An external-cavity (EC) quantum cascade (QC) laser using optical feedback from a partial-reflector is reported. With this configuration, the otherwise multi-mode emission of a Fabry-Perot QC laser was made single-mode with optical output powers exceeding 40 mW. A mode-hop free tuning range of 2.46 cm−1 was achieved by synchronously tuning the EC length and QC laser current. The linewidth of the partial-reflector EC-QC laser was measured for integration times from 100 μs to 4 seconds, and compared to a distributed feedback QC laser. Linewidths as small as 480 kHz were recorded for the EC-QC laser.
©2010 Optical Society of America
Quantum cascade (QC) lasers have emerged as a viable laser source in the mid-infrared spectral region due to their large wavelength design space, compact size, and high output power. Room temperature QC lasers are an attractive solution for trace gas sensing applications that require fast, portable, high-sensitivity measurements such as environmental monitoring and medical diagnostics [1,2]. However, the emission of a conventional Fabry-Perot (FP) QC laser is typically multi-mode  and is unsuitable for many desired applications requiring a single frequency and narrow-linewidth source, such as high-resolution spectroscopy or metrology . Distributed Feedback (DFB) QC lasers achieve single-mode emission but require complex fabrication techniques with specialized equipment often resulting in a low yield . Even when successful, conventional fabrication of DFB-QC lasers introduces a large inherent resonator loss, which increases the laser current threshold and limits their operating range and output power.
External cavity (EC) QC lasers, typically stabilized using optical feedback from a diffraction grating, provide single-mode emission and broad tunability across the QC laser gain bandwidth [6–8]. Here we present a simple alternative external cavity configuration to stabilize the wavelength of a FP-QC laser using feedback from an optical window serving as a partial reflector. This straightforward configuration uses simple components to induce single-mode operation in an otherwise multi-mode FP-QC laser with no antireflective facet coating. The approach is similar to one that has been successfully used in external cavity diode lasers [9–11]. The EC-QC laser was operated in continuous wave (CW) at room temperature and a mode-hop free tuning range of 2.46 cm−1 was obtained by synchronously tuning the EC length and FP-QC laser current. Output powers up to 40 mW were obtained, at currents up to 60% above threshold.
Key characteristics of a single-mode laser are its linewidth and frequency stability. Operating diode lasers in an external cavity configuration has been shown to reduce the linewidth by large factors ; however, there are relatively few discussions of the linewidths of EC-QC lasers reported in the literature. Measurements have been reported of grating-tuned EC-QC laser linewidths in the 4 − 50 MHz range for integration times typically greater than one second [6,7,12–14]. To our knowledge, the dependence of the EC-QC laser linewidth on integration time has not been presented in the literature, which is important because the contributions to the linewidth from current, vibration, and thermal noise sources are highly time dependent. We have characterized the EC-QC laser linewidth at integration times ranging from 100 µs to 4 seconds, allowing us to separate the electrical, acoustic, and thermal contributions to the linewidth. We measured a full-width at half-maximum (FWHM) linewidth of 480 kHz for short integration times and 1.08 MHz for a 1 second integration time. The EC-QC laser linewidth was compared to a DFB-QC laser with a similar wavelength measured using the same technique
2. Configuration and operation of the EC-QC laser
The EC-QC laser consisted of a thermoelectrically (TE) cooled, 3 mm long FP-QC laser from Maxion Technologies with a highly-reflective back-facet coating and an as-cleaved front facet, operating CW at 16° C while mounted on a water-cooled heat sink; a 10 mm, 0.85 NA Ge collimating lens; and a partial-reflector as the output mirror. The EC-QC laser is shown in Fig. 1(a) . Optical feedback from the partial-reflector forms an external cavity coupled with the FP resonator cavity, as shown in Fig. 1(b), with the lasing mode/s selected via the Vernier effect. Configurations using both an uncoated BaF2 window and an uncoated ZnSe wedged window as partial-reflectors were investigated. To power the FP-QC laser, a custom in-house-built low-noise current source was used. The collimating lens was placed in a fine-control tip-tilt lens mount, and secured on an X-Y-Z translation stage. The partial-reflector was placed in a fine-control tip-tilt lens mount and secured onto a linear translation stage. Adjustments to the tip-tilt of the partial-reflector were necessary in order to maximize the optical feedback into the FP-QC laser while the linear translation stage allowed the EC length to be varied with sub-micron resolution. The collimated EC-QC laser output beam was directed to either a Fourier Transform Infrared Spectrometer (FTIR) with a resolution of 0.5 cm−1 or to a Bristol Wavemeter, model 621b, with a resolution of 0.001 cm−1 for spectral measurements. A liquid nitrogen cooled Mercury-Cadmium-Telluride (MCT) detector followed by a low-noise preamplifier was used for high-bandwidth intensity measurements and a thermopile detector was used for CW power measurements.
