Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Dual-wavelength switchable, mid-infrared quantum cascade laser with two shallow-etched distributed Bragg reflectors

Open Access Open Access

Abstract

A dual-wavelength quantum cascade laser (QCL) with two shallow-etched distributed Bragg reflectors is designed and fabricated. Based on a heterogeneous active region within a single waveguide, single-mode emission at 7.6μm and 8.2μm was achieved. The two wavelengths can be independently controlled by selective current injection on different regions of the device, which are electrically isolated. High optical powers of about 275mW and 218mW at room temperature were obtained for the single-mode emission at 7.6μm and 8.2μm, respectively. The presented design concept for high power, dual-wavelength switchable, mid-infrared QCLs is significant in developing miniaturized multi-species gas detection systems.

© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

1. Introduction

With the rapid development in the past 20 years, quantum cascade lasers (QCLs) have now become powerful light sources in the mid- and far-infrared spectral region [1,2]. Since the fundamental ro-vibrational transition energies of most molecules lie in this region, QCL-based absorption spectroscopy is widely used for sensitive detection of trace amounts of gases [3,4]. For sensing applications, normally a single-mode laser with narrow line-width is required to achieve high sensitivity and selectivity. Distributed feedback (DFB), distributed Bragg reflector (DBR), and external cavity QCLs are commonly used to realize single-mode operation [58].

Different from traditional semiconductor lasers, QCL is based on intersubband transition of electrons in the conduction band, which has an atomic-like density of states peaked at the transition energy. Since the active region of a QCL is transparent on both the high-frequency and low-frequency sides of its transition energy, one can thus design a heterogeneous structure by stacking several active regions emitting at different wavelengths within the same waveguide to form a multi-wavelength emitter [9,10]. Based on a heterogeneous active region, different designs have been reported for the realization of dual-wavelength single-mode QCL, which greatly simplifies the sensor system for multi-species gas sensing. For example, Jagerska et al. [11]demonstrated a dual-wavelength QCL with electrically separated DFB sections to control the two wavelengths independently. Kapsalidis et al. [12]used two “neighbor” DFBs fabricated next to each other to achieve two wavelengths separately.

In this paper, using a different strategy we designed and fabricated a compact, dual-wavelength switchable, single-mode QCL with two shallow-etched DBRs. The two DBRs with different grating periods determined the two desired Bragg wavelengths, which are about 7.6μm and 8.2μm, respectively. Multiple electrically isolated sections were formed and different currents were injected to the front and rear gain sections to control the two wavelengths independently. Peak optical powers of about 275 mW and 218 mW at room temperature were obtained for the single-mode emission at 7.6μm and 8.2μm, respectively. Stable single-mode operation without mode hopping was observed at a large temperature range and the peak wavelength tuned with temperature linearly for both wavelengths. The presented design concept for high power, dual-wavelength switchable, mid-infrared QCLs is very useful in developing compact multi-species gas sensing systems.

2. Device design and fabrication

The designed laser structure includes the front and rear gain sections, two DBR sections (DBR1 and DBR2), with the four sections separated by three electrically isolated grooves. DBR1 and DBR2 with different grating periods are used to select the two desired Bragg wavelengths λ1 and λ2, respectively. In order to independently control the two wavelengths, DBR1 and DBR2 are also separated, with the rear gain section between them. The designed laser structure is schematically shown in Fig. 1. In our design, DBR1 is transparent for λ2 and DBR2 is transparent for λ1, that is, λ1 and λ2 are away from the reflection band of DBR2 and DBR1, respectively. For λ1 operation, two separated cavities exist, the first cavity is between the front facet and DBR1 using the front gain section, and the second cavity is between the DBR1 and the rear facet using the rear gain section. However, for the second cavity, since the DBR2 section near the rear facet is not injected with current in our case, which will cause absorption loss for the generated light, λ1 operation using the second cavity would be difficult. Therefore, we choose the first cavity between the front facet and DBR1 with current injection only on the front gain section to achieve λ1 operation. For λ2 operation, the cavity is between the front facet and DBR2 using both the front and rear gain sections.

 figure: Fig. 1.

