Open Access
Add to BibSonomyAbstract
We report on the design and characterization of a re-growth free InGaAsP/InP multiple quantum well two-electrode laterally coupled distributed feedback (LC-DFB) lasers. Third-order surface etched gratings have been defined on the ridge sidewalls along the laser cavity by means of stepper lithography. The lasers oscillate in single-mode around 1560 nm with high side mode suppression ratios (>52 dB), a wavelength tuning (≥ 3nm), an output power (≥ 6 mW), and narrow linewidth (<170 kHz) under various current injection ranges at room temperature. A minimum linewidth of 94 kHz has been recorded for 1500 µm-long two-electrode LC-DFB laser while providing non-uniform current injection through the two electrodes. The effect of the width of the inter-electrode gap on these different performance measures is also studied.
© 2014 Optical Society of America
1. Introduction
Although semiconductor distributed feedback (DFB) lasers have gained widespread applications [1–4], the formation of a well-defined periodic grating structure that provides the distributed feedback remains the main challenge in the DFB laser fabrication. Indeed, to ensure high single-mode laser yield with a high level of wavelength accuracy, it is of great importance to have a process that can simultaneously achieve low damage, uniform and reproducible gratings. If we look at the DFB fabrication, we find out that the gratings are defined during an interruption in the growth process, or through an etch and regrowth process. These multiple growth techniques require additional processing steps that make them more complex and difficult to implement; they are used routinely, in particular for high-performance InGaAsP/InP based lasers for telecommunications, but are a source of yield and reliability problems. Laterally-coupled distributed feedback (LC-DFB) lasers [5] combine the advantages of Fabry-Perot lasers (single growth) with the spectral purity of DFB lasers. In LC-DFB lasers, the periodic gratings structure is etched out of the upper growth layers that are used to form the ridge waveguide, as shown in Fig. 1. This process can be performed once all the epitaxy is done. In other words, there is no need for epitaxial regrowth or growth on the corrugated layers as in standard DFB laser fabrication, hence eliminating all the aforementioned shortcomings. Although an early demonstration of regrowth free DFB laser used gratings that had been etched on the ridge [6], lateral gratings can be etched exclusively in the ridge sidewalls (i.e. corrugated ridge waveguide lasers) [5], exclusively beside the ridge [7] or over both regions [8]. LC-DFB lasers that use metallic rather than etched semiconductor gratings on the cladding beside the ridge have also been investigated [9, 10].
Fig. 1 (a) 3D cutaway drawing for our two-electrode LC-DFB laser, (b) Scanning electron micrograph with different views for a fabricated ridge waveguide with third-order gratings.
Multi-electrode semiconductors lasers were originally introduced by S. Wang’s group [11-12], and primarily applied to DFB lasers by Yoshikuni and Motosugi [13]. The partition of the top electrode into two or more segments results in a laser cavity with electrically isolated sections. Accordingly, the control of current distribution through these sections allows an ‘artificial’ control of the carrier density distribution along the laser cavity. Changing the carrier density leads to gain peak and refractive index changes (i.e. non-uniformity of gain coefficient and refractive index). Indeed, these mechanisms are believed to be responsible for assuring both continuous tuning and frequency switching [12-14]. Additionally, by tailoring the injected current density distribution to that of the light intensity, the net carrier density, and thus the index profile, becomes more uniform. Therefore, the longitudinal spatial-hole-burning effect can be effectively suppressed [15]. This brings significant enhancements to the laser performance in terms of its frequency modulation (FM) and amplitude modulation (AM) characteristics, wavelength tunability, and spectral linewidth [14, 16-17].
