Chemosphere 77, 2009, 1455 1456 .pdf
Original filename: Chemosphere 77, 2009, 1455-1456.pdf
Title: Comment on •Ab initio study of the structural, electronic, and thermodynamic properties of linear perfluorooctane sulfonate (PFOS) and its branched isomersŽ by F.J. Torres, V. Ochoa-Herrera, P. Blowers, R. Sierra-Alvarez [Chemosphere (2009), doi:10.101
Author: "Sierra Rayne; Kaya Forest"
This PDF 1.4 document has been generated by Elsevier / Acrobat Distiller 8.1.0 (Windows), and has been sent on pdf-archive.com on 03/11/2015 at 04:11, from IP address 71.17.x.x.
The current document download page has been viewed 380 times.
File size: 125 KB (2 pages).
Privacy: public file
Download original PDF file
Chemosphere 77 (2009) 1455–1456
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/chemosphere
Letter to the Editor
Comment on ‘‘Ab initio study of the structural, electronic, and
thermodynamic properties of linear perﬂuorooctane sulfonate
(PFOS) and its branched isomers” by F.J. Torres, V. Ochoa-Herrera,
P. Blowers, R. Sierra-Alvarez [Chemosphere 76 (8) (2009) 1143–
Torres et al. (2009) conducted an ab initio computational study
on the anionic/acid forms and sodium/lithium salts of the linear
perﬂuorooctane sulfonate (n-PFOS) and its six monotriﬂuoromethyl branched isomers. The authors compare the gas phase relative
thermodynamic stabilities of these compounds using the
calculated Gibbs free energies and report that the linear, 1-monotriﬂuoromethyl branched (1-CF3-PFOS), and 6-monotriﬂuoromethyl (6-CF3-PFOS) branched congeners are signiﬁcantly more stable
(by up to 25 kJ mol 1) than the corresponding 2-(2-CF3-PFOS), 3(3-CF3-PFOS) 4-(4-CF3-PFOS), and 5-monotriﬂuoromethyl (5-CF3PFOS) branched congeners. With this data, Torres et al. conclude
that the ‘‘1-CF3- and 6-CF3-PFOS branched isomers and the linear
compound are the most favorable structures from the thermodynamic point of view in agreement with the experimental study
on the differences in the isomer composition of PFOS derivatives
(Vyas et al., 2007).” Previous work from this group (Ochoa-Herrera
et al., 2008) conducted similar ab initio calculations on the acid
forms of these compounds at the B3LYP/6–31++G(d,p) level using
GAUSSIAN03. In Ochoa-Herrera et al. (2008), the relative Gibbs free
energy values of the acid forms were similarly compared to a
technical mixture composition also obtained in this previous work.
In Vyas et al. (2007), only the linear and 6-CF3-PFOS congeners
were clearly quantiﬁed in various technical mixtures. We note that
this paper is not clear on whether quantiﬁcation of 1-CF3-PFOS is
included in the a-methyl branched subclass (perhaps along with
1,10 -CF3-PFOS), or whether 1-CF3-PFOS is included in the internally
monomethyl branched subclass. It is clear, though, that Vyas et al.
(2007) did not separate the 2-, 3-, 4-, and 5-CF3-PFOS congeners,
and instead, reported a summed contribution from these three
compounds as ‘‘[i]nternally monomethyl branched” (see Table 1
in Vyas et al. (2007)). Assuming that Vyas et al. (2007) included
1-CF3-PFOS in the a-methyl branched subclass, and that there
was a negligible contribution from the 1,10 -CF3-PFOS (an assumption consistent with other groups [see below] not reporting this dia-substituted congener in various other PFOS technical mixtures
analyzed to date), this yields the following reported range in isomeric contributions among the sulfonyl ﬂuoride, sulfonic acid,
and potassium sulfonate PFOS mixtures analyzed by Vyas et al.
(2007): n-PFOS, 71.8–82.2%; 1-CF3-PFOS, 0.1–3.5%; 2-, 3-, 4-, and
5-CF3-PFOS, 8.0–18.2%; and 6-CF3-PFOS, 9.4–10.8%.
Torres et al. fail to acknowledge, discuss, and incorporate their
current computational results into the following suite of previous
DOI of original article: 10.1016/j.chemosphere.2009.04.009
0045-6535/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
studies that report isomeric proﬁles of technical PFOS mixtures.
