|
Although step-scanning interferometers were developed concurrently with rapid-scanning interferomters, the latter dominates the commercial market. However, recently there has been a resurgence of interest in step-scanning interferometers as demonstrated by the recent publications 1,2 and the introduction of a commercial system by Bruker with a step-scanning option. Using a step-scanning interferometer is particularly advantageous when photoacoustic spectroscopy (PAS) is being performed. PAS has the ability to depth-profile a sample. This can be accomplished by varying the modulation frequency of the impinging radiation, which varies the penetration depth (more properly known as the thermal diffusion length, μs) by the equation:5 μs = (k/ πρCf)0.5 where k is the thermal conductivity, ρ is the density of the sample, C is the specific heat of the sample, and f is the modulation frequency of the radiation. In a rapid-scanning Michelson interferomter, the modulation frequency is equal to 2Vv Hz,6 where V is the optical velocity and v is the wavenumber. Since V is a constant, the modulation frequency is proportional to the wavenumber of the incident radiation. Thus, the penetration depth will vary across a spectrum. A step-scanning interferometer alleviates this limitation. Radiation modulation in a step-scanning interferometer is accomplished by either an external chopper (amplitude modulation) or by vibrating an interferometer mirror (phase modulation) and thus all of the radiation is modulation at the same frequency. Thus, penetration depth is uniform across a spectrum. The ability of the step-scanning interferometer/PAS combination to depth profile is demonstrated in this work. No consideration has ever been given to the effect of mirror positioning inaccuracies on the signal-to-noise ratio in PAS measurements. The interferogram is collected by stopping the moving mirror once every laser wavelength at the mid-points of the laser interference record. By analogy to a rapid scanning interferometer, these are known as the Laser Zero Crossings (LZC). The maximum signal-to-noise ratio allowed for by a positional error t is given by:3 SNR = (4/Ai5max) where 'max is the maximum spectral wavenumber. Positioning inaccuracies can be from the failure of the mirror to be moved between equally spaced intervals of retardation or from the effect of mirror drift once at these sampling points. The instrument used in the present study utilizes a mechanical drive that was observed to drift more than 10 nm from the desired retardation position over a period of a few seconds. To correct for this drift a piezoelectric transducer (PZT) was mounted between the drive and mirror; this is discussed more fully and the ability of the PZT to maintain the mirror position is demonstrated.
|
