Open Access
Add to BibSonomyAbstract
The distributed-coupling-coefficient and distributed-coupling-coefficient corrugation-pitch-modulated DFB lasers are experimentally demonstrated. The proposed lasers maintain good side mode suppression ratio over 50dBfrom 2.5 times to 12.5 times threshold current. The grating profiles of varying longitudinal parameters are equivalently obtained by specially designed sampled Bragg gratings and fabricated by conventional holographic exposure and μm-level photolithography.
© 2014 Optical Society of America
1. Introduction
Conventional quarter-wave shifted distributed feedback lasers are usually considered as single mode light sources for optical fiber communication systems. Stable single-longitude-mode (SLM) operation with high side mode suppression ratio (SMSR) is very important for practical applications. However, these lasers face the problems of mode hopping or SMSR degradation under high injection currents caused by longitudinal spatial hole burning (LSHB) [1,2]. There are two major approaches to reduce the sensitivity to LSHB [3]. The first is to flat the distribution of carrier and photon density along the laser by corrugation-pitch-modulated (CPM) structure [4] or multiple phase shifts (MPS) [5], etc. The second is to enlarge the gain margin, such as distributed-coupling-coefficient (DCC) structure [6]. Furthermore, DFB laser with longitudinal variation in both grating pitch and coupling coefficient profile has been studied in theory [3,7,8], which it has the advantages of the above methods and can offer improved static performance for DFB lasers. Such structure is usually named as distributed-coupling-coefficient corrugation-pitch-modulated (DCC-CPM). Although some methods have been proposed to fabricate the grating with varying coupling coefficient [9,10], they are all complicated in fabrication process, and therefore are not widely investigated and employed.
In this work, we use an equivalent method to realize the proposed varying longitudinal parameters based on the reconstruction equivalent chirp technique (REC) [11,12], which can equivalently realize complex phase shifts and chirps by specially designed sampled Bragg gratings (SBGs) with uniform seed grating. The grating profiles are fabricated by conventional holographic exposure and photolithography, both of which are flexible, fast and cost-effective. We experimentally demonstrate both DCC-DFB laser and DCC-CPMDFB laser based on REC technique, which have good SLM under high injection currents. To the best of our knowledge, the DCC-CPM DFB laser is experimentally reported for the first time.
2. Principle and design
Based on the Fourier analysis, the coupling coefficient of mth sub-grating κm in a SBG can be expressed as below [13],
Where κ0 is the coupling coefficient of uniform seed grating, γ is the duty cycle of the sampling structure. Hence the varying coupling coefficient κ ± 1 in ± 1st sub-grating can be obtained by changing γ. When γ = 0.5, κ ± 1 has the largest value. According to Ref [4], a phase-arranging-region (PAR) replaces the discrete phase shift in the center of the grating in CPM structure. Based on the REC technique, the sampling period P in the equivalent CPM structure of the mth sub-grating can be expressed as [14],Pc and Ps are the sampling periods in and out of the PAR, respectively. L is the length of the PAR, θ denotes the phase shift. The schematics of equivalent DCC and DCC-CPMDFB grating structures compared to a real DCC grating structure are shown in Fig. 1. In real DCC grating, the grating period is uniform along the cavity with a π phase shift inserted in the center. The grating depth is larger in the center region and smaller on both sides for a larger coupling coefficient in the center and a smaller coupling coefficient on the sides. In equivalent DCC and DCC-CPM structures, the seed grating Λ0 is uniform. According to Eq. (1), the duty cycle γ is less than 0.5 in side sections for smaller coupling coefficient κs. The sampling periods are the same and an equivalent π phase shift is inserted in the center of equivalent DCC structure. An equivalent PAR with π phase shift, in which the sampling period is different from the outside ones, locates in the center of equivalent DCC-CPM structure. The length of PAR corresponds to the center section Lc.
Fig. 1 The schematics of grating profile (a) real DCC, (b) equivalent DCC, (c) equivalent DCC-CPM(L: length,Λ0: seed uniform grating pitch, P: sampling period, γ: duty cycle of the sampling structure, and κ: coupling coefficient. The subscript c and s denote the parameters in the center and side section, respectively).