The concept of multi-electrode current injection has been applied to laterally coupled DFB lasers designs of different structures than the laser described herein. It has been mainly used with superimposed binary metal gratings at short wavelengths (960 nm) [18] or at longer wavelengths (1200 nm) [19], and (2800 nm) [20]. In this work, we demonstrate a two-electrode multiple quantum well InGaAsP/InP LC-DFB laser operating at a wavelength of 1560 nm. The laser cavity has been defined by a ridge waveguide supporting uniform third-order etched InP gratings, as can be clearly recognized in Fig. 1. No λ/4-phase-shift or any other sections have been integrated. To the authors’ best knowledge, this structure is novel and contributes to the advancement of the field by suggesting rooms for design and performance improvements for the LC-DFB lasers. In this paper, we report on the design, fabrication, and experimental characterizations of two-electrode LC-DFB lasers. Preliminary optical characteristics, such as light output and lasing wavelength, as well as linewidth determination have been carried out for different lasers with different inter-electrode spacing.-In addition to the easy monolithic integrability that these devices offer, the experimental characterization reveals that this laser would be a strong candidate for applications such as photonics integrated circuits and advanced optical coherent communications.
The device design including the epitaxial layer structure and the fabrication technique have been described in [21]. It should be noted that, in this work, the active region has been designed to have a maximum gain at 1550 nm [21] and the corrugated ridge waveguide has been defined with a gratings’ pitch allowing an emission at 1560 nm.
2. Device characterizations
In this work, we consider two 1500 µm-long LC-DFB devices with top electrodes divided into two equal-length electrically isolated sections, as shown in Fig. 1(a). The first device, L1, has two 747-µm-long sections separated by an inter-electrode spacing (ES) of 6 µm. The second device, L2, has an ES of 10 µm separating two 745-µm-long sections. L1 and L2 have the same series resistance of ~5 Ω, but with slightly different isolation resistances: 1300 Ω and 1500 Ω for L1 and L2, respectively. Both devices were fabricated with a corrugated ridge waveguide defined with third-order gratings uniformly etched along the whole cavity. The gratings are defined with a period of ~723 nm (i.e. emission wavelength around 1560 nm), a duty cycle of 0.55, a wide ridge width of 3 µm, a narrow ridge width of 1.5 µm, and an etch depth of 0.8 µm. A scanning electron micrograph of fabricated third-order gratings is provided in Fig. 1(b).
The optical output power has been measured from the front facet of as-cleaved devices that have been mounted p-up on a temperature controlled copper plate. Typical continuous wave (CW) light-current (L-I) characteristics are reported in Fig. 2 under different biasing conditions. First we consider the case of uniform injection (i.e. both electrodes are equally pumped (IF = IB)), with IF and IB are respectively the front- and back-electrode currents, as depicted in the inset of Fig. 2(a), both lasers exhibited kink-free L-I characteristic and almost the same threshold current (Ith) of ~2 × 35 mA at 25°C, as shown in Fig. 2(a). The maximum differential quantum efficiency is slightly higher when considering larger spacing ES: 0.04 and 0.05 for 6 µm (L1) and 10 μm (L2), respectively. Reasons behind this low efficiency have been discussed in [21]. Both devices can emit more than 6 mW CW power under single-mode operation up to a total biasing of ~4.85 × Ith.
Fig. 2 Light output/front section characteristics for 1500 μm-long as-cleaved devices L1 (ES = 6µm) and L2 (ES = 10µm) at 25°C versus biasing current (ES: inter-electrode spacing): (a) in the case of uniform injection (IF = IB), (b) while varying IF at a constant total current IT = 280mA, (c) when varying IF while keeping IB at 40 mA and 80 mA, and (d) when varying IB while keeping IF at 40 mA and 80 mA.
Considerable changes occur when we consider the case of non-uniform injection (i.e. both electrodes are not equally pumped). Indeed, Fig. 2(b) shows the variation of the optical output power against the current ratio IF/IT (at a constant total current IT = (IF + IB) = 280mA) at 25°C. In this case both devices showed almost the same characteristics, except with a higher power for the device with a larger inter-electrode spacing. Near the uniform injection (IF/IT = 0.5), devices L1 and L2 exhibit a maximum power of ~4.6 mW and ~5.3 mW, correspondingly. After small variations, the power starts to stabilize at 5 mW and 6.3 mW towards a current ratio of ~0.85 for both L1 and L2, respectively. Figure 2(c) shows the light output measured as a function of the current IF while keeping IB constant at different levels (40 mA and 80 mA). From this plot we can define a lasing threshold for the front electrode (IF,th~30 mA) after which the output power increases rapidly with a better efficiency for the device L2. Figure 2(d) shows the case where the output power is plotted against the current through the back section IB while keeping IF biased at 40 mA and 80 mA. When IF is at 40 mA, the output light from both devices start to increase the moment the threshold for the back section (IB,th~30mA) is reached. This is an indication of the presence of gain and stimulated emission in this section [22]. When IF is above 40 mA, both devices start to lase in single mode at null back current. Overall, we notice that in all cases exemplified by Fig. 2, device L2 shows slightly higher power than that of device L1.
