Conventional time domain Optical Coherence Tomography (OCT) relies on the detection of an interference pattern generated by the interference of backscattered light from the sample and a reference Optical Delay Line (ODL). By referencing the sample interference with the scan depth of the ODL, constructive interference indicates depth in the sample of a reflecting structure. Conventional ODLs used in time domain OCT require some physical movement of a mirror to scan a given depth range. This movement results in instrument degradation. Also in some situations it is necessary to have no moving parts. Stationary ODLs (SODLs) include dual Reflective Spatial Light Modulator (SLM) systems (Type I) and single Transmissive SLM with match-arrayed-waveguide systems (Type II).
In this paper, the method of fabrication and characterisation of a number of Stepped Mirrored Structures (SMS) is presented. These structures are intended for later use in proof-of-principle experiments that demonstrate Type II SODL: a six step, 2 mm step depth macro-SMS, an eight step 150 um deep micro-SMS with glue between steps, and a six step 150 um deep micro-SMS with no glue between steps.
These SMS are characterized in terms of their fabrication, step alignment and step height increment precision. The degree of alignment of each step was verified using half of a bulk Michelson interferometer. Step height was gauged using a pair of vernier callipers measuring each individual step. A change in notch frequency using an in-fibre Mach-Zhender interferometer was used to gauge the average step height and the result compared to the vernier calliper results.
The best aligned SMS was the micro-SMS prepared by method B with no glue between steps. It demonstrated a 95% confidence interval variation of 1% in reflected intensity, with the least variation in intensity within steps. This SMS also had the least absolute variation in step height increment: less than 8 um. Though less variation would be ideal, for producing micro-SMS for proof of principle experiments for Type II stationary ODL, of the method compared, method B, with no glue between steps, produced more reproducible step height increments and step alignment.