Laboratoire de Physique des Interfaces et des Couches Minces (LPICM), Centre National de la Recherche Scientifique (CNRS), Ecole Polytechnique, IP Paris, 91128 Palaiseau, France.
Laboratoire Instrumentation, Simulation et Informatique Scientifique (LISIS), Département Composants et Systèmes (COSYS), Université Gustave Eiffel, 77447 Marne-La-Vallée, France.
Sensors (Basel). 2021 Dec 29;22(1):218. doi: 10.3390/s22010218.
Carbon nanotubes (CNTs) combine high electrical conductivity with high surface area and chemical stability, which makes them very promising for chemical sensing. While water quality monitoring has particularly strong societal and environmental impacts, a lot of critical sensing needs remain unmet by commercial technologies. In the present review, we show across 20 water monitoring analytes and 90 references that carbon nanotube-based electrochemical sensors, chemistors and field-effect transistors (chemFET) can meet these needs. A set of 126 additional references provide context and supporting information. After introducing water quality monitoring challenges, the general operation and fabrication principles of CNT water quality sensors are summarized. They are sorted by target analytes (pH, micronutrients and metal ions, nitrogen, hardness, dissolved oxygen, disinfectants, sulfur and miscellaneous) and compared in terms of performances (limit of detection, sensitivity and detection range) and functionalization strategies. For each analyte, the references with best performances are discussed. Overall, the most frequently investigated analytes are H (pH) and lead (with 18% of references each), then cadmium (14%) and nitrite (11%). Micronutrients and toxic metals cover 40% of all references. Electrochemical sensors (73%) have been more investigated than chemistors (14%) or FETs (12%). Limits of detection in the ppt range have been reached, for instance Cu(II) detection with a liquid-gated chemFET using SWCNT functionalized with peptide-enhanced polyaniline or Pb(II) detection with stripping voltammetry using MWCNT functionalized with ionic liquid-dithizone based bucky-gel. The large majority of reports address functionalized CNTs (82%) instead of pristine or carboxyl-functionalized CNTs. For analytes where comparison is possible, FET-based and electrochemical transduction yield better performances than chemistors (Cu(II), Hg(II), Ca(II), HO); non-functionalized CNTs may yield better performances than functionalized ones (Zn(II), pH and chlorine).
碳纳米管（CNTs）结合了高导电性、高比表面积和化学稳定性，因此非常适合用于化学传感。虽然水质监测对社会和环境具有特别强的影响，但商业技术仍无法满足许多关键的传感需求。在本综述中，我们展示了 20 种水质监测分析物和 90 篇参考文献，证明基于碳纳米管的电化学传感器、化学电阻器和场效应晶体管（chemFET）可以满足这些需求。另外还有 126 篇参考文献提供了背景和支持信息。在介绍水质监测挑战之后，总结了 CNT 水质传感器的一般工作原理和制造原则。根据目标分析物（pH 值、微量元素和金属离子、氮、硬度、溶解氧、消毒剂、硫和其他）对它们进行分类，并在检测限、灵敏度和检测范围方面进行比较，并讨论了功能化策略。对于每个分析物，都讨论了性能最佳的参考文献。总的来说，研究最多的分析物是 H（pH 值）和铅（各占 18%），其次是镉（14%）和亚硝酸盐（11%）。微量元素和有毒金属占所有参考文献的 40%。电化学传感器（73%）的研究比化学电阻器（14%）或 FET（12%）更广泛。例如，使用带有基于多肽增强聚苯胺的 SWCNT 的液体门控 chemFET 检测 Cu(II)，或使用带有基于离子液体-二硫腙的 bucky-gel 的 MWCNT 检测 Pb(II)，可以达到 ppt 级别的检测限。大多数报告都涉及功能化 CNT（82%），而不是原始或羧基功能化 CNT。对于可以进行比较的分析物，基于 FET 的和电化学转换的传感器比化学电阻器具有更好的性能（Cu(II)、Hg(II)、Ca(II)、HO）；非功能化 CNT 可能比功能化 CNT 具有更好的性能（Zn(II)、pH 值和氯）。
