Robert Dumoulin(1), Stella M. L. Melo(1), Kimberly Strong(1), Scott Mabury(2), Naomi Stock(2), Michael Hurley(3), Jessica Misner(3), Timothy Wallington(3), Robert Waterland(4),(1)Department of Physics, University of Toronto, Toronto, ON, Canada, (2)Department of Chemistry, University of Toronto, Toronto, ON, Canada, (3)Ford Motor Company, Dearborn, MI, USA,

(4)Dupont Central Research and Development, E.I. DuPont de Nemours and Co. Inc., Wilmington, DE, USAAbstract - We report here on the measurements made at the University of Toronto using samples of pure fluorotelomer alcohols (FTOHs) in a 0.25-m-path cell at room temperature over the spectral region 500-4000 cm-1 using a Bomem DA8 Fourier transformspectrometer. We show that the FTOHs display strong absorption in the spectral region 1000-1500 cm-1. The resulting cross-sections are shown to be consistent with measurements done at Ford Research Laboratories using a Mattson Instruments Sirius 100

Fourier Transform Spectrometer and with theoretical calculations performed at DuPont Central Research and Development. These results will facilitate future laboratory studies and the spectroscopic detection of FTOHs in the atmosphere.

Results and DiscussionFigure 2 shows the spectrum of 1:2 FTOH obtained with 0.796 Torr of gas in the cell at 295 K. Alsoshown are the two empty cell spectra obtained before and after the 1:2 FTOH measurement. As can beseen from the figure, 1:2 FTOH has a strong and broad absorption feature in the spectral region 1100-1400 cm-1, and the stability of the system is seen in the similarity of the before and after empty cellspectra. Other features in the spectra are the water vapor absorption and CO2 lines.

Infrared absorbance cross sections were derived from the spectra using Beer’s Law:

(1)

where I0(ν) and I(ν) represent the intensity of a source incident and transmitted respectively through agas column of length L, P is the gas pressure in the cell at temperature T, P0 and T0 are the standardpressure and temperature, N0 is the Loschmidt number (2.687×1019 cm-3), and σ(ν) is the absorptioncross section at a given wavenumber ν.

Comparison of the University of Toronto (Ford) measurements shows that:

• 1:2 FTOH has strong absorption features at 1139 and 1260 cm-1 with cross-sections of 1.08x10-18 (1.08x10-18) and 9.39x10-19 (9.58x10-19) cm2 molec-1;

• 4:2 FTOH has strong absorption features at 1141 and 1246 cm-1 with cross-sections of 1.44x10-18 (1.52x10-18) and 3.81x10-18 (4.24x10-18) cm2 molec-1;

• 6:2 FTOH has strong absorption features at 1216 and 1251 cm-1 with cross-sections of 2.05x10-18 (1.75x10-18) and 5.69x10-18 (4.71x10-18) cm2 molec-1.

Time-dependent density functional theory (TD-DFT) is the standard methodfor rapid and accurate prediction of the properties of electronically excitedstates and for the estimation of UV and visible spectra. For the estimation ofIR spectra, DuPont re-optimized all molecular geometries and computedharmonic vibrational frequencies and intensities at the same level of theory onthe resultant optimized geometries. Vibrational frequencies were scaled by0.961 as recommended by Scott and Radom (1996). Figure 4 shows theresulting theoretical absorbance of 1:2, 4:2 and 6:2 FTOH. The theoreticalspectral peaks have been artificially broadened with Lorentzian lineshapefunctions of half-width at half-maximum of 10 cm-1 to allow comparison withthe experimental spectra.

Comparing Figures 3 and 4, we see that theory and experiment agree quitewell. Experimentally and theoretically, the spectra of the higher fluorinatedalcohols quite strongly resemble one another and are quite distinct from thespectrum of 1:2 FTOH. Theory and experiment have the highest absorbancepeaks in the same location and the relative intensities of the major spectralfeatures correspond reasonably well.

