Nanotechnology’s Role in Mid-Infrared Laser Development

Progress in developing improved semiconductor lasers with emission in the mid-IR spectral region (≈3 μm to ≈15 μm) has depended heavily on the use of nanometer-scaled structures. Mid-IR quantum cascade lasers (QCLs), for example, represent a “tour de force” of semiconductor nanotechnology where large band gap GaAs and InP based III-V semiconductor multiple quantum well (MQW) structures are used to engineer intersubband transition energies that enable mid-IR photon emission. First developed at Bell Labs and now demonstrated by many other groups, QCLs have offered great hope as a new mid-IR light source for applications such as trace gas sensing [1] and isotope ratio measurement [2]. However, from their first use [3], QCL operation has been complicated by high power inputs, typically a minimum of 5 watts, and associated high heat load in packaged systems. Considering the significant resources devoted to QCL development and the apparent lack of progress in reducing high power consumption levels over the last ten years, it is likely that this problem is fundamental to QCL design. QCLs require high applied voltages (>8 volts) to achieve the necessary band alignment and the cascade effect, so focusing on this contribution is not expected to be fruitful. The other contribution, high threshold current (≈300 mA), appears to be fundamental to all intersubband lasers where there are parallel energy versus momentum dispersion relationships for electrons associated with intraband laser transitions. Figure 1, which depicts E vs. k subband dispersion for a three-level QCL gain medium, shows that there is an efficient competing non-radiative relaxation pathway for excited electrons when they scatter with non-zone-center optical or acoustic phonons. Since low energy subband separation is required for mid-IR light emission and the sub-band dispersions are parallel, such electron-phonon scattering will always be an efficient upper laser state depopulation mechanism thus necessitating high electron currents to achieve population inversion. Note, as indicated in Figure 1, the deliberate use of electron-phonon resonance with longitudinal optical (LO) phonons in QCL designs to depopulate the lower laser transition subband states. Exploitation of such electrophonon resonance effects in reducing laser threshold currents will be discussed below within the context of interband IV-VI mid-IR lasers.

Figure 1: Intersubband transitions in a three-level quantum cascade laser (QCL) gain medium. Electron scattering with non-zone-center optical and acoustic phonons, which non-radiatively depopulate the upper laser transition state, is believed to be responsible for high QCL threshold currents.

A second cascade laser approach, pioneered by Rui Yang [4] and recently replicated at the Naval Research Laboratory [5], involves use of narrow band gap GaSb based heterostructures that employ a type II band alignment. In this case, lasing involves interband transitions between quantum confined conduction and valence subband states such that low energy mid-IR photons are generated. These lasers can benefit from the same cascade effect as QCLs, but since the lasing transitions are interband the subband dispersions for the relevant laser transition states are not parallel, and this helps to reduce significantly upper laser transition state depopulation effects associated with electron-phonon scattering. Low interband cascade laser (ICL) threshold currents of less than 20 mA (8 A/cm²) at 80 K [4], which are significantly smaller than those for QCLs, (e.g. 225 mA (400 A/cm²) at 80 K [6]), provide strong evidence for the negative impact that electron-phonon scattering can have on semiconductor lasers that utilize intraband transitions. It can thus be argued that interband mid-IR lasers have inherent advantages over QCL-type intraband mid-IR lasers.

Figure 2: Four-level laser design possible with [111]-oriented IV-VI semiconductor quantum well materials. Electrophonon resonance between subbands facilitates population inversion by rapidly populating upper laser transition states and depopulating lower laser transition states.

A third approach to mid-IR laser fabrication, which is just beginning to be explored theoretically and experimentally, involves the use of narrow bandgap IV-VI MQW materials with quantum confinement in the [111] direction. Interband transitions in these materials (PbSe, PbTe, and related alloys) occur at the four-fold degenerate (neglecting spin) L-point in k-space, and when there is quantum confinement in the [111] direction the quantum size effect removes this degeneracy [7,8] creating a subband dispersion relationship like that depicted in Figure 2. As with ICLs, radiative transitions are interband, and there is no parallel dispersion subband below the upper laser transition subband. Instead, there is a concentric subband above the upper laser transition subband. There is also a mirror image subband structure in the valence band due to an almost symmetric band structure near the L-point. In this design, mid-IR photon emission energy can be engineered by controlling QW band gap energy (i.e. material composition), while the subband degeneracy splitting energy can be engineered by controlling QW width. Theoretical analysis shows that degeneracy splitting energy can be designed to be resonant with either transverse optical (TO) or LO phonons for well thicknesses in the range of 5 nm to 30 nm depending on well material composition. Thus there is sufficient design flexibility to optimize electrophonon resonance to achieve rapid population of the upper laser transition subband and rapid depopulation of the lower laser transition subband. It is expected that such a four-level nanotechnology engineered laser gain medium will have low thresholds for population inversion and correspondingly low laser operation currents.

