CROATICA CHEMICA ACTA

CCACAA, ISSN 0011-1643, e-ISSN 1334-417X

Croat. Chem. Acta 84 (3) (2011) 399–406.
CCA-3488

Original Scientific Article

Application of Gas Chromatography to Determination of

Total Organic Fluorine after Defluorination of
Perfluorooctanoic Acid as a Model Compound†

Mariusz Koc,a Magdalena Anna Donten,a Jacek Musijowski,a Xinghua Guo,b

Alexander Fauland,b Ernst Lankmayr,b and Marek Trojanowicza,*
a
Department of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, Poland
b
Institute of Analytical Chemistry and Food Chemistry, Graz University of Technology, Graz, Austria

RECEIVED NOVEMBER 11, 2010; REVISED MAY 12, 2011; ACCEPTED MAY 13, 2011

Abstract. Because of the global presence of anthropogenic perfluorinated organic compounds in the envi-
ronment, foods and living organisms, and their large structural variety, it can be helpful to develop a me-
thod for determination of their total content at trace level in different matrices. In the developed method,
the defluorination was carried with sodium biphenyl, derivatization of released fluoride to triphenylfluoro-
silane and determination by gas chromatography. Three detection methods were compared: flame-
ionization detection, electron capture detection and mass spectrometry. Among them the MS detection
was found to be the most favorable one in terms of the instrumental limit of detection (LOD) , whereas the
flame-ionization detection was considered to be the most favorable in terms of the method limit of detec-
tion (MDL). For the initial sample volume of 1 L and performing the whole procedure of determination,
including preconcentration, the MDL value for perfluorooctanoic acid was evaluated as 0.043 ppb.
(doi: 10.5562/cca1798)
Keywords: perfluorinated organic compounds, surfactants, gas chromatography, total organic fluorine

INTRODUCTION the environmental samples and living organisms several

groups of compounds predominate, e.g. perfluorinated
The wide interest observed in the recent decades in the carboxylic and sulfonic acids, sulfonamides, and fluoro-
presence of fluorinated organic compounds in the envi- telomer alcohols.
ronment has different reasons and implications for the Due to a large number of environmentally occur-
alteration of various technologies, modification of nutri- ring compounds, their difficult identification and deter-
tional requirements, and diagnosis of some animal and mination at ultra-trace level, some attempts are focused
human diseases. Among commonly known reasons of on the determination of their total content. Reportedly,
this interest one can indicate e.g. problem of emission of so far in the analytical literature, attempts on determina-
freons to the atmosphere, significant impact of fluoride tion of the Total Organic Fluorine (TOF) follow two
ion on human health, and in the last decade the problem basic steps – release fluorine from organic compounds
of global proliferation of perfluorinated organic com- and determination of fluorine as fluoride ion employing
pounds. Although the presence of those fully anthropo- different analytical methods. In the proposed procedures
genic substances in human organisms was discovered at of the TOF determination defluorination step is carried
the end 1960s, only the studies conducted since 1990s out by combustion in oxygen in 900– 1000 °C,6,7 or
proved their global presence in environment and living by chemical reaction with sodium biphenyl reagent
organisms including most remote regions.1–4 Since this (SBP).8–10 Then the released fluoride are determined
discovery, very intense analytical, ecological and tox- with ion-chromatography,6,7 by the flow-injection anal-
icological studies were undertaken on their determina- ysis (FIA) with potentiometric (LOD = 103 µg L–1)8 or
tion in various materials, ways of their distribution in fluorescence9 detections, or after the transformation to
environment, their stability in environment, and also on triphenylfluorosilane (TPSiF) by revered-phase high
their interactions with living organisms.5 Among large performance liquid chromatography (RP-HPLC) with UV
number of different fluorinated substances identified in detection.10 The reported FIA methods had insufficient
†
Presented at the 10th International Symposium and Summer School on Bioanalysis within the CEEPUS Network CII-HU-0010-
04-0910, Zagreb, Croatia, July 2010.
* Author to whom correspondence should be addressed. (E-mail: trojan@chem.uw.edu.pl)
400 M. Koc et al., Application of Gas Chromatography to Determination of Total Organic Fluorine