Acknowledgements

We thank Prof. James R. Drummond and the NSERC Industrial Research Chair in Atmospheric Remote Sounding from Space (sponsored by COMDEV, Bomem, AES(now MSC), CSA, and NSERC) for the use of the Bomem DA8 Fourier transform spectrometer. We also thank Mr. Paul Chen for his technical assistance with theexperiment, and Matthew Toohey for setting up the foundation of the experiment. This work was supported through funds from NSERC.

References

Ellis, D.A., Mabury, S.A., Martin, J.W., Hurley, M.D., Sullbaek, M.P., Wallington, T.J., Atmospheric Lifetime of Fluortelomer Alcohols, Environ. Sci. Technol. 37,3816, 2003.

Martin, J.W., D.C.G. Muir, C.A. Moody, D.A. Ellis, K.W. C, K.R. Solomon, and S.A. Mabury. Collection of airborne fluorinated organics and analysis by gaschromatography/chemical ionization mass spectroscopy, Analytical Chemistry, 74 (3), 584-590, 2002.

Scott, A. P. and L. Radom, J. Phys. Chem., 100, 16502, 1996.

Stock, N., F, Lau, D. Ellis, J. Martin, D. Muir, and S. Mabury. Polyfluorinated Telomer Alcohols and Sulfonamides in the North American Troposphere, Environ. Sci.Technol, 38, .991-996, 2004.

Waterland, R., M. Hurley, J. Misner, T.J. Wallington, S.M.L. Melo, K. Strong, R. Dumoulin, L. Castera, N. Stock, and S. Mabury. Gas Phase UV and IR AbsorptionSpectra of CF3CH2CH2OH and F(CF2CF2)xCH2CH2OH (x=2,3,4), to be submitted to Journal of Fluorine Chemistry.

For more information contact Robert Dumoulin at dumoulin@atmosp.physics.utoronto.ca

IntroductionFluorotelomer alcohols (FTOH) are industrial chemicals widely used in a variety of industrialproducts, such as paints, coatings, polymers, adhesives, waxes, polishes, electronic materials,and caulks. FTOHs are similar to long-tailed polyfluorinated compounds called PFOSs. Eventhough PFOS-based products were phased out of production by their primary manufacturer (3MCorporation) in 2003 after their widespread detection in biota and concern about PFOSs asprimary environmental pollutants, FTOHs continue to be produced for comparable applications.FTOHs are primary fluorinated alcohols, characterized by two non-fluorinated carbon atomsadjacent to the hydroxyl function, and are normally referred as 1:2, 4:2, 6:2, 8:2 and 10:2 FTOHon the basis of the ratio of perfluorinated to nonfluorinated carbons. The telomerization processused in their manufacture results only in even-numbered carbon chain lengths.

In 2004, Stock et al. reported the measured tropospheric concentration of 6:2 FTOH from sixsites in North America. Concentrations of the FTOH compound varied from 0 pg/m3 in rural orisolated sites (such as Long Point, ON and Winnipeg, MN) to 82 pg/m3 at urban sites (such asCleves, OH or Toronto, ON). Studies of the atmospheric concentration of 8:2 and 10:2 FTOHhave been published, but we currently do not have a measure of the tropospheric concentrationof 1:2 or 4:2 FTOH.

To facilitate laboratory studies of the atmospheric chemistry of FTOHs and the spectroscopicdetection of FTOHs in the atmosphere, an experimental study of the infrared spectra of x:2FTOH (x=1,4,6) has been performed. We report here on the infrared absorption cross-sectionmeasurements of 1:2, 4:2 and 6:2 FTOH.

Figure 2 - 1:2 FTOH absorption spectrum measured at 0.796 Torr and 295 K.

ConclusionsThe peak absorption cross-sections for 1:2, 4:2 and 6:2 FTOH were found to be centered at 1139, 1246, and 1251) cm-1, respectively,with corresponding magnitudes of 1.08, 3.81, and 5.69 x 10-18 cm2 molec-1. The two independently obtained sets of measurements arein very good agreement. For 1:2, 4:2 and 6:2 FTOH, our results agree within experimental error in regions of maximum cross-section.Although we obtain slightly different absorption cross-section magnitudes for 6:2 FTOH, there is a large experimental uncertainty inthe vapor pressure (6:2 FTOH has a vapor pressure of only about 0.05 Torr), which makes it very difficult to measure.