Figure 3(a) discloses the threshold current variation as a function of the stage temperature for both devices while uniformly biased. Although, the threshold is slightly better for the device L2, the overall measured threshold is considered as higher than anticipated. This can be mainly attributed to the spectral detuning between the gain peak and the cavity resonance wavelengths: λgp and λRes ( = 1560 nm), respectively, as can be seen in Fig. 3(b). The modal optical gain, as shown in the inset of Fig. 3(b), has been extracted from a typical uncoated 500 μm-long Fabry-Perot (FP) device (with a threshold of ~30 mA at 25°C) cleaved from the same wafer as the fabricated LC-DFB lasers. Indeed, this extraction epitomizes a spectral shift of ~30 nm between λgp (~1590 nm) and λRes at 25°C. The reasons behind this observation could be the inherent limitations of the simulation software (LAS2D [23]) used for the epitaxial structure design. In fact, the Inherent Las2D software limitations reside in the fact that Las2D does not support lattice-matched barriers to the InP substrate. Alone, the latter limitation affects the electronic structure of the active material. It shifts the conduction and valence eigenenergy levels in opposite directions in such a way that the effective (engineered) bandgap energy shoots up by an order of a meV. In addition, LAS2D is optimized for an F-P cavity structure and the results of simulation have been adapted to the actual LC-DFB structure. This approach disregards the particularities of the gain gratings, coupled to the refractive index gratings, from the actual material gain calculations. In addition, some current leaking may occur as in any ridge waveguide laser [5], which could be another reason behind the large threshold values.
Fig. 3 (a) Threshold per electrode variations against temperature at uniform injection (IF = IB) for both devices (L1 and L2); (b) gain-peak wavelength (λgp) variations as a function of temperature (The inset shows the extracted modal gain from a 500 µm-long Fabry-Perot device fabricated on the same wafer as the LC-DFB lasers. The gain has been extracted using the Hakki-Paoli technique while the FP device has been biased below threshold at 22 mA under 25°C). (λRes: cavity resonance wavelength)
The single mode emission at 1560 nm is assured through a corrugated ridge waveguide defined with third-order gratings. In Fig. 4, we show the variations of the lasing wavelength and the side mode suppression ratios (SMSRs) under different injection conditions at 25°C. In the uniform injection case, devices L1 and L2 showed stable single-mode operation with an electronic tuning of 2.3 nm and 1.8 nm, respectively as can be seen in Fig. 4(a). The tuning rate (dλ/dI) is 0.019 nm/mA and 0.015 nm/mA for L1 and L2, respectively. The SMSR, the variations of which are shown in Fig. 4(b), is greater than 45 dB for wide biasing range for both devices with larger values for L1.
Fig. 4 Optical spectra characteristics for 1500 μm-long devices L1 (ES = 6μm) and L2 (ES = 10μm) at 25°C: Wavelength variations against (a) uniform injection, (c) front electrode current IF while IB fixed at 40 and 80 mA, and (e) back electrode current while IF is fixed at 40 and 80 mA. SMSR variations at 25°C as a function of: (b) IF = IB; (d) IF at IB = 40 and 80 mA, and (f) IB at IF = 40 and 80 mA.
When considering non-uniform injection some interesting behaviour has been observed either in the lasing wavelength or in the SMSRs. Figure 4(c) shows the lasing wavelength variations as a function of IF at constant IB values. As can be seen in the summary Table 1 both devices show almost the same average tuning rate (~0.013 nm/mA). Device L2 shows a slightly better tuning range. In the case where IB is varied while IF is fixed, device L1 shows better results for both the average rate and tuning range, as can be seen in Fig. 4(e) and Table 1.
