Open Access
Add to BibSonomy
Abstract
To overcome the limitation of the small tuning range of 1.3-µm-wavelength distributed Bragg reflector (DBR) lasers using the carrier-plasma effect, we designed a DBR structure with InAlAs carrier confinement layers and an InGaAlAs core layer. We found that the enhanced carrier density and small effective mass of electrons in the core layer of the DBR regions resulted in a wide Bragg wavelength shift. The enhanced refractive-index change due to the new structure enabled us to fabricate the world’s first 1.3-µm-wavelength superstructure-grating DBR laser with a quasi-continuous tuning range of over 30 nm.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
As data traffic rapidly increases, optical communication systems using wavelength division multiplexers (WDMs) are becoming wide-spread. The wavelength tunable semiconductor laser is a key technology for these systems, and various tunable lasers have been extensively studied [1,2]. Distributed Bragg reflector (DBR) lasers using the carrier-plasma effect [3] are attractive light sources for intensity modulation and direct detection (IMDD-) optical communication systems employing WDMs because they have intrinsically fast and wide wavelength-tuning characteristics induced by carrier injection [4–6]. Although electrically tuned DBR lasers have a relatively wide linewidth, this issue is not problematic for the IMDD application. Moreover, tunable DBR lasers are expected to be used as light sources for other applications such as gas sensing [7,8] and biomedical imaging using optical coherence tomography (OCT) [9]. A dramatic increase in the tuning range has been demonstrated using special DBR structures such as sampled grating (SGs) [10,11] and, superstructure grating (SSGs) [12–15]. The entire wavelength-tuning range of over 50 nm has been demonstrated using periodic multiple-reflection-peak characteristics of these special DBR structures and Vernier effect [16,17].
Although a great deal of effort has been devoted to studying tunable DBR lasers using the carrier-plasma effect, most studies have focused on the C- (1530–1565 nm) and L-band (1565–1625 nm) wavelength ranges. On the other hand, the use of such lasers in the O-band (1260–1360 nm) has not been sufficiently investigated even though the O band is extensively used by optical fiber communication systems because the distortion of the optical signal due to the chromatic dispersion of single-mode fiber is at a minimum in this band. This lack of studies is because the refractive-index change induced by the carrier-plasma effect for the 1.3-µm wavelength is rather small compared with that for the 1.55-µm wavelength; hence, the wavelength of the laser can only be tuned over an extremely small range. For example, one study in the 1980s reported a 1.3-µm-wavelength DBR laser with a tuning range of less than 4 nm with forward injection current for the DBR region [18]. Until recently, there has been no major progress on 1.3-µm-wavelength tunable DBR lasers. At the beginning of the 2010s, a Vernier tuned (VT-) DBR laser [19,20], having periodic-multiple-reflection-peak characteristics in the DBR region, exhibited outstanding performance over a tuning range of 30 nm. Although VT-DBR lasers attracted much attention as a potential light source for OCT imaging, the details of the device characteristics, such as the Bragg wavelength shift ΔλB and refractive-index change, were not described.
Thus, in an attempt to enlarge the tuning range of 1.3-µm wavelength DBR lasers, we designed a new DBR structure using InAlAs carrier confinement layers (CCLs) and an InGaAlAs core layer. The refractive-index change induced by the injected carriers was dramatically enhanced thanks to the strong carrier confinement in the core layer and the small effective mass of electrons in InGaAlAs. Accordingly, a wide ΔλB, the widest to the best of our knowledge for 1.3-µm-wavelength DBR lasers using the carrier-plasma effect, was achieved. We also fabricated the world’s first 1.3-µm-wavelength SSG-DBR laser with InAlAs CCLs and an InGaAlAs core layer. The enhanced refractive-index change enabled us to achieve quasi-continuous tuning over 30 nm by using multiple-reflection-peak characteristics of the SSG-DBR regions and the Vernier effect. The tuning characteristics of 58 consecutive 100-GHz-spacing channels for the fabricated SSG-DBR laser were also examined.