Fig. 1. Schematic diagram of the dual-wavelength QCL. The front gain section, DBR1, rear gain section, and DBR2 are electrically isolated by three grooves.

Download Full Size | PDF

The heterogeneous active core was grown on a semi-insulating InP substrate by solid-source molecular beam epitaxy (MBE), which comprises two active regions for emission centered at 8.0μm and 7.8μm, respectively, a 50nm-thick In0.53Ga0.47As separation layer was also grown between the two active regions. The first active region consists of 40 periods, with In0.53Ga0.47As and In0.52Al0.48As as the quantum wells and barriers, respectively. The layer sequence in one period is: 40/18/8/53/10/48/11/43/14/36/17/33/24/31/34/29. The second active region consists of 35 periods, with In0.58Ga0.42As and In0.47Al0.53As as the quantum wells and barriers, respectively. The layer sequence in one period as follows: 40/17/9/50.6/9/47/10/39/18/32/17/28/19/27/28/26. For both active layer sequences, the layer thickness is in angstroms, the bold represents InAlAs barriers and the underlined are layers with Si-doped to 2.5×1017 cm−3. The detailed layer sequence from substrate was arranged as follows: 0.2 μm highly doped (Si, 3×1018 cm−3) In0.53Ga0.47As contact layer, 3 μm low-doped (Si, 3×1016 cm−3) InP layer, 0.2 μm low-doped (Si, 3×1016 cm−3) In0.53Ga0.47As lower confinement layer, the heterogeneous active core, and 0.3 μm low-doped (Si, 3 × 1016 cm−3) In0.53Ga0.47As upper confinement layer. In order to fabricate two DBRs with different periods, we first deposited a 100nm-thick SiO2 layer on the upper In0.53Ga0.47As layer and selectively removed SiO2 in the region where DBR1 located by optical photolithography and wet chemical etching, then DBR1 with grating period of about 1.18μm was defined using holographic lithography and transferred to the exposed In0.53Ga0.47As region by wet chemical etching to the depth of about 282nm, the remaining 100nm-thick SiO2 layer was used as the hard mask to protect the other section of the upper In0.53Ga0.47As layer. After the fabrication of DBR1, the remaining 100nm-thick SiO2 layer was completely removed. By repeating the above process, DBR2 with grating period of about 1.28μm and depth of about 244nm was also defined on the upper In0.53Ga0.47As layer. Following the fabrication of the two DBRs, a 3μm low-doped (Si, 3×1016 cm-3) InP layer, a 0.15μm gradually doped (Si, 1×1017 to 3×1017 cm-3) InP layer, and a 0.85μm highly doped (Si, 5×1018 cm-3) InP contact layer were regrown in sequence as the upper cladding by metal organic vapor phase epitaxy (MOVPE). The wafer was then processed into 27μm-wide and 5mm-long chip by optical photolithography and wet chemical etching. The lengths of the front and rear gain sections are 1.7mm and 1.5mm, while the lengths of DBR1 and DBR2 are 0.8mm and 1mm, respectively. The three 20μm electrical isolation grooves were etched through the upper highly doped and gradually doped InP layers, leaving about 2.8μm-thick low-doped InP cladding layer on top of the upper In0.53Ga0.47As layer. A 450nm-thick SiO2 layer was deposited for electrical insulation, the top and bottom electrical contacts were formed by Ti/Au evaporation and lift-off. Finally, the wafer was cleaved into 5mm-long devices with both the front and rear facets uncoated, and soldered epi-side up on copper heat sinks with Indium solder and wire bonded.

