Wavelength-tunable prism-coupled external cavity passively mode-locked quantum-dot laser*
Wu Yan-Hua, Jian Wu, Jin Peng†, Wang Fei-Fei, Hu Fa-Jie, Wei Heng, Wang Zhan-Guo
Key Laboratory of Semiconductor Materials Science and Beijing Key Laboratory of Low-dimensional Semiconductor Materials and Devices, Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China

Corresponding author. E-mail: pengjin@semi.ac.cn

*Project supported by the National Natural Science Foundation of China (Grant No. 61274072) and the National High Technology Research and Development Program of China (Grant No. 2013AA014201).

Abstract

A wavelength-tunable mode-locked quantum dot laser using an InAs/GaAs quantum-dot gain medium and a discrete semiconductor saturable absorber mirror is demonstrated. A dispersion prism, which has lower optical loss and less spectral narrowing than a blazed grating, is used for wavelength selection and tuning. A wavelength tuning range of 45.5 nm (from 1137.3 nm to 1182.8 nm) under 140-mA injection current in the passive mode-locked regime is achieved. The maximum average power of 19 mW is obtained at the 1170.3-nm wavelength, corresponding to the single pulse energy of 36.5 pJ.

Keyword: 81.07.Ta; 42.60.Fc; 42.79.Bh; 81.16.Dn; quantum dot; mode-locked laser; prism-coupled external cavity; tunability
1. Introduction

External-cavity (EC) passively mode-locked quantum dot (QD) lasers are excellent candidates for generating ultra-short pulses, due to QD lasers’ inherent merits of low threshold current density, [13] temperature insensitivity, [46] and ultrafast carrier dynamics.[79] The broad gain spectrum due to the inherent dot size fluctuation during the growth process of self-assembled QDs is very promising for the development of broadly tunable laser[1021] and broadly tunable high-power picosecond source.[22, 23] Furthermore, EC configurations of such lasers offer some useful features when compared to monolithic devices. For instance, monolithic mode-locked lasers’ inability to generate relatively low repetition-rate pulses can be remedied easily by EC mode-locked lasers, since the EC configuration greatly extends the cavity length of the laser system. Appropriate optical elements can be introduced to realize wavelength tuning and dispersion compensation for pulse narrowing. In this context, broad wavelength tunability from an InAs/GaAs QD pulse laser has been reported using diffraction grating techniques. Kim et al. reported the first wavelength-tunable grating-coupled mode-locked EC QD laser in 2006; it demonstrated a continuous optical spectrum tuning range of 30  nm– 50.5  nm from ground-state and excited-state passive mode-locking, respectively.[21] The widest tunability range was demonstrated in Ref.  [22], where the gain chip used in the grating coupled external cavity was based on a QD structure that consisted of 10 non-identical InAs/GaAs QD layers. Owing to the inherent broad gain and absorption bandwidth afforded by QD structures, a broad (136.5  nm) wavelength tunability in a passive mode-locked regime has been demonstrated.

In this letter, we demonstrate a tunable mode-locked EC QD laser using the dispersion prism as the tuning element, instead of the traditional blazed grating. The former offers some advantages compared with the latter. First, the light loss caused by the prism is smaller and ensures a higher output power than the blazed grating. Second, the narrowing effect of the spectral width of the EC mode-locked laser with the prism is reduced significantly because of its relatively low light dispersion, resulting in a reduction in the broadening of the laser output optical pulse duration. By using a gain device with self-assembled InAs/GaAs QDs as its active region, we present 45.5-nm wavelength tunability between 1137.3  nm and 1182.8  nm in the mode-locked regime from the prism-coupled mode-locked EC QD laser. The maximum average power of 19  mW is achieved at the wavelength of 1170.3  nm at room temperature.

