Amy Johnson Building, Department of Automatic Control and Systems Engineering, University of Sheffield, Portobello St., Sheffield S1 3JD, UK.
Sir Robert Hadfield Building, Department of Materials Science and Engineering, University of Sheffield, Mappin St., Sheffield S1 3JD, UK.
Sensors (Basel). 2023 Jun 4;23(11):5326. doi: 10.3390/s23115326.
The requirement for alternatives in roll-to-roll (R2R) processing to expand thin film inspection in wider substrates at lower costs and reduced dimensions, and the need to enable newer control feedback options for these types of processes, represents an opportunity to explore the applicability of newer reduced-size spectrometers sensors. This paper presents the hardware and software development of a novel low-cost spectroscopic reflectance system using two state-of-the-art sensors for thin film thickness measurements. The parameters to enable the thin film measurements using the proposed system are the light intensity for two LEDs, the microprocessor integration time for both sensors and the distance from the thin film standard to the device light channel slit for reflectance calculations. The proposed system can deliver better-fit errors compared with a HAL/DEUT light source using two methods: curve fitting and interference interval. By enabling the curve fitting method, the lowest root mean squared error (RMSE) obtained for the best combination of components was 0.022 and the lowest normalised mean squared error (MSE) was 0.054. The interference interval method showed an error of 0.09 when comparing the measured with the expected modelled value. The proof of concept in this research work enables the expansion of multi-sensor arrays for thin film thickness measurements and the potential application in moving environments.
在卷对卷（R2R）处理中，需要替代方案来扩大更宽基底的薄膜检测范围，降低成本和缩小尺寸，并且需要为这些类型的工艺提供新的控制反馈选项，这为探索新型小型光谱仪传感器的适用性提供了机会。本文介绍了一种使用两种最先进的传感器进行薄膜厚度测量的新型低成本光谱反射率系统的硬件和软件开发。为了使用所提出的系统进行薄膜测量，需要调整两个 LED 的光强、两个传感器的微处理器积分时间以及薄膜标准到设备光通道狭缝的距离等参数。与使用 HAL/DEUT 光源相比，该系统通过两种方法（曲线拟合和干涉间隔）可以提供更好的拟合误差：曲线拟合和干涉间隔。通过启用曲线拟合方法，获得了最佳组件组合的最低均方根误差（RMSE）为 0.022，最低归一化均方误差（MSE）为 0.054。干涉间隔方法在比较测量值和预期模型值时显示误差为 0.09。这项研究工作的概念验证为薄膜厚度测量的多传感器阵列扩展以及在移动环境中的潜在应用奠定了基础。