The static characteristics of the equivalent DCC and DCC-CPM DFB lasers are simulated using the transfer matrix (TMM) method [15]. TMM method simulates the DFB structures with complicated grating profiles based on the analytical solution of coupled-mode equations. In this method, the laser cavity is divided into a number of smaller sub-section and the field propagation acts as a transfer matrix in each section. The calculated light intensity distributions along the cavities are shown in Fig. 2(a). The output powers of the lasers are both about 4mW at the same injection current of 60mA. As comparison, the ones with equivalent π-phase shift and CPM DFB laser are also calculated. Comparing with the π phase shift DFB laser, the light intensity of the DCC DFB laser is flattened out around the position of the phase shift in the grating, which means a less LSHB effect. In the DCC-CPM structure, the distribution of light intensity becomes even smoother than that of CPM structure. Hence, the LSHB is reduced more effectively. The gain margins under different injection currents are shown in Fig. 2(b). The gain margin is defined as the difference between the net gain of the main mode and that of the first side mode [7]. Although the gain margin of DCC-CPM DFB laser is less than the one of DCC DFB laser, it keeps larger than that of π phase shift. Considering the role for reducing LSHB, we believe that the DCC-CPM DFB laser can have better static characteristics. The simulated lasing spectra of the equivalent DCC and DCC-CPM DFB lasers at the injection current of 60mA are shown in Figs. 2(c) and 2(d) based on the model in Ref [16].
Fig. 2 The simulated static characteristics of the equivalent DCC and DCC-CPM DFB lasers: (a) light intensity distributions, (b) gain margins under different injection currents, (c) (d) the corresponding spectra.
In our design, the proposed two lasers are fabricated on the same laser bar with a length of 300μm. The Bragg wavelength λB of 0th SBG grating is designed at 1490nm and the seed grating period is about 232nm. The −1st sub-grating (used as resonance wavelength for lasing) is near the material gain curve peak at 1550 nm. The coupling coefficient κ of seed grating is about 140cm−1. In both lasers, the lengths of Lc and Ls are 150μm and 75μm with the duty circle 0.5 and 0.27, respectively. Hence the coupling coefficient κc to κs equals 4:3. The sampling period of DCC laser is designed as 6.012μm. The sampling period in the side sections of DCC-CPM laser is designed as 6.012μm and that in the center section (the PAR section) is 6.137μm.
The devices are fabricated by a conventional two-stage lower-pressure metal organic vapor phase epitaxial (MOVPE). A schematic illustration of the laser is shown in Fig. 3(a). An InP buffer layer, a lower gradual AlxGayInzAs separate-confinement-heterostructure (SCH) layer, a five pairs of compressively strained AlxGayInzAsMQW, an upper gradual AlxGayInzAs SCH and a 1.3Q InGaAsP grating layer are successively grown on an S-doped n-type InP (100)-oriented substrate. The seeding grating and sampling structures [Fig. 3(b)] are formed by holographic exposure and photolithography. A p-type cladding InP layer and a p-InGaAs contact layer are re-grown by MOCVD. Then 2.5μm ridge waveguides are etched and buried with SiO2. Ti-Au patterned p-contacts and AuGeNi n-contacts are formed on the p-side and the n-side, respectively. The front and rear facets are coated with antireflection film (AR 1%) and high-reflection (HR 90%) film, respectively.
3. Experimental results
The devices are tested as laser bar under CW operation. A thermoelectric cooler (TEC) is used to control the temperature. The light-current (L-I) characteristics of the DCC and DCC-CPM DFB laser at different ambient temperatures are shown in Fig. 4. The threshold currents of the lasers are both 16mA at 25°C and the slope efficiencies are about 0.25W/A. With the temperature adding to 45°C, the threshold currents increase to 22mA and the slope efficiencies keep about 0.23 W/A. The spectra of the proposed laser at 20mA and 200mA at room temperature (~25 þC) are shown in Fig. 5. The SMSRs of DCC and DCC-CPM DFB laser reach 41.5dB and 43.4dB at 20mA. The lasers both operate in good SLM with SMSR over 55dB at 200mA.The optical spectra in wide wavelength range, which are inset in Figs. 5(c) and 5(d), show that the 0th channel is suppressed completely. In comparison, the SMSR degradations are observed under 100mA in equivalent π-phase shift DFB lasers fabricated on the same wafer.
