Open Access
Add to BibSonomy
Abstract
A narrow linewidth electro-optically tuned multi-channel interference (MCI) widely tunable semiconductor laser based on carrier injection is demonstrated in this paper. The MCI laser with a common phase section and a semiconductor optical amplifier (SOA) is packaged into a 16-pin butterfly box. The laser is characterized by a strategy: shifting the longitudinal mode and then aligning the reflection peak, which obtains a quasi-continuous tuning range over 48 nm. The corresponding side mode suppression ratios (SMSRs) are higher than 40 dB and frequency deviations from ITU-grid are less than ± 1 GHz. Threshold currents are less than 28 mA. Fiber coupled output powers are higher than 20 mW and power variations with fixed gain and SOA currents are less than 0.8 dB over the whole tuning range. Lorentzian linewidths are less than 320 kHz over the entire tuning range, which is one of the lowest results for monolithic widely tunable semiconductor lasers tuned by carrier injection. These results demonstrate the potential prospects of the MCI laser with carrier injection in the field of optical sensing and optical communications.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Widely tunable semiconductor lasers are rapidly becoming key components in the dense wavelength division multiplexing system and next generation reconfigurable optical networks [1–3]. To tune wavelengths, three physical mechanisms are used to change the refractive index of semiconductors. These tuning mechanisms are voltage tuning [4], thermal tuning [5] and carrier injection tuning [6]. The voltage tuning has the advantages of high tuning speed and low power consumption, but only small effective index changes are achievable [7–10]. Thermal tuning yields larger index changes and is easy to implement, but has low response speed and requires high tuning power [11–18]. Carrier injection is the most frequently used physical mechanism and has yielded the largest tuning range. Several widely tunable semiconductor lasers with carrier injection tuning have been demonstrated with a wide quasi-continuous (QC) tuning range, including sampled grating distributed Bragg reflector (SG-DBR) laser [19–21], super structure grating (SSG) DBR laser [22–24], vertical-grating-assisted codirectional coupler laser with rear sampled grating reflector (GCSR) [25–27], modulated grating Y-branch (MGY) laser [28–30] and digital super-mode (DS) DBR laser [31–33]. Performance comparison of monolithic widely tunable semiconductor lasers is shown in Table 1. However, linewidths of these lasers with carrier injection are typically on the order of megahertz (MHz). The intrinsic linewidth arises from the white noise generated by spontaneous emission [34]. The excess linewidth broadening is mainly attributed to the injection-recombination shot noise and the surface-recombination 1/f noise in the tuning section [35,36]. The linewidth of widely tunable semiconductor lasers can be reduced by higher intra-cavity optical power, lower optical confinement factor, smaller linewidth enhancement factor or lower loss [37]. A low-resistance voltage source used to bias the tuning section can reduce the shot noise, and a shallowly-etched waveguide with current injection can reduce the 1/f noise [38,39].
Table 1. Performance comparison of monolithic widely tunable semiconductor lasers
The multi-channel interference tunable laser is a promising monolithic widely tunable semiconductor laser [40,41]. The MCI laser has no grating structure and can be fabricated by conventional photolithography. An independent phase shifter is fabricated on each arm, which increases the fabrication tolerance. Compared with other monolithic widely tunable lasers, the passive section of MCI lasers is very long and so the cavity length is long, which is beneficial for reducing the linewidth. However, for the reported MCI laser characterization, the common phase section is treated as the arm phase section or not involved [42,43]. The longitudinal mode behavior as controlled by the common phase section is not utilized.
In this paper, the fabricated MCI laser integrates an SOA and a common phase section at the same time. To reduce surface recombinations, all carrier-injected waveguides are shallowly etched. The chip is packaged into a butterfly box, and the MCI laser is quasi-continuously characterized by a shifting-aligning strategy. In the entire tuning range, the performance is good and especially the Lorentzian linewidth is quite good for monolithic widely tunable semiconductor lasers based on carrier injection tuning.
