We appreciate the interest from Titaley et al.(1) in our recent publication. To facilitate critical evaluation of our work, we provided more than 50 pages of Supporting Information including mass spectra of 32 compounds alongside the original article.(2) We are happy to see that the efforts to provide such detail has allowed for a critical review of our interpretation. Titaley et al. questioned the finding of gas phase or aerosolized perfluorooctanoic acid (PFOA) after agitation of an AFFF/water mixture, suggesting that thermal transformation of other PFASs present in AFFF may have occurred during analysis and that the products have the same chromatographic retention time and mass spectra. Titaley et al. reproduced our analytical approach by spiking standards of protonated PFOA (PFOA-H), deprotonated PFOA, and perfluoro-1-heptene (PFHP) into separate thermal desorption tubes (TD), which were analyzed using gas chromatography mass spectrometry (GC/MS). The instrumental conditions, excepting the desorption trap, were the same as in our work. Indeed, the retention times and fragmentation spectra for the three compounds were similar, potentially indicating that our quantification was of unresolved PFASs rather than PFOA. However, as noted by Titaley et al., PFOA may be thermally degraded in the thermal desorption system or the GC inlet, and therefore, its similarity in retention time and fragmentation to other PFASs does not preclude PFOA from being the aerosolized/gas phase compound. Quantification of compounds via surrogates is not specific to the question of PFOA; in fact, nearly all quantification by electron impact and/or tandem mass spectrometry is conducted via surrogates—usually fragment ions of the parent molecules. As stated in the original article, retention time and fragmentation matches to standards were the tools used by ourselves, and by others, to ensure the peak quantified was representative of assigned molecule. It is well-known that these tools are not infallible, and by showing that the retention time and spectra are similar for PFHP and PFOA, Titaley et al. have reinforced this but have not conclusively proven that the quantified peak is not PFOA. Thus, the suggestion that all PFCAs in the original article are misidentified is an overextension and may lead readers to doubt all GC/MS quantification based on retention time and fragmentation. Careful inspection of the spectra provided by Titaley et al. (Figure 1) in comparison to the spectra published in the original article (Figure S3) highlights the need for further investigation into this topic. First, as Titaley et al. correctly note, the spectra of the analyzed PFOA standard is not an exact match to the NIST spectra, lacking a prominent 45 m/z ion that is present in the NIST spectra.(3) However, this ion is also not present in the spectra provided by Titaley et al., which likely speaks to the NIST spectra more than it speaks to either of our measurements. Second, Titaley et al. attribute the presence of the 44 m/z ion to the decarboxylation of PFCAs, forming a CO₂ ion. Curiously, this ion is present in the spectra of both PFOA-H and PFHP, but PFHP does not contain a carboxyl group nor any O to form this ion. Additionally, the 44 m/z is not present in the reported PFOA spectra, which does contain a carboxyl group. The formation of this ion appears to be inconsistent with the presence or absence of a carboxyl group. Third, visual comparison of the reported spectra for PFOA provided in Figure S3 (original article(2)) to Figure 1 (Titaley et al.)(1) leads to the finding that the closest match is with PFOA (absence of 44 m/z, minor 50 m/z, absence of prominent 51 m/z), not with PFHP. Finally, the lack of a molecular ion by chemical ionization has limited or no implications to analysis performed by electron ionization (i.e., the lack of a molecular CI ion does not indicate that the molecule did not exist, only that it was not ionized). We find these fragmentation inconsistencies between instruments and between molecules fascinating and worthy of significant further study. Ultimately, we agree that in any GC/MS analysis based on fragmentation and retention time there is the possibility of misidentification, and it is impossible to entirely rule this out. The argument that MilSpec capped PFOA in AFFF at 800 ppb(4,5) is compelling but relies heavily on compliance rather than measurements. We have been intentionally forthcoming with spectra and supporting information in the original article to allow for such criticism. Whether the peak in question represents PFOA, another PFAS, or the sum of several PFASs, assuming similar ionization efficiency in the instrument and given the broad understanding that many PFASs are bioaccumulative and toxic,(6−9) the concentration is more than high enough to cause concern given the gentle agitation that was provided to these samples. We encourage additional research studying gas phase and aerosolized PFASs from AFFF handling and use to further our understanding of exposure. J.R. and I.A. contributed equally to this work. The authors declare no competing financial interest. This article references 9 other publications.

