conditions. Moreover, for a DML which induce the PPR effect to enhance modulation speed, the proposed design offers a feasible solution to improve its mode stability.
2. Device structure and simulation method
Figure 1 shows the laser structures of PG-AFDFB and UG-AFDFB. Each structure consists of a gain region (150 µm in this paper) located near the rear side and an AFW located near the front side, and their injected current is controlled by two separated electrodes. The main difference between above two structures is that the gain region of PG-AFDFB only has a grating on the side near the AFW (60 µm in this paper), whereas UG-AFDFB etches a grating throughout the entire gain region. The center wavelength of the grating is 1310 nm, with a corresponding grating period of 200 nm. The grating can be fabricated using holography or electron-beam lithography techniques, followed by etching through dry etching or wet etching processes. The grating coupling coefficient can be adjusted by varying the thickness of the waveguide and grating layer. Therefore, our proposed laser design does not require any additional processes compared to traditional laser designs. To achieve a high output power and sufficient feedback power, the reflectivity of the rear and front faces is set to 95% and 30%, respectively. In order to obtain a high modulation speed, the gain section have lengths of 150 µm, while the AFW has a length ranging from 50 µm to 150 µm. So, the total length of a laser is 200 to 300 µm, making it suitable for cleavage. To analyze the effects of cavity length, feedback optical intensity and phase on photon density, carrier density and lasing condition in the resonant cavity, a simple external optical feedback single mode stability condition is introduced [23].
C is feedback coefficient which characterizes the level of feedback in relation to how it affects the mode stability of laser [23]. ${f_{ext}}$ describes the ratio of optical feedback power to output power. ${\tau _L}$ and ${\tau _{ext}}$ represent the round-trip time of laser cavity and external cavity. α is the linewidth enhancement factor [24]. t and r are coefficients of transmission and reflection of the front face. As shown in formula (3), the amplitude and phase of optical feedback, as well as the length of external cavity, have a significant impact on laser mode stability and SMY. However, in the proposed structure, changing the injection current of AFW will simultaneously cause variations in the feedback intensity and phase. Therefore, to comprehensively analyze the performance of UG-AFDFB and PG-AFDFB lasers and account for the influence of various laser parameters, a commercial software called VPI photonics is utilized. This software is based on the time-domain transmission line laser model (TLLM), enabling a detailed examination of the lasers’ characteristics and behaviors [25,26].In the proposed laser, a PG plays a crucial role in ensuring a laser’s insensitive to the random phase of the rear facet. The guide between grating and HR rear face act as a phase shifter, effectively reducing the impact of the random phase of the HR face on the round-trip phase condition. In other words, this grating-free zone near the HR rear face which can partially compensate for the effect of the random phase through carrier perturbation. Based on the above-mentioned simulation software, the performance of UG-AFDFB and PG-AFDFB are simulated and these results are shown in Fig. 2. The structures of both lasers are illustrated in Fig. 1, with a gain section length of 150 µm, an AFW length of 50 µm, a grating couple coefficient of 10000 /m, and a grating length of 60 µm for PG-AFDFB. The injected current for the gain section is 60 mA and 0 mA for the AFW, while other laser parameters are listed in Table 1. Figure 2 depicts the power distribution (a) and carrier density distributions (b) for the UG-AFDFB laser and the power distribution (c) and carrier density distributions (d) for the PG-AFDFB laser. Due to the multi-mode lasering in other phases for UG-AFDFB, only the phase range of 120 deg to 280 deg is shown in Fig. 2. The results demonstrate that with changes in the rear facet phase, the power and carrier density distributions in the gain section exhibit significant fluctuations for the UG-AFDFB laser, whereas only slight fluctuations are observed for the PG-AFDFB laser. Additionally, it is evident that the phase changes near the HR face play a significant role in compensating for the random HR facet phase variation.
Fig. 2. (a) (c) The power distribution along cavity for UG-AFDFB and PG-AFDFB under different HR facet phase; (b) (d) the carrier density distribution along cavity for UG-AFDFB and PG-AFDFB under different HR facet phase.
