Noise in cavity ring-down spectroscopy caused by transverse mode coupling

In our continuous wave cavity ring-down spectroscopy (CW-CRDS) experiments, we have often observed that the decay time constant drops to a lower value at some cavity lengths or some intercavity pressures. The resulting instabilities lead to a reduction in the sensitivity of our CRDS system. We have deduced that the cause of this noise is the coupling between the TEM00 mode that the laser excites, and the higher order transverse modes of the cavity. The coupling will cause anti-crossings as the modes tune with cavity length. A consequence is that the decay of light intensity leaving the cavity is no longer a single exponential decay, but the signal can be quantitatively fit to a two-mode beating model. With a 4mm diameter intra-cavity aperture, the higher order modes are suppressed and the stability of the system improved greatly. One coupling mechanism is scattering from the mirror surfaces. This can explain some features of our data including the strength of this coupling and the relative tuning rate of the coupled modes. Remarkably, a scattering intensity between modes of similar to 10(-12) can produce observable changes in the cavity decay rate. However, the tuning rate between the TEM00 mode and the higher order modes in a cavity pressure scan is larger than predicted and is still not explained. Images of higher order transverse modes excited at certain cavity conditions were recorded by an Indium Gallium Arsenide (InGaAs) area camera. (c) 2007 Optical Society of America.