limits of detection, while the combustion ion-chromato- (system 1), HP 6890N GC with FID and ECD detectors
graphy (CIC) method6,7 and the HPLC with effective (system 2), HP 6890S GC with 5973 MS detector
preconcentration10 were satisfactory for analysis of the (system 3), Hewlett Packard (Palo Alto, CA, USA) and
environmental samples. For such determinations, the GC-17A with MS QP-5000 (system 4) and QP-5050
gas chromatography (GC) has not been employed as a (system 5) detectors (MS Shimadzu Kyoto, Japan).
method for determination of fluoride, as yet. The analytical columns used during measure-
The GC, commonly used in various areas of rou- ments: HP-1 (100 % dimethylpolysiloxane), 30 m; 0.32
tine analytical chemistry, has already found numerous mm i.d.; 1 µm film thickness, HP-5 (5 % phenyl 95 %
applications both for determination of fluoride as well dimethylpolysiloxane) 30 m; 0,25 mm i.d.; 0,25 µm
as fluorinated organic compounds with various detec- film thickness, Agilent J&W GC Columns (Santa Clara,
tion methods. GC determination of fluoride with flame- CA, USA) and ZB-5 (5 % phenyl 95 % dimethylpolysi-
ionization detection (FID) were developed in 1960s loxane) 30 m; 0,25 mm i.d.; 0,25 µm film thickness,
with derivatization of fluoride ion to volatile triethylflu- Zebron (Torrance, CA, USA).
orosilane11 or trimethylfluorosilane.12,13 The total con- The Vortex mixer type 16700 was obtained from
tent of fluorine in whole blood, serum/plasma and other Thermolyne (Thermo, Waltham, MA, USA).
biological samples was determined by GC using oxygen
bomb combustion and reaction with triethylsilanol.14 Chemicals and Samples
In the GC determination of fluorinated organic
compounds after suitable derivatization the electron All chemicals used in the experiments were of analytical
capture detection (ECD) can be used,15,16 but as it was purity grade, unless specific purity is given. Diethyl
recently showed in the determination of perfluorinated ether, heptane, toluene and perchloric acid (70 %) were
carboxylic acids (PFCA) C6 to C12 after the derivatiza- obtained from POCh (Gliwice, Poland), acetonitrile
tion with isobutyl chloroformate, evidently better values (gradient chromatography grade) was obtained from
of the LOD can be obtained for the mass spectrometry Merck (Darmstadt, Germany). Sodium fluoride (>99 %),
(MS) detection with electron impact ionization, than the triphenylsilyl fluoride (TPSiF) (97 %), triphenylsilanol
ECD.16 The GC measurements with various derivatiza- (TPSiOH) (>98 %), perfluorooctanoic acid (PFOA,
tion procedures is in fact the most frequently applied for C8 PFCA), perfluorodecanoic acid (C10 PFCA) and
this purpose. Without derivatization the GC was em- sodium biphenyl (SBP) were obtained from Sigma-
ployed with chemical ionization in the determination of Aldrich (Steinheim, Germany).
perfluorooctane sulfonate (PFOS) precursors and fluo- Deionized water used throughout the experiments
rotelomer alcohols,17 while with electron impact ioniza- was obtained from Waters Milli-Q system (Waters,
tion in the determination of 8:2 telomer alcohol in ani- Bedford, MA, USA)
mal plasma and tissues.18 Then with an appropriate deri-
vatization and chemical ionization the GC/MS was used Procedures
in the determination of PFCAs in aqueous matrices,19
and sediments,20 and also in isolating isomers of perflu- The defluorination procedure of perfluorinated com-
orocarboxylates in polar bears.21 pounds was conducted as follows: approximately 10–20
Recently in our research group we have developed μL of a given PFCA acetonitrile solution was mixed
a HPLC method as mentioned above for the determina- with 300 μL of SBP in a 1.5 mL glass vial. The vial was
tion of TOF, based on the transformation of fluoride kept closed for 10 minutes to conduct defluorination.
into TPSiF.10 As this compound is volatile, although the Then, 500 μL of water was added and the mixture was
separation requires a higher temperature than the earlier shaken vigorously. The organic layer was discarded and
used trialkylfluorosilanes,12,14 it can be also determined aqueous layer was washed once with 300 μL of diethyl
by the GC with various detection methods. The aim of ether. 400 μL of the aqueous phase containing fluoride
this work was to develop a GC method for the determi- was subjected to the derivatization procedure as de-
nation of TOF with the use of the earlier developed scribed below.
defluorination procedure with SBP and derivatization of The optimized procedure of fluoride derivatization
fluoride to TPSiF. In order to optimize this method, the was as follows: 400 μL of aqueous fluoride solution of
FID, ECD and MS detection have been compared. given concentration (water in case of blank) was placed
in a 2 mL polypropylene Eppendorf-type test tube,
EXPERIMENTAL mixed with 600 μL of concentrated perchloric acid and
50 μL of TPSiOH (typically 0.015 mol L–1 acetonitrile
Aparratus solution added for fluoride concentration range of 0.25–
75 μmol L–1). Then, 200 μL of heptane was added
The chromatographic measurements were perfor- to the mixture and the tube was vigorously agitated
med using: HP 5890 Series II GC with FID detector using the vortex for approximately 60 minutes at room