2. Design concept for enhancing refractive-index change
2.1 Refractive-index change due to carrier-plasma effect
The entire tuning range of a DBR laser can be significantly widened by using special grating structures such as SGs and SSGs that have periodic multiple-reflection-peak characteristics. However, it is important to have a large Bragg wavelength shift ΔλB for a single reflection peak to achieve a wide and quasi-continuous tuning range, even when these special DBR structures with multiple-reflection-peak characteristics are used. In other words, the large refractive-index change induced by the carrier-plasma effect is essential for DBR lasers with multiple-reflection-peak characteristics.
As the current in the DBR region increases, the refractive index of the core layer decreases due to the carrier-plasma effect. Consequently, the Bragg wavelength of the DBR region shifts to the shorter side. The amount of refractive-index change Δn due to the carrier-plasma effect can be expressed as [21]
In general, the refractive index change induced by current injection is much smaller in 1.3-µm-wavelength DBR lasers than in 1.55-µm-wavelength DBR lasers. As might be expected, one of the reasons is the small squared value of the 1.3-µm wavelength (λ2), which is only about 0.7 that of the 1.55-µm wavelength. Two other reasons originate from the physical properties of the InGaAsP/InP-based material: the large carrier overflow due to the small conduction band offset ΔEc and the increase in me. To overcome these limitations, we use InAlAs CCLs to suppress the carrier overflow and InGaAlAs as the core layer material to decrease me. The details of these measures are explained in the following.
2.2 Introduction of carrier confinement layers
The conduction band offset between the core layer and InP cladding layer for 1.3-µm-wavelength DBR lasers is relatively small compared with that for 1.55-µm-wavelength DBR lasers. In 1.55-µm-wavelength DBR lasers, InGaAsP, with a bandgap wavelength λbg of around 1.4 µm, is generally used as a core layer of the DBR regions. In this case, ΔEc is relatively large because of the significant difference in bandgap energy between the core layer (Eg = 0.89 eV) and InP cladding layers (Eg = 1.29 eV). In contrast, a material with a λbg of less than about 1.2 µm should be used for the core layer of 1.3-µm-wavelength DBR lasers to suppress optical absorption loss in the core layer. In other words, a material with a relatively large bandgap energy of more than 1.03 eV should be used as the core layer. In so doing, ΔEc between the core layer and InP cladding layers becomes smaller as the bandgap energy of the core layer increases. This increases the carrier overflow in the conduction band and decreases the carrier densities in the core layer. A schematic diagram of the band structure of the core layer in this case is shown in Fig. 1(a).
Fig. 1. Schematic band structure of (a) conventional DBR structure and (b) newly designed DBR structure with InAlAs CCLs.
To suppress the carrier overflow, we put CCLs between the core layer and each of the InP cladding layers. The CCLs were made from InAlAs material, which is lattice matched to InP and has a relatively large bandgap energy (1.45 eV) compared with that of InP (1.29 eV) [22]. As can be seen in Fig. 1(b), the two CCLs adjoining the p-InP and n-InP cladding consist of p- and n-doped InAlAs, and these layers strongly confine electrons and holes to the core layer. Δn of the core layer in the DBR region can be enhanced even in 1.3-µm-wavelength DBR lasers because the CCLs suppress the carrier overflow and increase the carrier densities in the core layer.
2.3 Reduction of effective mass of electrons
As mentioned above, the second reason for the small Δn of 1.3-µm-wavelengthDBR lasers is the increase in me in the core layer that arises from the different material composition of the core layer. Figure 2 shows the calculated effective electron mass in InGaAsP and InGaAlAs, which are lattice-matched to InP. me in InGaAsP changes by about 27% as λbg goes from 1.4 to 1.1 µm. Note as well that it has been shown that me increases as the composition of As becomes smaller [23]. Thus, to decrease me in the core layer and increase Δn, we decided to use InGaAlAs for the core layer. Here, me in InGaAlAs is relatively small compared with that in InGaAsP for all λbg [24]. me in InGaAlAs with a λbg of 1.1 µm is small enough to change the refractive index as much as when using InGaAsP core layer with a λbg of 1.4 µm in 1.55-µm-wavelength DBR lasers. Accordingly, we decided to fabricate a 1.3-µm-wavelength DBR laser consisting of an InGaAlAs lattice matched to InP with a λbg of 1.1 µm for the core layer.
