Fig. 1. Structure diagram of (a) traditional method and (c) interference-aided spectrum-fitting method. Simulated transmission spectra of (b) traditional method and (d) interference-aided spectrum-fitting method with different thickness undetermined films.

Fig. 2. (a) Simulated peak shift and amplitude caused by undetermined film vs. refractive index difference between interference layer and substrate, and (b) simulated peak shift and peak position versus thickness of interference layer. The refractive indexes and thicknesses of the substrate and undetermined film were set as 1.52 and 1.70, 1.0 and 5.0 nm in this simulation, respectively.

Fig. 3. (a) Measured (taken by Lambda 950) and fitted transmission spectra of interference layer Nb2O5 deposited on K9 glass; measured (taken by Lambda 950), fitted transmission spectra and AFM results of (b) 3.5, (c) 5.4, (d) 7.3, (e) 11.8, and (f) 15.8 nm ultra-thin films on interference layer.

Fig. 4. (a) Transmission spectra of interference layers n=1.9 and 2.1 and structure diagram of double interference-aided spectra fitting method, and (b) peak shift vs. thickness of undetermined films on interference layers n=1.9 and 2.1.

Table 1. Comparison of Simulated Thickness Determination Results Between Traditional and Our Methods with SNR of 1000∶1

Table 2. Results of Ultra-thin Film Thickness Experimentally Determined with Different Methods