Environment International 55 (2013) 25–32
Contents lists available at SciVerse ScienceDirect
Environment International
journal homepage: www.elsevier.com/locate/envint
Does wet precipitation represent local and regional atmospheric transportation by
perfluorinated alkyl substances?
Sachi Taniyasu a, Nobuyoshi Yamashita a,⁎, Hyo-Bang Moon b, Karen Y. Kwok c, Paul K.S. Lam c, Yuichi Horii d,
Gert Petrick e, Kurunthachalam Kannan f,⁎⁎
a
National Institute of Advanced Industrial Science and Technology (AIST), 16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan
Department of Marine Sciences and Convergence Technology, College of Science and Technology, Hanyang University, Ansan 426-791, Republic of Korea
Departmentof Biology and Chemistry, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong
d
Group of Chemical Substances, Center for Environmental Science in Saitama, 914 Kamitanadare, Kazo, Saitama 347-0115, Japan
e
Helmholtz Centre for Ocean Research (GEOMAR), Düsternbrooker Weg 20, 24105 Kiel, Germany
f
Wadsworth Center, New York State Department of Health and Department of Environmental Health Sciences, State University of New York, Albany, NY 12201-0509, USA
b
c
a r t i c l e
i n f o
Article history:
Received 31 December 2012
Accepted 17 February 2013
Available online 19 March 2013
Keywords:
Perfluorinated alkyl substances
Global transport
Precipitation
Snow core
Vertical profile
Long-range transport
a b s t r a c t
Perfluorinated alkyl substances (PFASs) have been found widely in the environment including remote marine locations. The mode of transport of PFASs to remote marine locations is a subject of considerable scientific interest.
Assessment of distribution of PFASs in wet precipitation samples (i.e., rainfall and snow) collected over an area
covering continental, coastal, and open ocean will enable an understanding of not only the global transport but
also the regional transport of PFASs. Nevertheless, it is imperative to examine the representativeness and suitability of wet precipitation matrixes to allow for drawing conclusions on the transport PFASs. In this study, we collected wet precipitation samples including rainfall, surface snow, and snow core from several locations in Japan to
elucidate the suitability of these matrixes for describing local and regional transport of PFASs. Rain water collected
at various time intervals within a single rainfall event showed high fluxes of PFASs in the first 1-mm deposition.
The scavenging rate of PFASs by wet deposition varied depending on the fluorocarbon chain length of PFAS. The
depositional fluxes of PFASs measured for continental (Tsukuba, Japan) and open ocean (Pacific Ocean, 1000 km
off Japanese coast) locations were similar, on the order of a few nanograms per square meter. The PFAS profiles in
“freshly” deposited and “aged” (deposited on the ground for a few days) snow samples taken from the same location varied considerably. The freshly deposited snow represents current atmospheric profiles of PFASs, whereas
the aged snow sample reflects sequestration of local sources of PFASs from the atmosphere. Post-depositional
modifications in PFAS profiles were evident, suggesting reactions of PFASs on snow/ice surface. Transformation
of precursor chemicals such as fluorotelomer alcohols into perfluoroalkylcarboxylates is evident on snow surface.
Snow cores have been used to evaluate time trends of PFAS contamination in remote environments. Snow collected at various depths from a core of up to 7.7 m deep, at Mt. Tateyama (2450 m), Japan, showed the highest concentrations of PFASs in the surface layer and the concentrations decreased with increasing depth for most PFASs,
except for perfluorobutanesulfonate (PFBS). Downward movement of highly water soluble PFASs such as PFBS,
following melting and freezing cycles of snow, was evident from the analysis of snow core.
© 2013 Elsevier Ltd. All rights reserved.
1. Introduction
Perfluorinated alkyl substances (PFASs) have been widely used in
commercial and industrial applications for over 60 years. PFASs have
emerged as global environmental pollutants (Butt et al., 2010; Giesy
and Kannan, 2001; Kannan, 2011; Shoeib et al., 2006; Tao et al., 2006).
In comparison with legacy persistent organic pollutants (POPs) such as
polychlorinated biphenyls, PFASs including perfluoroalkylcarboxylates
⁎ Corresponding author. Tel./fax: +81 29 861 8335.