Although 1:2, 4:2 and 6:2 FTOH have large infrared cross-sections, we estimate their global warming potentials to be negligible; themeasured lifetime of the studied FTOH molecules is estimated to be about 10-20 days (Ellis et al. 2003). The infrared spectra reportedherein should aid in future laboratory and field studies of the atmospheric chemistry of these species.

€

I(ν)=I0(ν)exp[−σ (ν)N0L PP0T0T ]

Experimental SetupThe experimental setup at the University of Toronto (see Figure 1) allowed the measurement ofthe spectrum of the light source through the evacuated cell, followed by a measurement throughthe cell with sample gas present. Stability was assessed by recording another spectrum of theempty cell after the full cell measurement. Details of the experiment are tabulated below.

Figure 4 - Theoretical absorbance of 1:2, 4:2 and6:2 FTOH.

0.435 < p (1:2 FTOH) < 0.996, 0.253 < p (4:2 FTOH) < 0.717, 0.021< p (6:2 FTOH) < 0.145 (all pressures in units of Torr)

Measured pressure range foreach FTOH compound

MKS baratron 627 B 10-Torr head equipped with CCR MKS SR-54.5 digital readout

Pressure gauge

Purity of greater than 99%. Several freeze-pump-thaw cycles wereused before taking measurements, in order to further improve purity.

FTOH samples

25-cm stainless steel cell equipped with ZnSe windows.All spectra were recorded at room temperature (295 K).

Gas Cell

350Co-ads (to improve SNR)

0.5 cm-1 (results of 0.017 cm-1 tests demonstrated broad chemicalspectra structures only)

Resolution

500-4000 cm-1Spectral range

Silicon carbon igniter global source/ KBr beamsplitter/ MCT detectorSource/beamsplitter/detector

Bomem DA8 Fourier transform spectrometerInstrument

Uncertainties in the measured pressure, temperature and length of the cellintroduced negligible errors in the cross sections. The primary sources ofmeasurement error are drift in the FTS baseline (measured to be ≤ 1.5%, andtypically less than 1%, from empty cell spectra recorded before and after eachsample run), and the difficulty of dealing with the sticky low-vapor pressurecompounds.

To ensure consistency of results, the infrared cross-section of each FTOHcompound obtained at the University of Toronto was measured on at least fourdifferent days at various pressures. From this, we obtained consistent resultsand confirmed that the FTOH absorption cross-sections are independent ofpressure. The final 1:2, 4:2, and 6:2 FTOH results are the averages of 8, 6, and14 cross-sections, respectively.

We estimate the accuracy of the cross-sections to be 5-10% for 1:2 FTOH (and4:2 FTOH, and 10-15% for 6:2 FTOH. Random errors in the regions ofmaximum absorption were calculated as the standard deviation on the averagecross-section derived from a number of independent measurements. Theywere found to be 2% for 1:2 FTOH (using the eight independent cross-sectionmeasurements), 2% for 4:2 FTOH (using six cross-sections), and 3% for 6:2FTOH (using 14 cross-sections).

Figure 3 shows the absorption cross-sections for 1:2, 4:2 and 6:2 FTOH ascalculated by applying Equat ion (1) to measurements conducted at theUniversity of Toronto and at Ford research laboratories in Michigan.

Figure 1 - The experimental setup at the University of Toronto.

The experimental setup at Ford consisted of a Mattson Instruments, Sirius 100 Fourier transforminf rare d s pectr omet er, int erfa ced to a 140-L, 2-m -l ong eva cuable Pyre x cha mber . Thi sspectrometer was operated at a spectral resolution of 0.50 cm-1. Infrared spectra were derived from32 co-added interferograms, and the spectra were recorded at 296 K in the presence of 700 Torr ofair diluent.

European GeosciencesUnion 

General Assembly 2005Vienna, Austria

Tuesday 26 April 2005Poster Area Hall X

X078

Figure 3 - 1:2, 4:2 and 6:2 FTOH absorption cross-sectionsmeasured at the University of Toronto (red) and at Ford (blue).