We demonstrate measurements of the ␣ factor of a distributed-feedback quantum cascade laser (QCL) by using a newly modified self-mixing interferometric technique exploring the laser itself as the detector. We find a strong dependence of the ␣ factor on the injection current, ranging from −0.44 at 120 mA to 2.29 at 180 mA, which can be attributed to the inherent physics of QCLs. © 2006 Optical Society of America OCIS codes: 140.5960, 140.3070, 040.5570, 120.3180. Today, quantum cascade lasers (QCLs) have evolved to be among the most important light sources in the mid-infrared (MIR) to far-infrared (FIR) wavelength range. The lasing transition in QCLs occurs between subsequently cascaded quantized energy levels within the conduction band. Hence the emitted wavelength does not depend on the energy bandgap as in conventional semiconductor lasers, but it is determined by the energy difference of discrete levels in the conduction band originating from the bandgapengineered design of the laser device.1 With the advent of this new type of laser structure, a series of questions on the laser performance and the relevant physical mechanisms have been brought up. For example, it has been found experimentally, and verified theoretically, that the scaling behavior of the intensity noise as a function of the emitted power is different with respect to interband lasers, resulting directly from the cascaded level scheme and the different carrier lifetime.2,3 The linewidth enhancement factor (␣ factor) determines the dynamics of semiconductor lasers, because it accounts for the coupling between amplitude and phase of the light as defined in Eq. (1),Here, r and i are the real and the imaginary parts of the susceptibility, respectively, and n is the carrier density. 4 The ␣ factor therefore describes how gain and refractive index change with respect to each other when the carrier density varies.5 From the very beginning of the first technological realizations of QCLs, there has been an urgent need to measure the ␣ factor of QCLs and to check whether the particular level scheme gives rise to a different physics, e.g., ifQCLs exhibit an ␣ factor close to zero as expected for a nondetuned atomiclike level scheme. The standard technique used for interband semiconductor lasers is the Hakki-Paoli method, which relies on measuring the wavelength shift of the longitudinal laser modes below threshold. However, this method gives only access to ␣ values below threshold, and for low ␣ values its accuracy may be affected by temperature effects. 6 An approach based on injection locking 7 is the only method, to date, that has been applied to QCLs above threshold.Our purpose in this paper is fourfold. First, we apply the self-mixing (SM) technique to the MIR spectral range using QCLs. Second, by directly using the laser itself as a detector instead of an external photodetector, we are able to obtain nice and clean SM signals. Third, by applying a modified analysis method, we are able to extract the ␣ factor from the measurement of these wavefo...