Table 1. Simulation parameters
To illustrate this further, let's consider the example of phase = 120 deg. When the rear facet phase of UG-AFDFB changes from 120 deg to 280 deg, to meet the new round-trip phase condition and the Bragg situation of grating simultaneously, the carrier density of the grating section near the rear face need a significant adjustment to compensate the phase difference. In contrast, for the PG-AFDFB laser, the required phase difference is distributed throughout the entire guide near the rear face, so the power and carrier density only have a slight fluctuation.
3. Simulation results and analysis
In our previous work, a 25 Ghz high-speed laser had been demonstrated [27], and we extracted the simulation parameters shown in Table 1 through parameter extraction [28]. In the follows, the performance of the PG-AFDFB and UG-AFDFB will be compared through simulations of their SMY, output power, small signal response, RIN, modulation eye pattern and frequency chirp.
Firstly, the relationship between the SMY of PG-AFDFB and UG-AFDFB with grating coupling coefficients (κ) and the injection current in AFW is discussed, and the other parameters is shown in Table 1. To ensure the generality of the simulation, the phase of the HR rear face is scanned from 0 deg to 350 deg, and the side mode suppression ratio (SMSR >35 dB) is used as a condition for single-mode operation. Figure 3(a) shows the change of SMY with various κ and injection currents in AFW. As depicted in Fig. 3(a), when the injection current of AFW is the same, the SMY of both lasers decreases with an increase in κ. Similarly, when keeping the same value of κ, the SMY of the lasers also decreases as the injection current increases. However, the SMY deterioration of UG-AFDFB would be more serious, while in contrast, the PG-AFDFB only exhibits slight degradation in SMY, which remains larger than 80% in all situations and shows better mode stability. And this difference in degradation of SMY between PG-AFDFB and UG-AFDFB becomes more significant with an increase in κ. When κ=12000 /m, the SMY of PG-AFDFB is ten times greater than that of UG-AFDFB. The reason for above degradation can be understood from the analysis of formulas (3) and (4). When the feedback length remains unchanged, the optical field passing through AFW can be attenuated or amplified by changing the amplitude of the injection current. If AFW is under gain, the intensity of the feedback would be amplified and affect the oscillation mode in the laser cavity, causing the laser wavelength to shift or even lase in multiple modes.
Fig. 3. The change of SMY: (a) with different κ and injection current in AFW; (b) with different κ and length of AFW for UG- AFDFB and PG- AFDFB.
In addition to the grating coupling coefficient κ and current of AFW, the mode stability is also related to the length of the AFW. Hence, Fig. 3(b) illustrates the change of SMY with different κ and length of feedback region for UG-AFDFB and PG-AFDFB. At this time the injection current in the feedback region is 0 mA, and the ratio of grating length is still 0.4, which is 60 µm. As shown in Fig. 3(b), although the SMY of PG-AFDFB decreases gradually with an increase in κ, the single mode characteristics of the proposed PG-AFDFB still maintain good performance. However, it should be noted that the variation of SMY with feedback cavity length is not monotonic. As formula (3) shows, the single mode stability condition is influenced by the combined effects of feedback region length, feedback intensity and phase. While increasing the length of the feedback region, the loss of the feedback region would increase, leading to a decrease in feedback intensity. Hence, the combined effect of these factors makes the change of SMY more complex. As mentioned earlier, one way to enhance modulation bandwidth is to introduce the PPR effect, and to obtain an appropriate PPR frequency, a longer feedback external cavity is necessary [14]. Thus, discussing the single mode characteristics under different lengths of the feedback region can help optimize laser structure and improve stability while increasing bandwidth.
In actual production, the SMY of lasers can impact their yield and cost, while higher output power directly affects the transmission distance and quality of communication in communication systems. The output power of a laser is related to the phase of the face, so the next step is to analyze the change of the output power of UG-AFDFB and PG-AFDFB with different rear face phases and κ. The laser parameters remain the same as those shown in Table 1, changing the current of gain section and the injection current in the feedback area is 0 mA. The PI curves of the laser under different κ and phases (0-360 deg) are shown in Fig. 4. It can be observed that when κ=6000 /m and the injection current is 60 mA, the output power of PG-AFDFB fluctuates by 2 mW, while the power fluctuation of UG-AFDFB is relatively large, reaching 6 mW. As κ increases to 10000 /m, the power fluctuation of PG-AFDFB increases to 3 mW, and that of UG-AFDFB increases to 8 mW. This means that regardless of whether κ is large or small, the fluctuation range of PG-AFDFB is still smaller than that of UG-AFDFB. Hence, it can be concluded that PG-AFDFB has better resistance to phase fluctuation of facet phase, which can help stabilize the output power and ensure consistent performance of the laser manufacturing.