Table 1. Summary results regarding the rate of increase with bias current (dλ/dI) and tuning range (Δλ) for devices L1 and L2 in the non-uniform case at 25°C (in connection with Figs. (c) and (e))
From these results, we see that the tuning range and rate depend not only on the inter-electrode spacing (i.e. the separating electrical resistance) but also on the biasing conditions. The observed tunability can be explained as the consequence of the artificial control of the carrier density distribution achieved through the injection current control via the electrodes [13, 24]. Figure 4(d) shows the SMSR variations as a function of IF while maintaining IB at constant levels. Once IF reaches 90 mA and IB is 40 mA, we recorded SMSR ≥ 50 dB whilst L2 shows better side mode discrimination than L1. When IB is increased to 80 mA device L1 shows better discrimination while IF < 120 mA. Above this level device L2 performs better. In this range (IF>120mA and IB = 80mA) some side mode suppression degradation occurs for both devices: at IF = 180 mA for L1 and at IF = 150 and 220 mA for L2. The inset of Fig. 4(d) shows the optical spectra at IF = 140 and 150 mA for L2, where in the latter case a multi-peak spectrum appears causing the SMSR degradation. Figure 4(f) shows the SMSR variations against IB while leaving the front electrode current fixed. When IF is fixed at 40 and 80 mA, device L2 performs better as IB exceeds 110 mA. A degradation of side mode suppression is observed for L1 around IB = 140 and 200 mA while IB = 80 mA. The optical spectrum for L1 at IB = 140 mA is shown in the inset of Fig. 4(f). In these two cases, the optical spectra shown at the SMSR degradation points indicate that small mode hopping instability occurs at these particular section currents.
Figure 5 shows the variations of the lasing wavelength and SMSR against the stage temperature. Two different injection current schemes have been used while maintaining the total current IT (≡IF + IB) fixed at 280 mA. In the first case the total current is evenly injected amongst the two electrodes (i.e. current ratio (R) = 50%). In the second case, 78.5% of the total current is injected into the front electrode. In the former case, devices L1 an L2 show respectively a tuning range of 4.45 and 4.74 nm and a red-shift temperature rate (dλ/dT) of 0.093 and 0.099 nm/°C as can be inferred from Fig. 5(a). The latter case shows that devices L1 and L2 demonstrate respectively a tuning range of 4.41 and 4.53 nm and a tuning rate of 0.092 and 0.093 nm/°C. Measurements of side mode discrimination are shown in Fig. 5(b); device L1 performs well if the case of uniform injection is considered, whereas device L2 performs better if the non-uniform injection is considered. In the uniform injection case, L1 shows better results than L2. In contrast, L2 shows better results in the non-uniform injection case.
Fig. 5 Variations of: (a) the lasing wavelength and (b) the SMSR against the temperature for devices L1 (ES = 6µm) and L2 (ES = 10µm). Two injection current distributions have been used according to the current ratio R = IF/(IF + IB) = (50 and 78.5%).
In order to determine the optical linewidth of these devices, we have used the delayed self-heterodyne interferometric (DSHI) technique [25]. The experimental setup consists of an interferometer where the beam is split into two paths. In one path the optical field is delayed by a 20 km fiber-optic spool, and in the other path it is frequency shifted by an acousto-optic modulator (AOM), which is driven at a constant frequency (110 MHz). The fiber delay line used in this work allows a measurement resolution of 10 kHz. At the output of the DSHI setup, the mixing of the two beams is detected by a photo-detector (PD). The resulted photo-current power spectrum is then analyzed by an electrical spectrum analyzer (ESA) in order to extract the linewidth. In addition, an optical isolator has been used to minimize the back-reflections towards the laser under test. Besides, a commercial battery powered, ultra-low noise current source (model LDX-3620B) with a low noise filter (ILX LNF–320) have been used to drive the devices under test (in order to minimize the effect of source noise on the measured linewidth). In this section, due to the time-consuming nature of these measurements, we report only on the measurements that have been performed at 25°C.