3. Enhancement of refractive-index change in fabricated 1.3-µm-wavelength InGaAlAs-based DBR lasers
3.1 Device structure of InGaAlAs-based DBR lasers
To confirm the validity of our concept to enlarge Δn of 1.3-µm-wavelength DBR lasers, we fabricated a basic DBR laser with single-reflection-peak characteristics. Figure 3 shows a schematic diagram of this laser. The laser has a ridge waveguide structure. The laser cavity consists of an active region, phase control region, and front/rear DBR regions. The front and rear DBR regions have an InGaAlAs bulk core layer with a λbg of 1.1 µm and are butt-jointed to the active region. Their lengths are 340 and 650 µm, respectively. Moreover, a semiconductor optical amplifier (SOA) is integrated with the DBR laser. To enhance the carrier density in the core layer, 10-nm-thick InAlAs CCLs that are lattice matched to InP are put at both ends of the core layer, as shown in Fig. 4(a) (Sample A). The doping concentrations of the p- and n-InAlAs CCLs are 1.0 × 1018 cm-3. To confirm the effect of the CCLs on ΔλB, we also fabricated a DBR laser without InAlAs CCLs in the DBR regions by using the same fabrication process (Sample B). The cross-sectional structure of the DBR region of Sample B is shown in Fig. 4(b). The only difference between this laser and Sample A is the absence of CCLs in the DBR regions.
Fig. 4. Cross sectional structure of DBR region of (a) DBR laser with InAlAs CCLs (Sample A) and (b) DBR laser without InAlAs CCLs (Sample B).
3.2 Evaluation of refractive-index change in 1.3-µm-wavelength InGaAlAs-based DBR lasers
The measured lasing spectra of Sample A under various DBR current conditions are shown in Fig. 5(a). The currents injected into the active region and the SOA were 90 and 40 mA, respectively. First, the currents injected into the front and rear DBR regions were set to zero. Then, stable single-mode operation was confirmed for both lasers. Under these conditions, the lasing wavelength of Sample A was 1296.7 nm, as shown in Fig. 5(a). To change the lasing wavelength, the current injected into the front DBR Ifront was increased from 0 to 81 mA. The current injected into the rear DBR Irear was tuned to have the same current density as that of the front DBR region. In other words, the Bragg wavelengths of the DBRs were tuned to be the same. The injection current for the phase-control region Iphase was also tuned to make the lasing wavelength coincide with the Bragg wavelength of the DBR regions. Therefore, the wavelength shift shown in Fig. 5(a) is equal to ΔλB. As Ifront increases, the lasing wavelength becomes shorter because the equivalent refractive index of the DBR regions decreases due to the carrier-plasma effect. As can be seen in Fig. 5(a), stable single-mode operation was maintained over the entire wavelength tuning range of Sample A. The measured lasing wavelength shifts of Samples A and B are plotted in Fig. 5(b). For Sample B, the maximum ΔλB was about 3.0 nm. The measured wavelength shift is slightly larger than that of the InGaAsP-based 1.3-µm-wavelength DBR laser reported in the 1980s [18], but it is significantly smaller than that for 1.55-µm-wavelength DBR lasers [5,25]. The small me in the InGaAlAs core layer with a λbg of 1.1 µm can be attributed to the relatively large Δn even for the 1.3-µm wavelength. Furthermore, we can clearly see that ΔλB of Sample A was much larger than that of Sample B. The maximum ΔλB was 5.1 nm, confirming that the carrier density in the InGaAlAs core layer can be effectively increased using InAlAs CCLs. For both samples, the wavelength shift was very small when Ifront1/2 is less than 2 mA1/2. This is probably due to non-radiative surface recombination on the sidewall of the ridge mesa structure. In this small current region, the injected carriers fill the non-radiative recombination center and do not contribute to the carrier density in the core layer. The non-radiative recombination on the surface can be suppressed, and the wavelength shift in small current region can be improved by optimizing the dry and wet etching processes used to form the ridge mesa structure.