Fig. 4. PI of UG-AFDFB and PG-AFDFB under different grating couple coefficient κ: (a) PG-AFDFB, κ=6000 /m; (b) UG-AFDFB, κ=6000 /m; (c) PG-AFDFB, κ=10000 /m; (c) UG-AFDFB, κ=10000 /m.
To evaluate the high-speed transmission performance of the chip, the small signal response of the two types of lasers is compared in Fig. 5. The bias current of gain region is set to 60 mA, and the S21 of the lasers is measured at different phases with feedback region currents of 0 mA and 5 mA. When the current is 0 mA, the 3 dB bandwidth fluctuation of UG-AFDFB reaches 10 GHz as the HR end phase increases from 0 to 180 degrees, while PG-AFDFB shows a smaller fluctuation of 7 GHz. The bandwidth fluctuation is not only due to the influence of the rear facet phase on the output power of the laser but also related to the DL effect caused by the AFW, which is equivalent to a Fabry-Perot (FP) cavity. Figure 6 shows the change of the lasing wavelength and optical power with different rear facet phase for PG-AFDFB. It can be observed that as the phase increases, the lasing wavelength undergoes a redshift, while the output power first decreases and then increases. This indicates that when the phase is 60∼180 degrees, the lasing wavelength is located on the short wavelength side of the reflection spectrum of the FP cavity, and when it is greater than 180 degrees, it is located on the long wavelength side. Hence, when the lasing wavelength is on the long wavelength side of the reflection spectrum, the effective differential gain of the laser increases due to the DL effect, thereby enhancing the modulation bandwidth [29]. When the injection current of AFW increases to 5 mA, it causes a change in the refractive index of the AFW, which results in a change in the phase of the feedback light. Consequently, the lasing wavelength and position of DL will change, leading to a change in the phase corresponding to the maximum bandwidth. Overall, it is evident that the bandwidth fluctuation of PG-AFDFB is smaller than UG-AFDFB under different injection currents. Additionally, it can be observed that increasing the injection current in AFW not only does not degrade the modulation bandwidth of the laser but also enhances the output optical power due to reduced losses in the AFW.
Fig. 5. The S21 of UG-AFDFB and PG-AFDFB at different injected current of AFW (I2): (a) I2 = 0 mA; (b) I2 = 5 mA, κ=6000 /m.
Fig. 6. The change of lasing wavelength and output power of PG-AFDFB with different rear facet phase.
To achieve a higher modulation bandwidth for DML, a shorter cavity length is generally used. However, to maintain a lower threshold current, a larger grating coupling coefficient is often used. In this study, the κ was increased to 10000 /m to investigate its effect on the S21 at different phases for both UG-AFDFB and PG-AFDFB with a bias current of 60 mA, as shown in Fig. 7. To ensure fairness in comparison and considering that the SMY of UG-AFDFB is only 55% at κ=10000 /m shown in Fig. 3, S21 under four phases is selected. It can be observed that the laser's bandwidth has improved comparing with κ=6000 /m, and PG-AFDFB still outperforms UG-AFDFB in terms of the stability of bandwidth variation with different phase. In other words, when designing DML with IAL, the proposed PG-AFDFB has superior resistance to random rear facet phase fluctuations.
The above S21 results show that both structures can support modulation rates of over 25 Gbps, without considering the SMY. However, for optical communication systems, communication quality is not solely determined by bandwidth, the RIN of the optical carriers also has a significant effect on it. Therefore, the static RIN of PG-AFDFB and UG-AFDFB lasers with varying grating coupling coefficients κ and rear facet phase is analyzed. Figure 8 shows that when κ=6000 /m and the AFW current is 0 mA, the RIN of both structures is similar as the phase changes. As the κ of the grating increases to 10000 /m, although the peak of RIN of both structures is reduced, the RIN noise in some phases of the UG-AFDFB laser deteriorates due to multimode lasing. The effect of different AFW injection currents on laser RIN is analyzed in Fig. 9. At κ=6000 /m, both UG-AFDFB and PG-AFDFB exhibit good RIN characteristics as the injection current of AFW increases. And when κ increases to 10000 /m, the RIN noise of UG-AFDFB deteriorates, similar to SMY. However, it is important to note that increasing AFW injection current can compensate for the losses it brings. Figure 10 demonstrates the distribution of forward transmission optical power in the cavity direction, showing that when the AFW current is 0 mA, the optical filed in the AFW undergoes power reduction due to loss. And, when the AFW current is increased to 5 mA, the gain of the AFW compensates for the loss, resulting in no attenuation in optical power.