⁎⁎ Corresponding author. Tel.: +1 518 474 0015.
E-mail addresses: [email protected] (N. Yamashita), [email protected]
(K. Kannan).
0160-4120/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.envint.2013.02.005
(PFCAs) and perfluoroalkylsulfonates (PFSAs) have higher water solubilities on the order of a few to several thousands of milligrams per
liter (Rayne and Forest, 2009). A few studies have suggested that hydrosphere can be the ultimate sink for PFASs (Ahrens et al., 2009; Kwok et
al., 2010; Yamashita et al., 2005, 2008). Several studies have reported
the occurrence of PFCAs and PFASs in wet precipitation from several locations across the globe (Barton et al., 2007; Kim and Kannan, 2007;
Kwok et al., 2010; Liu et al., 2009; Scott et al., 2006; Taniyasu et al.,
2006; Yamashita, 2005; Yamashita et al., 2004). Occurrence of notable
concentrations of PFASs in wet precipitation implies that wet deposition
is an effective scavenger for removal of PFASs from the atmosphere
(Barton et al., 2007; Kim and Kannan, 2007). Furthermore, wet precipitation can be a major pathway for the migration of PFASs from
26
S. Taniyasu et al. / Environment International 55 (2013) 25–32
atmospheric to hydrospheric compartments (Scott et al., 2006;
Stemmler and Lammel, 2010). Franco et al. (2011) suggested that clouds
could play a role in the persistence of perfluorooctanoate (PFOA) in the
troposphere. Although several studies have suggested the role of wet
precipitation matrixes in determining transport of PFASs, representativeness and reliability of these matrixes can be an issue depending on
the time and location of collection. The partitioning and atmospheric
transportation of PFASs can be influenced by several factors including
physicochemical properties and deposition rates of chemicals as well
as intensity and frequency of wet deposition, meteorological conditions,
among others (Cai et al., 2012; Dreyer et al., 2010; Franco et al., 2011;
Rayne and Forest, 2009; Simcik and Dorweiler, 2005).
Although precipitation samples can be useful in understanding the
sources, transport, and deposition of PFASs between atmospheric and
hydrospheric compartments, several factors can influence the concentrations and profiles of PFASs in precipitation samples (Kwok et
al., 2010). To elucidate some of the variables associated with PFAS
levels in precipitation samples, in this study, we collected rain
water, surface snow, and snow core (at various depths) from several
locations in Japan and determined concentrations and profiles of
PFASs. Surface seawater samples were collected from a region that receives drains of melt snow water, to elucidate variations in PFAS profiles between wet deposition and seawater samples. The objectives of
this study were to elucidate the representativeness and reliability of
wet deposition matrixes for understanding atmospheric transportation of PFASs.
2. Materials and methods
2.1. Sample collection
Wet precipitation samples, such as rain water and snow were collected during 2005–2012. Rain water samples were collected from
Tsukuba city during seven rainfall events (1 July 2006, 12 December
2006, 16 April 2007, 10 June 2007, 14 June 2007, 19 October 2007
and 12 January 2008) and in the Pacific Ocean (between 29°47N,
149°20E and 29°58N, 149°20E, January 20–21, 2012; approximately
1000 km off Japanese coast) during the MR11-8 research cruise on
the research vessel (RV) Mirai. Details of the wet precipitation sampling have been reported elsewhere (Scott et al., 2006; Yamashita,
2005; Yamashita et al., 2004) and the sampling locations are shown
in Fig. 1. Rain water samples were collected consecutively at every
1-mm interval, for up to 5-mm wet deposition, during each of the
rainfall events by use of a rain water sampler (W-2S, W-102, Shibata
Scientific Technology Ltd., Saitama, Japan). To avoid potential for contamination by PFASs during sample collection, the rain water sampler
was made up of a fluoropolymer-free material and consisted of a
stainless funnel with polypropylene collection tubes; the surface
area of the funnel was 0.1 m 2. Samples were collected from each sampling event, so that the volume of rain water collected for every 1-mm
of deposition was 100 mL.
