**3. Result and discussion**

#### **3.1. PFC concentrations**

**Chemical Name Abbreviation Formula**

60 Emerging Pollutants in the Environment - Current and Further Implications

Perfluorobutanoic acid PFBA C3F7COO– 214.0 1.3 5.0 0.060 Perfluoropentanoic acid PFPeA C4F9COO– 264.0 2.1 5.3 0.54 Perfluorohexanoic acid PFHxA C5F11COO- 314.1 3.1 5.6 1.1 Perfluoroheptanoic acid PFHpA C6F13COO– 364.1 2.8 5.9 1.6 Perfluorooctanoic acid PFOA C7F15COO– 414.1 3.6 6.3 2.3 Perfluorononanoic acid PFNA C8F17COO– 464.1 4.5 6.6 2.9 Perfluorodecanoic acid PFDA C9F19COO– 514.1 5.4 6.8 3.5 Perfluoroundecanoic acid PFUnA C10F21COO– 564.1 6.4 7.1 4.2 Perfluorododecanoic acid PFDoA C11F23COO– 614.1 7.1 7.4 5.0 Perfluorotetradecanoic acid PFTA C13F27COO– 714.1 8.8 8 6.1 Perfluorobutane sulfonic acid PFBS C4F9SO3– 300.1 N/A N/A -0.53 Perfluorohexane sulfonic acid PFHxS C6F13SO3– 400.1 N/A N/A 0.54 Perfluorooctane sulfonic acid PFOS C8F17SO3– 500.1 4.3 7.8 1.7 Perfluorodecane sulfonic acid PFDS C10F21SO3– 600.1 N/A N/A 3.1 Perfluorooctane sulfonamide PFOSA C8F17SO2NH2 499.1 6.3 8.4 -

Perfluorooctanoic acida 13C2–PFOA - - - - - Perfluorodecanoic acidb 13C2–PFDA - - - - - Perfluorododecanoic acidc 13C2–PFDoA - - - - - Perfluorooctane sulfonic acidd 13C4–PFOS - - - - - Perfluorooctanoic acide 13C4–PFOA - - - - -

\*Log *K*OW and *K*OA values of individual PFCs were compiled from published values calculated using SPARC general

**Table 1.** List of target perfluoroalkyl chemicals (PFCs) and radiolabeled surrogates monitored using LC/MS/MS.

**Target Analytes**

**Mass Labeled Standards**

MW: molecular weight

partitioning model [33].

used to quantify PFDA and PFUnA

used to quantify PFDoA and PFTA

*a*

*b*

*c*

*d*

*e*

\*\*Log *D* values were calculated at pH = 7.5 and T = 21°C using SPARC.

used to quantify PFBA, PFPeA, PFHxA, PFHpA, PFOA, PFNA

used to quantify PFBS, PFHxS PFOS, PFDS, PFOSA

used to quantify recovery of mass labeled surrogates

**MW**

**(g/mol) Log\* KOW Log\* KOA Log\*\* D**

Several PFC compounds showed concentrations above the MLOQ, as shown in Table 3. Perfluorotetradecanoic acid (PFTA), a chemical with a high *KOW* and high *KOA* that will persist for decades in humans, was measured in 60% of sediment samples, but undetected or below the MLOQ in lichens, feces, and feathers. Perfluoroheptanoic acid (PFHpA) was detected in all seabird feather samples (range = 1.60−2.85 ww ng/g; Table 3), and in 47% of penguin feces, ranging 0.37−22 ng/g ww. All lichen samples exhibited concentrations of perfluorohexanesul‐ fonate (PFHxS), ranging 0.20−1.20 ng/g ww, while perfluorobutyric acid (PFBA), perfluoro-npentanoic acid (PFPeA), and PFHpA were measured in 80%, 60%, and 60% of lichen samples, respectively. PFOA and PFOS were not quantified in most samples (i.e., < MLOQ or ND; Table 3), except for the detection of PFOS in two penguin feces samples (2.8 and 3.14 ng/g ww), and PFOA in a single fecal sample (2.0 ng/g ww) and lichen (4.7 ng/g ww). The lack of PFOS and PFHxS detection in Antarctic seabird feathers contrasts with the levels of PFOS and PFHxS found in feathers of grey herons (PFOS: 247 ng/g dw; PFHxS: ≈ 20 ng/g dw) and herring gulls (PFOS: 79 ng/g dw; PFHxS: > 30 ng/g dw) from the Northern Hemisphere (Flanders, Belgium) [28]. However, the absence of PFOA in our feather samples is consistent with the lack of detection of this compound in bird feathers from the same region [28]. For comparison purposes, the PFOA concentration detected in a sample of gentoo penguin feces was 14 times lower than the PFOA concentration (28.2 ng/g ww) detected in a single herring gull liver sample from Belgium [28]. Despite samples from other parts of the world that indicate a continued increase or no change in PFOS levels following the 2002 phase-out [34–37], a fast decline in PFOS concentrations has been observed in wildlife over the past decade [38,39]. PFDA, PFUnA, PFDoA, and PFOSA were not detected (ND) or < MDL. Except for the com‐ pound PFHpA, lack of detection of most analytes in samples and small sample sizes preclude undertaking robust statistical analyses for multisite or/and inter-species comparisons.