Figure 6(a) shows the detected RF beat note spectrum for 1500 µm-long laser L2 while uniformly pumped. The broadening observed in the lineshape near the center of the beat note spectrum is non-Lorentzian. It is rather a Gaussian broadening, which is due to extrinsic (technically avoidable) noise introduced by the immediate environment [26, 27]. In such situations it is important to know the intrinsic or the real linewidth of the laser under testing. The Voigt profile – a convolution between Gaussian and Lorentzian profiles – allows simultaneous quantification of the effect of extrinsic noise (mapped by the Gaussian part) and the intrinsic linewidth (mapped by the Lorentzian part). We therefore apply a Voigt fitting procedure [27, 28] to the measured RF beat spectrum as illustrated by Fig. 6(a). In this figure, the spectrum has been recorded at the output of the DSHI setup for the laser L2 pumped under the uniform injection case (at 145/145 mA). The Voigt fitting gives a full width at half maximum (FWHM) of 1.068 MHz, from which the extracted Lorentzian and Gaussian FWHMs are 0.339 MHz and 0.873 MHz, respectively. In the case of DSHI method the intrinsic linewidth (δυ) of the laser is the half of the Lorentzian part FWHM of the beat note; hence the final linewidth is 169.5 kHz. Applying this procedure we have determined the linewidth δυ for devices L1 and L2 in the uniform case (IF = IB). Figure 6(b) shows the variations of the linewidth δυ for both devices as a function of the inverse power. For output powers ≥ 3 mW, both devices show linewidths ≤ 0.4 MHz. For instance, at a total injection level of 260 mA (i.e. IF = IB = 130mA), device L1 (1/P1 = 0.234 mW−1) show a δν1 of 0.264 MHz whereas device L2 (1/P2 = 0.208 mW−1) show a δν2 of 0.19 MHz. The minimum recorded linewidth δν is 184 kHz for device L1 at IT = 290mA and 164 kHz for device L2 at IT = 285 mA.
Fig. 6 (a) Measured RF beat note spectrum for device L2 at 145/145 mA and 25°C (The experimental data has been fitted with a Voigt profile); (b) The variations of the intrinsic linewidth as a function of the inverse of output power for both devices in the uniform case (IF = IB) at 25°C. The inset shows the driving current scheme.
Additionally, following the above fitting procedure, we have studied the variations of the intrinsic linewidth in the case of non-uniform injection, the results of which are summarized in Fig. 7. When varying the front current IF while fixing the back current IB, the linewidth decreases for both devices when IB increases from 40 to 80 mA at 25°C, as depicted in Fig. 7(a). At IB = 40mA, device L2 shows lower linewidths than device L1 for most of the IF injection range. When IB = 80mA both devices are alternately better from each other with a minimum δν of 144.5 kHz at IF = 150 mA for L1 and 93.5 kHz at IF = 210 mA for L2. In this latter case, an interesting phenomenon has been observed: two jumps occur in the linewidth variations at IF = 180 and 150 mA for device L1 and L2, respectively. These jumps can be due to the degradation of the side mode suppression as depicted in Fig. 4(d). This is confirmed by the experimental observation made by Kuo and Dutta while they were characterizing standard semiconductor two-electrode DFB lasers [22]. It is likely that a small mode jump occurs at these particular section currents (see for example Figs. 4(e) and 4(f) when IB = 150 mA and IF = 80 mA for device L1) that can be behind the linewidth rise [29]. In Fig. 7(b) we show the variations of δυ against the back current IB while maintaining IF at constant levels. In this case the linewidth decreases for both devices when IF increases from 40 to 80 mA at 25°C. At IF = 40mA, device L1 shows lower linewidths than L2 for IB ≤ 125 mA whereas the linewidth of L2 becomes lower once IB > 125 mA. When IF = 80mA both devices show almost similar linewidths up to IB = 130mA, after which L2 shows narrower and more stable linewidths with a minimum δυ of 0.302 MHZ at IB = 190 mA for L1 and 0.22 MHz at IB = 150 mA for L2. In this case, a jump occurs for L1 at IB = 140mA while IF is fixed at 80mA.