Fig. 8. RIN of PG-AFDFB and UG-AFDFB with different rear facet phase: (a) κ=6000 /m, (b) κ=10000 /m.
Fig. 9. RIN of PG-AFDFB and UG-AFDFB with different injected current of AFW: (a) κ=6000 /m, (b) κ=10000 /m.
Fig. 10. Power distribution of UG-AFDFB: (a) I2 = 0 mA; (b) I2 = 50 mA, the rear facet phase = 90 deg. I2: the inject current of AFW.
To further compare the mode stability of the two structures, the relationship between RIN and the feedback strength and length of the external cavity is discussed below. In order to consider the impact of external cavity feedback, an external cavity feedback with an equivalent length of 15 cm was introduced into the simulation model to simulate the laser RIN state under different feedback intensities. Figure 11 shows the relationship between the RIN and rear facet phase of the two structures at a feedback intensity of -25 dB. It can be observed that when the external feedback intensity is -25 dB, the R IN of UG-AFDFB deteriorates more significantly, reaching -125 dBc/Hz, while the maximum RIN of PG-AFDFB is only -135 dBc/Hz. Furthermore, as the phase changes, PG-AFDFB accounts for 59.5% of RIN < -135 dBc/Hz, while UG-DFG only accounts for 32.4% as shown in Fig. 11(c). Moreover, if considering SMY, this value of UG-DFB will be even smaller. The relationship between the RIN of the two structures and the injection current and feedback intensity of AFW is shown in Fig. 12 (the RIN of both structures remains below -145 dBc/Hz without injection current and feedback). It can be observed that as the injection current of AFW increases, the RIN of UG-AFDFB continues to deteriorate, even approaching -110 dBc/Hz, while the RIN of PG-AFDFB still remains below -145 dBc/Hz at the resonant peak. Therefore, the above results further demonstrate that the resistance of PG-AFDFB to random phase of rear face and external cavity optical feedback is much higher than that of UG-AFDFB.
Fig. 11. The RIN of two laser at different rear facet phase: (a) UG-AFDFB; (b) PG-AFDFB. (c) The distribution of RIN with rear facet phase for two structures, κ=10000 /m and feedback power is -25dB.
Fig. 12. When feedback power is -30 dB and -25 dB, the RIN of two laser at different I2: (a) UG-AFDFB, (b) PG-AFDFB, κ=10000 /m. I2: the inject current of AFW.
To visually observe the transmission performance of the two structured lasers, NRZ modulation with a modulation rate of 25 Gbps was used as an example. The external cavity feedback intensity was set to -25 dB, and the rear facet phase was set at 150 degrees, 180 degrees, 210 degrees, and 240 degrees (due to the single mode limitation of UG-AFDFB). The modulation current was adjusted to let the extinction ratio of both structured lasers greater than 8.5 dB (the low-level and high-level currents were 10 mA and 60 mA, respectively). The modulation eye diagram of PG-AFDFB and UG-AFDFB was obtained as shown in Fig. 13. It can be seen, the amplitude of high-level of UG-DFB’s gradually increased with phase, and the ratio of power overshoot to high-level power (ROH) also increased from 30% to 57%. In contrast, the high-level power chirp of PG-AFDFB was maintained in a constant range, with an ROH of about 27% to the high level. Moreover, the noise at the low level of the UG-AFDFB eye diagram will be more severe. For DML, amplitude modulation is accompanied by frequency chirp [30].
Fig. 13. The eye diagram of two lasers at the rear facet phases of 150°, 180°, 210°, 240°: (a) PG-AFDFB, (b) UG-AFDFB.