2. Working principle and packaged laser
Figure 1(a) and (b) show the top-view and cross-sectional view of the MCI laser (schematic diagram). The MCI laser mainly consists of four sections: the SOA section which amplifies the light emitted from the laser cavity; the gain section which provides gain for the laser oscillation; the common phase section which can shift the longitudinal mode position; the multi-channel interference section which provides wavelength sensitive reflection therefore mode selection [40]. The multi-channel interference section contains cascaded 1 × 2 multi-mode interferometers (MMIs), eight arms with different lengths and multi-mode interference reflectors (MIRs). The cascaded MMIs split the light equally, and the one-port MIR provides reflection at each waveguide end. Eight arms interfere with each other to form a reflection spectrum dominated by a single reflection peak at which the eight arms have the same phase. The arm-length difference determines the full width at half maximum (FWHM) of the main reflection peak and the suppression characteristics of the other minor reflection peaks, i.e. determines the shape of the whole reflection spectrum. Each arm has its own independent phase shifter, and its phase is tuned by carrier injection. The free carrier absorption induced by carrier injection generates additional loss and the phase-current relationship is not linear, however, we can approximately assume that the phase depends linearly on the square root of the tuning current.
Fig. 1. Top-view (a) and cross-sectional view (b) of the schematic diagram of the MCI laser; (c) Schematic diagram of the mode selection. CP: common phase. AP: arm phase. m1: rear facet. m2: front facet.
One arm is selected as the reference arm from the multi-channel interference section whose phase shifter does not need to work. The other seven arms have to adjust their phases through injecting current into their phase shifters in order to align their phases with the reference arm at a specific wavelength. Once all eight arms have the same phase at a specific wavelength, the shape of the reflection spectrum is as shown in Fig. 1(c). Now, the effect of the multi-channel interference section is equivalent to a DBR mirror. The common phase section can shift the longitudinal mode position relative to the reflection peak and finely tune the laser output wavelength, which has no difference from a standard DBR laser. The wavelength tuning of the MCI laser is similar to that of a DBR laser in which the DBR section determines the reflection peak and the phase section fine-tunes the lasing mode. Therefore, one challenge of the MCI laser is to control seven arm phase shifters to align their phases with the reference arm at any designated wavelength.
The active-passive integration is based on the offset quantum well scheme [41]. The active section contains seven compressively strained indium gallium arsenide phosphide (InGaAsP) multiple quantum wells. In the passive section, the core layer is InGaAsP and its bandgap is 1.3 µm. In order to reduce surface recombination, shallowly-etched ridge waveguides are used in the carrier-injected sections and deeply-etched waveguides are used in other sections [42]. A shallow-deep transition structure is fabricated to reduce the loss caused by the mode mismatch between the shallowly-etched and deeply-etched waveguides. The fabrication processes are described in detail in Ref. [42,44]. Figure 2(a) shows a microscope image of the fabricated chip which has a size of 2300 µm × 500 µm. The front facet is anti-reflection (AR) coated. The SOA section is 500 µm long and 7° tilted. The two-port MIR which serves as a front mirror, is 9 µm wide and 56.8 µm long. The gain section is 400 µm long, and the phase section is 150 µm long. The 1 × 2 MMI is 6 µm wide and 38.5 µm long. The radius of the bend waveguide is 100 µm. The one-port MIR is 6 µm wide and 38.2 µm long. The equivalent cavity length of the MCI laser is 1558.90 µm, and the longitudinal mode interval is 0.21 nm. The laser chip is soldered onto an aluminum nitride (AlN) heat sink and packaged into a 16-pin butterfly package. Figure 2(b) shows a photograph of the inside of the butterfly package. The output light of the laser chip is collimated by an aspheric lens. A free-space isolator is then inserted to prevent the reflected light from affecting the lasing characteristics. Finally, the light is coupled into a single-mode fiber through a collimating lens. The total fiber coupling efficiency is estimated to be about 60%. The thermistor and thermoelectric cooler (TEC) are also packaged into the butterfly box. The 16 pins correspond to the SOA, the gain, the common phase section anode, eight-arm phase section anodes, TEC +, TEC -, thermistor +, thermistor -, and the common ground. The butterfly package is fixed on a custom-made mount, and a photograph is shown in Fig. 2(c).