Fig. 7 Linewidth variations at non-uniform injection at 25°C for both devices L1 and L2: (a) variations of Δν against IF at fixed IB, and (b) variations of Δν against IB at fixed IF.
In general, the measured linewidths in the non-uniform case were narrower (≤ 150 kHz) than in the case of uniform injection for both devices. Table 2 shows a comparison between the linewidths obtained for devices L1 and L2 at different injection conditions. Depending on the biasing scheme, a narrow linewidth may be obtained. The narrow linewidth measured in this study could be attributed to the large detuning between the lasing wavelength and the gain peak wavelength. In this case an important reduction in the linewidth would be expected [30]. In addition, it is well known that the linewidth is inversely proportional to the laser output power, which has been found to be relatively low in this study. However, the low measured linewidth enhancement factor (α) (≤1.79) around the lasing wavelength (1560 nm) compensates for the low output powers. This low α-factor is in reasonable agreement with the observed linewidth given the large detuning of ~-30 nm (i.e. the LC-DFB lasing wavelength is at the shorter side of the gain peak) [30, 31].
Table 2. Comparison of the linewidth δν (in MHz) for both devices L1 and L2 at uniform and non-uniform injection at different constant total current IT at 25°C
3. Conclusion
We have demonstrated narrow linewidth and single-mode two-electrode InGaAsP/InP LC-DFB lasers at 1560 nm. The ridge waveguide has been defined with uniform third-order InP etched gratings, which have been written by using I-line stepper lithography. The structure is very simple avoiding hence the recourse to the complexities of integrating other sections (such as λ/4 phase-shift) within the cavity. The top electrode has been partitioned into two equally and electrically isolated sections. We have studied the effect of the inter-spacing between the electrodes on some device performances under uniform and non-uniform injection. We have demonstrated narrow linewidth (<170 kHz) for wide injection range and under different biasing conditions. The lasers exhibited stable single mode operation with side mode suppression ratio in excess of 52 dB, a wavelength tuning range over 3 nm, and an output power over 6 mW. In addition to these preliminary features, the ease of fabrication and monolithic integrate-ability potential make our laser a distinctive source for advanced coherent optical communications as well as in other photonic-based applications where narrow linewidth is required. Further characterization such as the relative intensity noise and frequency modulation response measurements needs to be performed as a forward work in order to further assess the capability of these devices.
Acknowledgments
The authors would like to thank Joe Seregelyi for his helpful discussions and for providing both the isolator and the ultra-low noise source. The authors are very thankful to Michel Poulin of Teraxion for fruitful discussions and his suggestions. The authors are also grateful to the Natural Sciences and Engineering Research Council (NSERC) for its support of this research; CMC Microsystemsfor its support of the device fabrication. Dr. Trevor Hall is a Canada Research Chair Tier I.