Croat. Chem. Acta 84 (2011) 399.

M. Koc et al., Application of Gas Chromatography to Determination of Total Organic Fluorine 401

(a) (b)

Figure 1. Chromatograms obtained in GC / FID system HP 5890 (system 1) with HP-1 column for 100 µmol L–1 TPSiF solution in
heptane for different modes of injection in isothermal conditions at 225 °C (a) and with temperature gradient conditions (b). Sam-
ple injection volume 2 L. In (a): split injection (1), and splitless injection with split closure time 0.1 (2), 0.2 (3), 0.3 (4) and 0.4
(5) min. In (b) temperature gradient program 40 °C for 0.5 min, 40 – 250 °C with 30 °C min–1 ramp, 250 °C for 7.5 min; split injection
(1), and splitless injection with split closure time 0.1 (2), 0.2 (3), 0.3 (4) and 0.4 (5) min. Injection at 325 °C, detection at 325 °C.

temperature. Finally, 100 μL of heptane layer was aspi- for both. Mass spectra were recorded across the range of
rated and subjected to the GC measurement. 40– 700 m/z and for SIM mode 154, 201 and 278 m/z
The chromatographic analysis was performed were selected.
using HP 5890 Series II with FID detector (system 1)
and GC-17A with MS QP-5000 (EI) detector (system 4)
RESULTS AND DISCUSSION
on HP-1 column with helium as the carrier gas and
the flow rate of 30 cm s–1. Split injection was carried out
Optimization of Chromatographic Conditions
with ratio 1 :20. Splitless injection (2 µL, 0.2 min) were
performed at an initial oven temperature 40 °C for 0.5 As method for the TOF determination should allow the
min, ramped at 30 °C min–1 to 250 °C and hold 7.5 min. analysis of trace level of perfluorinated compounds in
Injector temperature was 325 °C. In case of FID detector different matrices, the main parameter for the optimiza-
temperature was 325 °C and the flow rate of hydrogen tion should be LOD and a satisfactory separation of
and air was 40 and 450 mL min–1 respectively. Mass TPSiF signal from the other solutes, which may result
spectra were recorded across the range of 40–700 m/z from a defluorination or derivatization procedure of
and for SIM mode m/z 154, 201 and 278 were selected. fluoride ion. The initial measurements for the standard
The chromatographic analysis also was perfor- TPSiF solutions in heptane were carried with the FID
med using GC-17A with MS QP-5050 (EI) detector detection (system 1) and with column Agilent J&W
(system 5) on HP-5 column with helium as the carrier HP-1. To obtain a satisfactory sensitivity of measure-
gas and the flow rate of 30 cm s–1. Splitless injection (1 ments, a significant factor is the mode of sample intro-
µL, 0.2 min) were performed at an initial oven tempera- duction, hence the signal magnitude was examined
ture 40 °C for 0.5 min, ramped at 30 °C min–1 to 250 °C using splitless injection for different periods of closure
and hold 7.5 min. Injector temperature was 325 °C. time of the splitting (0.1 to 0.5 min). In the isothermal
Mass spectra were recorded across the range of 40–700 conditions of separation at 225 °C, for splitting closed
m/z and for SIM mode 154, 201 and 278 m/z were selected. for 0.1 min almost twice increase of signal was noticed
The chromatographic analysis was performed us- in comparison to injection with open splitting, while the
ing HP 6890N with FID and ECD detectors (system 2) prolongation of closure of splitting practically does not
and HP 6890S with MS (EI) detector (system 3) on influence the signal magnitude, and deteriorate the peak
ZB-5 column with helium as the carrier gas and 30 cm s–1 shape (Figure 1a).
flow rate. Splitless injection (1 µL, 0.6 min) were per- Another important factor affecting the signal mag-
formed at an initial oven temperature 40 °C for 2 min, nitude is the employed temperature program for chro-
ramped at 40 °C min–1 to 180 °C hold 0.5 min, subse- matographic run. The most favorable results were ob-
quently ramped at 10 °C min–1 to 280 °C followed by tained for introduction of the sample to the capillary
2 min hold. The injector temperature was 290 °C. column at 40 °C, and after 0.5 min the possible quickest
In case of FID detector temperature was 280 °C and temperature increase was employed up to 250 °C, in
290 °C in case of ECD detector and the flow rate of order to obtain the shortest possible retention time for
hydrogen and air was 40 and 450 mL min–1 respectively TPSiF. The final temperature was achieved after 8.8 min,