Fig. 2. (a) Microscope image of the fabricated MCI laser chip; (b) Photograph of the inside of the butterfly package; (c) Photograph of the butterfly box fixed on a custom-made mount.
3. Laser characterization process and results
Figure 3 shows the experimental setup for the laser characterization. In the following experiment, the working temperature of the MCI laser is set at 20 °C by a temperature controller. The currents injected into the gain and the SOA sections are fixed at 100 mA, respectively, unless otherwise specified. The output light of the laser is split by 10:90 couplers. Then they are connected to an optical spectrum analyzer (OSA), a power detector (PD), a wavelength meter (WM), and a linewidth measurement system. The linewidth measurement system is based on the time-delayed self-homodyne coherent detection method [45–47].
Fig. 3. Schematic diagram of the experimental setup. OSA: optical spectrum analyzer, PD: power detector, WM: wavelength meter.
For laser characterization, the previously reported method is based on particle swarm optimization (PSO) algorithm, which used a tunable optical filter to set the target output wavelength and relied on the PSO algorithm to optimize eight arm phase shifters so as to maximize the output power, which reaches the maximum when eight arms have their phases aligned [43]. This method can only obtain the control information at those discrete output wavelengths and cannot quasi-continuously tune the laser.
Unlike using the PSO algorithm, here a characterization scheme relying on shifting the longitudinal mode and aligning the reflection peak (shifting-aligning strategy) is proposed, which can achieve quasi-continuous tuning.
In this work we used the Hill Climbing (HC) algorithm to align the eight arm phases at a specific output wavelength. Aligned phases for eight arms correspond to the maximum output power. A parabolic behavior occurs between the phase and the output power when one arm has its phase deviated. The main idea of the arm phase alignment process is to perturb the phase of each arm in turn near its initial value (square root of the current perturbed by ± 0.2 mA1/2), find the maximum output power by parabolic fitting and update its phase iteratively. When one arm phase is perturbed, the output power is monitored. This can be done with an external PD or with the SOA reverse biased as a PD. The curve between the phase and the output power is fitted with a parabola to find the maximum output power, and the arm phase is set to the peak power position. Then another arm phase is perturbed the same way until seven arm phases are all adjusted, which is called an iteration. Repeat the process, i.e. implement the HC algorithm until the convergence has been reached. The convergence condition is defined as that the square root of the phase shifter currents between two adjacent iterations is less than a pre-described value (0.04 mA1/2). Because the perturbation range is relatively small, mode hopping generally does not occur. Even if mode hopping occurs occasionally, the mode-hopping which causes the sudden change of the output power, will not affect the parabolic fitting very much and the HC algorithm can still continue. When the same phase state for all eight arms has been reached, parabolic curves can be observed as seen from Fig. 4(a), where the seven arm phase sections are perturbed in turn for one output wavelength. Once we know how to set seven arm phase sections for the specific output wavelength, the MCI laser works no different from a standard DBR laser except that here the DBR mirror can be tuned over a wide wavelength range.
Fig. 4. (a) Normalized output power versus the square root of the tuning current applied on each arm phase section. Over a range, wavelength versus square root of the tuning current applied on the common phase section (b) and seven arm phase sections (c)∼(i).
Because this arm phase alignment strategy is based on power measurement and applied around the same phase state, it generally converges quickly after less than ten iterations. If using standard GPIB connections between the computer and the measurement instruments, the alignment time for a single wavelength is about several seconds. Compared with the PSO algorithm [43], the phase alignment accuracy achieved in this new strategy is better which results in more accurate current settings, and the overall characterization speed is improved several times.