References and links
1. M. Funabashi, H. Nasu, T. Mukaihara, T. Kimoto, T. Shinagawa, T. Kise, K. Takaki, T. Takagi, M. Oike, T. Nomura, and A. Kasukawa, “Recent advances in DFB lasers for ultradense WDM applications,” IEEE J. Sel. Top. Quantum Electron. 10(2), 312–320 (2004). [CrossRef]
2. F. A. Fish, D. Welch, R. Nagarajan, J. L. Pleumeekers, V. Lal, M. Ziari, A. Nilsson, M. Kato, S. Murthy, P. Evans, S. W. Corzine, M. Mitchell, P. Samra, M. Missey, S. DeMars, R. P. Schneider, M. S. Reffle, T. Butrie, J. T. Rahn, M. V. Leeuwen, J. W. Stewart, D. J. H. Lambert, R. C. Muthiah, H.-S. Tsai, J. S. Bostak, A. Dentai, K.-T. Wu, H. Sun, D. J. Pavinski, J. Zhang, J. Tang, J. McNicol, M. Kuntz, V. Dominic, B. D. Taylor, R. A. Salvatore, M. Fisher, A. Spannagel, E. Strzelecka, P. Studenkov, M. Raburn, W. Williams, D. Christini, K. J. Thomson, S. S. Agashe, R. Malendevich, G. Goldfarb, S. Melle, C. Joyner, M. Kaufman, and S. G. Grubb, “Current Status of Large-Scale InP Photonic Integrated Circuits,” IEEE J. Sel. Top. Quantum Electron. 17(6), 1470–1489 (2011). [CrossRef]
3. R. M. Briggs, C. Frez, M. Bagheri, C. E. Borgentun, J. A. Gupta, M. F. Witinski, J. G. Anderson, and S. Forouhar, “Single-mode 2.65 µm InGaAsSb/AlInGaAsSb laterally coupled distributed-feedback diode lasers for atmospheric gas detection,” Opt. Express 21(1), 1317–1323 (2013). [CrossRef] [PubMed]
4. H. Nasim and Y. Jamil, “Diode lasers: From laboratory to industry,” Opt. Laser Technol. 56, 211–222 (2014). [CrossRef]
5. L. M. Miller, J. T. Verdeyen, J. J. Coleman, R. P. Bryan, J. J. Alwan, K. J. Beernink, J. S. Hughes, and T. M. Cockerill, “A distributed feedback ridge waveguide quantum well heterostructure laser,” IEEE Photon. Technol. Lett. 3(1), 6–8 (1991). [CrossRef]
6. Z.-L. Liau, D. C. Flanders, J. N. Walpole, and N. DeMeo, “A novel GaInAsP/InP distributed feedback laser,” Appl. Phys. Lett. 46(3), 221–223 (1985). [CrossRef]
7. R. D. Martin, S. Forouhar, S. Keo, R. J. Lang, R. G. Hunsperger, R. C. Tiberio, and P. F. Chapman, “CW performance of an InGaAs-GaAs-AlGaAs laterally-coupled distributed feedback (LC-DFB) ridge laser diode,” IEEE Photon. Technol. Lett. 7(3), 244–246 (1995). [CrossRef]
8. N. Chen, Y. Watanabe, K. Takei, and K. Chikuma, “InGaAsP/InP laterally coupled distributed feedback laser,” Jpn. J. Appl. Phys. 39(Part 1, No. 3B), 1508–1511 (2000). [CrossRef]
9. M. Kamp, J. Hofmann, A. Forchel, F. Schafer, and J.-P. Reithmaier, “Low-threshold high-quantum-efficiency laterally gain-coupled InGaAs/AlGaAs distributed feedback lasers,” Appl. Phys. Lett. 74(4), 483–485 (1999). [CrossRef]
10. M. Müller, M. Kamp, A. Forchel, and J.-L. Gentner, “Wide-range-tunable laterally coupled distributed feedback lasers based on InGaAsP–InP,” Appl. Phys. Lett. 79(17), 2684–2686 (2001). [CrossRef]
11. A. Antreasyan and S. Wang, “Electronic wavelength tuning with semiconductor integrated etalon interference lasers,” Appl. Phys. Lett. 43(6), 530–532 (1983). [CrossRef]
12. Z.-J. Fang and S. Wang, “Longitudinal mode behavior and tunability of separately pumped (GaAl)As lasers,” Appl. Phys. Lett. 44(1), 13–15 (1984). [CrossRef]
13. Y. Yoshikuni, K. Oe, G. Motosugi, and T. Matsuoka, “Broad wavelength tuning under single-mode oscillation with a multi-electrode distributed feedback laser,” Electron. Lett. 22(22), 1153–1154 (1986). [CrossRef]