Croat. Chem. Acta 84 (2011) 399.

402 M. Koc et al., Application of Gas Chromatography to Determination of Total Organic Fluorine

well as the time of its duration from 10 to 60 min has

not showed influence of these factors on the reaction
yield. Also the solvent used for PFCAs for defluorina-
tion (acetonitrile, toluene and diethyl ether) had no
effect on the defluorination yield. Figure 2 shows chro-
matograms recorded in the GC/FID system (system 1)
for 75 µmol L–1 fluoride solution after derivatization,
and for 125 µmol L–1 solution of C8 PFCA (PFOA)
after defluorination with SBP and derivatization of
fluoride to TPSiF. One can see in both chromatograms
that TPSiF signal is well resolved from all interferences
from both steps of the applied procedures.
As it has recently been reported21 the efficient
Figure 2. Comparison of chromatograms recorded for 75 method of defluorination of PFCAs in aqueous solution
µmol L–1 fluoride solution after derivatization (A) and for 125 can be the reaction with microwave-induced persulfate.
µmol L–1 C8 PFCA (PFOA) solution after defluorination and It was confirmed also in this study for the decomposi-
derivatization (B). Conditions: GC/FID system HP 5890 (sys- tion of the PFOA solution at submillimolar solutions by
tem 1) with HP-1 column, sample injection volume 2 L, split- carrying decomposition process at 160 °C by 60 min.
less injection with split closure time 0.2 min, temperature gradi- The recovery calculated based on the determined fluo-
ent program 40 °C for 0.5 min, 40–250 °C with 30 C min–1 ride ion concentration was above 80 %. Due to the ne-
ramp, 250C for 7.5 min, injection at 325 °C, detection at 325 °C. cessary step of PFCs preconcentration for trace level of
determined analytes and elution with acetonitrile, similar
while the retention time for TPSiF in these conditions process was examined in acetonitrile solutions. Due to
was 9.3 min. For the samples analyzed after the com- scavenging of persulfate radicals by acetonitrile the yield
plete procedure of defluorination with SBP and deri- of decomposition in acetonitrile was less than 55 %,
vatization to TPSiF, in order to elute all the other pre- hence this method was not used in further experiments.
sent components the time of chromatographic run was The derivatization of fluoride with TPSiOH is car-
15 min. For such a temperature gradient the injection ried out in strong acidic conditions. In our recent work10
conditions were also optimized. As it is showed in Fig- we reported the optimization of kind of acids used,
ure 1b, the optimum time for closing splitting was 0.2 concentration of acid, taken volume of derivatizing
min, where TPSiF signal was 20 times higher than for reagent TPSiOH solution, reaction time and type of
solvent used for extraction of derivatization product.
injection with splitting.
First four of the mentioned parameters are independent
In such conditions, with 2 l injected sample vo-
of the detection method used for determination of TPSiF
lume, the calibration for TFSiF was made in the concen-
and can be identical for HPLC and GC determination.
tration range 0.1 to 75 µmol L–1. In the whole range a
Regarding solvent taken for the extraction for HPLC
satisfactory linearity of response was obtained with
heptane was selected due to a low absorption of UV
precision not worse than 5 % RSD. For S/N = 3 the
radiation as compared to toluene and benzene, used
instrumental LOD was evaluated as 0.02 µmol L–1.
earlier in the GC determinations.12 In this work we did
not consider benzene due to its toxicity and the optimi-
Conditions of Defluorination and Derivatization of
zation was carried out for heptane and toluene. Figure 3
Fluoride Ions
shows results obtained for the yield of derivatization
In order to release fluorine from the examined model of conditions. Also in this case as an optimum solvent
perfluorinated carboxylic acids for the TOF determina- heptane was selected, which for 60 min derivatization
tion, the reactions with SBP or with persulfate radicals time in ambient temperature allows to obtain a satisfac-
enhanced with microwave irradiation were employed. tory yield of derivatization.
The first one was optimized earlier for the HPLC The optimized procedure of derivatization was
determinations,10 and the earlier developed procedure employed for calibration with fluoride in concentration
was employed in this work for C8, C10 and C12 range from 0.2 to 80 µmol L–1 in the initial solution. In
PFCAs. In the present GC determinations for SBP reac- whole range the obtained calibration was linear with
tion 300 µl volume of reagent was applied, which was a correlation coefficient 0.9999 and precision not worse
compromise between obtained yield of defluorination than 5 % RSD. The blank in such a procedure was eva-
and blank signal, resulting from trace contamination of luated as 0.15 µmol L–1 fluoride, and its main source is
commercial SBP reagent with fluoride reported earlier.10 fluoride contamination is concentrated perchloric acid.
The examination of effect of the temperature for carry- Obviously, it is only a part of the total blank for com-
ing the reaction with SBP in the range 20–40 °C, as plete procedure in TOF determination.