#### **3.2. PFC patterns**

Figure 2 shows the composition of PFCs observed in biotic and abiotic samples. PFHpA was the only compound detected in feathers of both petrels and penguins, accounting for 100% of total PFCs, while PFTA was equal to 100% of PFCs in sediment samples (Figure 2). PFHpA was also found in feces and lichen samples making up 24.5% and 23% of total PFCs, respec‐ tively. PFDS contributed to 54% and 17% of PFCs in feces and lichens, contrasting with PFPeA and PFHxS, which accounted for 3.4% and 50%, and for 4% and 21% of PFCs in feces and lichen samples, respectively. PFNA accounted for 38% of the PFCs in feces. These patterns clearly show that both perfluorinated sulfonates (PFSAs) and carboxylates (PFCAs) exhibit different fractions in seabirds, reflecting the potential role of biotransformation in shaping the accumu‐ lation of these compounds.

#### **3.3. Bioaccumulation of PFCs**

The biomagnification factor (BMF) [40] for PFHpA (i.e., BMF = C B/CD, where CB is the PFHpA concentration detected in the predator, the giant petrel, and CD is the PFHpA concentration observed in the diet/prey, gentoo penguin) was calculated using feather concentrations, as this was the only PFC compound readily detected in 100% of feathers samples. Hence, the concentrations of PFHpA in the petrel feathers (i.e., mean ±SD = 2.6 ± 0.60 ng/g ww; *n* = 5) and that of the penguin (1.60 ng/g ww; *n* =1; Table 3) were used as surrogates for concentrations in the tissues of the whole organism, assuming that the birds had been exposed to the com‐ pound for a sufficiently long time to allow the concentrations to reach steady state [40]. The criterion applied to indicate that PFHpA was biomagnified in petrels was a BMF > 1, such that a BMF greater than 1 indicates that the chemical is a bioaccumulative substance [41]. Here, we found that the BMF was close to 2 (i.e., 1.6), indicating that PFHpA biomagnifies in petrels. Although the concentrations of PFHpA in feces appear to be relatively higher than the concentrations found in lichen and feathers, comparisons of the PFHpA concentrations among


**Table 3.** Quantification data of PFCs (ng/g ww) in feather, feces, lichen, and sediment samples collected in the Antarctic Peninsula. Data taken and modified from reference [63].

3), except for the detection of PFOS in two penguin feces samples (2.8 and 3.14 ng/g ww), and PFOA in a single fecal sample (2.0 ng/g ww) and lichen (4.7 ng/g ww). The lack of PFOS and PFHxS detection in Antarctic seabird feathers contrasts with the levels of PFOS and PFHxS found in feathers of grey herons (PFOS: 247 ng/g dw; PFHxS: ≈ 20 ng/g dw) and herring gulls (PFOS: 79 ng/g dw; PFHxS: > 30 ng/g dw) from the Northern Hemisphere (Flanders, Belgium) [28]. However, the absence of PFOA in our feather samples is consistent with the lack of detection of this compound in bird feathers from the same region [28]. For comparison purposes, the PFOA concentration detected in a sample of gentoo penguin feces was 14 times lower than the PFOA concentration (28.2 ng/g ww) detected in a single herring gull liver sample from Belgium [28]. Despite samples from other parts of the world that indicate a continued increase or no change in PFOS levels following the 2002 phase-out [34–37], a fast decline in PFOS concentrations has been observed in wildlife over the past decade [38,39]. PFDA, PFUnA, PFDoA, and PFOSA were not detected (ND) or < MDL. Except for the com‐ pound PFHpA, lack of detection of most analytes in samples and small sample sizes preclude undertaking robust statistical analyses for multisite or/and inter-species comparisons.