Fig. 5. (a) Lasing spectra for DBR laser with CCLs (Sample A), and (b) measured ΔλB due to the carrier injection for DBR lasers with and without CCLs.
In addition, a further increase in the amount of wavelength shift can be expected by increasing λbg of the InGaAlAs core layer. A λbg around 1.2 µm would produce a large refractive index change and sufficiently suppress the optical absorption loss. In this case, the difference in energy between λbg of the core layer and the lasing wavelength is almost the same as that of conventional 1.55 µm wavelength DBR lasers using a core layer with a λbg of 1.4 µm. Furthermore, there is still room to enhance the carrier confinement effect by optimizing the thickness and strain of InAlAs CCLs.
The change in ΔλB and the equivalent refractive-index change Δneq/neq of the fabricated DBR lasers were compared with those of conventional tunable lasers using the carrier-plasma effect (Table 1). The relationship between ΔλB and Δneq can be written as,
where λ is the lasing wavelength. Δneq/neq of Sample A is about 1.6 times larger than that of Sample B. This change is more than double that of the 1.3-µm-wavelength DBR laser reported in the 1980s [18]. Although another 1.3-µm-wavelength VT-DBR laser was demonstrated in 2014, the details of its structure and tuning characteristics, including ΔλB, were not described [20]. Therefore to the best of our knowledge, ΔλB of Sample A is the widest difference reported so far for 1.3-µm-wavelength DBR lasers using the carrier-plasma effect. A tuning range of over 12 nm was recently demonstrated for 1.55-µm-wavelength DBR lasers with single-reflection-peak characteristics [25], which gives a Δneq/neq about two times larger than that of Sample A. As described above, the large conduction band offset and small effective mass for 1.55-µm-wavelength DBR lasers are attributed to the large Δn. A 1.55-µm-wavelength SSG-DBR laser with a tuning range of over 30 nm was demonstrated; its ΔλB was 4.5 nm [13]. Thus, we conclude that ΔλB of Sample A is comparable to those of the previously reported 1.55-µm-wavelength DBR lasers and is sufficient for a DBR laser that has periodic multiple-reflection-peak characteristics over a tuning range exceeding 30 nm.
Table 1. Comparison of Bragg wavelength shift ΔλB and equivalent refractive index change Δneq/neq between Sample A and conventional tunable lasers using carrier-plasma effect.
4. 1.3-µm-wavelength SSG-DBR laser
4.1 Design of 1.3-µm-wavelength SSG-DBR laser
On the basis of the above results, we designed and fabricated a 1.3-µm-wavelength SSG-DBR laser. An SSG-DBR is a special DBR with periodic multiple-reflection-peak characteristics. It enables us to dramatically enhance the entire wavelength tuning range because the lasing wavelength can be adjusted by selecting an SSG mode among the reflection peaks. The SSG has a diffraction grating, the phase (or frequency) of which is periodically modulated. Theoretical analyses of SSGs are presented in our previous studies [13–15].
