AM and FM modulation spectroscopies1,2 have been used to study the properties of matter as well as for applications such as reading out optical memories.3 Advances in the accuracy of these methods have turned what used to be an academic subject, quantum noise, into a practical consideration. For example, memory readout fields must be very weak to keep the medium from heating up too much and quantum noise puts a lower limit on the readout field strength. Similarly if weak modulation sidebands are used to study coherent dips via AM spectroscopy, quantum noise becomes relevant. We use our quantum theory of multiwave mixing4 to predict the effects of quantum noise on these problems. We find that, for modulation frequencies about equal to the Rabi frequency for the strong wave, spontaneous emission contributes a pronounced AM component that might be confused with the appearance of a hole in reading out optical memories. In AM spectroscopy we find that the coherent dips are partly filled in by spontaneous emission.
We present a detailed analysis of an application of frequency modulation (FM) spectroscopy in the high modulation depth limit. We have recently completed and reported a measurement of the Stark shift in the indium
5p1/2→6s1/2 410 nm transition using this spectroscopy method [Ranjit, et al. Phys. Rev. A 87, 032506 (2013)]. FM spectroscopy proved essential to resolve spectroscopic features in an atomic beam where the optical depth was
∼10−3. A dual-modulation scheme is described which ensures truly background-free FM signals even in the low-density limit. Lock-in detection of the FM signal was accomplished using both the fundamental (1f) and second-harmonic (2f) modulation frequencies. A derivation of both the 1f and 2f signal line shapes in the high-modulation-depth limit is presented. These line shapes form the basis for quantitative fits to a wide variety of experimental FM spectra.