The EC length was set to 34 mm, resulting in an approximate EC mode spacing of 0.147 cm−1. It was observed that at long EC lengths it was more difficult to achieve single-mode operation due in part to a tighter EC mode spacing. By adjusting the alignment of the partial-reflector and collimating lens and monitoring the reduction in lasing threshold, we were able to maximize the amount of optical feedback. Single-mode emission was achieved from the otherwise multi-mode FP-QC laser with a side-mode suppression ratio (SMSR) greater than 15 dB measured from the FTIR spectrum, as shown in Figs. 2(a) and (b) .
Figure 2(c) shows how the single-mode output power and EC-QC laser current threshold depend on the reflectivity of the EC partial-reflector (the single-surface reflectivities are 3% and 17% for BaF2 and ZnSe, respectively). The current threshold of the EC-QC laser based on a ZnSe wedged window was lowered by ~12% while the maximum output power was reduced by ~50% to 43 mW, as compared to the FP-QC laser. The current threshold and maximum output power of the BaF2-based EC-QC laser were similar to that of the FP-QC laser but with the laser emission now made single-mode. Although the BaF2-based EC-QC laser emitted more single-mode power, the ZnSe wedged window provided better mode stability while tuning the EC-QC laser wavelength. For this reason, the ZnSe wedged window was used for the rest of the work. Adjusting the reflectivity of the partial-reflector allows a certain amount of design flexibility to optimize for either single-mode power or tuning.
Tuning of the EC-QC laser wavelength was achieved by varying the operating current and the EC length. An increase in current results in a higher active region temperature for the QC laser: increasing the effective index of refraction and the cavity length and consequently tuning the FP resonator modes. Similarly, the EC resonator modes are tuned by changing the EC length. For example, at an EC length of 34mm and a FP-QC laser current of 1150 mA, the EC-QC laser lased on a mode at 1038.5 cm−1. For this lasing mode, the EC length was varied to fine-tune the EC-QC laser wavelength over a range of ~0.032 cm−1, but any further increase in length resulted in a hop to another mode. Similarly, adjusting only the FP-QC laser current gave a range of ~0.034 cm−1. Tuning both the EC modes and the FP modes using synchronized length and current tuning at matched tuning rates achieved a mode-hop free tuning range of 2.46 cm−1. By adjusting the EC length and/or the FP-QC laser current by larger amounts in order to induce mode-hops, different lasing modes were selected. A total tuning range of 4.68 cm−1 for all starting lasing modes was measured, as seen in Fig. 3 . The mode-hop free tuning range of the EC-QC laser is comparable with typical values of 2 − 4 cm−1 for DFB-QC lasers .
The overall tuning range of the partial-reflector EC-QC laser is limited by using a coupled-cavity effect for wavelength selection because the coupled cavity modes are periodic in wavelength, lasing will occur preferentially on the mode closest to the peak of the QC laser gain spectrum. The tuning range could potentially be increased by using a higher reflectivity partial reflector or by making the FP and EC optical lengths nearly equal to maximize the spacing of the coupled cavity modes. Additionally, a broader tuning range could be obtained by using a highly-reflective partial-reflector coupled to the back-facet and a front-coupled partial-reflector creating three coupled cavities. Heat-sink temperature tuning would also increase the tuning range; however, temperature tuning is not as rapid as EC length and FP-QC laser current tuning. A broader overall tuning range can be obtained by using additional wavelength-selective elements in the cavity such as diffraction gratings. While the tuning range of the partial-reflector EC-QC laser is limited to regions near the peak of the QC laser gain spectrum, its simplicity may offer advantages for many applications in which a narrow-linewidth tunable laser source is needed.