The SSGs for the fabricated 1.3-µm-wavelength SSG-DBR laser were designed in consideration of the ΔλB of Sample A described above. The calculated reflection spectra of the front and rear SSG-DBR regions are shown in Fig. 6. The lengths of the front and rear regions were 340 and 650 µm, respectively. The coupling coefficient κi was set to 40 cm-1 for both regions. Note that the reflection-peak spacing for both regions (Δλfront and Δλrear) should be designed to be smaller than the maximum wavelength shift for a single SSG-mode. Otherwise, quasi-continuous tuning characteristics using all SSG modes cannot be achieved because wavelength gaps would be generated between SSG modes. The wavelength shift for a single SSG mode ΔλSSG is equal to ΔλB for the basic DBR laser with single-reflection-peak characteristics. Therefore, the reflection-peak spacing was designed to be 4.3 nm for the front SSG-DBR region and 3.7 nm for the rear region by considering the measured maximum ΔλB (5.1 nm). In addition, there needs to be a slight difference in the reflection-peak spacing between the front and rear SSG-DBR regions in order to use the Vernier effect. Accordingly, the quasi-continuous tuning range of the SSG-DBR laser Δλqc can be estimated as [15]
where N is the number of reflection peaks, and ΔλS (=Δλfront - Δλrear) is the offset between the front and rear reflection-peak spacings. As is clear from Eq. (3), Δλqc increases with N. However, there is a limitation on Δλqc because the reflectivity of each reflection peak becomes smaller as N increases. For an SSG-DBR with N reflection peaks, the reflectivity of each peak is 1/N1/2 times smaller than that for a uniform-grating DBR with the same coupling coefficient [15]. A large N may degrade lasing characteristics, such as by causing instability in the oscillation wavelength. To ensure sufficient lasing characteristics together with a wide quasi-continuous tuning range of more than 30 nm, the front and rear SSG-DBR regions were designed to have nine reflection peaks (N = 9). The estimated quasi-continuous tuning range for this laser was about 35 nm.4.2 Wavelength-tuning characteristics
The basic device structure of the fabricated SSG-DBR laser is the same as the DBR laser with single-reflection-peak characteristics shown in Fig. 2. The main difference is the grating structure. The newly designed SSGs explained above were formed on both front and rear DBR regions. We again fabricated SSG-DBR lasers with and without CCLs to confirm the effect of using CCLs. The cross-sectional structures of the DBR regions were the same as the DBR lasers described in Section 3 (Fig. 3). The active and SOA regions have InGaAlAs multi-quantum wells (MQWs) with a photoluminescence (PL) wavelength of 1.28 µm. The lengths of the active and SOA regions were 350 and 420 µm, respectively. The only difference in device structure between these devices was the presence or absence of the InAlAs CCLs in the DBR and phase control regions.
First, we evaluated the lasing characteristics of the fabricated SSG-DBR laser with InAlAs CCLs. The operation temperature of the SSG-DBR laser chip was set to 15°C. The threshold current was 9 mA when Ifront, Irear and Iphase were set to 0 mA. Under these conditions, a fiber coupled light intensity of 11.0 dBm was achieved when the currents in the active region (Iact) and SOA (ISOA) were set to 90 and 40 mA, respectively. Then, the basic wavelength tuning characteristics were measured. In this measurement, Ifront and Irear were varied independently; i.e., when one was swept, the other was set to 0 mA. The measured lasing wavelengths are plotted in Fig. 7. The dependence on Ifront is on the left side, and the dependence on Irear is on the right. When Ifront and Irear were both set to 0 mA, the lasing wavelength was 1294.0 nm. As Ifront was increased, the lasing wavelength hopped to a longer SSG mode. In contrast, the lasing wavelength hopped to a shorter mode when Irear was increased. In total, nine SSG modes were confirmed, as shown in the figure. This indicates that the 1.3-µm-wavelength SSG-DBR laser performed well and all SSG modes could be used for wavelength tuning.
