The artificial neural networks (ANNs) have been often used for thin-film thickness measurement, whose performance evaluations were only conducted at the level of simple comparisons with the existing analysis methods. However, it is not an easy and simple way to verify the reliability of an ANN based on international length standards. In this article, we propose for the first time a method by which to design and evaluate an ANN for determining the thickness of the thin film with international standards. The original achievements of this work are to choose parameters of the ANN reasonably and to evaluate the training instead of a simple comparison with conventional methods. To do this, ANNs were built in 12 different cases, and then trained using theoretical spectra. The experimental spectra of the certified reference materials (CRMs) used here served as the validation data of each trained ANN, with the output then compared with a certified value. When both values agree with each other within an expanded uncertainty of the CRMs, the ANN is considered to be reliable. We expect that the proposed method can be useful for evaluating the reliability of ANN in the future.
In this study, an optical method that allows simultaneous thickness measurements of two different layers distributed over a broad thickness range from several tens of nanometers to a few millimeters based on the integration of a spectroscopic reflectometer and a spectral-domain interferometer is proposed. Regarding the optical configuration of the integrated system, various factors, such as the operating spectral band, the measurement beam paths, and the illumination beam type, were considered to match the measurement positions and effectively separate two measurement signals acquired using both measurement techniques. Furthermore, for the thickness measurement algorithm, a model-based analysis method for high-precision substrate thickness measurements in thin-film specimens was designed to minimize the measurement error caused by thin films, and it was confirmed that the error is decreased significantly to less than 8 nm as compared to that when using a Fourier-transform analysis. The ability to undertake simultaneous thickness measurements of both layers using the proposed system was successfully verified on a specimen consisting of silicon dioxide thin film with nominal thicknesses of 100 nm and 150 nm and a 450 μm-thick silicon substrate, resulting in the exact separation between the two layers. From measurement uncertainty evaluation of a thin-film, a substrate in a thin-film specimen, and a single substrate, the uncertainties were estimated to be 0.12 nm for the thin-film, 0.094 μm for the substrate in a thin-film specimen, and 0.076 μm for the substrate. The measurement performance of thicknesses distributed on multi-scale was verified through comparative measurements using standard measurement equipment for several reference samples.
The importance of dimensional metrology has gradually emerged from fundamental research to high-technology industries. In the era of the fourth industrial revolution, absolute distance measurements are required to cope with various applications, such as unmanned vehicles, intelligent robots, and positioning sensors for smart factories. In such cases, the size, weight, power, and cost (SWaP-C) should essentially be restricted. In this paper, sub-100 nm precision distance measurements based on an amplitude-modulated continuous-wave laser (AMCW) with an all-fiber photonic microwave mixing technique is proposed and realized potentially to satisfy SWaP-C requirements. Short and long target distances of 0.879 m and 8.198 m were measured by detecting the phase delay of 15 GHz modulation frequencies. According to our measurement results, the repeatability could reach 43 nm at an average time of 1 s, a result not previously achieved by conventional AMCW laser distance measurement methods. Moreover, the performance by the proposed method in terms of Allan deviation is competitive with most frequency-comb-based absolute distance measurement methods, even with a simple configuration. Because the proposed method has a simple configuration such that it can be easily utilized and demonstrated on a chip-scale platform using CMOS-compatible silicon photonics, it is expected to herald new possibilities, leading to the practical realization of a fully integrated chip-scale LIDAR system.