CLASSICAL THEORY OF LASER LINEWIDTH

J. Arnaud

Tutorial Review, Optical and Quantum Electronics, November 1996, Volume 28, Issue 11, pp 1589-1615

ABSTRACT

The classical theory of light fluctuations rests on the intuitive concept that jumps between atomic states occur at independent times when the optical field has a prescribed value. The statistical properties of phase-noise sources are obtained in the present paper by applying this principle to detuned atoms. Formulae for amplitude and phase fluctuations coincide with quantum-theory results even when ‘non-classical’ states of light are generated. Theories employing semiclassical or quantum concepts are reviewed. We consider particularly the linewidth of laser oscillators operating below and above threshold when the atomic polarization cannot be adiabatically eliminated. Quantum-theory results by Lax (1966) are recovered from classical theory in a straightforward manner. More general results are given for dispersive loads, applicable to external-cavity lasers and relevant to gain guidance. It is emphasized that the K-factor as calculated by Petermann is applicable only below threshold. When more than one emitting element is present, population rate equations need to be considered and the linewidth decreases when the pump fluctuations are suppressed. The role of gain compression relating to semiconductor lasers is discussed. It is shown that at low and moderate powers gain compression reduces the effective phase-amplitude coupling factor, α. But at high power a number of mechanisms contribute to linewidth rebroadening. One of them is the statistical (quasi-thermal equilibrium) fluctuation of the refractive index. General concepts applicable to broadband light are outlined in an appendix.

LIEN PAYANT VERS  L’ARTICLE : CLASSICAL THEORY OF LASER LINEWIDTH