The wavelength tuning of the MCI laser is achieved by adjusting the longitudinal mode and the reflection peak. Therefore, this laser scheme is essentially quasi-continuous tuning. Furthermore, the lasing wavelength can be finely tuned by shifting the longitudinal mode. The tuning current in the common phase section slightly decreases, and the longitudinal mode shifts towards longer wavelengths. The arms are not at the same phase state at this new wavelength, and the arm phase alignment procedure is then carried out to align the phases. At this time, the relative position of the longitudinal mode to the reflection peak returns to the aligned state. The process can be repeated over and over again, extending the quasi-continuous tuning range to the entire tuning range. However, the tuning current of each arm cannot increase or decrease indefinitely. Due to the consistent characteristics of all arm phase shifters, the applied tuning currents are set to have the same upper and lower limits. Once the tuning current applied on one phase shifter reaches the set limit, the tuning current is adjusted back to within the set range by changing several 2π round-trip phase shifts. After this, the arm phase alignment procedure can be carried out to align the phases again. The wavelength extension process described above can then be continued. In addition, the strategy ensures that the lasing mode is basically aligned with the loss minimum, so that the linewidth is relatively low compared to those at the mode boundaries where frequency jumps occur [48].
Figure 4(b)∼(i) show the square root of eight tuning currents versus wavelength over a range. The tuning current in the common phase section changes periodically, which reflects the quasi-continuous shifting of the longitudinal mode, as shown in Fig. 4(b). The reflection peak follows the longitudinal mode shifting, which makes the seven arm phase tuning currents change periodically, as shown in Fig. 4(c) ∼ (i). The difference in the changing period reflects the arm length difference. Longer arms require more phase shift when shifting wavelength by the same amount. Relative to the arm phase shifters, the changing period of the common phase section is much smaller because the common phase section represents the phase shift of the longitudinal mode which is determined by the equivalent cavity length.
By applying the shifting-aligning strategy, the MCI laser is characterized at a 0.1-nm wavelength interval. The lasing spectrum overlays are plotted in Fig. 5(a) and the quasi-continuous tuning range covers 48.8 nm. Figure 5(b) shows that the corresponding SMSRs are higher than 40 dB and frequency deviations from the ITU-grid are less than ± 1 GHz over the entire tuning range. For adjacent calibrated wavelengths, wavelengths with smaller interval can be obtained by linearly interpolating the tuning current in the common phase section, as shown in Fig. 5(c). Ten wavelengths are interpolated by 0.01 nm between 1549.9 nm and 1550 nm. The corresponding SMSRs decrease gradually but does not exceed 1 dB degradation, which is caused by the gradual deviation of the longitudinal mode from the reflection peak but the lasing mode still dominates. This shows that denser target wavelengths can be interpolated without worrying about spectral degradation. Therefore, a look-up table for any wavelength can be obtained quickly over the entire tuning range.
Fig. 5. (a) Lasing spectrum overlays over the entire tuning range; (b) Corresponding SMSRs and frequency deviations from the ITU-grid; (c) Measured interpolated wavelengths and SMSRs between adjacent wavelength channels.
LI curves at different lasing wavelengths are measured by reversely biasing the SOA section as a detector, as shown in Fig. 6(a). The fiber coupled output power is shown in Fig. 6(b) when the SOA current varies from 0 to 200 mA and the gain current keeps 100 mA. The fiber coupled output powers are higher than 20 mW for different lasing wavelengths when the SOA current is biased above 170 mA. Figure 6(c) shows that threshold currents are all less than 28 mA and power deviations are less than 0.8 dB when the SOA current is fixed at 200 mA. The threshold current on the long-wavelength side is higher than that on the short-wavelength side with the same detuning from the gain peak, which is related to the lower differential gain of the quantum wells on the long wavelength side.
Fig. 6. (a) LI curves at different lasing wavelengths; (b) Fiber coupled output power when the SOA current varies from 0 to 200 mA; (c) Corresponding threshold current versus wavelength and fiber coupled output power when the SOA current is fixed at 200 mA.