Croat. Chem. Acta 84 (2011) 399.

M. Koc et al., Application of Gas Chromatography to Determination of Total Organic Fluorine 403

Figure 3. Effect of solvent used, used time of reaction, and

temperature on yield of derivatization of fluoride solution
(c = 10 µmol L–1) with TPSiOH.

Table 1. Instrumental LOD values obtained for determination

of TPSiF and MDL values obtained for fluoride ion involving
the derivatization procedure for GC measurements with differ-
ent detections for different columns

Measuring LOD (TPSiF)

Method Column
system(a) µmol L–1
FID HP-1 System 1 0.02 t / min
ZB-5 System 2 0.08 Figure 4. Chromatograms recorded in different GC systems
ECD ZB-5 System 2 0.3 for 0.1 µmol L–1 TPSiF solution in heptane: GC/FID system
MS HP-1 System 4 0.1 HP 5890 (system 1) with HP-1 column, sample injection
HP-5 System 5 0.02 volume 2 µL, splitless injection with split closure time
0.2 min, temperature gradient program 40 °C for 0.5 min,
ZB-5 System 3 0.003
40 – 250 °C with 30 °C min–1 ramp, 250 °C for 7.5 min, injec-
(a)
Numbering according to descriptions in experimental. tion at 325 °C, detection at 325 °C (A); GC/MS system QP-
5000 (system 4) with HP-1 column, SIM mode m/z: 278, 201,
Comparison of Detection Methods 154, sample injection volume 2 µL, splitless injection with
For the GC determination of TPSiF employed for de- split closure time 0.2 min, temperature gradient program
termination of TOF, besides the FID detection, also the 40 °C for 0.5 min, 40 – 250 °C with 30 °C min–1 ramp, 250 °C
detection with ECD detector and the MS with electron for 7.5 min, injection at 325 °C (B); GC/MS system QP-5050
impact ionization were investigated. Additionally for (system 5) with HP-5 column, SIM mode m/z: 278, 201, 154,
better illustration of the effect of different factors LOD sample injection volume 1 µL, splitless injection with split
values were determined in case of the FID and the MS closure time 0.2 min, temperature gradient program 40 °C for
detections using different measuring systems equipped 0.5 min, 40 – 250 °C with 30 °C min–1 ramp, 250 °C for 7.5
with different columns, in order to examine how type of min, injection at 325 °C (C).
instrument may affect such determinations.
The values of limit of the detection were evaluated Chromatogram B (system 4) has the smallest signal and
in two different meanings. The instrumental LOD value the largest noise. Chromatogram A (system 1) has a
was evaluated taking into account the signal magnitude smaller signal than the chromatogram C, but it also has
S for injected TPSiF solutions, and the noise amplitude a smaller noise so in this case two particular systems
N of the base-line. The LOD was calculated as a con- with FID and MS Detections (system 1 and 5) exhibit
centration corresponding to S/N = 3. In this case one the same LOD values. The MDL values were evaluated
may expect that the LOD values significantly depend on taking into account magnitude and precision of blank
an employed instrument, type of column, and conditions measurements, obtained for fluoride detection with
of chromatographic run. Some comparison of the ob- derivatization, or for C8 PFCA used as model perfluo-
tained recording for three systems with the FID and the rinated substance for TOF determination using com-
MS detectors are showed in Figure 4. The aim of pre- plete procedure with defluorination and derivatization.
senting Figure 4 is to show the differences in signal to Table 1 shows the obtained results of the instrumental
noise ratio, because it determines the instrumental LOD. LOD values for all configurations of the GC systems