3. Results and discussion

We calculated the reflection spectrum of the two DBRs by using transfer matrix method, as shown in Fig. 2. In order to estimate the peak reflectivity at the two Bragg wavelengths (7.6μm and 8.2μm), we use the following equations, which are derived from Ref. [7]:

$$R = {\left( {\frac{{\frac{\kappa }{\gamma }\sinh ({\gamma {L_{DBR}}} )}}{{\cosh ({\gamma {L_{DBR}}} )+ \frac{\alpha }{\gamma }\sinh ({\gamma {L_{DBR}}} )}}} \right)^2},$$
$${\gamma ^2} = {\kappa ^2} + {\alpha ^2},$$
where κ is the coupling strength of the grating and κ≈2Δneff/λ; Δneff is the modulation of the effective index; λ is the Bragg wavelength; α is the waveguide loss, LDBR is the length of the DBR section, R is the reflectivity at the Bragg wavelength. We used the finite element method to calculate Δneff for each DBR, the coupling strength κ for DBR1 and DBR2 were estimated to be about 27 cm-1 and 21 cm-1, respectively. The waveguide loss α was also calculated using the finite element method, which is about 2.8 cm-1 and 3.5 cm-1 for 7.6μm and 8.2μm wavelengths, respectively. Then using Eqs. (1) and (2), the reflectivity at the two Bragg wavelengths (7.6μm and 8.2μm) was estimated to be about 78% and 68%, respectively.

 figure: Fig. 2.

Fig. 2. The calculated reflection spectrum of DBR1 and DBR2.

Download Full Size | PDF

For spectral and electrical characterization, the device was mounted on a holder with a thermoelectric cooler (TEC) and a thermistor to regulate and monitor the temperature, respectively. Two pulsed power supplies at a duty cycle of 1% with 2μs pulses were used for the current injection on the front and rear gain sections separately. In case where both power supplies were needed, it is important to keep the two pulses from the two power supplies synchronized, although the amplitudes of the two pulses might be different. The emission spectra were measured by a Fourier transform infrared (FTIR) spectrometer with 0.25 cm-1 resolution. The output optical power from the front facet was measured by a calibrated thermopile detector.

When only the front gain section is injected with current, only the short wavelength λ1 at 1318 cm-1 (7.6μm) is emitted with a threshold current (Ith) about 1.8A. Figure 3(a) shows the light-current curve of the 7.6μm emission when only the front gain section is injected with current, peak optical power of about 275mW is achieved. A typical spectrum for the 7.6μm emission is shown in the inset of Fig. 3(a). However, when only the rear gain section is injected with current, no light emission is observed because the light generated from the rear gain section has to propagate through the unbiased front section and gets attenuated. This issue is partially resolved by applying a subthreshold current at the front gain region, thereby reducing the propagation loss of the longer wavelength λ2. Figure 3(b) displays the peak optical power varying with the current applied on the rear gain section Irear while the front gain section is kept at a subthreshold current of 1.6A (Ifront=1.6A). At Irear=1.4A, only the longer wavelength λ2 at 1220cm-1 (8.2μm) starts to lase. At Irear=2.8A, the peak optical power reaches up to 46mW. The inset of Fig. 3(b) gives a typical spectrum of the 8.2μm emission measured at Ifront=1.6A and Irear=1.9A. For both wavelengths, single-mode emission with side-mode suppression ratio (SMSR) above 25dB is achieved. The experimental results indicate that the independent control of the two wavelengths can be realized by selective current injection on the front and rear gain sections, either by injection on the front gain section alone to get the 7.6μm emission or by injection on both the front and rear gain sections synchronously with different currents to select the 8.2μm emission. From Fig. 3 we can also see that the 8.2μm emission has a lower threshold current density than the 7.6μm emission. The reason is that the 8.2μm emission has a much longer cavity length than that of the 7.6μm emission, which leads to a much smaller mirror loss for the 8.2μm emission.

 figure: Fig. 3.

Fig. 3. (a) the light-current curve of the 7.6μm emission when only the front gain section is biased. (b) The light-current curve of the 8.2μm emission measured by varying the rear current Irear while fixing the front current Ifront at 1.6A (< Ith of λ1). The insets of Fig. 3 (a) and Fig. 3 (b) show the typical spectrum for the 7.6μm and 8.2μm emission, respectively, single-mode is obtained with SMSR above 25dB.

