The processes of absorption and stimulated emission are investigated via (i) energy conservation in a Fabry-Perot resonator, (ii) the Lorentz-oscillator model, and (iii) the Kramers-Kronig relations. Vacuum fluctuations are added and the relation between stimulated and spontaneous emission is clarified. It is shown that previous models start from incorrect assumptions about the physical foundations of spontaneous emission. The non-orthogonality of longitudinal optical modes in a linear resonator is investigated and its consequences for the laser linewidth are discussed. Based on these results, a semi-classical approach to the laser linewidth below, around, and above the laser threshold which is consistent with resonator theory is presented. These results also shed new light on the usually considered linewidth-narrowing and -broadening factors.
The processes of absorption and stimulated emission are investigated via (i) energy conservation in a Fabry-Perot resonator, (ii) the Lorentz-oscillator model, and (iii) the Kramers-Kronig relations. Vacuum fluctuations are added and the relation between stimulated and spontaneous emission is clarified. It is shown that previous models start from incorrect assumptions about the physical foundations of spontaneous emission. The non-orthogonality of longitudinal optical modes in a linear resonator is investigated and its consequences for the laser linewidth are discussed. Based on these results, a semi-classical approach to the laser linewidth below, around, and above the laser threshold which is consistent with resonator theory is presented. These results also shed new light on the usually considered linewidth-narrowing and -broadening factors.
In the literature one finds several conflicting accounts of the phase difference of stimulated and spontaneous emission, as well as absorption, with respect to an existing (triggering) electromagnetic field. One of these approaches proposes that stimulated emission and absorption occur in phase and out of phase with their driving field, respectively, whereas spontaneous emission occurs under an arbitrary phase difference with respect to an existing field. It has served as a basis for explaining quantum-mechanically the laser linewidth, its narrowing by a factor of 2 around the laser threshold, as well as its broadening due to amplitude-phase coupling, resulting in Henry’s α-factor. Assuming the validity of Maxwell’s equations, all three processes would, thus, violate the law of energy conservation. In semi-classical approaches, we investigate stimulated emission in a Fabry-Pérot resonator, analyze the Lorentz oscillator model, apply the Kramers-Kronig relations to the complex susceptibility, understand the summation of quantized electric fields, and quantitatively interpret emission and absorption in the amplitude-phase diagram. In all cases, we derive that the phase of stimulated emission is 90o in lead of the driving field, and the phase of absorption lags 90o behind the transmitted field. Also spontaneous emission must obey energy conservation, hence it occurs with 90o phase in lead of an existing field. These semi-classical findings agree with recent experimental investigations regarding the interaction of attosecond pulses with an atom, thereby questioning the physical explanation of the laser linewidth and its narrowing or broadening.
The laser linewidth manifests the spectral coherence of a lasing resonator mode. It is of fundamental importance, theoretically as much as practically. For more than half a century we have been told that the laser linewidth is a result of quantum fluctuations due to spontaneous emission, which induce amplitude and phase fluctuations, whereas stimulated emission occurs in phase with the incident field. Unfortunately, in such a scenario each emission process would violate the law of energy conservation. We show that a (spontaneous or stimulated) emission process occurs 90 degrees in lead of the incident field, converts one atomic excitation into one photon in the lasing mode and, therefore, conserves the energy, whereas the mentioned fluctuations are the result of vacuum energy and occur at an orders-of-magnitude faster time scale than an emission process.
In the literature one finds several conflicting accounts of the phase difference of stimulated and spontaneous emission, as well as absorption, with respect to an existing (triggering) electromagnetic field. One of these approaches proposes that stimulated emission and absorption occur in phase and out of phase with their driving field, respectively, whereas spontaneous emission occurs under an arbitrary phase difference with respect to an existing field. It has served as a basis for explaining quantum-mechanically the laser linewidth, its narrowing by a factor of 2 around the laser threshold, as well as its broadening due to amplitude–phase coupling, resulting in Henry’s α-factor. Assuming the validity of Maxwell’s equations, all three processes would, thus, violate the law of energy conservation. In semi-classical approaches, we investigate stimulated emission in a Fabry–Perot resonator, analyze the Lorentz oscillator model, apply the Kramers–Kronig relations to the complex susceptibility, understand the summation of quantized electric fields, and quantitatively interpret emission and absorption in the amplitude–phase diagram. In all cases, we derive that the phase of stimulated emission is 90° in lead of the driving field, and the phase of absorption lags 90° behind the transmitted field. Also spontaneous emission must obey energy conservation, hence it occurs with 90° phase in lead of an existing field. These semi-classical findings agree with recent experimental investigations regarding the interaction of attosecond pulses with an atom, thereby questioning the physical explanation of the laser linewidth and its narrowing or broadening.
We review the physics underlying the process of spontaneous emission, with a special focus on spontaneous emission into a resonator mode. We define the mode volume, verify the fundamental modal dimensions, present the spectral mode profile, the coherence time, the Q-factor, the Füchtbauer-Ladenburg equation, and the Purcell factor, and discuss their influence on different types of lasers. We obtain the relation between peak emission cross section, radiative lifetime, and emission linewidth. By interpreting spontaneous emission as stimulated emission driven by vacuum fluctuations, we derive the spontaneous-emission rate into a resonator mode and establish physical expressions for the fractions of spontaneous emission and total decay from the upper laser level into this mode. Furthermore, we discuss coupling of the atomic system with the coherent field inside a lasing resonator mode, resulting in the formation of a Mollow triplet, and demonstrate that it leads to a reduction of the spontaneous-emission rate by a factor of 2.
We theoretically investigate the quantum dynamics of an unstable resonator laser. Compared to a stable cavity laser of equal gain and loss it exhibits a K-fold enhanced linewidth. This excess noise factor K is a measure of the nonorthogonality of the resonator eigenmodes and amounts to an enlargement of the quantum vacuum fluctuations. Using a quantum treatment starting from first principles based on the nonorthogonal eigenmodes, we put previous theoretical predictions onto a more firm ground. While we find a position-dependent enhancement of the spontaneous emission rate into an empty mode of only K, the constructive quantum interference of the spontaneous emission with a single oscillating mode lets the Petermann excess noise factor K reappear in the phase diffusion (laser linewidth). Hence locally enhanced spontaneous emission as well as noise enhanced by interference (amplified spontaneous emission) play an equal role in the origin of excess noise.
This lecture describes the operation principle of a continuous-wave (cw) laser. In order to keep the photon rate equation and its solution simple, usually the spontaneous-emission rate is neglected with the argument that it is so much smaller than the stimulated-emission rate. The direct consequence is that in a cw laser the gain would equal the losses. Yet, additional implications are that the light emitted by such a laser would be a pure sine wave with an infinite coherence length, its linewidth would become a delta function, its Q-factor would assume an infinite value, the coherent photon number would build up and coherence would manifest itself inside the resonator only when pumping above the laser threshold, and the threshold inversion would depend only on the total resonator losses. None of these implications holds true for any laser that mankind has ever created. Starting from vacuum fluctuations, we consider spontaneous emission directly in the photon rate equation, thereby a priori avoiding all these inconsistencies. It is then straight-forward to see that in a cw laser the gain is smaller than the losses, a cw laser does not only have a finite linewidth that can be derived in a very simple manner, but also a finite Q-factor, as well as two laser thresholds which are both different from the commonly assumed “threshold inversion‿. This work was performed in collaboration with Dr. Marc Eichhorn from the French-German Research Institute of St. Louis, France.