CPM Seminar

Quantum cascade lasers: from principles to THz lasers

Emmanuel Dupont

Institute For Microstructural Sciences
NRC

Of all semiconductor devices, Quantum cascade lasers (QCLs) provide some of the best illustrations of the power of electronic structure engineering offered by advanced epitaxial technologies such as Molecular Beam Epitaxy. The first proposal of superlattice based unipolar laser has been put forward in 1971 by Kazarinov and Suris [1], just a year after Esaki and Tsu in their seminal work [2] showed that periodic potential - which should form in the conduction band of hypothetical at that time epitaxial heterostructure superlattice - will lead to strong modification of the vertical electronic transport through such structure, with well-defined negative differential resistance (NDR) regions. The spontaneous emission between quantized levels in the conduction was demonstrated much later in 1989, and curiously in the far-infrared region [3]. The first QCL was emitting in the mid-infrared and demontrated in 1994 by J. Faist in Bell Labs [4]. This discovery was made possible by a careful control of the electrical stability of the devices, i.e., by forcing the optical gain to occur concomitantly with a positive differential resistance (PDR), hence the importance of the injector region designed to link the active (lasing) quantum wells. The THz version of QCLs was demonstrated with a chirped superlattice design by Tredicucci's group at Scuola Normale Superiore in Pisa [5]. In this seminar, we will describe the principle of operation of these devices in the framework of wave mechanics, rate equations and a simplified density matrix (SDM) model [6]. The tunnel effect is the heart of operation of these devices and we will propose a wave-packet illustration [7]. Despites its inherent limitations, the SDM model captures most of the physics of transport and optical gain. We will describe how the electrical characteristics of these devices are rich of information. Even though the SDM is versatile from mid-infrared to far-infrared (QCL wavelength spans from 2.65 to 250 um), this model proved to be well suited for structures with few states per module, such as the THz QCLs based on resonant phonon scattering. It was applied to these devices in order to optimize the gain at high temperature and recently, devices at 3.22 THz with a record maximum operating temperature ~200 K were demonstrated [8]. An alternative design based on resonant phonon scattering to pump the upper lasing state will be also presented. NRC is now working on QCL in the 3-5 um atmospheric window with low effective mass material such as InAs.

References:
[1] R. F. Kazarinov and R. A. Suris, Sov. Phys. Semicond. 5 (1971) 707.
[2] L. Esaki and R. Tsu, IBM J. Res. Dev. 14 (1970) 61.
[3] M. Helm, P. England, E. Colas, F. DeRosa, and S. J. Allen, Jr., Phys. Rev. Lett. 63 (1989) 74.
[4] J. Faist, F. Capasso, D. L. Sivco, C. Sirtori, A. L. Hutchinson, and A. Y. Cho, Science 264 (1994) 553.
[5] R. Kohler, A. Tredicucci, F. Beltram, H. E. Beere, E. H. Linfield, A. G. Davies, D. A. Ritchie, R. C. Iotti, and F. Rossi, Nature 417 (2002) 156.
[6] E. Dupont, S. Fathololoumi, and H. C. Liu, Phys. Rev. B 81(2010) 205311.
[7] H. Callebaut and Q. Hu, J. Appl. Phys. 98 (2005) 104505.
[8] S. Fathololoumi, E. Dupont, C. W. I. Chan, Z. R. Wasilewski, S. R. Laframboise, D. Ban, A. Matyas, C. Jirauschek, Q. Hu, and H. C. Liu, Opt. Express 20 (2012) 3866.