Croat. Chem. Acta 84 (2011) 399.

404 M. Koc et al., Application of Gas Chromatography to Determination of Total Organic Fluorine

(a)

Figure 5. Mass spectra obtained from measurements in GC/MS system HP 6890S (system 3) with ZB-5 column for 1 µL injec-
tion of 90 µmol L–1 TPSiF solution in heptane (a), and for solution obtained after complete procedure of defluorination and deri-
vatization carried out for 20 µL of 7.7 µmol L–1 solution of C8 PFCA (PFOA) (b). Conditions: splitless injection with split closure
time 0.6 min, temperature gradient program 40 °C for 2 min, 40 – 180 °C with 40 °C min–1 ramp, 180 °C for 0.5 min, 180 – 280 °C
with 10 °C min–1 ramp, 280 °C for 2 min, injection at 290 °C, m/z range from 40 to 700.

TPSiF

t / min

Figure 6. Chromatogram recorded in GC/MS system HP 6890S (system 3) with ZB-5 column for blank (20 µL acetonitrile) after
carrying the complete procedure of defluorination and derivatization. Conditions: sample injection volume 1 µL, splitless injec-
tion with split closure time 0.6 min, temperature gradient program 40 °C for 2 min, 40 – 180 °C with 40 °C min–1 ramp, 180 °C for
0.5 min, 180 – 280 °C with 10 °C min–1 ramp, 280C for 2 min, injection at 290 °C, SIM mode m/z: 278, 201, 154.

used. In comparison to the GC/FID system reported Taking into account the earlier works on applica-
above, for the other one (system 2) used with ZB-5 tion of the GC in the determination of PFCs, the im-
column, the LOD value was found as 0.08 µmol L–1. provement of limit of the detection and the selectivity of
The observed difference resulted mostly from the determinations can be expected from application of the
different noise amplitude for both systems, regardless MS detection with electron impact ionization. Based on
the column used. the mass spectra recorded for TPSiF solution in heptane
The application in the same GC (system 2) setup (Figure 5a), which was in good agreement with spec-
with ZB-5 column of the ECD detector provides disap- trum from NIST catalogue, for examination of sensitivi-
pointing results. For such a system the poorest LOD ty of TPSiF determination three signals were used. Sig-
value 0.3 µmol L–1 was obtained with the worst efficiency nal at m/z 278 corresponds to monovalent molecular
of separation, which can be attributed probably to the ion, and signals for m/z 210 and 154 result from frag-
presence of non-resolved halogenated organic contami- mentation of TPSiF by subtraction of one or two phenyl
nants for used solvents. The obtained calibration line also groups. The LOD determined for total ion current (TIC)
showed poorer linearity (r2 = 0.987) and precision in the mode for TPSiF was 0.045 µmol L–1, while for single
range of 2.2 to 10.8 % RSD for measured concentration ion monitoring (SIM) mode calculated for sum of three
range. The precision of the both GC/FID used was better mentioned ions was 0.003 µmol L–1, which is listed in
than 3.0 % RSD in whole concentration range. Table 1.

Croat. Chem. Acta 84 (2011) 399.