62 Emerging Pollutants in the Environment - Current and Further Implications

Figure 2 shows the composition of PFCs observed in biotic and abiotic samples. PFHpA was the only compound detected in feathers of both petrels and penguins, accounting for 100% of total PFCs, while PFTA was equal to 100% of PFCs in sediment samples (Figure 2). PFHpA was also found in feces and lichen samples making up 24.5% and 23% of total PFCs, respec‐ tively. PFDS contributed to 54% and 17% of PFCs in feces and lichens, contrasting with PFPeA and PFHxS, which accounted for 3.4% and 50%, and for 4% and 21% of PFCs in feces and lichen samples, respectively. PFNA accounted for 38% of the PFCs in feces. These patterns clearly show that both perfluorinated sulfonates (PFSAs) and carboxylates (PFCAs) exhibit different fractions in seabirds, reflecting the potential role of biotransformation in shaping the accumu‐

The biomagnification factor (BMF) [40] for PFHpA (i.e., BMF = C B/CD, where CB is the PFHpA concentration detected in the predator, the giant petrel, and CD is the PFHpA concentration observed in the diet/prey, gentoo penguin) was calculated using feather concentrations, as this was the only PFC compound readily detected in 100% of feathers samples. Hence, the concentrations of PFHpA in the petrel feathers (i.e., mean ±SD = 2.6 ± 0.60 ng/g ww; *n* = 5) and that of the penguin (1.60 ng/g ww; *n* =1; Table 3) were used as surrogates for concentrations in the tissues of the whole organism, assuming that the birds had been exposed to the com‐ pound for a sufficiently long time to allow the concentrations to reach steady state [40]. The criterion applied to indicate that PFHpA was biomagnified in petrels was a BMF > 1, such that a BMF greater than 1 indicates that the chemical is a bioaccumulative substance [41]. Here, we found that the BMF was close to 2 (i.e., 1.6), indicating that PFHpA biomagnifies in petrels. Although the concentrations of PFHpA in feces appear to be relatively higher than the concentrations found in lichen and feathers, comparisons of the PFHpA concentrations among

**3.2. PFC patterns**

lation of these compounds.

**3.3. Bioaccumulation of PFCs**

**Figure 2.** Composition pattern of PFC compounds detected in biotic (feathers and feces of seabirds, and lichen) and abiotic (sediment) samples from the Antarctic Peninsula. Abbreviations for PFC chemical names are defined in Table 1.

biota samples show lack of significant differences (Welch's ANOVA, *p* > 0.05; Tukey–Kramer HSD (honest significant difference) test, *p* >0.05), as shown in Figure 3.

To further illustrate the behavior of PFC concentrations in these samples, detected PFC compounds were plotted as a function of log *D* and log *KOA*, as shown in Figure 4. The majority of PFCs concentrations observed in biotic samples (i.e., feces and lichens) fall within log *D* values between 0 and 3, as seen in Figure 4A. While concentrations of PFCs tend to increase with increasing log *D* values from log *D* of 0 to log *D* of 3 in feces, PFC concentrations appear to decrease as the log *D* increases within the same range of log *D* values in lichens (Figure 4A). This observation may be an indication that both ionized and unionized forms of PFC compounds with low log *D* values (i.e., PFBA, PFPeA, PFHxA, PFHxS, PFHpA, PFOS, PFOA, PFNA, PFDS) are present in some organisms residing in this region and prone to potential transportation by oceanic currents (e.g., Antarctic Circulation Current) from either continental/ regional or local sources (i.e., international military bases and research stations) to the Antarctic Peninsula. Similarly, most PFCs concentrations observed in these samples, especially in lichens, fall within log *KOA* values of 5.0 and 6.5 (Figure 4B). Although concentrations for some PFC compounds show a tendency to decrease with increasing log *KOA* in feces (i.e., PFBS, PFHxS, PFPeA, PFHxA, PFHpA, PFOA, PFNA, PFOS), concentrations for similar PFCs seem to increase as log *KOA* increases in lichens (i.e., PFDS, PFHxS, PFBA, PFPeA, PFHpA, PFOA), as seen in Figure 4B. These trends may support the notion that low molecular weight com‐ pounds (e.g., 214-414 g/mol) with low log *KOA* are likely to be subject to long-range atmospheric transport and potentially reaching the region, where these compounds accumulate in biotic compartments, mainly in natural air samplers such as lichens and secondary in air-breathing organisms such as seabirds (petrels, penguins).