The frequency-modulation (FM) noise spectrum is defined as the power spectrum of the instantaneous frequency and represents the amplitude of each instantaneous frequency fluctuation. Figure 7(a) plots the measured FM noise spectrum of the MCI laser near 1550 nm with or without the electrical noise filter circuit. In the original circuit (red line), the electrical noise causes unusual fluctuations over the entire frequency range, which greatly broadens the linewidth. Therefore, a low-noise source or a filter circuit is essential. After connecting a capacitor of 2.2 µF to each injected current (black line), the electrical noise is significantly suppressed and the spectrum becomes normal and smooth. The 1/f noise exists in the low frequency range below 100 MHz, and the rise at the high frequency end is due to the differential additive white Gaussian noise (AWGN) from the receiver [46]. The white noise can be observed at 200 ∼ 400 MHz, with the minimum value of 3.2 × 104 Hz2/Hz. The Lorentzian linewidth can be calculated from the white noise. Figure 7(b) shows Lorentzian linewidths are less than 320 kHz over the entire tuning range. Lorentzian linewidths on the short wavelength side are less than 200 kHz, and the long wavelength has a larger linewidth. This is because the linewidth enhancement factor is wavelength-dependent, longer wavelengths have higher linewidth enhancement factors. The maximum Lorentzian linewidth of the MCI laser is 316.6 kHz over the entire quasi-continuous tuning range, which as far as we know is one of the lowest results for monolithic widely tunable semiconductor lasers based on carrier injection [21,22,27,30,33]. This is resulted from the following factors such as a long cavity length, low loss, and the lasing mode aligned with the cavity loss minimum.
Fig. 7. (a) FM noise spectrum of the MCI laser near 1550 nm; (b) Lorentzian linewidths over the entire tuning range.
4. Conclusion
In this paper, we report a narrow linewidth MCI widely tunable semiconductor laser with carrier injection. The fabricated chip integrates an SOA section and a common phase section. The chip is packaged into a 16-pin butterfly box. The MCI laser is characterized by a shifting-aligning strategy. This strategy relies on shifting the longitudinal mode through tuning the common phase section and then aligning the reflection peak through tuning the seven arm phase sections. The performance metrics achieved include: quasi-continuous tuning range > 48 nm, over the entire tuning range SMSRs > 40 dB, frequency deviations from ITU-grid < ± 1 GHz, threshold currents < 28 mA, fiber coupled output power > 20 mW, and power deviations < 0.8 dB with fixed SOA currents. Due to the long cavity length, low loss and alignment of the lasing mode with the cavity loss minimum, Lorentzian linewidths are less than 320 kHz over the tuning range of 48.8 nm, which is one of the lowest results for monolithic widely tunable semiconductor lasers based on carrier injection.
Funding
National Key Research and Development Program of China (2022YFB2802901); National Natural Science Foundation of China (61875066); National Natural Science Foundation of China (61904064).
Disclosures
The authors declare no conflicts of interest.
Data availability
The 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. J. Buus, M. C. Amann, and D. J. Blumenthal, Tunable Laser Diodes and Related Optical Sources, 2nd ed. (Wiley-IEEE Press, 2005).