Fig. 5 The measured lasing spectra at 25 °C (a) DCC laser at 20mA, (b) DCC-CPM laser at 20mA, (c) DCC laser at 200mA, (d) DCC-CPM laser at 200mA (inset: the optical spectrum in wide wavelength range under the same conditions).
Figure 6 shows the lasing wavelengths and SMSRs of the proposed lasers under different injection currents. Similar red-shifts of both lasers are observed.
Fig. 6 The measured lasing wavelengths and SMSRs of the DCC and DCC-CPM DFB laser when injection current changes from 20mA to 200mA.
According to Ref [17], when the lasers operate below thresholds, the wavelengths show blue shift with growing injection current, due to the carrier densities in the active region increase dramatically which causes refractive index decreasing, while the thermal effect is not significant. When the lasers operate above threshold, in the case of continuous current injection, current heating results in the temperature rising and the thermal effect dominants the refractive index increasing. Hence, the lasing wavelengths shift towards longer wavelengths. Although a thermoelectric cooler (TEC) is used, heating cannot be fully compensated. Both lasers maintain large SMSRs over 50dB under injection currents changing from 2.5 times to 12.5 times threshold current. The largest SMSRs of DCC and DCC-CPM DFB lasers reach to 58.8dB and 59.21dB at 10 times threshold current. In general, the SMSR of the DCC-CPM DFB laser is a little larger than that of DCC DFB laser at the same injection current, due to the better elimination of LSHB.
The lasing characteristics at different ambient temperatures are tested. Figure 7(a) shows the spectra of the proposed lasers under 100mA at 25°C, 35°C and45°C and good SLM with SMSRs>50dB is maintained [Fig. 7(b)]. When ambient temperature reaches to 45°C, the largest injection currents for DCC-/DCC-CPM DFB laser keeping SMSRs>45dB are 180mA and 190mA, respectively. The SMSR will decrease rapidly with bigger injection current.
Fig. 7 The measured spectra (a) DCC and DCC-CPM DFB laser at 100mA with different ambient temperatures, (b) the corresponding SMSRs, (c) DCC DFB laser at 180mA (~45°C), (d) DCC-CPM DFB laser at 190mA (~45°C).
4. Conclusion
We report experimental results of equivalent DCC and DCC-CPM DFB lasers based on REC technique. The proposed lasers both operate in good SLM with high SMSRs under large injection current range and different ambient temperatures. The static characteristics of DCC-CPM DFB laser are slightly better than that of DCC DFB laser, due to the LSHB is reduced more effectively. The grating profiles of varying longitudinal parameters are equivalently obtained by specially designed SBGs. The proposed method offers an easy fabrication and cost effective way for high-end DFB laser manufactures.
Acknowledgments
This research was supported by the National Nature Science Foundation of China under Grant 61090392, the Key Programs of the Ministry of Education of China under Grant 20100091110005, National “863” project under Grand 2011AA010300, the Fundamental Research Funds for the Central Universities and PAPD, Jiangsu, China, the Research and Innovation Program for Postgraduates in Universities of Jiangsu (CXZZ12-0049), and the Scientific Research Foundation of Graduate School of Nanjing University (2012CL03).