Langlois et al. (2007) analyzed a technical PFOS mixture by
GC-MS and were unable to identify and quantitate any monomethyl branched isomers with the exception of 6-CF3-PFOS, which
they qualitatively reported as the major non-linear congener. Subsequent work by Arsenault et al. (2008a,b) using 19F NMR on a
variety of technical PFOS mixtures provided the following range
of linear and monomethyl branched isomer contributions: n-PFOS,
62.3–78.9%; 1-CF3-PFOS, 1.0–3.2%; 2-CF3-PFOS, 0.6–2.2%; 3-CF3PFOS, 1.9–6.5%; 4-CF3-PFOS, 2.2–5.3%; 5-CF3-PFOS, 4.5–8.0%; and
6-CF3-PFOS, 9.0–11.4% (the br-PFOSK and L-PFOSK samples are
not included as they represented sequentially recrystallized
samples of a commercial mixture). Houde et al. (2008), using
LC-MS/MS, found the following technical PFOS standard isomer
proﬁles: n-PFOS, 76.9 ± 0.1%; 1-/5-CF3-PFOS, 10.6 ± 0.3%; 3-/4CF3-PFOS, 5.1 ± 0.2%; and 6-CF3-PFOS, 3.9 ± 0.2%. Most recently,
Chu and Letcher (2009) using GC–MS have reported the following
linear and monomethyl branched isomer contributions in a technical PFOS product: n-PFOS, 65.0 ± 1.8%; 1-CF3-PFOS, 0.9 ± 0.1%; 2CF3-PFOS, 1.1 ± 1.1%; 3-CF3-PFOS, 7.2 ± 7.2%; 4-CF3-PFOS, 5.6 ±
0.3%; 5-CF3-PFOS, 6.9 ± 0.3%; and 6-CF3-PFOS, 11.3 ± 0.6%. Finally,
we note with interest that Torres et al. (2009) fail to cite their prior
analyses of a technical PFOS mixture in Ochoa-Herrera et al.
(2008), where the authors ﬁrst discussed their hypotheses on
linking the ab initio calculated Gibbs free energies to the following
isomer patterns using LC–MS/MS they reported at that time:
n-PFOS, 75.4%; 1-CF3-PFOS, 8.7%; 3-/4-CF3-PFOS, 2.7%; and 5-/6CF3-PFOS, 13.2%. This literature database highlights the wide
variability in isomer contributions within technical PFOS mixtures,
a complexity also not addressed by Torres et al. in their thermodynamic analysis.
According to the ab initio calculated Gibbs free energies values
and logical arguments put forward by Torres et al. (2009), technical
PFOS mixtures should be dominated by n-PFOS, 1-CF3-PFOS, and
6-CF3-PFOS. As set out in the prior literature above, this is not the
case. Torres et al. (2009) either predict that the 1-CF3-PFOS is the
most thermodynamically stable congener (by up to 10 kJ mol 1 for
the anion, and between 3.5 and 4.0 kJ mol 1 for the sodium and lithium salts, respectively), or they predict that n-PFOS, 1-CF3-PFOS, and
6-CF3-PFOS have approximately equal Gibbs free energies for the
acid forms (within 2.1 kJ mol 1 of each other, which is likely within
the error range [ca. 1 kcal mol 1 = 4.2 kJ mol 1] for current computational methods). If the perﬂuoroalkyl chain rearrangements from
the straight chain hydrocarbon starting material are under thermodynamic control during synthesis of technical mixtures via electrochemical ﬂuorination, a logical prerequisite for a Gibbs free energy
based analysis such as that conducted by Torres et al. (2009), then
the isomeric distribution of the ﬁnal product should obey a Boltzmann distribution based on the respective free energies of each congener. If this was the case, n-PFOS, 1-CF3-PFOS, and 6-CF3-PFOS
would be by far the most dominant congeners (with 1-CF3-PFOS
Letter to the Editor / Chemosphere 77 (2009) 1455–1456
possibly dominating some technical mixtures even above the nPFOS isomer), and corresponding negligible contributions from the
far less thermodynamically stable 2- through 5-CF3-PFOS would
In all studies to date, with the exception of the prior work by
this group (Ochoa-Herrera et al., 2008) that is not cited in Torres
et al. (2009) as a source of technical PFOS composition data, the
thermodynamically more stable 1-CF3-PFOS (according to Torres
et al., 2009) is only a minor contributor, with higher contributions
from the 2- through 5-CF3-PFOS isomers that Torres et al. (2009)
claim are less thermodynamically stable, and which should therefore not be present in any signiﬁcant amount in the technical
mixtures. In their conceptual framework, this well established
presence of 6-CF3-PFOS as the major monomethyl branched congener, with lesser amounts of 1-CF3-PFOS than the 2- through 5-CF3PFOS isomers (and whose summative contributions are most often
greater than that of 6-CF3-PFOS), is inconsistent with their computational data and their postulate that the synthesis of technical
PFOS mixtures is under thermodynamic control. The general presence of the 2- through 5-CF3-PFOS isomers at individually higher
quantities than the 1-CF3-PFOS isomer in several prior studies to
date unequivocally shows that the combination of ab initio computational results and thermodynamic control assumptions put
forward by Torres et al. is not valid for understanding the synthetic
conditions that govern isomer distributions in PFOS mixtures.