Surface snow samples were collected at five locations within
Tsukuba City and Mt. Tsukuba, Japan, on 21 January 2006, when a snowfall amount of 16 cm occurred. Five locations, denoted as Stations A to E,
were selected for sampling, as follows; Station A: Institute's (AIST)
ground, Station B: Institute's lawn, Station C: rooftop of a building
(approximately 8 m above the ground), Station D: residential area
ground, and Station E: Mt. Tsukuba (altitude 260 m above sea level; approximately 10 km from Tsukuba city center). Snow samples were collected on day 0 (immediately after the snow on January 21), day 4
(4 days after the snow), and days 6 or 7 (6–7 days after the snow).
There was no snowfall between the days 0 and 7. The snow samples collected on days 4, 6, and 7 (from the day of snow fall) are referred in this
study as “aged” or “deposited” snow.
Snow core samples were collected from Mt. Tateyama (2450 m
above sea level in Toyama Prefecture) in 15 May 2005, Toyama City
(approximately 23 m above the sea level and 35 km from the foot
of Mt. Tateyama) in 12 February 2006, and in Mt. Zao (1736 m;
Yamagata Prefecture; two sampling points denoted as ZA, ZB and
ZC) in 19 February 2005 and 22 February 2009. Mt. Tateyama, at
2450 m altitude, receives snowfall every year from October to May.
The snow cover in Mt. Tateyama at the time of sampling in May
2005 was 16 m deep. Snow samples were collected by digging a
hole of up to 7.7 m and sampled at several depths (surface, 50, 100,
180, 600 and 770 cm) of the core.
Surface seawater was collected from the Japan Sea (39°19′N,
136°24′E) in 20 May 2005 during a research cruise of KT-05-11 at RV
Tansei-maru, and from Toyama Bay (36°82′N,137°01′E) in 16 May
2005, to compare the concentration and composition of PFASs with precipitation samples. Toyama Bay receives discharges from Toyama City
and is a part of the Japan Sea (Fig. 1).
2.2. Extraction and analysis of PFASs
Snow samples were thawed at room temperature prior to analysis. Analysis of PFASs in different types of water samples was
performed using a solid phase extraction (SPE) method as described
in previous studies (Taniyasu et al., 2008; Yamashita et al., 2008). In
brief, 1 ng of 13C-labeled internal standards ( 13C2-PFBA, 13C4-PFOA,
13
C5-PFNA, 13C2-PFDA and 13C4-PFOS) was spiked into water samples prior to extraction by the SPE method. The extraction was
performed using Oasis®WAX SPE cartridges, which were preconditioned by passage of 4 mL of 0.1% ammonia/methanol, 4 mL
of methanol, and 4 mL of Milli-Q-water. Precipitation samples
(100 mL for rain samples, 200 mL for snow samples from Tsukuba
city and station ZB and station ZC at Mt. Zao, 400 mL for snow samples from Toyama, 500 mL for snow samples from Mt. Tateyama and
station ZA at Mt. Zao and seawater from Toyama Bay and 860 mL for
seawater from Japan Sea) were passed through the SPE cartridges,
and rinsed with 4 mL of acetate buffer (pH = 4), which was
discarded. The moisture remaining in the cartridges was completely
removed by centrifugation. The target analytes were eluted into
4 mL of methanol and 4 mL of 0.1% ammonia/methanol. The sample
bottles were rinsed with methanol, and the rinsate was eluted
through the SPE cartridge. This step was necessary to recover long
chain PFCAs sorbed onto sample containers. The eluents from both
fractions were concentrated under a gentle stream of nitrogen to
0.2 mL to 1 mL.