To visually compare the frequency chirp under high-speed modulation, the frequency chirp was presented in the form of an eye diagram as shown in Fig. 14. It can be seen that the frequency chirp of the two structures in the first three phases was mainly adiabatic chirp, with an amplitude around 180 GHz. However, the frequency chirp of UG-AFDFB suddenly deteriorated at 240 degrees, resulting in severe transient chirp, which will seriously affect the quality of communication transmission. In summary, during the high-speed modulation process, PG-AFDFB has better resistance to rear facet phase disturbances, and the frequency chirp is basically adiabatic chirp. If combined with a chirp-managed laser, the transmission distance can be further improved [31].
Fig. 14. The frequency chip of two lasers at the rear facet phases of 150°, 180°, 210°, 240°: (a) UG-AFDFB, (b) PG-AFDFB.
4. Conclusion
In order to meet the growing demand for high-speed PON and Ethernet transmission, the bandwidth of DML needs to be further enhanced. Meanwhile, to simplify the epitaxy and chip fabrication process while maintaining high SMY, a partial corrugated grating DFB with integrated active optical feedback waveguide based on IAL process has been proposed.
The simulation results indicate that the overall performance of PG-AFDFB with IAL design is superior to that of the UG-AFDFB. As the grating coupling coefficient increases, the SMY of PG-AFDFB under different injection currents in AFW remains above 80%, whereas the SMY of UG-AFDFB drops below 70% even without injection current in the AFW. Moreover, with an increase in AFW injection current, the SMY of UG-AFDFB declines rapidly and drops below 40% at 20 mA when κ > 8000 /m. These results suggest that PG-AFDFB offers a wider range of effective current adjustment, which is crucial for adjusting the feedback phase of AFW to achieve a suitable PPR frequency. Furthermore, the power fluctuation of PG-AFDFB remains at 2 mW throughout all phases, while UG-AFDFB exhibits 6 mW power fluctuation when κ=6000 /m and current of gain section is 60 mA. When κ increases to 10000 /m, the power fluctuation of PG-AFDFB and UG-AFDFB increases to 3 mW and 8 mW, respectively. This suggests that PG-AFDFB has stronger robustness in output power. For S21, the bandwidth of both structures varies between 20 GHz and 30 GHz with different phases when κ=6000 /m. However, the S21 fluctuation of UG-AFDFB is greater (10 GHz), while that of PG-AFDFB is smaller (only 7 GHz). Even with an increase in AFW injection current, PG-AFDFB still has a smaller fluctuation range. When κ=6000 /m, both structures exhibit the same RIN, and the maximum value is maintained below -145 dBc/Hz. However, when κ increases to 10000 /m, the RIN of UG-AFDFB deteriorates due to multimode lasing, while that of PG-AFDFB still remains below -150 dBc/Hz. Particularly, when the external feedback is -25 dB, PG-AFDFB accounts for 59.5% of RIN < -135 dBc/Hz, while UG-DFG only accounts for 32.4%. When under high-speed modulation, the ‘1’ level of the modulation signal for PG-AFDFB remains unchanged at 11 mW regardless of the phase changes, whereas for UG-AFDFB, it increases from 10 mW to 14 mW. Meanwhile, the adiabatic chirp of both lasers is approximately 180 Ghz, but UG-AFDFB exhibits severe transient chirp at some phases, which is detrimental to long-distance communication. In summary, the proposed PG-AFDFB in this paper has less fluctuation in output power, S21, and overshoot of the eye diagram with the changes in facet phase, indicating excellent random rear facet phase resistance, which is beneficial for getting a suitable PPR frequency and high-speed optical transmission.
Funding
Chinese National Key Basic Research Special Fund (2018YFE0201200); Strategic Priority Research Program of Chinese Academy of Sciences (XDB43000000).