2. L. A. Coldren, “Monolithic tunable diode lasers,” IEEE J. Select. Topics Quantum Electron. 6(6), 988–999 (2000). [CrossRef]
3. L. A. Coldren, G. A. Fish, Y. Akulova, J. S. Barton, L. Johansson, and C. W. Coldren, “Tunable semiconductor lasers: a tutorial,” J. Lightwave Technol. 22(1), 193–202 (2004). [CrossRef]
4. N. Susa and T. Nakahara, “Enhancement of change in the refractive index in an asymmetric quantum well,” Appl. Phys. Lett. 60(20), 2457–2459 (1992). [CrossRef]
5. S. L. Woodward and U. Koren, “A DBR laser tunable by resistive heating,” IEEE Photon. Technol. Lett. 4(12), 1330–1332 (1992). [CrossRef]
6. B. R. Bennett and R. A. Soref, “Carrier-induced change in refractive index of InP, GaAs and InGaAsP,” IEEE J. Quantum Electron. 26(1), 113–122 (1990). [CrossRef]
7. M. Pantouvaki, C. C. Renaud, P. Cannard, M. J. Robertson, R. Gwilliam, and A. J. Seeds, “Fast Tuneable InGaAsP DBR Laser Using Quantum-Confined Stark-Effect-Induced Refractive Index Change,” IEEE J. Select. Topics Quantum Electron. 13(5), 1112–1121 (2007). [CrossRef]
8. S. Latkowski, A. Hänsel, N. Bhattacharya, T. D. Vries, L. Augustin, K. Williams, M. Smit, and E. Bente, “Novel Widely Tunable Monolithically Integrated Laser Source,” IEEE Photonics J. 7(6), 1–9 (2015). [CrossRef]
9. Y. Ueda, T. Shindo, S. Kanazawa, N. Fujiwara, and M. Ishikawa, “Electro-optically tunable laser with ultra-low tuning power dissipation and nanosecond-order wavelength switching for coherent networks,” Optica 7(8), 1003–1006 (2020). [CrossRef]
10. Y. Saito, Y. Ueda, T. Shindo, S. Kanazawa, H. Matsuzaki, and M. Ishikawa, “Burst-tolerant tuning of reflection-type transversal filter laser with single active region,” IEEE Photon. Technol. Lett. 34(1), 23–26 (2022). [CrossRef]
11. M. Larson, Y. Feng, P. C. Koh, X. Huang, M. Moewe, A. Semakov, A. Patwardhan, E. Chiu, A. Bhardwaj, K. Chan, J. Lu, S. Bajwa, and K. Duncan, “Narrow linewidth high power thermally tuned sampled-grating distributed Bragg reflector laser,” in Optical Fiber Communications Conference (OFC) (2013), paper OTh3I.4. [CrossRef]
12. M. C. Larson, A. Bhardwaj, W. Xiong, Y. Feng, X. D. Huang, K. P. Petrov, M. Moewe, H. Y. Ji, A. Semakov, C. W. Lv, S. Kutty, A. Patwardhan, N. Liu, Z. M. Li, Y. J. Bao, Z. H. Shen, S. Bajwa, F. H. Zhou, and P. C. Koh, “Narrow linewidth sampled-grating distributed Bragg reflector laser with enhanced side-mode suppression,” in Optical Fiber Communications Conference (OFC) (2015), paper M2D.1.
13. S. C. Davies, R. A. Griffin, A. J. Ward, N. D. Whitbread, I. Davies, L. Langley, S. Fourte, J. Mo, Y. Xu, and A. Carter, “Narrow linewidth, high power, high operating temperature digital supermode distributed Bragg reflector laser,” in European Conference on Optical Communication (ECOC) (2013), pp. 1–3.
14. U. Eriksson, J.-O. Wesström, Y. Liu, S. Hammerfeldt, M. Hassler, B. Stoltz, N. Carlsson, S. Siraj, E. Goobar, and Y. Matsui, “High performance narrow linewidth thermally tuned semiconductor laser,” in European Conference on Optical Communication (ECOC) (2015), pp. 1–3.
15. Y. Matsui, U. Eriksson, J. -O. Wesström, Y. Liu, S. Hammerfeldt, M. Hassler, B. Stoltz, N. Carlsson, S. Siraj, and E. Goobar, “Narrow linewidth tunable semiconductor laser,” in Compound Semiconductor Week (CSW) (2016), pp. 1–2.
16. T. Kaneko, H. Matsuura, K. Tanizawa, and K. Uesaka, “High-power and narrow-linewidth tunable distributed-reflector laser,” in Proc. SPIE9773, 97730A (2016).