Concentrations of PFASs were determined by use of Agilent HP1100
liquid chromatograph interfaced with a Micromass Quattro UltimaPt
mass spectrometer that was operated in an electrospray negative ionization mode. All of the samples were injected onto both RSpak JJ-50 2D
(2.0 mm inner diameter × 150 mm length; Shodex, Showa Denko K.K.,
Kawasaki, Japan) and Keystone Betasil C18 column (2.1 mm inner diameter × 50 mm length, 5 μm, 100 Å pore size, end-capped) separately to
confirm the accuracy of identification (Taniyasu et al., 2005, 2008). If
the results from the two columns did not match, samples were
re-analyzed. Five PFSAs (PFEtS, PFPrS, PFBS, PFHxS and PFOS), twelve
PFCAs (PFBA, PFPeA, PFHxA, PFHpA, PFOA, PFNA, PFDA, PFUnDA,
PFDoDA, PFTeDA, PFHxDA and PFOcDA), PFOSA, and 8:2FTUCA were
analyzed. Limit of detection (LOD) ranged from 0.05 to 0.25 ng/m2 (or
ng/L; based on the area of rain water sampler which was 0.1 m2 or
1000 cm2 and a collection volume of 100 mL, which equates to 1 mm
rain) for rain water samples, from 0.05 to 0.25 ng/L for snow samples
from Tsukuba city, from 0.01 to 0.2 ng/L for snow samples from Toyama,
from 0.05 to 0.25 ng/L for snow samples from Mt. Zao (station ZB and
ZC), from 0.0008 to 0.02 ng/L for snow samples from Mt. Tateyama and
Mt. Zao (station ZA), from 0.002 to 0.05 ng/L for seawater from Toyama
Bay and 0.0004 to 0.01 ng/L for seawater from Japan Sea. Average recoveries of internal standards spiked into water samples were 92%, 101%,
117%, 99% and 107% for 13C2-PFBA, 13C4-PFOA, 13C5-PFNA, 13C2-PFDA
and 13C4-PFOS, respectively.
S. Taniyasu et al. / Environment International 55 (2013) 25–32
27
Fig. 1. Map of Japan showing sampling locations.
The flux (X) of PFASs through wet deposition was calculated from
sample concentration (C; ng/L), sample volume (L) and surface area
of rain water sampler (m 2) using the following equation;
3
2
C ðng=LÞ 1ðmmÞ ¼ X ng=0:001 m 0:001ðmÞ ¼ X ng=m
in Japan, as reported earlier (Fig. S1; modified from Kwok et al.,
2010). Our findings provide a clear evidence of scavenging of PFASs
from the atmosphere by wet deposition. On the basis of the total
fluxes of PFASs calculated for 5-mm deposition, we determined that,
approximately 75% of PFASs were scavenged in the first 3-mm deposition, indicating that most of the removal of PFASs from the atmosphere occurred at the beginning of a rainfall event (Fig. 3). A
similar scavenging effect was reported for POPs such as dioxins and
furans (Koester and Hites, 1992). The scavenging rate (i.e., relative
flux) of individual PFASs during the rainfall event was calculated as
follows;
3. Results and discussion
Scavenging rate = Flux at X th mm (ng/m 2)/Flux at the first mm
(ng/m 2).
2
2
C ðng=LÞ 0:1ðLÞ=0:1 m ¼ X ng=m
The flux (X) can also be calculated from the sample concentration
(C; ng/L) and deposition (mm), using the following equation;
3.1. Rain water
The fluxes of PFASs in rain water samples collected at seven discrete rainfall events in Tsukuba are summarized in Table S1. The
total flux of rainfall for each event was 5 mm except for one rainfall
event which was 4 mm. The total flux of sum of all PFASs in rainfall
ranged from 40.8 to 186 ng/m 2. The major PFASs found in rain
water were perfluorobutanoate (PFBA), perfluorononanoate (PFNA),
and PFOA, in all the samples. The highest flux of PFASs was found in
the first 1-mm of deposition, and decreased gradually after the first
1-mm of deposition (Fig. 2A). A similar decreasing trend of PFAS
fluxes was found in rainfall samples collected from Kawaguchi city
The scavenging rates of individual PFASs were different during a
rainfall event (Fig. S2). It is interesting to note that PFBA and short
chain PFCAs (up to PFNA) were scavenged rapidly within the first
3-mm deposition. In contrast, PFDA and long chain PFCAs were scavenged relatively slowly. The relative proportion of long chain PFCAs in
rain water collected at 3- to 5-mm depositions was higher than that
for short chain PFCAs. In particular, the flux of PFDoDA, a long chain
PFCA, in rainfall samples increased gradually from 0 to 3 mm and
then decreased. This suggests that scavenging efficiencies of individual
PFASs by wet precipitation vary widely. Furthermore, higher concentration of PFOS was found in the 1–2 mm deposition than in 0–1 mm deposition. This indicates differences in scavenging rates between PFCAs
28
S. Taniyasu et al. / Environment International 55 (2013) 25–32
Fig. 2. Average deposition flux (ng/m2) of PFASs at every 1-mm rainfall from 0 to 5-mm in (A) Tsukuba, Japan (n = 7) from June 2006 to January 2008 and (B) the Pacific Ocean
(n = 1) from January 2012, 1000 km off Japanese coast. PFEtS, PFPrS, PFBS and PFHxS were below the LOD in all the samples.