Perfluorinated Chemicals Assessment in Sediments, Lichens, and Seabirds from the Antarctic Peninsula… http://dx.doi.org/10.5772/60205 65

**Figure 3.** Box plots showing log transformed concentrations of PFHpA (ng/g ww) detected in lichen (*n* =3), and feather (*n* =6) and feces (*n* =9) of seabirds from the Antarctic Peninsula. The internal line across the box is the median; the ends of the box are the 25% and 75% quartiles; and the whisker bars are the minimum and maximum values. Because of unequal variances (i.e., heteroscedasticity; Bartlett test, *p* < 0.005), a Welch's ANOVA, followed by a Tukey–Kramer HSD test, was used for the multicomparison, showing no significant differences in PFHpA concentrations among the biotic samples (*p* > 0.05).

#### **3.4. PFC health risks**

biota samples show lack of significant differences (Welch's ANOVA, *p* > 0.05; Tukey–Kramer

**Feathers Feces Lichen Sediment**

**Figure 2.** Composition pattern of PFC compounds detected in biotic (feathers and feces of seabirds, and lichen) and abiotic (sediment) samples from the Antarctic Peninsula. Abbreviations for PFC chemical names are defined in Table 1.

PFBA PFPeA PFHxA PFHpA PFOA PFNA PFTA PFBS PFHxS PFOS PFDS FHUEA FOUEA FDUEA

To further illustrate the behavior of PFC concentrations in these samples, detected PFC compounds were plotted as a function of log *D* and log *KOA*, as shown in Figure 4. The majority of PFCs concentrations observed in biotic samples (i.e., feces and lichens) fall within log *D* values between 0 and 3, as seen in Figure 4A. While concentrations of PFCs tend to increase with increasing log *D* values from log *D* of 0 to log *D* of 3 in feces, PFC concentrations appear to decrease as the log *D* increases within the same range of log *D* values in lichens (Figure 4A). This observation may be an indication that both ionized and unionized forms of PFC compounds with low log *D* values (i.e., PFBA, PFPeA, PFHxA, PFHxS, PFHpA, PFOS, PFOA, PFNA, PFDS) are present in some organisms residing in this region and prone to potential transportation by oceanic currents (e.g., Antarctic Circulation Current) from either continental/ regional or local sources (i.e., international military bases and research stations) to the Antarctic Peninsula. Similarly, most PFCs concentrations observed in these samples, especially in lichens, fall within log *KOA* values of 5.0 and 6.5 (Figure 4B). Although concentrations for some PFC compounds show a tendency to decrease with increasing log *KOA* in feces (i.e., PFBS, PFHxS, PFPeA, PFHxA, PFHpA, PFOA, PFNA, PFOS), concentrations for similar PFCs seem to increase as log *KOA* increases in lichens (i.e., PFDS, PFHxS, PFBA, PFPeA, PFHpA, PFOA), as seen in Figure 4B. These trends may support the notion that low molecular weight com‐ pounds (e.g., 214-414 g/mol) with low log *KOA* are likely to be subject to long-range atmospheric transport and potentially reaching the region, where these compounds accumulate in biotic compartments, mainly in natural air samplers such as lichens and secondary in air-breathing

HSD (honest significant difference) test, *p* >0.05), as shown in Figure 3.

**0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%**

64 Emerging Pollutants in the Environment - Current and Further Implications

**% Of total PFCs**

organisms such as seabirds (petrels, penguins).