3. EC-QC laser: linewidth measurement
To further characterize the EC-QC laser, a measurement of the wavelength stability and linewidth was performed for integration times ranging from 100 µs to 4 seconds. In order to measure the linewidth, the steep slope of a low-pressure NH3 absorption line was used to convert the wavelength fluctuations of a QC laser into detector voltage fluctuations . The EC-QC laser output beam was directed through a low-pressure (<1 Torr) NH3 gas cell and onto an MCT detector with the detector voltage values collected via a high-speed DAQ at a rate of 1 MS/s using Labview. The heat-sink temperature of the EC-QC laser was raised from 16° C to 29° C and the operating current increased to 1630 mA to tune the lasing wavelength near the NH3 absorption line of 1034.2448 cm−1. The EC-QC laser was enclosed in soundproofing foam to reduce vibration-induced wavelength fluctuations. The enclosure also helped in reducing thermal disturbances, but the TE cooler alone did not stabilize the heat sink temperature so additional active chilling of the cooling water was needed to stabilize the EC-QC laser within a few mK. For comparison, the linewidth of a Hamamatsu DFB-QC laser operating in CW at 910 mA with a wavelength of ~1034.27 cm−1 at 21° C was also measured using the same technique. The DFB-QC laser was mounted in a custom HHL package with an integrated TE cooler. The HHL package was mounted to a water-cooled heat sink and placed within a foam enclosure.
To calibrate the EC-QC laser frequency, a 10 Hz, 8 mA current ramp was applied to the FP-QC laser to scan the wavelength across a Doppler-broadened NH3 absorption line. The conversion factor from time to frequency was obtained by fitting the measured absorption lineshape to a Gaussian function with a FWHM of 92 MHz as calculated for the Doppler-broadened NH3 line. To increase the slope of the absorption line and increase the sensitivity of the laser linewidth measurements, the NH3 pressure was increased to saturate the absorption slightly. Using the previously determined frequency calibration, the slope of the saturated line was determined to be 0.042 V/MHz, as shown in Fig. 4(a) .
For the laser linewidth measurements, the current ramp was removed and the EC-QC laser frequency was current tuned to the side of the absorption. A 4 second data set of detector voltage fluctuations was collected; Fig. 4(b) shows the collected time series of the detector voltage for 1 second duration. The time series of voltages was converted to laser frequency fluctuations using the slope of the absorption feature determined as described above. To determine the laser linewidth, histograms of the fluctuations over different integration times were computed; the laser frequency fluctuation distributions for 1 ms and 1 second integration times are shown in Fig. 4(c).
The laser linewidth was determined from the FWHM of the laser frequency distributions at integration times ranging from 100 μs to 4 seconds. Results for the EC-QC laser and DFB-QC laser are presented in Fig. 5 . For integration times of 100 µs, the EC-QC and DFB-QC lasers have measured linewidths of 480 kHz and 660 kHz, respectively. For these integration times, current noise is the dominant contribution to the linewidth. We attribute the small laser linewidths observed to the low-noise current supplies, as observed in previous studies [15,16]. At intermediate integration times from 1 ms to 1 second, corresponding approximately to acoustic frequencies, the EC-QC laser becomes sensitive to vibrations resulting in an increased linewidth of 1 MHz while the DFB-QC laser linewidth increases only slightly to 690 kHz. At longer integration times, thermal fluctuations become dominant, which increased both the EC-QC laser and DFB-QC laser linewidths to 1.5 MHz and 4 MHz, respectively. The difference in linewidth increase is attributed to the different thermally induced effective refractive index drifts for the EC-QC laser and the DFB-QC laser. As a result, an EC-QC laser frequency has a reduced sensitivity to both thermal and current fluctuations in the active region than a comparable DFB or FP-QC laser due to the presence of the EC. The reduction in an EC-QC laser scales with the ratio of the FP-QC laser optical length to the total cavity optical length; for this EC-QC laser a 0.23 reduction factor is calculated.