Next, the lasing wavelength for the various combinations of Ifront and Irear were evaluated. Ifront was changed from 0 to 50 mA in 1.0-mA steps, and Irear was changed from 0 to 80 mA in 1.0-mA steps. Iphase was set to 0 mA. The lasing wavelengths were measured for all combinations of Ifront and Irear, and are plotted in Fig. 8, where (a) is for the SSG-DBR laser without CCLs and (b) is for the laser with CCLs. Fitting curves are also plotted. We can clearly see that the nine curves correspond to the nine SSG modes. Several points are shifted from the trend of the fitting curves. These points can be accurately tuned to coincide with the trend of the fitting curves by varying Iphase. For the SSG-DBR laser without CCLs, wavelength gaps were generated between SSG modes because the lasing wavelength shift for a single SSG mode due to the carrier-plasma effect was relatively small. In addition, the wavelength shift for a single SSG mode saturated when Ifront reached 50 mA. These results mean that quasi-continuous tuning characteristics using multiple-reflection-peak characteristics cannot be achieved by increasing Ifront and Irear. On the other hand, there are no wavelength gaps between SSG modes in the case of the SSG-DBR laser with CCLs. The measured wavelength shift ΔλSSG for the SSG-DBR laser with CCLs was 4.5 nm when Ifront was changed from 0 to 50 mA. Because ΔλSSG for a single SSG-mode was sufficiently larger than the measured SSG-mode spacing for the front SSG-DBR Δλfront of 4.0 nm, there was a wide wavelength-tuning range without gaps, due to the use of all nine SSG modes. The estimated total tuning range of the SSG-DBR laser with CCLs was 32.1 nm. These results indicate that placing InAlAs CCLs in DBR regions enables 1.3-µm wavelength SSG-DBR lasers using multiple SSG-modes to achieve quasi-continuous wavelength-tuning characteristics.
Fig. 8. Lasing wavelength under various current conditions for DBR laser (a) with CCLs and (b) without CCLs.
Figure 9 shows a contour plot of the measured lasing wavelength for all combinations of Ifront and Irear for the SSG-DBR laser with InAlAs CCLs. The contour plot is divided into several areas. Each area corresponds to a different SSG mode. The nine SSG modes are visible in the plot. Some areas, such as those of the 7th or 9th SSG modes, are relatively small compared with the other areas because of their unstable oscillation conditions. As shown in Fig. 6, these SSG-DBRs have some fluctuation in reflectivity between each reflection peak. The relatively small reflectivity of several reflection peaks causes the corresponding SSG modes to have unstable oscillation characteristics. This problem can be alleviated by optimizing the design of the SSG.
Finally, we evaluated the lasing characteristics of the fabricated SSG-DBR laser with InAlAs CCLs for 58 consecutive channels with a 100-GHz spacing. The measured spectra for all channels are plotted in Fig. 10. Iact and ISOA were 90 and 40 mA, respectively. For each channel, Ifront, Irear and Iphase were varied in order to tune the lasing wavelength and maximize the side mode suppression ratio (SMSR). The lasing wavelength could be accurately tuned from the wavelength of 1280.2 nm for the 1st channel to 1312.0 nm for the 58th channel by using all nine SSG modes. As shown in Fig. 10, single-mode operations was possible for all channels. The lasing characteristics for all channels are summarized in Fig. 11. The upper graph shows the required Ifront and Irear to tune the lasing wavelength for all channels. The measured lasing wavelength, fiber coupled light intensity, and SMSRs for all channels are plotted in the lower graph. The required Ifront and Irear to tune the lasing wavelength were less than 50 mA for almost all channels. The SMSRs of all channels were over 35 dB, and they deteriorated for the channels on the shorter wavelength side because of the relatively small optical gain at those wavelengths. As can be seen from Fig. 10, the estimated optical gain peak of the active region is over 1300 nm. Therefore, the optical gain of the channels on the shorter wavelength side is smaller than that of the channels on the longer side. These small SMSRs on the short wavelength side can be improved by optimizing the design of the wavelength detuning between the center wavelength of the tuning range and the optical gain peak. The average light intensity of all channels was 8.9 dBm, and the fluctuation in light intensity across all channels was less than 2 dB. These results indicate that the new DBR structure enabled us to increase Δn due to the carrier-plasma effect and fabricate, for the first time, 1.3-µm-wavelength SSG-DBR lasers with comparable performance to conventional 1.55-µm-wavelength SSG-DBR lasers.