2. Experiment

The tunable mode-locked EC QD laser, as shown in Fig.  1, was realized by employing a simple linear EC configuration, incorporating a QD gain device, a collimating lens, a focusing lens, a SESAM and an equilateral prism as the dispersion element. The detailed description of the scheme of the epitaxial structure of the QD gain device can be found in our previous papers.[24, 25] Then the QD epitaxial chip was processed to fabricate the gain device, using standard photolithography and a wet etching process. The gain device had a ridge waveguide structure with a waveguide width of 5  μ m and a device length of 1.5  mm. Both facets were coated: a single λ /4 SiO2 anti-reflection (AR) layer designed to minimize the reflectivity at 1.16  μ m was deposited on one facet; SiO2/Ta2O5 double layers were deposited on the other facet with designed reflectivity of 20% at the same wavelength. Finally, the gain device was mounted epitaxial-side down on a copper heat sink.

Light emission from the AR-coated facet was coupled into the EC using an 8-mm focal-length collimating lens, and then passed through the prism, and finally focused onto the SESAM using another 8-mm focal-length focusing lens. The saturation fluence of the SESAM (Batop SAM-1150-6) was around 60  μ J/cm2 and the reflectivity exceeded 94% for wavelengths ranging from 1.11  μ m to 1.19  μ m. As the SESAM was fixed, the light through the prism had a deflection angle of a certain value (a calculated deflection angle fixed at 63° ), the different wavelengths of light were matched to the deflection angle by rotating the prism, and hence the wavelength tuning was achieved. Mode-locked pulses were detected by a high-speed InGaAs photodiode detector whose output signal was connected to a radio-frequency (RF) spectrum analyzer with a bandwidth of 3  GHz. Measurements of emission spectrum, light power, and RF spectrum were all performed under continuous-wave (CW) current injection at room temperature. In this work, the power value measured by the pyro-electric probe is the average power of the mode-locked pulses, not the peak one.

Fig.  1. Configuration of the mode-locked EC QD laser.

3. Results and discussion

A fundamental mode-locking wavelength tuning range of 40.9  nm (1141.9  nm– 1182.8  nm) is achieved for the mode-locked EC QD laser under 120-mA current applied to the gain device, and an extended tuning of 45.5  nm (1137.3  nm– 1182.8  nm) is demonstrated at the current injection of 140  mA, as shown in Fig.  2. The full-width at half maximum (FWHM) of the optical spectra of the laser varies between 1.2  nm and 2.6  nm across the tuning range, which is smaller than the feedback bandwidth (2.8  nm) of the prism. This proves that the prism used in the laser could not result in spectrum narrowing. The tuning bandwidth of the mode-locked EC QD laser is mainly affected by the reflection bandwidth of SESAM and operating current. The reflection bandwidth of SESAM is 1110  nm– 1190  nm, so the tuning bandwidth of the mode-locking is limited in this range.

In Fig.  3, the RF characteristics for one representative operation wavelength of 1166.2  nm are presented. The repetition rate of 520  MHz is realized for the EC optical length. The − 3-dB RF linewidth of 23  kHz is obtained and the signal-to-noise ratio in the RF spectra is larger than 30  dB, showing the low noise and robust stability for mode-locked operation, as shown in Fig.  3(b).

Fig.  2. Optical spectra of the wavelength-tunable EC mode-locked QD laser under 140-mA current applied to the gain device.

From the PI measurements at different tuning wavelengths, we obtain the dependence of the threshold current and of the output power on the lasing wavelength, as shown in Fig.  4. The lowest threshold current of 52  mA is obtained at a lasing wavelength of 1170.3  nm; meanwhile, the average power of the mode-locked pulses reaches a maximum value of 19  mW, corresponding to the single pulse energy of 36.5  pJ. Compared to the average power of 20  mW of the mode-locked pulses in the mode-locked EC QD laser with a SESAM employing the same QD gain device without the prism at the same wavelength and with the same injection current in our previous paper, [26] the output power is only slightly reduced due to the introduction of the prism, indicating that the dispersion prism has not induced a significant degree of light loss.

Fig.  3. RF spectral trace for mode locking under 120-mA current at 1166.2-nm wavelengths: (a) 3-GHz scan range, measured with a resolution bandwidth of 100  kHz; (b) 0.6-MHz scan range measured with a resolution bandwidth of 30  kHz.

Fig.  4. Threshold current and average power under 120-mA and 140-mA injection currents, each as a function of the tuning wavelength of the mode-locked EC QD laser.