17. K. Wang, Q. Chen, C. Jiang, Z. Chen, Q. Lu, and W. Guo, “Narrow linewidth and low thermal tuning power thermally tuned multi-channel interference widely tunable semiconductor laser,” in Optical Fiber Communication Conference (OFC) (2021), paper W1B.5. [CrossRef]
18. C. Jiang, Q. Chen, K. Wang, Q. Lu, M. Lu, and W. Guo, “Narrow-linewidth thermally tuned multi-channel interference laser integrated with a SOA and spot size converter,” Opt. Express 29(9), 13246–13255 (2021). [CrossRef]
19. V. Jayaraman, Z. M. Chuang, and L. A. Coldren, “Theory, design, and performance of extended tuning range semiconductor lasers with sampled gratings,” IEEE J. Quantum Electron. 29(6), 1824–1834 (1993). [CrossRef]
20. M. C. Larson, Y. A. Akulova, C. W. Coldren, T. Liljeberg, G. A. Fish, S. Nakagawa, A. Dahl, P. Kozodoy, D. Bingo, M. Bai, N. Ramdas, S. Penniman, T. Wipiejewski, and L. A. Coldren, “High performance widely-tunable SG-DBR lasers,” Proc. SPIE 4995, 66–80 (2003). [CrossRef]
21. S. Nakagawa, G. Fish, A. Dahl, P. Koh, and R. Yu, “Phase noise of widely-tunable SG-DBR laser,” in Optical Fiber Communications Conference (OFC) (2003), paper ThF2. [CrossRef]
22. F. Kano, H. Ishii, and Y. Tohmori, “Characteristics of super structure grating (SSG) DBR lasers under broad range wavelength tuning,” IEEE Photon. Technol. Lett. 5(6), 611–613 (1993). [CrossRef]
23. H. Ishii and H. Tanobe, “Broad-range wavelength coverage (62.4 nm) with superstructure-grating DBR laser,” Electron. Lett. 32(5), 454 (1996). [CrossRef]
24. H. Ishii, H. Tanobe, F. Kano, Y. Tohmori, Y. Kondo, and Y. Yoshikuni, “Quasicontinuous wavelength tuning in super-structure-grating (SSG) DBR lasers,” IEEE J. Quantum Electron. 32(3), 433–441 (1996). [CrossRef]
25. M. Oberg, S. Nilsson, K. Streubel, J. Wallin, L. Backbom, and T. Klinga, “74 nm wavelength tuning range of an InGaAsP/InP vertical grating assisted codirectional coupler laser with rear sampled grating reflector,” IEEE Photon. Technol. Lett. 5(7), 735–737 (1993). [CrossRef]
26. P. J. Rigol, S. Nilsson, L. Bäckbom, B. Stålnacke, E. Berglind, J. P. Weber, and B. Stoltz, “Quasi-continuous tuning range from 1560 to 1520 nm in a GCSR laser, with high power and low tuning currents,” Electron. Lett. 32(25), 2352–2354 (1996). [CrossRef]
27. A. A. Saavedra, P. J. Rigole, E. Goobar, R. Schatz, and S. Nilsson, “Relative intensity noise and linewidth measurements of a widely tunable GCSR laser,” IEEE Photon. Technol. Lett. 10(4), 481–483 (1998). [CrossRef]
28. J. O. Wesstrom, S. Hammerfeldt, J. Buus, R. Siljan, R. Laroy, and H. de Vries, “Design of a widely tunable modulated grating Y-branch laser using the additive Vernier effect for improved supermode selection,” in IEEE International Semiconductor Laser Conference (ISLC) (2002), pp. 99–100.
29. J. Wesström, G. Sarlet, S. Hammerfeldt, L. Lundqvist, P. Szabo, and P. Rigole, “State-of-the-art performance of widely tunable modulated grating Y-branch lasers,” in Optical Fiber Communications Conference (OFC) (2004), paper TuE2.