Fig. 11. (a) Required Ifront and Irear and (b) measured lasing wavelength, light intensity, and SMSRs for all 58 consecutive channels with 100-GHz spacing.
5. Summary
The tuning range of 1.3-µm-wavelength DBR lasers using the carrier-plasma effect was successfully widened by using our new DBR structure. We used InAlAs CCLs to suppress the carrier overflow and increase the carrier density in the core layer in the DBR regions. We also used InGaAlAs for the core layer to decrease the effective mass of electrons and increase the refractive-index change due to the carrier-plasma effect. To confirm the effectiveness of the new DBR structure, we first evaluated the Bragg wavelength shift of a basic DBR laser with single-reflection-peak characteristics. The amount of Bragg wavelength shift due to the carrier-plasma effect was 5.1 nm. The estimated refractive-index change was about 0.39%, a significant improvement compared with the value for a DBR laser without CCLs. Next, we fabricated a 1.3-µm-wavelength SSG-DBR laser using the new DBR structure. We designed the front and rear SSGs with nine reflection peaks and reflection-peak spacings of 4.3 nm for the front and 3.7 nm for the rear SSG-DBR. Then, we evaluated the lasing characteristics of the fabricated SSG-DBR laser for 58 consecutive channels with 100-GHz spacing. The lasing wavelength could be accurately tuned with SMSRs of over 35 dB and fluctuations in light intensity of less than 2 dB for all channels. These results indicate that the wavelength-tuning range was dramatically widened by using InAlAs CCLs and an InGaAlAs core layer. This is the world’s first 1.3-µm-wavelength SSG-DBR laser with a 32-nm quasi-continuous wavelength tuning range.
Disclosures
The authors declare no conflicts of interest.
References
1. 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]
2. J. Buus and E. J. Murphy, “Tunable lasers in optical networks,” J. Lightwave Technol. 24(1), 5–11 (2006). [CrossRef]
3. B. R. Bennett, R. A. Soref, and J. A. Del Alamo, “Carrier-induced change in refractive index of InP, GaAs, and InGaAsP,” IEEE J. Quantum Electron. 26(1), 113–122 (1990). [CrossRef]
4. Y. Tohmori, Y. Suematsu, Y. Tushima, and S. Arai, “Wavelength tuning of GaInAsP/InP integrated laser with butt-jointed built-in distributed Bragg reflector,” Electron. Lett. 19(17), 656–657 (1983). [CrossRef]
5. S. Murata, I. Mito, and K. Kobayashi, “Tuning range for 1.5 µm wavelength tunable DBR lasers,” Electron. Lett. 24(10), 577–579 (1988). [CrossRef]
6. N. Fujiwara, T. Kakitsuka, M. Ishikawa, F. Kano, H. Okamoto, Y. Kawaguchi, Y. Kondo, Y. Yoshikuni, and Y. Tohmori, “Inherently mode-hop-free distributed Bragg reflector (DBR) laser array,” IEEE J. Sel. Top. Quantum Electron. 9(5), 1132–1137 (2003). [CrossRef]
7. A. S. Diba, F. Xie, B. Gross, L. C. Hughes, C. Zah, and F. Moshary, “Application of broadly tunable SG-DBR QCL for multi-species trace gas spectroscopy,” Opt. Express 23(21), 27123–27133 (2015). [CrossRef]
8. M. Abe, T. Kanai, N. Fujiwara, Y. Ohiso, H. Ishii, M. Shimokozono, H. Mastuzaki, R. Kasahara, and M. Itoh, “4-nm continuous rapid sweeping spectroscopy in 2-µm band using distributed Bragg reflector laser,” Appl. Phys. B 123(10), 260 (2017). [CrossRef]