Optically pumped IV-VI mid-IR lasers have been made using (111)-oriented PbSe/PbSrSe MQW active region material [9], and measured pumping thresholds were 7x lower than what were measured for comparable device structures without MQW active regions even though subband degeneracy splitting was not optimized for electrophonon resonance. Using typical threshold currents of about 200 mA for double heterostructure IV-VI mid-IR lasers, which are grown on (100)-oriented substrates and do not incorporate MQWs in the active region, as a reference, a similar 7x reduction in threshold current will allow continuous wave operation of mid-IR lasers with injection currents in the 50 mA range. With applied voltages for IV-VI diode lasers in the 300 mV range, the total power inputs for IV-VI MQW mid-IR lasers are expected to be in the range of 15 mW, 300 times lower than typical power input levels for QCLs and 10 times lower than those for ICLs. This analysis offers a compelling case for the further development of nanostructure-engineered IV-VI semiconductor materials [10] since it appears that they offer the most viable technical approach to development of battery-powered sensors based on mid-IR laser absorption spectroscopic methods.

This article was written by Patrick McCann, Ph.D., of the University of Oklahomaís (Norman, OK) School of Electrical and Computer Engineering. For more information, contact Dr. McCann at This email address is being protected from spambots. You need JavaScript enabled to view it., or visit http://info.hotims.com/10976-202 .

References

C. B. Roller, B. P. Holland, G. McMillen, D. L. Step, C. R. Krehbiel, K. Namjou, and P. J. McCann, “Measurement of exhaled nitric oxide in beef cattle using tunable diode laser absorption spectroscopy”, Applied Optics 46, 1333 (2007).
J. Yi, K. Namjou, Z. N. Zahran, P. J. McCann, G. B. Richter-Addo, “Specific detection of gaseous NO and 15NO in the headspace from liquid- phase reactions involving NO-generating organic, inorganic, and biochemical samples using a mid-infrared laser”, Nitric Oxide Biology and Chemistry 15, 154 (2006).
K. Namjou, S. Cai, E. A. Whittaker, J. Faist, C. Gmachl, F. Capasso, D. L. Sivco, and A. Y. Cho, “Sensitive absorption spectroscopy with a room-temperature distributed-feedback quantum-cascade laser,” Optics Letters. 23, 219 (1998).
R. Q. Yang, C. J. Hill, and B. H. Yang, “High- temperature and low-threshold midinfrared interband cascade lasers”, Applied Physics Letters 87, 151109 (2005).
C. S. Kim, C. L. Canedy, E. H. Aifer, M. Kim, W. W. Bewley, J. G. Tischler, D. C. Larrabee, J. A. Nolde, J. H. Warner, I. Vurgaftman, E. M. Jackson, and J. R. Meyer, “Molecular beam epitaxy growth of antimonide type-II “W” high-power interband cascade lasers and long-wavelength infrared photodiodes”, J. Vac. Sci. Technol. B 25, 991 (2007).
Gangyi Xu, Aizhen Li, Yaoyao Li, Lin Wei, Yonggang Zhang, Chun Lin, and Hua Li, “Low threshold current density distributed feedback quantum cascade lasers with deep top gratings”, Appl. Phys. Lett. 89, 161102 (2006).
H. Z. Wu, N. Dai, M. B. Johnson, P. J. McCann, Z. S. Shi, “Unambiguous Observation of Subband Transitions from Longitudinal Valley and Oblique Valleys in IV—VI multiple Quantum Wells”, Applied Physics Letters 78, 2199 (2001).
L. A. Elizondo, Y. Li, A. Sow, R. Kamana, H. Z. Wu, S. Mukherjee, F. Zhao, Z. Shi, and P. J. McCann, “Optically pumped mid-infrared light emitter on silicon”, Journal of Applied Physics 101, 104504 (2007).
C. L. Felix, W. W. Bewley, I. Vurgaftman, J. R. Lindle, J. R. Meyer, H. Z. Wu, G. Xu, S. Khosravani, and Z. Shi, “Low-threshold optically pumped l= 4.4 μm vertical-cavity surface-emitting laser with a PbSe quantum-well active region”, Applied Physics Letters 78, 3770 (2001).
P. J. McCann and Y. Selivanov, “IV-VI Semiconductor Mid-IR Lasers”, Progress in Semiconductor Materials V - Novel Materials and Electronic and Optoelectronic Applications, edited by Linda J. Olafsen, Robert M. Biefeld, Michael C. Wanke, Adam W. Saxler (Mater. Res. Soc. Symp. Proc. 891, Warrendale, PA, 2006) EE01-05.