中文翻译：

---

对“从成膜泡沫中释放挥发性全氟和多氟烷基物质”评论的回应

我们感谢Titaley等人（1）在我们最近的出版物中的关注。为了便于对我们的工作进行严格的评估，我们在原始文章的旁边提供了超过50页的支持信息，其中包括32种化合物的质谱图。（2）我们很高兴看到提供此类详细信息的努力使得我们可以对本文进行严格的审查。我们的解释。Titaley等。质疑在搅动AFFF /水混合物后发现气相或雾化的全氟辛酸（PFOA），表明在分析过程中可能发生了AFFF中存在的其他PFAS的热转化，并且产品具有相同的色谱保留时间和质谱图。Titaley等。通过加标质子化PFOA（PFOA-H），去质子化PFOA，和全氟-1-庚烯（PFHP）分别放入热解吸管（TD）中，然后使用气相色谱质谱（GC / MS）进行分析。除解吸阱外，仪器条件与我们的工作相同。的确，这三种化合物的保留时间和碎片光谱相似，这可能表明我们的定量是未解析的PFAS而不是PFOA。但是，如Titaley等人所述，PFOA在热脱附系统或GC进样口中可能会发生热降解，因此，其保留时间的相似性和与其他PFAS的碎裂并不排除PFOA成为雾化/气相化合物。 。通过替代物对化合物进行定量对PFOA的问题不是特定的。事实上，几乎所有通过电子碰撞和/或串联质谱法进行的定量分析都是通过替代物进行的，替代物通常是母体分子的碎片离子。如原始文章所述，保留时间和与标准品匹配的碎片是我们自己和其他人使用的工具，以确保定量峰代表指定的分子。众所周知，这些工具并非万无一失，并且通过显示PFHP和PFOA的保留时间和光谱相似，Titaley等人。已经加强了这一点，但尚未最终证明量化峰不是PFOA。因此，关于原始文章中所有PFCA均被错误识别的建议过分延伸，可能会使读者怀疑所有基于保留时间和碎片化的GC / MS定量分析。仔细检查由Titaley等提供的光谱。（图1）与原始文章（图S3）中发布的光谱相比，强调了对该主题进行进一步研究的必要性。首先，如Titaley等人所述。正确地指出，所分析的PFOA标准品的光谱与NIST光谱不完全匹配，缺少显着的45NIST光谱中存在m / z离子。（3）但是，Titaley等人提供的光谱中也不存在该离子，这可能比我们测量的任何一种都更能说明NIST光谱。第二，Titaley等。归因于44 m / z离子的存在与PFCA的脱羧反应形成了CO ₂离子。奇怪的是，该离子同时存在于PFOA-H和PFHP的光谱中，但PFHP不含羧基或任何O形成该离子。此外，44 m / z在报告的PFOA光谱中不存在，它确实含有一个羧基。该离子的形成似乎与羧基的存在与否不一致。第三，将图S3（原始文章（2））和图1（Titaley等人）（1）中提供的PFOA的光谱进行直观比较，发现与PFOA的光谱最接近（不存在44 m / z，较小的50 m / z，不存在突出的51 m / z），而不是使用PFHP。最后，通过化学电离缺少分子离子对通过电子电离进行的分析具有有限的影响或没有影响（即，缺少分子CI离子并不表明该分子不存在，仅表明该分子未被电离）。我们发现，这些仪器之间以及分子之间的碎片不一致令人着迷，值得进行深入研究。最终，我们同意，在任何基于碎片和保留时间的GC / MS分析中，都有可能被误识别，并且不可能完全排除这种情况。关于MilSpec将AFFF中的PFOA限制为800 ppb（4,5）的论点是有说服力的，但在很大程度上取决于依从性而不是测量值。我们故意在原始文章中介绍了光谱和支持信息，以允许进行此类批评。假设仪器中的电离效率相似，并且广泛理解许多PFAS具有生物蓄积性和毒性，（6-9）所讨论的峰是代表PFOA，另一个PFAS还是多个PFAS的总和，（6-9）浓度足够高考虑到这些样品的轻微搅动，引起了关注。我们鼓励进行更多关于AFFF处理和使用中气相和气化PFAS的研究，以进一步了解暴露。JR和IA对这项工作做出了同样的贡献。作者宣称没有竞争性的经济利益。本文引用了其他9种出版物。假设仪器中的电离效率相近，并且广泛理解许多PFAS具有生物蓄积性和毒性，（6-9）浓度高到足以引起人们的关注，因为仪器提供了轻微的搅动这些样本。我们鼓励进行更多关于AFFF处理和使用中的气相和雾化PFAS的研究，以进一步了解暴露。JR和IA对这项工作做出了同样的贡献。作者宣称没有竞争性的经济利益。本文引用了其他9种出版物。假设仪器中的电离效率相近，并且广泛理解许多PFAS具有生物蓄积性和毒性，（6-9）浓度高到足以引起人们的关注，因为仪器提供了轻微的搅动这些样本。我们鼓励进行更多关于AFFF处理和使用中的气相和雾化PFAS的研究，以进一步了解暴露。JR和IA对这项工作做出了同样的贡献。作者宣称没有竞争性的经济利益。本文引用了其他9种出版物。（6-9）浓度高到足以引起关注，因为这些样品提供了轻微的搅动。我们鼓励进行更多关于AFFF处理和使用中的气相和雾化PFAS的研究，以进一步了解暴露。JR和IA对这项工作做出了同样的贡献。作者宣称没有竞争性的经济利益。本文引用了其他9种出版物。（6-9）浓度高到足以引起关注，因为这些样品提供了轻微的搅动。我们鼓励进行更多关于AFFF处理和使用中的气相和雾化PFAS的研究，以进一步了解暴露。JR和IA对这项工作做出了同样的贡献。作者宣称没有竞争性的经济利益。本文引用了其他9种出版物。