30. F. Liu, Y. Lin, Y. Liu, A. P. Anthur, Y. Yu, and L. P. Barry, “Investigation into the phase noise of modulated grating Y-branch lasers,” IEEE J. Select. Topics Quantum Electron. 23(6), 1–9 (2017). [CrossRef]
31. A. J. Ward, D. J. Robbins, G. Busico, E. Barton, L. Ponnampalam, J. P. Duck, N. D. Whitbread, P. J. Williams, D. C. J. Reid, and A. C. Carter, “Widely tunable DS-DBR laser with monolithically integrated SOA: design and performance,” IEEE J. Select. Topics Quantum Electron. 11(1), 149–156 (2005). [CrossRef]
32. L. Ponnampalam, D. J. Robbins, A. J. Ward, N. D. Whitbread, J. P. Duck, G. Busico, and D. J. Bazley, “Equivalent performance in C- and L-bands of digital supermode distributed bragg reflector lasers,” IEEE J. Quantum Electron. 43(9), 798–803 (2007). [CrossRef]
33. A. J. Ward, G. Busico, N. D. Whitbread, L. Ponnampalam, J. P. Duck, and D. J. Robbins, “Linewidth in widely tunable digital supermode distributed Bragg reflector lasers: comparison between theory and measurement,” IEEE J. Quantum Electron. 42(11), 1122–1127 (2006). [CrossRef]
34. C. Henry, “Theory of the linewidth of semiconductor lasers,” IEEE J. Quantum Electron. 18(2), 259–264 (1982). [CrossRef]
35. M. C. Amann, R. Hakimi, B. Borchert, and S. Illek, “Linewidth broadening by 1/f noise in wavelength-tunable laser diodes,” Appl. Phys. Lett. 70(12), 1512–1514 (1997). [CrossRef]
36. M. C. Amann and R. Schimpe, “Excess linewidth broadening in wavelength-tunable laser diodes,” Electron. Lett. 26(5), 279–280 (1990). [CrossRef]
37. M. C. Larson, “Widely tunable semiconductor lasers,” in Optical Fiber Communications Conference (OFC) (2014), paper Tu2H.1. [CrossRef]
38. M. C. Amann and S. Illek, “Linewidth reduction in wavelength tunable laser diodes by voltage control,” Electron. Lett. 27(6), 531–532 (1991). [CrossRef]
39. R. A. Griffin and G. Busico, “Laser FM noise impact on DCF-free transmission utilising electronic dispersion compensation,” in European Conference on Optical Communication (ECOC) (2005), pp. 261–262.
40. Q. Chen, Q. Lu, and W. Guo, “Theory and simulation of multi-channel interference (MCI) widely tunable lasers,” Opt. Express 23(14), 18040–18051 (2015). [CrossRef]
41. Q. Chen, X. Ma, W. Sun, Y. Liu, G. Liu, G. Zhao, Q. Lu, and W. Guo, “Demonstration of multi-channel interference widely tunable semiconductor laser,” IEEE Photon. Technol. Lett. 28(24), 2862–2865 (2016). [CrossRef]
42. Q. Chen, C. Jiang, X. Ma, Y. Liu, D. Yang, Q. Lu, and W. Guo, “1 × 8 MMI based multi-channel interference laser integrated with SOA through a 2-port multimode interference reflector,” Opt. Express 26(16), 19940–19949 (2018). [CrossRef]
43. Q. Chen, G. Liu, Q. Lu, and W. Guo, “Optimization algorithm based characterization scheme for tunable semiconductor lasers,” Opt. Express 24(18), 20982–20992 (2016). [CrossRef]
44. Q. Chen, K. Wang, C. Jiang, X. Ma, Y. Liu, Q. Lu, and W. Guo, “Butterfly-packaged multi-channel interference widely tunable semiconductor laser with improved performance,” Opt. Express 29(5), 6344–6355 (2021). [CrossRef]
45. T. N. Huynh, L. Nguyen, and L. P. Barry, “Phase noise characterization of SGDBR lasers using phase modulation detection method with delayed self-heterodyne measurements,” J. Lightwave Technol. 31(8), 1300–1308 (2013). [CrossRef]
46. T. N. Huynh, S. P. Ó. Dúill, L. Nguyen, L. A. Rusch, and L. P. Barry, “Simple analytical model for low-frequency frequency-modulation noise of monolithic tunable lasers,” Appl. Opt. 53(5), 830–835 (2014). [CrossRef]
47. K. Wang, M. Zhang, Q. Chen, C. Jiang, Q. Lu, and W. Guo, “Lorentzian linewidth of electro-optically tuned multi-channel interference widely tunable laser,” in Asia Communications and Photonics Conference /International Conference on Information Photonics and Optical Communications (ACP/IPOC) (2020), paper M4A.163. [CrossRef]
48. T. L. Koch, U. Koren, R. P. Gnall, C. A. Burrus, and B. I. Miller, “Continuously tunable 1.5µm multiple-quantum-well GaInAs/GaInAsP distributed-Bragg-reflector lasers,” Electron. Lett. 24(23), 1431–1433 (1988). [CrossRef]