14. Y. Yoshikuni and G. Motosugi, “Multielectrode distributed feedback laser for pure frequency modulation and chirping suppressed amplitude modulation,” J. Lightwave Technol. 5(4), 516–522 (1987). [CrossRef]
15. M. Usami and S. Akiba, “Suppression of longitudinal spatial hole-burning effect in λ/4-shifted DFB lasers by nonuniform current distribution,” IEEE J. Quantum Electron. 25(6), 1245–1253 (1989). [CrossRef]
16. D. Leclerc, J. Jacquet, D. Sigogne, C. Labourie, Y. Louis, C. Artigue, and J. Benoit, “Three-electrode DFB wavelength tunable FSK transmitter at 1.53 μm,” Electron. Lett. 25(1), 45–47 (1989). [CrossRef]
17. M. Fukuda, K. Sato, Y. Kondo, and M. Nakao, “Continuously tunable thin active layer and multisession DFB laser with narrow linewidth and high power,” J. Lightwave Technol. 7(10), 1504–1509 (1989). [CrossRef]
18. M. Muller, F. Klopf, M. Kamp, J.-P. Reithmaier, and A. Forchel, “Wide range tunable laterally coupled distributed-feedback lasers based on InGaAs-GaAs quantum dots,” IEEE Photon. Technol. Lett. 14(9), 1246–1248 (2002). [CrossRef]
19. N. A. Naderi, F. Grillot, K. Yang, J. B. Wright, A. Gin, and L. F. Lester, “Two-color multi-section quantum dot distributed feedback laser,” Opt. Express 18(26), 27028–27035 (2010). [CrossRef] [PubMed]
20. T. Lehnhardt, S. Hofling, M. Kamp, L. Worschech, and A. Forchel, “Tunable long wavelength (~2.8 μm) GaInAsSb–GaSb quantum-well binary superimposed grating lasers,” IEEE Photon. Technol. Lett. 22, 1662–1664 (2010).
21. K. Dridi, A. Benhsaien, J. Zhang, and T. J. Hall, “Narrow linewidth 1550 nm corrugated ridge waveguide DFB lasers,” IEEE Photon. Technol. Lett. 26(12), 1192–1195 (2014). [CrossRef]
22. C. Y. Kuo and N. K. Dutta, “Characteristics of two-electrode DFB lasers,” Electron. Lett. 24(15), 947–949 (1988). [CrossRef]
23. LAS2D software, Laboratoire d’Optoélectronique, École Polytechnique de Montréal, Montréal, QC, Canada (2006).
24. M. Kuznetsov, “Theory of wavelength tuning in two-segment distributed feedback lasers,” IEEE J. Quantum Electron. 24(9), 1837–1844 (1988). [CrossRef]
25. T. Okoshi, K. Kikuchi, and A. Nakayama, “Novel method for high resolution measurement of laser output spectrum,” Electron. Lett. 16(16), 630–631 (1980). [CrossRef]
26. L. B. Mercer, “1/f frequency noise effects on self-heterodyne linewidth measurements,” J. Lightwave Technol. 9(4), 485–493 (1991). [CrossRef]
27. S. Spiessberger, M. Schiemangk, A. Wicht, H. Wenzel, O. Brox, and G. Erbert, “Narrow Linewidth DFB Lasers Emitting Near a Wavelength of 1064 nm,” J. Lightwave Technol. 28(17), 2611–2616 (2010). [CrossRef]
28. J. Telkkälä, J. Viheriälä, A. Aho, P. Melanen, J. Karinen, M. Dumitrescu, and M. Guina, “Narrow linewidth laterally-coupled 1.55 μ m DFB lasers fabricated using nanoimprint lithography,” Electron. Lett. 47(6), 400–401 (2011). [CrossRef]
29. S. Murata, I. Mito, and K. Kobayashi, “Spectral characteristics for 1.5 µm DBR laser with frequency-tuning region,” IEEE J. Quantum Electron. 23(6), 835–838 (1987). [CrossRef]
30. S. Ogita, M. Yano, H. Ishikawa, and H. Imai, “Linewidth reduction in DFB laser by detuning effect,” Electron. Lett. 23(8), 393–394 (1987). [CrossRef]
31. K. Y. Liou, N. K. Dutta, and C. A. Burrus, “Linewidth narrowed distributed feedback injection lasers with long cavity length and detuned Bragg wavelength,” Appl. Phys. Lett. 50(9), 489–491 (1987). [CrossRef]
