Absorption strength in QCLs

Initially I didn't find the experiment result of the optical properties of GaInAs/AlInAs, and I used the result of AlGaAs instead. Since they are also (III-III) semiconductor alloys, I expect their index and absorption to be at least of the same order. In all the data available, the wavelength is either shorter than 1 micron or longer than 25 microns. But at these two edges the absorption corresponds to an imaginary part of the index of ~1E-2, and that's the value based on which I did the estimate for the 5 microns experiment. Yesterday while I was reading a few earlier papers by the Bell lab group, I'm surprised to find out that they claim Im[n] is of the order of 10E-4, which is way smaller than I expected. With this absorption value I doubt the bowties would never turn on. I am doing the tweaking to see whether I can get some other librational modes to turn on with the same directional emission. If that's possible, that may explain why the threshold spacing is of the same order of the value of the first threshold.