Download Full Size | PDF

Figure 4 shows the emission spectra of the dual-wavelength QCL in pulsed mode operation at different heat sink temperatures from 10℃ to 60℃ with a step of 2℃. When only the front gain section is injected with currents of 1.05Ith of λ1, the peak wavelength of the λ1 emission tunes from 1317.84 to 1312.64cm-1 (7.588-7.618μm), as shown in Fig. 4(a). When the front current was kept below Ith of the λ1 emission, the temperature-dependent spectra were measured at different rear currents (1.05Ith of λ2). The peak wavelength of the λ2 emission varied from 1220.99 to 1216.90cm-1 (8.190-8.218μm), as depicted in Fig. 4(b). For both wavelengths, stable single-mode operation without mode hopping was observed at a large temperature range. The peak wavelength varied with temperature linearly with a tuning coefficient of about -0.104cm-1K-1 and -0.085 cm-1K-1 for the 7.6μm and 8.2μm emission, respectively. We also measured the optical spectra at different currents while keeping the heat sink temperature at 20℃.

 figure: Fig. 4.

Fig. 4. The temperature-dependent spectra of the dual-wavelength QCL from 10℃ to 60℃ with a step of 2℃. (a) When only the front gain section is injected with currents of 1.05Ith of λ1; (b) both the front and rear gain sections are injected synchronously with different currents (Ifront < Ith of λ1 and Irear = 1.05Ith of λ2).

Download Full Size | PDF

Figure 5 shows the emission spectra of the dual-wavelength QCL in pulsed mode operation at heat sink temperature of 23℃ at different injection currents. For Fig. 5(a), only the front gain section is injected with currents varied from 1.9A to 3.5A with a step of 80mA. Stable single mode operation at 7.6μm was observed up to the maximum injection current. For Fig. 5(b), when fixing the front current Ifront at 1.6A (< Ith of λ1), single mode operation at 8.2μm was observed for rear current Irear varied from 1.9A to 2.5A. When fixing Ifront at 2.7A (1.5 Ith of λ1) and changed Irear from 2.5A to 3.0A, single mode operation at 8.2μm was also achieved. Thus, single mode operation at 8.2μm can also be achieved up to the maximum rear current except we have to increase the Ifront when the Irear is large. More detailed investigation was also performed and discussed next. From Fig. 5, for both wavelengths we noticed that the shift of peak wavelength with increasing current is very small, which is because the thermal accumulation is not obvious in pulsed mode operation with a low duty cycle of 1%.

 figure: Fig. 5.

Fig. 5. The emission spectra of the dual-wavelength QCL at different injection currents. (a) Only the front gain section is injected with currents varied from 1.9A to 3.5A; (b) both the front and rear gain sections are injected synchronously. When Irear varied from 1.9A to 2.5A, the front current was fixed at 1.6A; when Irear varied from 2.5A to 3.0A, the front current was fixed at 2.7A.

Download Full Size | PDF

To further characterize the lasing properties of the dual-wavelength device, we also measured the peak optical powers at different injection currents on the rear gain section Irear while fixing the front current Ifront=2.7A (1.5 Ith of λ1). The optical power curve shows a little complicated behavior. With increasing rear current Irear, the optical power slightly increases at first, then decreases a little, and finally shows a fast increase up to about 218mW, as illustrated in Fig. 6(a). Detailed spectral analysis was also performed at Ifront=2.7A and variable Irear, which was depicted in Fig. 6(b). When the rear current Irear is small (≤ 1A), only the 7.6μm emission exists. As Irear is increased (from 1.1A to 2.2A), both 7.6μm and 8.2μm emission can be observed, the intensity of the 7.6μm emission gradually decreases and that of the 8.2μm emission becomes stronger with increasing Irear. The state of dual-wavelength coexistence leads to the power fluctuations observed in Fig. 6(a). When Irear is further increased (>2.2A), only the 8.2μm emission occurs and its intensity increases with current before reaching the rollover point at 2.5A. This means that at larger Irear only the 8.2μm emission contributes to the measured optical power, which reaches about 218mW at 2.5A. In comparison, when the bias current on the front gain section increases from 1.6A (subthreshold) to 2.7A (1.5Ith of λ1), the optical power of the 8.2μm emission gets enhanced by more than 5 times. The optical power drops when Irear >2.5A, which is due to the decrease of the tunnel coupling between the ground state of the injector and the upper laser level [13]. It should be noted that more measurements were also carried out for different front current Ifront (> Ith of λ1) while changing the rear current Irear, similar behaviors were observed. This phenomenon can be explained by the mode competing between the 7.6μm and 8.2μm emission. In our case, the gain curves of two active regions are centered at 8.0μm (155meV) and 7.8μm (159meV), respectively, which are highly overlapped. With a fixed high current at the front section (for example Ifront=2.7A), when the rear current Irear is small, only the 7.6μm light emits. When the rear current Irear reaches the threshold current for the 8.2μm emission, both 7.6μm and 8.2μm emissions exist and the two modes compete with each other for the inverted carrier population. When the rear current is further increased, the intensity of the 8.2μm emission gets enhanced. Because of the consumption of the inverted carriers, the overall gain gradually decreases, and finally falls below the threshold for the 7.6μm emission. The 7.6μm emission then disappears and only the 8.2μm emission left to lase.