M. Koc et al., Application of Gas Chromatography to Determination of Total Organic Fluorine 405

The recorded GC/MS (system 3) chromatogram of PFCA, which belong besides C9 PFCA to most com-
blank in the SIM mode showed in Figure 6 indicated from monly occurring PFCAs in environmental samples.4
one side the selectivity and sensitivity of the GS/MS The measuring system, with the use of complete proce-
determination, but on another one a significant role of dure of defluorination and derivatization was calibrated
contamination in the blank for the TOF determination in using standard fluoride solutions in concentration range
whole procedure including defluorination and derivatiza- from 0.2 to 80 µmol L–1 fluoride. Then the same mea-
tion. For the illustration of possible application of MS surements were performed with C8 and C10 PFCAs
detection for the determination of TOF, the mass spectrum acetonitrile solutions in the concentration range from
for PFOA, after complete procedure of defluorination and 3 to 133 µmol L–1 in 20 µL sample volume taken for
derivatization, is showed in Figure 5b. The similarity defluorination reaction with SBP and further derivatiza-
of the both spectra showed in Figure 5 indicates the lack tion. The signal values obtained in the GC/FID mea-
of evident interferences affecting the signal of analyte in surements allowed to get the linear calibration plots,
such procedure. For these instrumental conditions the line- with relative slope compared to fluoride plot 99.8 and
arity of calibration curve was satisfactory (r2 = 0.9991), 75.5 % for C8 and C10 PFCAs, respectively. It means
while precision of measurements was poorer than in the obtaining a satisfactory recovery of whole procedure
measurements with FID detection (0.6–6.7 % RSD). for given range of PFCAs concentrations.
Similar measurements were carried out with two The detection limits for C8 PFCA, for 20 µL of
other GC/MS instruments (system 5) equipped with HP- solution taken for defluorination and derivatization was
5 MS column (nominally the same as ZB-5), and with evaluated as 5.2 µmol L–1, however when it is taken as
HP-1 (system 4) of larger bore and thicker layer. In the initial sample volume for preconcentration on carbon
system with HP-5 column (system 5) about 6-fold poor- sorbent 1000 mL,10 and preconcentrated species will be
er LOD was observed for the SIM mode, while for HP-1 eluted with acetonitrile and evaporated to obtain 20 µL
column, essentially not suitable for the MS detection, for further processing, the real MDL: value can be eva-
about 30 times worse LOD values were obtained. As it luated as 0.104 nmol L–1 (43 ppt), which means that the
was mentioned (and illustrated in Figure 4) the main developed the GC/FID method for TOF determination
factor contributing to this effect is the base-line noise can be employed for analysis of environmental samples.
amplitude and conditions of chromatographic run. For
comparison it can be admitted that instrumental LOD
CONCLUSION
for the determination of TPSiF by HPLC with UV de-
tection was 0.09 µmol L–1.10
The purpose of this article is to present the possibilities
In the determination of TOF with complex defluo-
of the TOF determination with the equipment available
rination procedure the crucial factor for the MDL is the
in each laboratory (chemical defluorination with the
level of blank measured in each instrumental setup and
SBP and determination by the GC) as an alternative to
precision of blank measurements from multiple repeti-
the CIC method.6,7 The reagent used for derivatization -
tions of derivatization. Generally, it was observed that
TPSiOH has a higher boiling point than the previously
precision of TPSiF determination is much better for FID
used reagents11–13 so it is easier to prepare the samples
detectors (0.2 to 3.0 % RSD). The MDL values were
as well as conduct separation by GC with a suitable
compared for system 1 and 3, exhibiting LOD values as
temperature program. The presence of three phenyl
shown in Table 1.
groups makes TPSiOH more universal as a derivatiza-
The best value of MDL for fluoride determination
tion reagent because the reaction product may be deter-
0.4 µmol L–1 was obtained for system 1 with FID detec-
mined by the GC and the HPLC. Compared to the CIC
tor, while for the GC/MS instrument (system 3) the
method, and similarly do RP-HPLC with UV detection,
MDL was 1.3 µmol L–1 fluoride. The reason of this
present method is instrumentally simpler, as it does not
difference is evidently worse precision observed in
require a combustion oven. In comparison of three dif-
measurements with the MS detection, exceeding in
ferent detection methods, for ECD the worst LOD val-
some cases even 10 % RSD. The MDL in the GC/FID
ues were observe. The best LOD value was found for
system is then practically the same as obtained for HPLC
the MS detection, but the MDL values were better for
determinations with UV detection 0.36 µmol L–1,10 and
the GC system with FID detection. In the optimized
more favorable than measured for GC/MS measurements.
conditions, taking into account the contamination of
reagents and the reported earlier possibility of efficient
GC Determination of TOF with Flame-Ionization
preconcentration combined on activated carbon sorbents
Detection
with separation of inorganic fluoride, the MDL value
The possibility of TOF determination with developed for C8 PFCA (PFOA) was evaluated as 0.104 nmol L–1
the GC/FID method was examined for two model per- (43 ppt), allowing the use of this method for the analysis
fluorinated compounds C8 PFCA (PFOA) and C10 of environmental samples.