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. Matsuda, A. Uetake, T. Simoyama, S. Okumura, K. Takabayashi, M. Ekawa, and T. Yamamoto, “1.3-µm-Wavelength AlGaInAs Multiple-Quantum-Well Semi-Insulating Buried-Heterostructure Distributed-Reflector Laser Arrays on Semi-Insulating InP Substrate,” IEEE J. Sel. Top. Quantum Electron. 21(6), 241–247 (2015). [CrossRef]
2. K. Naoe, T. Nakajima, Y. Nakai, Y. Yamaguchi, Y. Sakuma, and N. Sasada, “Advanced InP laser technologies for 400 G and beyond hyperscale interconnections,” Proc. SPIE 11356, 1135604 (2020). [CrossRef]
3. Z. Zhou, M. Bi, S. Xiao, Y. Zhang, and W. Hu, “Experimental Demonstration of Symmetric 100-Gb/s DML-Based TWDM-PON System,” IEEE Photonics Technol. Lett. 27(5), 470–473 (2015). [CrossRef]
4. F. Saliou, G. Gaillard, G. Simon, S. L. Huérou, J. Potet, and P. Chanclou, “Triple Coexistence of PON Technologies: Experimentation of G-PON, XGS-PON and 50 G(S)-PON over a Class C+ ODN,” in European Conference on Optical Communication (2022), pp. 1–4.
5. Y. Matsui, T. Pham, T. Sudo, G. Carey, B. Young, J. Xu, C. Cole, and C. Roxlo, “28-Gbaud PAM4 and 56-Gb/s NRZ Performance Comparison Using 1310-nm Al-BH DFB Laser,” J. Lightwave Technol. 34(11), 2677–2683 (2016). [CrossRef]
6. Y. Matsui, H. Murai, S. Arahira, S. Kutsuzawa, and Y. Ogawa, “30-GHz bandwidth 1.55-µm strain-compensated InGaAlAs-InGaAsP MQW laser,” IEEE Photonics Technol. Lett. 9(1), 25–27 (1997). [CrossRef]
7. W. Kobayashi, T. Tadokoro, T. Ito, T. Fujisawa, T. Yamanaka, Y. Shibata, and M. Kohtoku, “High-speed operation at 50 Gb/s and 60-km SMF transmission with 1.3-µm InGaAlAs-based DML,” in International Semiconductor Laser Conference (2012), pp. 50–51.
8. G. Liu, G. Zhao, J. Sun, D. Gao, Q. Lu, and W. Guo, “Experimental demonstration of DFB lasers with active distributed reflector,” Opt. Express 26(23), 29784 (2018). [CrossRef]
9. J. Kreissl, V. Vercesi, U. Troppenz, T. Gaertner, W. Wenisch, and M. Schell, “Up to 40 Gb/s Directly Modulated Laser Operating at Low Driving Current: Buried-Heterostructure Passive Feedback Laser (BH-PFL),” IEEE Photonics Technol. Lett. 24(5), 362–364 (2012). [CrossRef]
10. Y. Matsui, T. Pham, W. A. Ling, R. Schatz, G. Carey, H. Daghighian, T. Sudo, and C. Roxlo, “55-GHz bandwidth short-cavity distributed reflector laser and its application to 112-Gb/s PAM-4,” in Optical Fiber Communications Conference and Exhibition (2016), pp. 1–3.
11. Z. Liu, Y. Matsui, R. Schatz, F. Khan, M. Kwakernaak, and T. Sudo, “50-GHz Repetition Gain Switching Using a Cavity-Enhanced DFB Laser Assisted by Optical Injection Locking,” J. Lightwave Technol. 38(7), 1844–1850 (2020). [CrossRef]
12. Y. Matsui, R. Schatz, D. Che, F. Khan, M. Kwakernaak, and T. Sudo, “Low-chirp isolator-free 65-GHz-bandwidth directly modulated lasers,” Nat. Photonics 15(1), 59–63 (2021). [CrossRef]
13. S. Matsuo, N.-P. Diamantopoulos, S. Yamaoka, and H. Nishi, “Direct Modulation of Membrane Distributed Reflector Lasers using Optical Feedback,” in Optical Fiber Communications Conference and Exhibition (2021), pp. 1–4.
14. M. Radziunas, A. Glitzky, U. Bandelow, M. Wolfrum, U. Troppenz, J. Kreissl, and W. Rehbein, “Improving the Modulation Bandwidth in Semiconductor Lasers by Passive Feedback,” IEEE J. Sel. Top. Quantum Electron. 13(1), 136–142 (2007). [CrossRef]
15. M. Theurer, C. Kottke, R. Freund, F. Ganzer, P. Runge, M. Moehrle, U. Troppenz, A. Sigmund, and M. Schell, “200 Gb/s Uncooled EML with Single MQW Layer Stack Design,” in European Conference on Optical Communication (2022), pp. 1–4.