References and links
1. P. Correc, “Stability of phase-shifted DFB lasers against hole burning,” IEEE J. Quantum Electron. 30(11), 2467–2476 (1994). [CrossRef]
2. J. E. A. Whiteaway, G. H. B. Thompson, A. J. Collar, and C. J. Armistead, “The design assessment of λ/4 phase-shifted DFB laser structures,” IEEE J. Quantum Electron. 25(6), 1261–1279 (1989). [CrossRef]
3. T. Fessant, “Influence of a nonuniform coupling coefficient on the static and large signal dynamic behavior of Bragg-detuned DFB lasers,” J. Lightwave Technol. 16(3), 419–427 (1998). [CrossRef]
4. M. Okai, N. Chinone, H. Taira, and T. Harada, “Corrugation-pitch-modulated phase-shifted DFB laser,” IEEE Photonics Technol. Lett. 1(8), 200–201 (1989). [CrossRef]
5. G. P. Agrawal, J. E. Geusic, and P. J. Anthony, “Distributed feedback lasers with multiple phase-shift regions,” Appl. Phys. Lett. 53(3), 178–179 (1988). [CrossRef]
6. B. S. K. Lo and H. Ghafouri-Shiraz, “Spectral characteristics of distributed feedback laser diodes with distributed coupling coefficient,” J. Lightwave Technol. 13(2), 200–212 (1995). [CrossRef]
7. T. Fessant, “Threshold and above-threshold analysis of corrugation-pitch-modulated DFB lasers with inhomogeneous coupling coefficient,” IEE Proc. Optoelectron. 144(6), 365–376 (1997).
8. T. Fessant, “Influence of a nonuniform coupling coefficient on the static and large signal dynamic behavior of Bragg-detuned DFB lasers,” J. Lightwave Technol. 16(3), 419–427 (1998). [CrossRef]
9. A. Talneau, J. Charil, A. Ougazzaden, and J. C. Bouley, “High power operation of phase-shifted DFB lasers with amplitude modulated coupling coefficient,” Electron. Lett. 28(15), 1395–1396 (1992). [CrossRef]
10. S. Nilsson, T. Kjellberg, T. Klinga, J. Wallin, K. Streubel, and R. Schatz, “DFB laser with nonuniform coupling coefficient realized by double-layer buried grating,” IEEE Photonics Technol. Lett. 5(10), 1128–1131 (1993). [CrossRef]
11. J. Li, H. Wang, X. Chen, Z. Yin, Y. Shi, Y. Lu, Y. Dai, and H. Zhu, “Experimental demonstration of distributed feedback semiconductor lasers based on reconstruction-equivalent-chirp technology,” Opt. Express 17(7), 5240–5245 (2009). [CrossRef] [PubMed]
12. Y. Shi, S. Li, L. Li, R. Guo, T. Zhang, R. Liu, W. Li, L. Lu, S. Tang, Y. Zhou, J. Li, and X. Chen, “Study of the multiwavelength DFB semiconductor laser array based on the reconstruction-equivalent-chirptechnique,” J. Lightwave Technol. 31(20), 3243–3250 (2013). [CrossRef]
13. V. Veerasubramanian, G. Beaudin, A. Giguere, B. LeDrogoff, V. Aimez, and A. G. Kirk, “Design and demonstration of apodizedcomb filters on SOI,” IEEE Photonics J. 4(4), 1133–1139 (2012). [CrossRef]
14. S. Li, R. Li, L. Li, R. Liu, L. Gao, and X. Chen, “Dual wavelength semiconductor laser based on reconstruction-equivalent-chirp technique,” IEEE Photonics Technol. Lett. 25(3), 299–302 (2013). [CrossRef]
15. T. Makino, “Transfer-matrix analysis of the intensity and phase noise of multisection DFB semiconductorlasers,” IEEE J. Quantum Electron. 27(11), 2404–2414 (1991). [CrossRef]
16. W. Fang, A. Hsu, S. L. Chuang, T. Tanbun-Ek, and A. M. Sergent, “Measurement and modeling of distributed-feedback lasers with spatial holeburning,” IEEE J. Sel. Top. Quantum Electron. 3(2), 547–554 (1997). [CrossRef]
17. X. Li and W.-P. Huang, “Simulation of DFB semiconductor lasers incorporating thermal effects,” IEEE J. Quantum Electron. 31(10), 1846–1855 (1995).
