Collaborative Research: Environmental Sensing of Per and Polyfluoroalkyl Substances in Water Utilizing a Microelectrode Sensor Array Platform and Machine Learning Enabled Detection

全氟烷基和多氟烷基物质（PFAS）是一组“永久化学品”，用于许多消费品和工业产品，包括不粘炊具，油漆，衣服，清洁产品，食品包装和消防泡沫。这些产品要么被释放到环境中，要么被填埋，因此有可能污染天然沃茨和饮用水源。医学研究表明，暴露于极低水平的PFAS可能导致婴儿长期发育障碍，增加不孕症和癌症风险。因此，PFAS污染是一个重要的环境问题，低成本的快速和可靠的监测方法是必要的。目前的PFAS检测方法需要昂贵的设备和专门的培训来维护复杂的仪器。该项目的总体目标是建立一种低成本的PFAS传感方法来监测水中的PFAS污染。这一目标将通过开发微电极传感器阵列与机器学习算法相结合来实现，以检测不同水源中PFAS的混合物。该项目的成功完成将通过开发一种低成本的方法来监测PFAS，从而造福社会。通过对学生的教育和宣传，包括对芝加哥伊利诺斯大学的两名研究生和普渡大学的一名本科生的指导，将为社会带来更多的好处。目前的PFAS分析依赖于与昂贵且笨重的质谱检测器耦合的色谱方法。虽然这些方法对于PFAS的精确低水平定量是有用的，但是它们不是移动的，并且它们需要专门的培训来维护复杂的仪器。该项目的总体目标是创建一个自下而上的框架，用于开发移动的、低成本的PFAS传感平台，这些平台可以在现场和使用点用于监测水中的PFAS污染。所提出的框架将通过功能化微电极传感器阵列（梅萨）平台的开发，结合机器学习算法，用于检测和定量PFAS在不同水基质中的一系列物理性质的混合物来证明。本项目的具体研究目标是：（1）利用电化学石英晶体微天平实验装置表征PFAS在吸附材料上的基本吸附/脱附机理：（2）利用计算密度泛函理论计算揭示控制PFAS在不同吸附材料上吸附/脱附的比表面相互作用;（3）整合实验-计算结果，以指导选择用于各种PFAS的选择性可逆吸附剂;以及（4）制造和测试用于PFAS检测的机器学习梅萨平台。该项目的成功完成有可能通过开发一种传感器产生变革性影响，该传感器可选择性检测PFAS混合物中的单个化合物，其检测限在低ng/L浓度范围内，并在不同的水源基质中具有可靠的性能。通过每年为本科生提供的暑期研究体验，以及创建一个为期四周的研讨会，向高中生介绍机器学习概念，将进一步造福社会。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

Per- and polyfluoroalkyl substances (PFAS) are a group of “forever chemicals” that are used in numerous consumer and industrial products including non-stick cookware, paints, clothes, cleaning products, food packaging, and firefighting foams. These products are either released to the environment or disposed of in landfills and, therefore, have the potential to contaminate natural waters and drinking water sources. Medical studies suggest that exposure to very low levels of PFAS could result in long-term developmental disabilities in infants, increased infertility, and risk of cancer. As such, PFAS contamination is an important environmental problem, and low-cost methods for rapid and reliable monitoring are necessary. Current PFAS detection methods require expensive equipment and specialized training to maintain the complicated instrumentation. The overall objective of this project is to create a low-cost PFAS sensing method to monitor PFAS contamination in water. This objective will be accomplished by developing a microelectrode sensor array coupled with machine learning algorithms to detect a mixture of PFAS in diverse water sources. The successful completion of this project will benefit society through the development of a low-cost method to monitor PFAS. Additional benefits to society will be achieved through student education and outreach including the mentoring of two graduate students at the University of Illinois at Chicago and an undergraduate student at Purdue University.Low-cost methods for rapid and reliable monitoring of per- and polyfluoroalkyl substances (PFAS) are greatly needed. Current PFAS analysis relies on chromatographic methods coupled to expensive and bulky mass spectrometric detectors. While these methods are useful for accurate low-level quantification of PFAS, they are not mobile, and they require specialized training to maintain the complicated instrumentation. The overarching objective of this project is to create a bottom-up framework for the development of mobile, low-cost PFAS sensing platforms that can be used in-situ and at the point-of-use to monitor PFAS contamination in water. The proposed framework will be demonstrated through the development of a functionalized microelectrode sensor array (MESA) platform, coupled with machine learning algorithms, for the detection and quantification of a mixture of PFAS with a range of physical properties in diverse water matrices. The specific research objectives of this project are to: (1) characterize the fundamental adsorption/desorption mechanisms of PFAS on sorbent materials using an electrochemical quartz crystal microbalance experimental setup; (2) utilize computational density functional theory calculations to reveal the specific surface interactions that control PFAS adsorption/desorption on different sorbent materials; (3) integrate the experimental-computational results to guide the selection of selective, reversible adsorbents for various PFAS; and (4) fabricate and test a machine-learning enabled MESA platform for PFAS detection. The successful completion of this project has potential for transformative impact through the development of a sensor for the selective detection of individual compounds within a PFAS mixture with detection limits in the low ng/L concentration range and reliable performance in varying source water matrices. Further benefits to society will be accomplished through an annual summer research experience for undergraduates and by creating a four-week workshop to introduce machine-learning concepts to high school students.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.