The timely detection of aqueous analytes is critical to decision-makers in agriculture, industry, and municipalities. However, nearly all aqueous sensor systems rely on single-point measurements, often taken at an instantaneous point in time and in one location, which can limit their ability to detect analytes passing through the aqueous solution at other locations or times. In this work, we present the concept of employing a mass-manufactured nanotextured diffraction surface as a variable-area sensor system capable of providing spectrophotometric information about aqueous analytes across multiple locations over time. We show that by placing the nanotextured surface of the sensor system under or behind a water sample, the water can be scanned by simply changing the location or angle of the light source and detector. We demonstrate the detection and quantification of a variety of aqueous analytes, including visible and ultraviolet (UV)-absorbing dyes, dust particles, and microalgae species, at accuracies similar to those of commercial equipment. A machine-learning algorithm was used to lower the limit of detection of dye from 5 to 3 μg/mL as well as automate the classification of distinct analyte types. These results demonstrate that using a mass-manufactured, textured surface can offer benefits as aqueous sensors, facilitating widely deployable aqueous analyte monitoring in a variety of applications.

中文翻译：

---

可变区域传感器允许对水性生物和化学分析物进行近乎连续的多点测量


及时检测水性分析物对于农业、工业和市政当局的决策者至关重要。然而，几乎所有的水溶液传感器系统都依赖于单点测量，通常在一个位置的瞬间时间点进行测量，这可能会限制它们检测在其他位置或时间通过水溶液的分析物的能力。在这项工作中，我们提出了将大规模制造的纳米纹理衍射表面用作可变面积传感器系统的概念，该系统能够随时间提供有关多个位置的水性分析物的分光光度信息。我们表明，通过将传感器系统的纳米纹理表面放置在水样下方或后面，只需改变光源和检测器的位置或角度即可扫描水。我们展示了各种水性分析物的检测和定量，包括可见光和吸收紫外线 （UV） 的染料、灰尘颗粒和微藻种类，其精度与商用设备相似。使用机器学习算法将染料的检测限从 5 μg/mL 降低到 3 μg/mL，并自动对不同分析物类型进行分类。这些结果表明，使用大规模制造的纹理表面可以作为水性传感器，有助于在各种应用中广泛部署水性分析物监测。