The measured EC-QC laser linewidth is smaller than previously reported 4 − 50 MHz linewidths of grating-tuned EC-QC lasers [6,7,12–14], even at long integration times. Contributions to the 4 second experimentally-obtained linewidth due to detector noise were measured by blocking the beam at the detector and measuring the detector-induced voltage fluctuations. These were found to be less than 1%. Laser intensity noise contributions were also measured by tuning the wavelength away from the absorption line and measuring the intensity-induced voltage fluctuations. The DFB-QC laser intensity noise represented 4% of the total voltage fluctuations while the EC-QC laser intensity noise accounted for 18% of the total voltage fluctuations. Due to the higher laser threshold, the EC-QC laser required a different current controller with twice the maximum current, and corresponding higher noise levels than the current controller used for the DFB-QC laser. However, despite the higher noise from the current controller the observed linewidth of the EC-QC laser at short integration times was measured to be smaller than the DFB linewidth. As discussed above, this is due in part to the reduction in current tuning rate caused by the EC: the measured DFB laser current tuning rate was 285 MHz/mA, while the EC-QC laser current tuning rate was 47.5 MHz/mA.
The measured values of the EC-QC laser and DFB-QC laser linewidths are significantly higher than the intrinsic QC laser linewidth , as expected for the integration times investigated where current, mechanical, and thermal contributions to the linewidth are dominant. It is predicted that the intrinsic linewidth for an EC-QC laser will be smaller than for a DFB QC laser; however, we did not investigate the differences in intrinsic linewidths for this study. The intrinsic linewidth is also influenced by differences in design, fabrication and operating conditions of the two devices . Further studies will be required to investigate fully the factors affecting the EC-QC laser linewidth and to determine the minimum achievable linewidths. Additionally, the laser linewidth for the EC-QC laser could be reduced further by using active stabilization of the current or the external cavity length to lock the frequency to a molecular transition or optical cavity [16,19].
To summarize, a simple EC-QC laser configuration that uses optical feedback from a partial-reflector to change the multi-mode emission of a FP-QC laser into a narrow and single-mode emission is presented. Single-mode emission with mode-hop free tuning of 2.46 cm−1 was achieved by synchronously tuning the EC length and QC laser current with optical output powers exceeding 40 mW. Direct linewidth measurements of the partial-reflector EC-QC laser showed a linewidth of 480 kHz for short integration times, and 1.08 MHz for a 1 second integration time, making partial-reflector EC-QC lasers a viable single-mode narrow-linewidth source.
The authors would like to thank Fred J. Towner of Maxion Technologies Inc. for wafer growth, Claire Gmachl for discussions, and MIRTHE (NSRF-ERC Grant No.EEC-0540832). The Pacific Northwest National Laboratory is operated for the U.S. Department of Energy (DOE) by the Battelle Memorial Institute under Contract No. DE-AC05-76RL01830.
References and links
1. A. Kosterev, G. Wysocki, Y. Bakhirkin, S. So, R. Lewicki, M. Fraser, F. Tittel, and R. Curl, “Applications of quantum cascade lasers to trace gas analysis,” Appl. Phys. B 90(2), 165–176 (2008). [CrossRef]
2. J. H. Shorter, D. D. Nelson, J. B. McManus, M. S. Zahniser, and D. K. Milton, “Multicomponent Breath Analysis With Infrared Absorption Using Room-Temperature Quantum Cascade Lasers,” IEEE Sens. J. 10(1), 76–84 (2010). [CrossRef] [PubMed]