9. D. Choi, R. Yoshimura, and K. Ohbayashi, “Tuning of successively scanned two monolithic Vernier-tuned lasers and selective data sampling in optical comb swept source optical coherence tomography,” Opt. Express 4(12), 2962–2987 (2013). [CrossRef]
10. V. Jayaraman, D. A. Cohen, and L. A. Coldren, “Demonstration of broadband tenability in a semiconductor laser using sampled gratings,” Appl. Phys. Lett. 60(19), 2321–2323 (1992). [CrossRef]
11. B. 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]
12. Y. Tohmori, Y. Yoshikuni, H. Ishii, F. Kano, T. Tamamura, Y. Kondo, and M. Yamamoto, “Broad-range wavelength-tunable superstructure grating (SSG) DBR lasers,” IEEE J. Quantum Electron. 29(6), 1817–1823 (1993). [CrossRef]
13. H. Ishii, Y. Tohmori, F. Kano, Y. Yoshikuni, and Y. Kondo, “Multiple-phase-shift super structure grating DBR lasers,” IEEE Photonics Technol. Lett. 5(6), 613–615 (1993). [CrossRef]
14. H. Ishii, Y. Tohmori, T. Tamamura, and Y. Yoshikuni, “Super structure grating (SSG) for broadly tunable DBR lasers,” IEEE Photonics Technol. Lett. 5(4), 393–395 (1993). [CrossRef]
15. 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]
16. H. Ishii, H. Tanobe, F. Kano, Y. Tohmori, Y. Kondo, and Y. Yoshikuni, “Broad-range wavelength coverage (62.4 nm) with superstructure-grating DBR laser,” Electron. Lett. 32(5), 454–455 (1996). [CrossRef]
17. L. A. Coldren, “Monolithic tunable diode lasers,” IEEE J. Sel. Top. Quantum Electron. 6(6), 988–999 (2000). [CrossRef]
18. M. Yamaguchi, M. Kitamura, S. Murata, I. Mito, and K. Kobayashi, “Wide range wavelength tuning in 1.3 µm DBR-DC-PBH-LDs by current injection into the DBR region,” Electron. Lett. 21(2), 63–65 (1985). [CrossRef]
19. M. P. Minneman, J. Ensher, M. Crawford, and D. Derickson, “All-semiconductor high-speed akinetic swept-source for OCT,” Proc. SPIE 8311, 831116 (2011). [CrossRef]
20. M. Bonesi, M. P. Minneman, J. Ensher, B. Zabihian, H. Sattmann, P. Boschert, E. Hoover, R. A. Leitgeb, M. Crawford, and W. Drexler, “Akinetic all-semiconductor programmable swept-source at 1550 nm and 1310 nm with centimeters coherence length,” Opt. Express 22(3), 2632–2655 (2014). [CrossRef]
21. C. H. Henry, R. A. Logan, and K. A. Bertness, “Spectral dependence of the change in refractive index due to carrier injection in GaAs lasers,” J. Appl. Phys. 52(7), 4457–4461 (1981). [CrossRef]
22. D. K. Gaskill, N. Bottka, L. Aina, and M. Mattingly, “Band-gap determination by photoreflectance of InGaAs and InAlAs lattice matched to InP,” Appl. Phys. Lett. 56(13), 1269–1271 (1990). [CrossRef]
23. R. J. Nicholas, S. J. Session, and J. C. Portal, “Cyclotron resonance and the magnetophonon effect in GaxIn1−xAsyP1−y,” Appl. Phys. Lett. 37(2), 178–180 (1980). [CrossRef]
24. R. F. Kopf, H. P. Wei, A. P. Perley, and G. Livescu, “Electron effective mass and band-gap dependence on alloy composition of AlyGaxIn1−y−xAs, lattice matched to InP,” Appl. Phys. Lett. 60(19), 2386–2388 (1992). [CrossRef]
25. L. Han, S. Liang, H. Wang, L. Qiao, J. Xu, L. Zhao, H. Zhu, B. Wang, and W. Wang, “Electroabsorption-modulated widely tunable DBR laser transmitter for WDM-PONs,” Opt. Express 22(24), 30368–30376 (2014). [CrossRef]