Concentrations of PFCs detected in feathers and feces of the two seabird species studied here are well below the toxicological reference value (TRV) of PFOS (600 ng/g ww), calculated as an exposure threshold value for birds in nature, especially for apex avian predators [42]. This comparison indicates that gentoo penguins and petrels are not at risk by PFOS toxic effects.

#### **3.5. Transport mechanisms, global and local sources of PFCs to the Antarctic**

There is still a degree of uncertainty surrounding the dominant pathway of PFC movement to the Antarctic, though researchers have highlighted two primary mechanisms generally accepted as the major modes of PFC transportation to the Antarctic: atmospheric and oceanic. Neutral, volatile precursor compounds, such as perfluorinated sulfonamide alcohols (FOSEs), perfluorooctane sulfonamides (FOSAs), and fluorotelomer alcohols (FTOHs), referred to as "flyers", are capable of being delivered to the Antarctic via fast, direct transport of contami‐ nated wind, as opposed to cold trapping, as is common for many legacy POPs [12, 43–45]. Following deposition, these compounds are degraded via oxidation to form ionic PFCs, including PFSAs and PFCAs [12, 44–48]. Evidence supporting this mechanism of travel includes measurements of FTOHs from Europe to the Antarctica showing declining concen‐ trations in the atmosphere with increased distance from sources in the Northern Hemisphere [12]. Given the far distances PFCs must travel to reach the Antarctic, in combination with short atmospheric residence times (ranging on average from 10 to 50 days), the level of effectiveness associated with atmospheric delivery of PFCs is relatively low. Additionally, the yield of ionic

**Figure 4.** Concentrations of PFC compounds (logarithmic scale in ng/g wet weight) measured in sediments, lichens, and seabird feces and feathers from the Antarctic Peninsula as a function of log *D* (A) and log *KOA* (B). In Figure 4A, the solid line shows the behavior of detected PFC concentrations versus log *D* in lichen; the dashed line indicates the trend of detected PFC concentrations versus log *D* values in feces. In Figure 4B, the solid line indicates the behavior of detect‐ ed PFC concentrations versus log *KOA* in lichen, while the dashed line shows detected PFC concentrations versus log *KOA* in feces. Abbreviations for PFC chemical names are defined in Table 1.

PFCs produced via oxidation of precursor compounds once transported to the Antarctic is often low [10, 11, 43–45, 47].

It is therefore expected that most PFCs are delivered to the Antarctic in their ionic, watersoluble state via the oceans [12, 14, 49]. Oceanic transport functions on a slower time scale for the Antarctic (in the order of decades, compared to days or weeks for atmospheric transport) because of the circulation patterns of the Southern Ocean, protecting the Antarctic from immediate fluxes in PFC concentrations as they are released elsewhere in the world. As time progresses, however, contamination from oceanic sources is anticipated to increase [10, 11, 45]. Slow oceanic transport is cited as the reason for increasing PFC concentrations in the Arctic since the 1950s [14]. Models designed for Arctic research show that if oceanic transport to the Arctic ceased, the quantity of PFCs and their precursors delivered to the Arctic via atmospheric transport could not account for the concentrations measured in water, and thus marine transport is considered to be more important than atmospheric transport [15, 43]. It is also important to note that atmospheric and oceanic transport may be difficult or impossible to discern. For instance, PFCs found in the ocean are made up of three inputs: direct emissions to water, atmospheric deposition into water, and precursor compounds into water followed by degradation to ionic PFCs [14].

Among the compounds found in this study, PFHpA, PFBA, and PFPeA are byproducts of stain/grease-proof coatings on food packaging, couches, and carpets, while PFHxS was used in fire-fighting foams and carpet treatments and phased out of consumer products along with PFOS and PFOA by the major manufacturer (3M Co.) in the early 2000s due to health risks. While long-range atmospheric and oceanic transport of PFCs may partially explain the ubiquitous nature of these contaminants in the Antarctic Peninsula, military bases and infrastructure of nations established there may also be contributing sources of PFCs in Antarctic ecosystems. Atmospheric long-range transport of PFAAs as marine aerosols and degradation of PFCA and PFSA precursors such as low molecular weight FTOHs and acrylates/acids (FTAs) or perfluoroalkyl sulfonamids (FASA) and sulfonamido ethanols (FASE), which are more volatile and released to the atmosphere during fluoropolymer production processes, can be considered as other major pathways [11, 14, 44, 50, 51] to reach and deposit on the Antarctic Peninsula.