3. K. Fujita, T. Edamura, S. Furuta, and M. Yamanishi, “High-performance, homogenous broad-gain quantum cascade lasers based on dual-upper-state design,” Appl. Phys. Lett. 96(24), 241107 (2010). [CrossRef]
4. G. Stancu, N. Lang, J. Röpcke, M. Reinicke, A. Steinbach, and S. Wege, “In Situ Monitoring of Silicon Plasma Etching Using a Quantum Cascade Laser Arrangement,” Chem. Vap. Deposition 13(6-7), 351–360 (2007). [CrossRef]
5. A. Wittmann, Y. Bonetti, M. Fischer, J. Faist, S. Blaser, and E. Gini, “Distributed-Feedback Quantum Cascade Lasers at 9 μm Operating in Continuous Wave Up to 423K,” IEEE Photon. Lett. 21(12), 814–816 (2009). [CrossRef]
6. G. Wysocki, R. Curl, F. Tittel, R. Maulini, J. Bulliard, and J. Faist, “Widely tunable mode-hop free external cavity quantum cascade laser for high resolution spectroscopic applications,” Appl. Phys. B 81(6), 769–777 (2005). [CrossRef]
7. R. Maulini, D. A. Yarekha, J. M. Bulliard, M. Giovannini, J. Faist, and E. Gini, “Continuous-wave operation of a broadly tunable thermoelectrically cooled external cavity quantum-cascade laser,” Opt. Lett. 30(19), 2584–2586 (2005). [CrossRef] [PubMed]
8. G. Wysocki, R. Lewicki, R. Curl, F. Tittel, L. Diehl, F. Capasso, M. Troccoli, G. Hofler, D. Bour, S. Corzine, R. Maulini, M. Giovannini, and J. Faist, “Widely tunable mode-hop free external cavity quantum cascade lasers for high resolution spectroscopy and chemical sensing,” Appl. Phys. B 92(3), 305–311 (2008). [CrossRef]
9. R. Lang and K. Kobayashi, “External optical feedback effects on semiconductor injection laser properties,” IEEE J. Quantum Electron. 16(3), 347–355 (1980). [CrossRef]
10. H. Kakiuchida and J. Ohtsubo, “Characteristics of a semiconductor laser with external feedback,” IEEE J. Quantum Electron. 30(9), 2087–2097 (1994). [CrossRef]
12. N. Mukherjee, R. Go, C. Kumar, and N. Patel, “Linewidth measurement of external grating cavity quantum cascade laser using saturation spectroscopy,” Appl. Phys. Lett. 92(11), 111116 (2008). [CrossRef]
13. N. Mukherjee and C. Patel, “Molecular fine structure and transition dipole moment of NO2 using an external cavity quantum cascade laser,” Chem. Phys. Lett. 462(1-3), 10–13 (2008). [CrossRef]
14. G. Hancock, J. van Helden, R. Peverall, G. Ritchie, and R. Walker, “Direct and wavelength modulation spectroscopy using a cw external cavity quantum cascade laser,” Appl. Phys. Lett. 94(20), 201110 (2009). [CrossRef]
15. T. L. Myers, R. M. Williams, M. S. Taubman, C. Gmachl, F. Capasso, D. L. Sivco, J. N. Baillargeon, and A. Y. Cho, “Free-running frequency stability of mid-infrared quantum cascade lasers,” Opt. Lett. 27(3), 170–172 (2002). [CrossRef]
16. M. S. Taubman, T. L. Myers, B. D. Cannon, R. M. Williams, F. Capasso, C. Gmachl, D. L. Sivco, and A. Y. Cho, “Frequency stabilization of quantum-cascade lasers by use of optical cavities,” Opt. Lett. 27(24), 2164–2166 (2002). [CrossRef]
17. M. Yamanishi, T. Edamura, K. Fujita, N. Akikusa, and H. Kan, “Theory of the Intrinsic Linewidth of Quantum-Cascade Lasers: Hidden Reason for the Narrow Linewidth and Line-Broadening by Thermal Photons,” IEEE J. Quantum Electron. 44(1), 12–29 (2008). [CrossRef]
18. S. Bartalini, S. Borri, P. Cancio, A. Castrillo, I. Galli, G. Giusfredi, D. Mazzotti, L. Gianfrani, and P. De Natale, “Observing the intrinsic linewidth of a quantum-cascade laser: beyond the Schawlow-Townes limit,” Phys. Rev. Lett. 104(8), 083904 (2010). [CrossRef] [PubMed]
19. F. Bielsa, A. Douillet, T. Valenzuela, J. P. Karr, and L. Hilico, “Narrow-line phase-locked quantum cascade laser in the 9.2 microm range,” Opt. Lett. 32(12), 1641–1643 (2007). [CrossRef] [PubMed]