Croat. Chem. Acta 84 (2011) 399.

406 M. Koc et al., Application of Gas Chromatography to Determination of Total Organic Fluorine

Further studies on this subject will be focused on 7. Y. Miyake, N. Yamahsita, P. Rostkowski, M. K. So, S. Taniyasu, P.
examination of the developed method for different fluo- K. S. Lam, and K. Kannan, J. Chromatogr., A 1143 (2007) 98–104.
8. J. Musijowski, M. Trojanowicz, B. Szostek, J. L. Costa Lima,
rinated organic analytes expected in environmental R. Lapa , H. Yamashita, T. Takayanagi, and S. Motomizu,
samples, and on the determination of TOF in real sam- Anal. Chim. Acta 600 (2007) 147–154.
ples of natural surface waters. 9. T. Takyanagi, H. Yamashita, S. Motomizu, J. Musijowski, and
M. Trojanowicz, Talanta 74 (2008) 1224–1230.
Acknowledgements. The financial support for M. A. Donten 10. J. Musijowski, B. Szostek, M. Koc, and M. Trojanowicz,
and M. Koc for their study stays at Graz University of Tech- J. Sep .Sci. 33 (2010) 2636–2644.
nology, research group of Professor E. Lankmayr, by CEE- 11. R. Bock and H. J. Semmler, Z. Anal. Chem. 230 (1967) 161–184.
12. J. A. Fresen, F. H. Cox, and M. J. Witter, Pharm. Weekblad 103
PUS Program CII-HU-II is kindly acknowledged
(1968) 909–914.
13. J. H. Glerum, A. Van Dijk, and S. W. Klein, Pharm. Weekblad
Sci. Ed. 6 (1984) 75–79.
REFERENCES 14. J. Belisle and D. F. Hagen, Anal. Biochem. 87 (1978) 545–555.
15. J. Belisle and D. H. Hagen, Anal. Biochem. 101 (1980) 369–376.
1. D. R. Taves, Nature 217 (1968) 1050–1051. 16. V. Dufkova, R. Cabala, D. Maradova, and M. Sticha, J. Chroma-
2. B. D. Key, R. D. Howell, and C. S. Criddle, Environ. Sci. togr., A 1216 (2009) 8659–8664.
Technol. 31 (1997) 2445–2454. 17. J. W. Martin, D. C. G. Muir, C. A. Moody, D. A. Ellis, W. C.
3. J. P. Giesy and K. Kannan, Environ. Sci. Technol. 35 (2001) Kwan, K. R. Solomon, and S. A. Mabury, Anal. Chem. 74 (2002)
1339–1342. 584–590.
4. K. Prevedouros, I. T. Cousins, R. C. Buck, and S. H. Korzeni- 18. B. Szostek and K. B. Prickett, J. Chromatogr., B: Anal. Technol.
owski, Environ. Sci. Technol. 40 (2006) 32–44. Biomed. Life Sci. 813 (2004) 313–321.
5. Analysis of Fluorinated Surfactants in: E. Kissa (Ed.), Fluori- 19. R. Alzaga and J. M. Bayona, J. Chromatogr., A 1042 (2004)
nated Surfactants and Repellants. Chp. 9 Marcel Dekker, New 155–162.
York 2001, pp. 390–450. 20. R. Alzaga, C. Salgado-Petinal, E. Jover, and J. M. Bayona,
6. Y. Miyake, N. Yamashita, M. K. So, P. Rostkowski, S. Taniyasu, J. Chromatogr., A 1083 (2005) 1–6.
P. K. S. Lam, and K. Kannan, J. Chromatogr., A 1154 (2007) 21. Y. C. Lee, S. L. Lo, P. T. Chiueh, and D. G. Chang, Water Res.
214–221. 43 (2009) 2811–2816.

Croat. Chem. Acta 84 (2011) 399.