Determination of 31 perfluoroalkyl and polyfluoroalkyl substances in water by ultra performance liquid chromatography-triple quadrupole mass spectrometry combined with direct injection

Perfluoroalkyl and polyfluoroalkyl substances （PFAS） possess desirable properties， including hydrophobicity， oleophobicity， surface activity， and thermal and chemical stability. Their extensive production and widespread application have resulted in the pervasive presence of PFAS in diverse environmental media. However， accumulating evidence indicates that PFAS are persistent， capable of long-range transport， bioaccumulative， and toxic； consequently， their adverse effects on ecosystems and humans are of widespread concern. Aquatic environments serve as a major transport pathway and contamination route for PFAS， making accurate measurement of PFAS levels in water crucial for assessing associated environmental and health risks. However， accurate quantification requires multi-step procedures， including sample filtration， enrichment， nitrogen blow-down concentration， and reconstitution， such as solid-phase extraction （SPE） and accelerated solvent extraction （ASE）. These methods are often labor-intensive and time-consuming. Although research on fully automated SPE technology is increasing， it necessitates installation of online SPE systems， which entail high costs and may present limitations in sample throughput per run. With continuous advancements in mass spectrometry， instrumental sensitivity has improved considerably， making direct injection of water samples for multi-analyte analysis technically feasible. However， reports on the use of direct injection methods for detecting PFAS in water remain limited， and the number of target analytes covered in such studies is relatively small. In this study， a direct injection-ultra performance liquid chromatography-triple quadrupole mass spectrometry （UPLC-MS/MS） method was developed for the determination of 31 PFAS in water. To optimize the chromatographic separation， enhance the detection sensitivity of target analytes， and minimize undesirable adsorption losses， the method was meticulously optimized with respect to solvent selection， injection volume， and syringe filter type. Our method involves the following procedure： 0.5 mL of water is aliquoted， mixed with 0.5 mL of methanol spiked with 2 ng of internal standard， and filtered through a 0.22 μm polypropylene membrane. The PFAS were analyzed by UPLC-MS/MS with an injection volume of 35 µL. The analytes were ionized in electrospray ionization negative mode （ESI-） with scheduled multiple-reaction monitoring （sMRM）. The MS parameters， including precursor and product ions， collision energy， and declustering voltage were optimized. Through optimization of the analytical column and mobile phases， the analytes were separated on an RSLC 120 C18 column with a gradient of methanol and 5 mmol/L ammonium acetate aqueous solution as the mobile phase in a gradient elution program. The results were quantified by the internal standard method. The method demonstrated excellent linearity （R²>0.994） across a defined concentration range. The limits of detection （LODs） and quantification （LOQs） were 0.007 1–3.0 ng/L and 0.024–10 ng/L， respectively. Recoveries at spiked levels of 2， 10， and 500 ng/L ranged from 67.2% to 130.2%， with relative standard deviations （RSDs） of 0.30% to 18%. To quantify the effective equivalence between the enrichment efficiency of SPE and the sensitivity of direct injection methods， a comparative analysis of analyte recovery rates was performed for both approaches. Furthermore， for long-chain PFAS， direct injection demonstrated consistent and favorable recovery performance. The method was applied to analyze PFAS in groundwater samples. The results showed that 24 PFAS were detectable with the total PFAS content （∑PFAS） ranging from 20.6 to 521 ng/L， with perfluorooctanoic acid （PFOA） and perfluorobutanoic acid （PFBA） being the primary pollutants. This approach is simple， rapid， highly sensitive， and provides broad coverage of target analytes， making it suitable for the quantitative analysis of PFAS in urban groundwater. It offers an efficient and reliable technical solution for determining trace-level PFAS in environmental water samples.

全氟和多氟烷基物质（Perfluoroalkyl and polyfluoroalkyl substances，PFAS）具备诸多优异特性，包括疏水性、疏油性、表面活性以及热与化学稳定性。其大规模生产与广泛应用，使得PFAS在各类环境介质中普遍存在。然而，越来越多的研究证据表明，PFAS具有持久性、长距离迁移能力、生物累积性与毒性，因此其对生态系统和人类健康的不利影响受到了广泛关注。水环境是PFAS的主要迁移路径与污染途径，因此精准测定水体中的PFAS水平，对评估相关环境与健康风险至关重要。但精准定量往往需要多步前处理流程，包括样品过滤、富集、氮吹浓缩与复溶，例如固相萃取（solid-phase extraction，SPE）和加速溶剂萃取（accelerated solvent extraction，ASE）。这类方法通常劳动强度大且耗时较长。尽管全自动固相萃取技术的研究日益增多，但该技术需要安装在线固相萃取系统，不仅成本高昂，还可能存在单批次样品处理量有限的局限。随着质谱技术的持续进步，仪器灵敏度得到了大幅提升，使得直接进样水体样品进行多分析物检测在技术上具备可行性。不过，目前采用直接进样法检测水体中PFAS的相关报道仍较为有限，且此类研究覆盖的目标分析物数量相对较少。本研究开发了一种直接进样-超高效液相色谱-三重四极杆质谱（direct injection-ultra performance liquid chromatography-triple quadrupole mass spectrometry，UPLC-MS/MS）方法，用于测定水体中的31种PFAS。为优化色谱分离效果、提升目标分析物的检测灵敏度并减少不必要的吸附损失，研究人员针对溶剂选择、进样体积与注射器过滤器类型进行了细致优化。本方法的具体流程如下：量取0.5 mL水样，与0.5 mL添加了2 ng内标的甲醇混合，经0.22 μm聚丙烯膜过滤。进样体积为35 μL，采用UPLC-MS/MS对PFAS进行分析。分析物在电喷雾负离子模式（electrospray ionization negative mode，ESI-）下电离，并采用预设多反应监测（scheduled multiple-reaction monitoring，sMRM）模式。研究人员对包括前体离子、产物离子、碰撞能量与去簇电压在内的质谱参数进行了优化。通过优化分析色谱柱与流动相，最终采用RSLC 120 C18色谱柱，以甲醇与5 mmol/L乙酸铵水溶液作为流动相，通过梯度洗脱程序实现分析物分离。采用内标法进行结果定量。该方法在设定的浓度范围内展现出优异的线性关系（决定系数R²>0.994）。方法的检出限（limits of detection，LODs）与定量限（limits of quantification，LOQs）分别为0.0071~3.0 ng/L与0.024~10 ng/L。在2、10与500 ng/L三个加标水平下，回收率介于67.2%~130.2%之间，相对标准偏差（relative standard deviations，RSDs）为0.30%~18%。为对比固相萃取的富集效率与直接进样法的灵敏度的等效性，研究对两种方法的分析物回收率进行了比较分析。此外，对于长链PFAS，直接进样法展现出稳定且良好的回收性能。该方法被应用于地下水样品中的PFAS分析。结果显示，24种PFAS可被检出，总PFAS含量（∑PFAS）介于20.6~521 ng/L之间，全氟辛酸（perfluorooctanoic acid，PFOA）与全氟丁酸（perfluorobutanoic acid，PFBA）为主要污染物。该方法简便、快速、灵敏度高，且覆盖的目标分析物范围广泛，适用于城市地下水中PFAS的定量分析，为环境水样中痕量PFAS的检测提供了一种高效可靠的技术解决方案。