 figure: Fig. 6.

Fig. 6. (a) The light-current curve measured by varying the rear current Irear while fixing the front current Ifront at 2.7A (1.5 Ith of λ1). (b) The corresponding optical spectra measured at different rear current Irear while fixing the Ifront at 2.7A.

Download Full Size | PDF

4. Conclusion

In summary, we have demonstrated a mid-infrared QCL with switchable dual-wavelength by using two shallow-etched DBRs. The two wavelengths can be independently controlled by selective current injection on the front and rear gain sections. Stable single-mode emission at 7.6μm and 8.2μm with peak power of about 275mW and 218mW were obtained, respectively. By optimizing the physical length of each section on the device and improving the thermal management, better laser performance is expected. The design concept presented here can be used to fabricate dual-wavelength QCLs at other spectral regions, which are significant in developing compact multi-species gas sensing systems.

Funding

National Natural Science Foundation of China (61874110, 61734006, 61991430, 61835011); Chinese Academy of Sciences Key Project (XDB43000000).

Acknowledgments

The authors would like to thank Ping Liang and Ying Hu for their help in the device processing.

Disclosures

The authors declare no conflicts of interest.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

1. M. Razeghi, W. Zhou, S. Slivken, Q. Y. Lu, D. Wu, and R. McClintock, “Recent progress of quantum cascade laser research from 3 to 12 mum at the Center for Quantum Devices [Invited],” Appl. Opt. 56(31), H30–H44 (2017). [CrossRef]  

2. Y. Yao, A. J. Hoffman, and C. F. Gmachl, “Mid-infrared quantum cascade lasers,” Nat. Photonics 6(7), 432–439 (2012). [CrossRef]  

3. S. Bartalini, M. S. Vitiello, and P. De Natale, “Quantum cascade lasers: a versatile source for precise measurements in the mid/far-infrared range,” Meas. Sci. Technol. 25(1), 012001 (2014). [CrossRef]  

4. R. F. Curl, F. Capasso, C. Gmachl, A. A. Kosterev, B. McManus, R. Lewicki, M. Pusharsky, G. Wysocki, and F. K. Tittel, “Quantum cascade lasers in chemical physics,” Chem. Phys. Lett. 487(1-3), 1–18 (2010). [CrossRef]  

5. Q. Y. Lu, Y. Bai, N. Bandyopadhyay, S. Slivken, and M. Razeghi, “2.4 W room temperature continuous wave operation of distributed feedback quantum cascade lasers,” Appl. Phys. Lett. 98(18), 181106 (2011). [CrossRef]  

6. P. Fuchs, J. Friedl, S. Hofling, J. Koeth, A. Forchel, L. Worschech, and M. Kamp, “Single mode quantum cascade lasers with shallow-etched distributed Bragg reflector,” Opt. Express 20(4), 3890–3897 (2012). [CrossRef]  