16. Y. Mao, W. Zhao, J. Wang, H. Wang, Y. Huang, D. Lu, C. Ji, Q. Kan, and W. Wang, “Modulation Bandwidth Enhancement and Chirp Reduction in DFB Lasers with Active Optical Feedback,” in Conference on Lasers and Electro-Optics (2019), pp. 1–2.
17. M. Theurer, M. Moehrle, U. Troppenz, H.-G. Bach, A. Sigmund, G. Przyrembel, M. Gruner, and M. Schell, “4 × 56 Gb/s High Output Power Electroabsorption Modulated Laser Array With up to 7 km Fiber Transmission in L-Band,” J. Lightwave Technol. 36(2), 181–186 (2018). [CrossRef]
18. Y. Mao, W. Wang, Z. Ren, L. Guo, H. Wang, R. Zhang, Y. Huang, D. Lu, Q. Kan, and C. Ji, “Modulation Bandwidth Enhancement in Distributed Reflector Laser Based on Identical Active Layer Approach,” IEEE Photonics J. 10(3), 1–8 (2018). [CrossRef]
19. T. Okuda, H. Yamada, T. Torikai, and T. Uji, “Novel partially corrugated waveguide laser diode with low modulation distortion characteristics for subcarrier multiplexing,” Electron. Lett. 30(11), 862 (1994). [CrossRef]
20. Y. Huang, H. Yamada, T. Okuda, T. Torikai, and T. Uji, “External optical feedback resistant characteristics in partially-corrugated-waveguide laser diodes,” in Optical Fiber Communications (1996), pp. 34–36.
21. S. Sulikhah, S. L. Lee, and H. W. Tsao, “Improvement on Direct Modulation Responses and Stability by Partially Corrugated Gratings Based DFB Lasers With Passive Feedback,” IEEE Photonics J. 13(1), 1–14 (2021). [CrossRef]
22. P. D. Pukhrambam, S. L. Lee, and G. Keiser, “Electroabsorption Modulated Lasers With High Immunity to Residual Facet Reflection by Using Lasers With Partially Corrugated Gratings,” IEEE Photonics J. 9, 1 (2017). [CrossRef]
23. L. A. Coldren and S. W. Corzine, Diode Lasers and Photonic Integrated Circuits (Wiley, 1995).
24. C. Henry, “Theory of the linewidth of semiconductor lasers,” IEEE J. Quantum Electron. 18(2), 259–264 (1982). [CrossRef]
25. A. J. Lowery, “New dynamic semiconductor laser model based on the transmission-line modelling method,” IEE Proc. J Optoelectron. 134(5), 281 (1987). [CrossRef]
26. A. J. Lowery, P. C. R. Gurney, X.-H. Wang, L. Nguyen, Y. C. Chan, and M. Premaratne, “Time-domain simulation of photonic devices, circuits, and systems,” Proc. SPIE 2693, 624 (1996). [CrossRef]
27. X. Shijun, X. Borui, X. Pengfei, B. Shuai, W. Renfan, Z. Yao, L. Wei, and Z. Ninghua, “1.3 µm High-Speed Directly Modulated Semiconductor Laser,” Acta Opt. Sin. 42, 1614001 (2022). [CrossRef]
28. J. C. Cartledge and R. C. Srinivasan, “Extraction of DFB laser rate equation parameters for system simulation purposes,” J. Lightwave Technol. 15(5), 852–860 (1997). [CrossRef]
29. K. Vahala and A. Yariv, “Detuned loading in coupled cavity semiconductor lasers—effect on quantum noise and dynamics,” Appl. Phys. Lett. 45(5), 501–503 (1984). [CrossRef]
30. K. Zhang, Q. Zhuge, H. Xin, W. Hu, and D. V. Plant, “Performance comparison of DML, EML and MZM in dispersion-unmanaged short reach transmissions with digital signal processing,” Opt. Express 26(26), 34288 (2018). [CrossRef]
31. D. Mahgerefteh, Y. Matsui, X. Zheng, and K. McCallion, “Chirp Managed Laser and Applications,” IEEE J. Sel. Top. Quantum Electron. 16(5), 1126–1139 (2010). [CrossRef]
