|
Résumé |
Quantum Cascade Lasers [1] (QCLs) have become an important source for
infrared spectroscopy within the last decade. In addition, new structures
operating in the THz-region have been developed, suggesting a variety of new
applications including the fields of medicine and security. These
semiconductor heterostructure devices are based on optical transitions between
electronic subband states, where the population inversion is caused by
specifically designed tunneling processes at the operation bias. While the
essential concept of QCLs stems from semiconductor superlattices [2], the
first QCLs realized exhibited very complicated structures with plenty of
layers in each period. For THz QCLs the design could be radically simplified
during the last years resulting in a recently realized operating device with
only two barrier and wells per period [3]. Such simplified structures allow
for a detailed study of the operating principles and raise the questions, how
many levels and transitions are ultimately needed for a QCL as well as the
relation to the gain in superlattice structures.
In the recent years we have developed a simulation scheme based on
nonequilibrium Green's functions, which allows to understand the underlying
coherent transport processes [4] and the details of the gain spectrum
[5]. Here we focus on the relation between the gain spectrum and the electric
conductance. As our approach systematically includes all possible couplings
with the alternating field without referring to the rotating wave
approximation, the gain spectrum can be calculated over the entire frequency
range. This establishes the link between gain and negative differential
conductance, which is the main hinder for observation of gain in
superlattices. We argue, that a further tunneling resonance, not being related
to the main gain peak, is the main ingrediaent of a QCL. Its main property is to
provide positive conductance under operating conditions in order to stabilize
the field distribution. Taking this into account, at least three different
levels must be present in each period of a QCL strcuture.
Finally, an alternative type of structure is discussed, where the injection
into the upper laser level is provided by a scattering transition.
Calculations show that this might raise the operating temperature above 200 K.
[1] J. Faist, F. Capasso, D.L. Sivco, C. Sirtori, A.L. Hutchinson, and
A.Y. Cho, Science 264, 553 (1994).
[2] R.F. Kazarinov and R.A. Suris, Sov. Phys. Semicond. 5, 707 (1971).
[3] S. Kumar, C.W.I. Chan, Q.Hu, and J.L. Reno, Appl. Phys. Lett. 95, 141110
(2009).
[4] S.-C. Lee, F. Banit, M. Woerner, and A. Wacker, Phys. Rev. B 73, 245320
(2006).
[5] F. Banit, S.-C. Lee, A. Knorr, and A. Wacker, Appl. Phys. Lett. 86, 41108
(2005).
[6] A. Wacker, Appl. Phys. Lett. 97, 081105 (2010). |