7. F. Xie, C. G. Caneau, H. P. LeBlanc, M. T. Ho, J. Wang, S. Chaparala, L. C. Hughes, and C. E. Zah, “High power and high temperature continuous-wave operation of distributed Bragg reflector quantum cascade lasers,” Appl. Phys. Lett. 104(7), 071109 (2014). [CrossRef]  

8. A. Hugi, R. Maulini, and J. Faist, “External cavity quantum cascade laser,” Semicond. Sci. Technol. 25(8), 083001 (2010). [CrossRef]  

9. C. Gmachl, D. L. Sivco, J. N. Baillargeon, A. L. Hutchinson, F. Capasso, and A. Y. Cho, “Quantum cascade lasers with a heterogeneous cascade: Two-wavelength operation,” Appl. Phys. Lett. 79(5), 572–574 (2001). [CrossRef]  

10. C. Gmachl, D. L. Sivco, R. Colombelli, F. Capasso, and A. Y. Cho, “Ultra-broadband semiconductor laser,” Nature 415(6874), 883–887 (2002). [CrossRef]  

11. J. Jagerska, P. Jouy, A. Hugi, B. Tuzson, H. Looser, M. Mangold, M. Beck, L. Emmenegger, and J. Faist, “Dual-wavelength quantum cascade laser for trace gas spectroscopy,” Appl. Phys. Lett. 105(16), 161109 (2014). [CrossRef]  

12. F. Kapsalidis, M. Shahmohammadi, M. J. Suess, J. M. Wolf, E. Gini, M. Beck, M. Hundt, B. Tuzson, L. Emmenegger, and J. Faist, “Dual-wavelength DFB quantum cascade lasers: sources for multi-species trace gas spectroscopy,” Appl. Phys. B 124(6), 107 (2018). [CrossRef]  

13. S. Menzel, L. Diehl, C. Pflügl, A. Goyal, C. Wang, A. Sanchez, G. Turner, and F. Capasso, “Quantum cascade laser master-oscillator power-amplifier with 1.5 W output power at 300 K,” Opt. Express 19(17), 16229–16235 (2011). [CrossRef]  

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (6)

Fig. 1.
Fig. 1. Schematic diagram of the dual-wavelength QCL. The front gain section, DBR1, rear gain section, and DBR2 are electrically isolated by three grooves.
Fig. 2.
Fig. 2. The calculated reflection spectrum of DBR1 and DBR2.
Fig. 3.
Fig. 3. (a) the light-current curve of the 7.6μm emission when only the front gain section is biased. (b) The light-current curve of the 8.2μm emission measured by varying the rear current Irear while fixing the front current Ifront at 1.6A (< Ith of λ1). The insets of Fig. 3 (a) and Fig. 3 (b) show the typical spectrum for the 7.6μm and 8.2μm emission, respectively, single-mode is obtained with SMSR above 25dB.
Fig. 4.
Fig. 4. The temperature-dependent spectra of the dual-wavelength QCL from 10℃ to 60℃ with a step of 2℃. (a) When only the front gain section is injected with currents of 1.05Ith of λ1; (b) both the front and rear gain sections are injected synchronously with different currents (Ifront < Ith of λ1 and Irear = 1.05Ith of λ2).
Fig. 5.
Fig. 5. The emission spectra of the dual-wavelength QCL at different injection currents. (a) Only the front gain section is injected with currents varied from 1.9A to 3.5A; (b) both the front and rear gain sections are injected synchronously. When Irear varied from 1.9A to 2.5A, the front current was fixed at 1.6A; when Irear varied from 2.5A to 3.0A, the front current was fixed at 2.7A.
Fig. 6.
Fig. 6. (a) The light-current curve measured by varying the rear current Irear while fixing the front current Ifront at 2.7A (1.5 Ith of λ1). (b) The corresponding optical spectra measured at different rear current Irear while fixing the Ifront at 2.7A.

Equations (2)

Equations on this page are rendered with MathJax. Learn more.

R = ( κ γ sinh ( γ L D B R ) cosh ( γ L D B R ) + α γ sinh ( γ L D B R ) ) 2 ,
γ 2 = κ 2 + α 2 ,
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.