Self-quenching of fundamental phase and amplitude noise in semiconductor lasers with dispersive loss

Abstract

We show theoretically that the incorporation of a frequency-dependent loss mechanism in a semiconductor laser can lead, in concert with the amplitude-to-phase coupling, to major reductions of the fundamental intensity and phase noise. A loss dispersion of the wrong sign, on the other hand, leads to an increase of the noise and, at a certain strength, to instability. The optical field is taken aswhere 5(t) and OM(t) are the noise-driven amplitude and phase excursions, respectively. The loss is represented by a frequency-dependent photon lifetime Tp,Tp rpo (2) with rnm the average frequency and xi ( 3 ) and Xr(3) the imaginary and real part of the third-order amplifying medium susceptibility so that a is the amplitude-tophase coupling factor. A formal solution of Eqs. (3) with 6(0) = 0, q(0) = 0 leads toW' WI (1 + Ca), This is the key ensatz. It stipulates that the loss rate where Ai and Ar are, respectively, the real and imaginary parts of the Langevin noise sources representing the spontaneous emission, and (5) (6) (7) where co' is seen to play the role of the fundamental decay rate of the fluctuations. This rate increases for Ca >O. Taking first the case of col'> O we can neglect the decaying terms, i.e., terms with cl'(t -X), in the exponent of Eq. (6) A~k Xr(3)°1---

Amplitude noise reduction in semiconductor lasers with weak, dispersive optical feedback

“…7 Second, noise that is due to vacuum fluctuations (spontaneous emission into the lasing mode) and dipole moment fluctuations, which are the dominant noise sources at low pump rates, can be reduced by a maximum of 1 + a 2 in accordance with the semiclassical theory. 4 Thus we expect to obtain large reductions in the amplitude noise at injection currents near threshold, with the reduction decreasing at higher injection currents. Such behavior is shown in Fig.…”

“…2 3 Under different conditions this correlation can also be used to reduce the laser amplitude noise.' Although amplitude noise reduction by use of weak optical feedback has been both predicted 4 and observed, 2 a thorough understanding of this subject is still lacking. It has been reported 5 that optical feedback tends to increase the noise rather than reduce it, and the conditions under which noise reduction is possible remain unclear.…”

Self-quenching of the semiconductor laser linewidth below the Schawlow–Townes limit by using optical feedback

“…In a previous Letter it was pointed out that the amplitude-to-phase coupling mechanism can be exploited to quench the semiconductor laser linewidth even below the Schawlow-Townes limit. 4 This can be achieved by including a frequency-dependent loss mechanism in the laser cavity. In this Letter we show that a frequency-dependent optical feedback, produced by a narrow Doppler-free Faraday resonance from Cs vapor, can give rise to a few orders of magnitude reduction of the quantum-limited laser linewidth.…”

“…4 This can be achieved by including a frequency-dependent loss mechanism in the laser cavity. In this Letter we show that a frequency-dependent optical feedback, produced by a narrow Doppler-free Faraday resonance from Cs vapor, can give rise to a few orders of magnitude reduction of the quantum-limited laser linewidth.…”

“…This can be accounted for by allowing C to be a complex quantity, C = Cr + iC 1 , where Cr represents a frequency-dependent loss and C 1 represents a frequency-dependent index change. The modified Van der Pol equations are therefore given by 4 6 + &w15 + AoCr = Ai/2(om,…”