Furthermore, the literature to date has identiﬁed more than just
monomethyl branched PFOS isomers in technical mixtures. Dimethyl branched congeners have also been structurally elucidated
and quantiﬁed in various PFOS mixtures. If, as Torres et al. argue,
the isomeric signature of PFOS technical mixtures can be understood by ﬁrst principles thermodynamic calculations, then their
conceptual framework should also apply to, and need to include
for logical consistency, all 89 potential geometrical isomers of
PFOS (Rayne et al., 2008), and not just the limited set of six monomethyl branched isomers under consideration. To validate their
approach, Torres et al. (2009) would have had to calculate the
Gibbs free energies for all 89 PFOS isomers and show that the more
branched derivatives are less thermodynamically stable than the
corresponding more linear derivatives, and then match these free
energy proﬁles to the best available PFOS technical mixture signatures from the literature (likely the proﬁles given in Arsenault et al.
(2008a) and Chu and Letcher (2009)). However, as we have demonstrated above, the limited computational analysis conducted
by Torres et al. does not match the reported monomethyl branched
proﬁles, nor the composite mono- and dimethyl-branched proﬁles
currently available. As such, it seems unlikely that the chain rear-
rangements that occur during the electrochemical ﬂuorination
synthetic methods for technical PFOS mixtures are under thermodynamic control; and thus, computational methods based on a free
energy analytical framework will not likely provide a rigorous
understanding of the isomer proﬁles of PFOS technical mixtures.
Arsenault, G., Chittim, B., McAlees, A., McCrindle, R., Riddell, N., Yeo, B., 2008a. Some
issues relating to the use of perﬂuorooctanesulfonate (PFOS) samples as
reference standards. Chemosphere 70, 616–625.
Arsenault, G., Chittim, B., Gu, J., McAlees, A., McCrindle, R., Robertson, V., 2008b.
Separation and ﬂuorine nuclear magnetic resonance spectroscopic (19F NMR)
perﬂuorooctanesulfonic acid (PFOS). Chemosphere 73, S53–S59.
Chu, S., Letcher, R.J., 2009. Linear and branched perﬂuorooctane sulfonate isomers
in technical product and environmental samples by in-port derivatization-gas
chromatography-mass spectrometry. Anal. Chem.. doi:10.1021/ac8027273.
Houde, M., Czub, G., Small, J.M., Backus, S., Wang, X., Alaee, M., Muir, D.C.G., 2008.
Fractionation and bioaccumulation of perﬂuorooctane sulfonate (PFOS) isomers
in a Lake Ontario food web. Environ. Sci. Technol. 42, 9397–9403.
Langlois, I., Berger, U., Zencak, Z., Oehme, M., 2007. Mass spectral studies of
perﬂuorooctane sulfonate derivatives separated by high-resolution gas
chromatography. Rapid Commun. Mass Spectrom. 21, 3547–3553.
Ochoa-Herrera, V., Sierra-Alvarez, R., Somogyi, A., Jacobsen, N.E., Wysocki, V.H.,
Field, J.A., 2008. Reductive deﬂuorination of perﬂuorooctane sulfonate. Environ.
Sci. Technol. 42, 3260–3264.
Rayne, S., Forest, K., Friesen, K.J., 2008. Congener-speciﬁc numbering systems for the
environmentally relevant C4 through C8 perﬂuorinated homologue groups of
alkyl sulfonates, carboxylates, telomer alcohols, oleﬁns, and acids, and their
derivatives. J. Environ. Sci. Health A 43, 1391–1401.
Torres, F.J., Ochoa-Herrera, V., Blowers, P., Sierra-Alvarez, R., 2009. Ab initio study of
the structural, electronic, and thermodynamic properties of linear
perﬂuorooctane sulfonate (PFOS) and its branched isomers. Chemosphere 76
Vyas, S.M., Kania-Korwel, I., Lehmler, H.J., 2007. Differences in the isomer
composition of perﬂuorooctanesulfonyl (PFOS) derivatives. J. Environ. Sci.
Health A 42, 249–255.
Ecologica Research, Penticton, British Columbia,
Canada V2A 8J3
Tel.: +1 250 490 0166
E-mail address: email@example.com.
Department of Chemistry, Okanagan College, Penticton,
Canada V2A 8E1
Available online 8 August 2009