Additional and potential sources of PFAAs in the Antarctic Peninsula include aqueous filmforming foams (AFFF) and emissions of a current use insecticide, sulfluramid (N-ethyl perfluorooctane sulfonamide), to control leaf-cutting and fire ants in South America [20, 52; J. Benskin, pers. comm., June 2012). AFFF formulations have consisted of perfluoroalkyl sulfonates (PFHxS, PFOS, PFDS) and more recently, fluorotelomer sulfonamide-based surfactants. While these latter materials can degrade down to short-chain perfluoroalkylcar‐ boxylates (typically C4, C5, C6 PFCAs), sulfluramid can degrade to PFOS, FOSA, and PFCAs by abiotic and/or biological processes [53, 54]. Sulfluramid is manufactured in Brazil (≈30 tons/ year in 2007), and, in 2006, about 12 tons was exported to 13 other Central and South American countries [23, 55]. Because this insecticide is a semivolatile substance, it could be transported atmospherically to the Antarctic. Sulfluramid degradation products include PFOSA, PFOS, and potentially PFOA [52]. Despite high concentrations of PFOS, PFOA, and PFOSA measured off the Atlantic coast of South America (South Atlantic), increasing from Brazil to near Rio de la Plata (Argentina–Uruguay), attributed to the use of this substance [52], PFOS and PFOA were not detected in these Antarctic samples, with the exception of two feces samples and a lichen sample. This indicates that these two compounds have not yet fully reached the Antarctic Peninsula region, or local sources are not significant. The detection of several PFCA compounds in the present study is of particular importance as increasing trends of PFCA precursors (i.e., FTOHs) was observed in the Arctic with doubling times of 2.3–3.3 years from 2006 to 2012 [6].

PFCs produced via oxidation of precursor compounds once transported to the Antarctic is

**Figure 4.** Concentrations of PFC compounds (logarithmic scale in ng/g wet weight) measured in sediments, lichens, and seabird feces and feathers from the Antarctic Peninsula as a function of log *D* (A) and log *KOA* (B). In Figure 4A, the solid line shows the behavior of detected PFC concentrations versus log *D* in lichen; the dashed line indicates the trend of detected PFC concentrations versus log *D* values in feces. In Figure 4B, the solid line indicates the behavior of detect‐ ed PFC concentrations versus log *KOA* in lichen, while the dashed line shows detected PFC concentrations versus log

**PFHpA**

**PFOA**

**PFNA**

**PFDS**

**-1.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00**

**Log** *D*

**PFNA**

**PFHx**S **PFOS**

**4.00 5.00 6.00 7.00 8.00 9.00**

**Log** *KOA*

**PFTA**

**Feathers Feces Lichen Sediment**

**PFTA**

**PFDS**

**PFHpA**

**PFOA**

**PFHpA**

**PFHpA PFOA**

**PFOA**

**PFHpA**

**PFPeA PFHxA**

**PFPeA**

**PFBS**

**PFHx**S

**PFBA** 

**PFDS**

*KOA* in feces. Abbreviations for PFC chemical names are defined in Table 1.

**PFHxA**

**PFHxS**

**PFDS**

**PFHxS PFOS**

**PFHpA**

**PFPeA**

**Feathers Feces Lichen Sediment**

66 Emerging Pollutants in the Environment - Current and Further Implications

**PFPeA**

**PFBA**

**PFBS**

**0.1**

**100**

**B**

**0.1**

**1**

**10**

**PFCs (ng/g wet weight)**

**1**

**10**

**PFCs (ng/g wet weight)**

**100**

**A**

It is therefore expected that most PFCs are delivered to the Antarctic in their ionic, watersoluble state via the oceans [12, 14, 49]. Oceanic transport functions on a slower time scale for the Antarctic (in the order of decades, compared to days or weeks for atmospheric transport) because of the circulation patterns of the Southern Ocean, protecting the Antarctic from immediate fluxes in PFC concentrations as they are released elsewhere in the world. As time

often low [10, 11, 43–45, 47].