According to the calculation of the saturation parameter S in our previous paper, [27] we obtain saturation parameters S of the SESAM at different lasing wavelengths, as shown in Fig.  5. Typically, the SESAM operates with a saturation parameter S from 3 to 5[28] as marked in Fig.  5, since the mode-locked laser can produce the shortest pulse duration under such conditions.[26] The saturation parameters S of the SESAM at different lasing wavelengths and different injection currents are close to ideal.

Fig.  5. Saturation parameters S of the SESAM at the different lasing wavelengths. The two dashed lines indicate the ideal operating range of the SESAM.

4. Conclusion

In conclusion, we demonstrate a wavelength-tunable mode-locked EC QD laser using the prism as the wavelength tuning element. A broad wavelength tuning range of 45.5  nm (from 1137.3  nm to 1182.8  nm) under 140-mA injection current in a passive mode-locked regime is achieved. The maximum average power of 19  mW is obtained at the wavelength of 1170.3  nm, corresponding to the single pulse energy of 36.5  pJ. The saturation parameters S of the SESAM at the different lasing wavelengths and the different injection currents are close to ideal. These results reinforce the view that as the tuning element, a dispersion prism outperforms a grating for a wavelength-tunable mode-locked EC laser due to lower optical loss and less spectral narrowing.

Reference
1 Rafailov E U, Cataluna M A and Sibbett W 2007 Nat. Photon. 1 395 DOI:10.1038/nphoton.2007.120 [Cited within:1] [JCR: 27.254]
2 Lv X Q, Jin P, Wang W Y and Wang Z G 2010 Opt. Express 18 8916 DOI:10.1364/OE.18.008916 [Cited within:1] [JCR: 3.546]
3 Lin G, Su P Y and Cheng H C 2012 Opt. Express 20 3941 DOI:10.1364/OE.20.003941 [Cited within:1] [JCR: 3.546]
4 Cataluna M A, Rafailov E U, McRobbie A D, Sibbett W, Livshits D A and Kovshet A R 2006 IEEE Photon. Technol. Lett. 18 1500 DOI:10.1109/LPT.2006.877589 [Cited within:1]
5 Shchekin O B, Ahn J and Deppe D G 2002 Electron. Lett. 38 712 DOI:10.1049/el:20020509 [Cited within:1] [JCR: 1.038]
6 Mikhrin S S, Kovsh A R, Krestnikov I L, Kozhukhov A V, Livshits D A, Ledentsov N N, Shernyakov Y M, Novikov I I, Maximov M V, Ustinov V M and Alferov Z A 2005 Semicond. Sci. Technol. 20 340 DOI:10.1088/0268-1242/20/5/002 [Cited within:1]
7 Liu J R, Lu Z G, Raymond S, Poole P J, Barrios P J and Poitras D 2008 Opt. Lett. 33 1702 DOI:10.1364/OL.33.001702 [Cited within:1] [JCR: 3.385]
8 Lu Z G, Liu J R, Raymond S, Poole P J, Barrios P J and Poitras D 2008 Opt. Express 16 10835 DOI:10.1364/OE.16.010835 [Cited within:1]
9 Nikitichev D, Ding Y, Ruiz M, Calligaro M, Michel N, Krakowski M, Krestnikov I, Livshits D, Cataluna M and Rafailov E 2011 Appl. Phys. B 103 609 DOI:10.1007/s00340-010-4290-5 [Cited within:1] [JCR: 1.782]
10 Lv X Q, Jin P and Wang Z G 2010 Chin. Phys. B 19 018104 DOI:10.1088/1674-1056/19/1/018104 [Cited within:1] [JCR: 1.148] [CJCR: 1.2429]
11 Wu J, Lv X Q, Jin P, Meng X Q and Wang Z G 2011 Chin. Phys. B 20 064202 DOI:10.1088/1674-1056/20/6/064202 [Cited within:1] [JCR: 1.148] [CJCR: 1.2429]
12 Lv X Q, Jin P and Wang Z G 2010 IEEE Photon. Technol. Lett. 22 1799 DOI:10.1109/LPT.2010.2086052 [Cited within:1]
13 Varangis P M, Li H, Liu G T, Newell T C, Stintz A, Fuchs B, Malloy K J and Lester L F 2000 Electron. Lett. 36 1544 DOI:10.1049/el:20001080 [Cited within:1] [JCR: 1.038]
14 Eliseev P, Li H, Stintz A, Liu G T, Newell T C, Malloy K J and Lester L F 2000 IEEE J. Quantum Electron. 36 479 DOI:10.1109/3.831026 [Cited within:1] [JCR: 1.83]
15 Biebersdorf A, Lingk C, Giorgi M D, Feldmann J, Sacher J, Arzberger M, Ulbrich C, Böhm G, Amann M C and Abstreiter G 2003 J. Phys. D: Appl. Phys. 36 1928 DOI:10.1088/0022-3727/36/16/302 [Cited within:1] [JCR: 2.528]
16 Allen C Nì, Poole P J, Barrios P, Marshall P, Pakulski G, Raymond S and Fafard S 2005 Phys. E 26 372 DOI:10.1016/j.physe.2004.08.009 [Cited within:1]
17 Ortner G, Allen C Nì, Dion C, Barrios P, Poitras D, Dalacu D, Pakulski G, Lapointe J, Poole P J, Render W and Raymond S 2006 Appl. Phys. Lett. 88 121119 DOI:10.1063/1.2187431 [Cited within:1] [JCR: 3.794]
18 A Tierno and T Ackemann 2007 Appl. Phys. B 89 585 DOI:10.1007/s00340-007-2817-1 [Cited within:1] [JCR: 1.782]
19 Nevsky A Yu, Bressel U, Ernsting I, Eisele Ch, Okhapkin M, Schiller S, Gubenko A, Livshits D, Mikhrin S, Krestnikov I and Kovsh A 2008 Appl. Phys. B 92 501 DOI:10.1007/s00340-008-3113-4 [Cited within:1] [JCR: 1.782]
20 Li H, Liu G T, Varangis P M, Newell T C, Stintz A, Fuchs B, Malloy K J and Lester L F 2000 IEEE Photon. Technol. Lett. 12 759 DOI:10.1109/68.853491 [Cited within:1]
21 Wei H, Jin P, Luo S, Ji H M, Yang T, Li X K, Wu J, An Q, Wu Y H, Chen H M, Wang F F, Wu J and Wang Z G 2013 Chin. Phys. B 22 094211 DOI:10.1088/1674-1056/22/9/094211 [Cited within:2] [JCR: 1.148] [CJCR: 1.2429]
22 Nikitichev D I, Fedorova K A, Ding Y, Alhazime A, Able A, Kaenders W, Krestnikov I, Livshits D and Rafailov E U 2012 Appl. Phys. Lett. 101 121107 DOI:10.1063/1.4751034 [Cited within:2]
23 Alhazime A, Ding Y, Nikitichev D I, Fedorova K A, Krestnikov I L, Krakowski M and Rafailov E U 2013 Electron. Lett. 49 5 DOI:10.1364/OE.20.014308 [Cited within:1] [JCR: 1.038]
24 Li X K, Jin P, An Q, Wang Z C, Lv X Q, Wei H, Wu J, Wu J and Wang Z G 2011 Nanoscale Res. Lett. 6 625 DOI:10.1186/1556-276X-6-625 [Cited within:1] [JCR: 2.524]
25 Wang Z C, Jin P, Lv X Q, Li X K and Wang Z G 2011 Electron. Lett. 47 1191 DOI:10.1049/el.2011.2852 [Cited within:1]
26 Paschotta R and Keller U 2001 Appl. Phys. B 73 653 DOI:10.1007/s003400100726 [Cited within:2] [JCR: 1.782]
27 Wu J, Jin P, Li X K, Wei H, Wu Y H, Wang F F, Chen H M, Wu J and Wang Z G 2013 Chin. Phys. B 22 104206 DOI:10.1088/1674-1056/22/10/104206 [Cited within:1] [JCR: 1.148] [CJCR: 1.2429]
28 Keller U 2003 Nature 424 831 DOI:10.1038/nature01938 [Cited within:1] [JCR: 38.597]