and PFSAs, and this might be related to physiochemical properties.
PFSAs have relatively lower water solubilities than PFCAs of corresponding carbon chain length and the former may be sorbed to atmospheric particles more strongly than the latter (Rayne and Forest,
2009). Long chain PFCAs have lower water solubilities and stronger
sorption potentials than short chain PFCAs (Rayne and Forest, 2009).
Our results suggest that the wet deposition flux of PFASs is associated
with the amount of rainfall and physicochemical properties of PFASs
(Barton et al., 2007; Dreyer et al., 2010). The difference in scavenging
efficiencies can affect the composition of PFASs in precipitation
Fig. 3. Average composition of individual PFASs to total concentrations at every 1-mm
deposition for up to 5-mm rainfall collected from Tsukuba, Japan.
depending on the amount of rainfall and sampling time. For instance,
if the rain water samples had been collected after a first few mm deposition, it would not represent the actual concentrations and profiles
found in the atmosphere. Our finding also suggests that high concentrations of PFASs in rainfall samples can be expected when the amount of
rainfall is low. Therefore, comparison of PFAS concentrations in precipitation samples between studies should take into account of rainfall
amount, and the time of sampling during a rainfall event. It is more appropriate to present PFAS data for precipitation as flux (i.e., amount per
square area) with the total amount of rainfall rather than concentration.
In addition to rain water samples collected from Tsukuba, we also
collected rain water samples from the Pacific Ocean (between 29°47N,
149°20E and 29°58N, 149°20E; 1000 km off Japan) on January 20–21,
2012, during a research cruise. The rain water samples were collected
at every 1-mm interval for up to 5-mm deposition. The total flux of
PFASs ranged from 4.18 to 25.8 ng/m2 at each 1-mm deposition. The
major PFASs found in rain water from the Pacific Ocean were PFOA,
PFOS, and PFNA. PFBA and PFPeA were not detected (b 0.25 ng/m2) in
rain water samples from the Pacific Ocean, although these compounds
were found at high levels in Tsukuba. This suggests a short half-life of
these two compounds in the atmosphere. It is interesting to note that
rain water samples collected from continental location (Tsukuba,
Japan) and approximately 1000 km off Japan, in the open Pacific
Ocean, showed a same order magnitude in fluxes of PFASs, except for
PFBA and PFPeA (Fig. 2B). These results suggest the movement of air
and clouds by the prevailing westerly winds carry contaminants from
the continental Asia to the mid-latitude Pacific Ocean. The environmental levels of PFASs have changed over the time period. Therefore, it is also
expected that the differences in PFASs in rainwater collected at 1000 km
off Japan versus rain in Tsukuba could be because of the changes in
S. Taniyasu et al. / Environment International 55 (2013) 25–32
29
profiles between 2006–2008 and 2012. There were no information on
the trends of PFASs in rain after 2009 in Tsukuba, although the estimated
annual fluxes for PFBA in 2006–2007 (1280 ng/m2) and 2007–2008
were similar (2030 ng/m2) (Kwok et al., 2010). Further studies are needed on the trends in the fluxes of PFASs in continental and open ocean
locations.
The PFAS fluxes between continental (Tsukuba) and open ocean
locations (Pacific Ocean) were on the same order magnitude, although the concentrations of PFASs in inland waters were several orders of magnitude higher than the concentrations measured in open
ocean waters (Yamashita et al., 2005, 2008). A similar flux of PFASs
between rain water collected from inland, continental locations to
open oceans suggests that PFASs are transported by jet stream and
cloud movements on a regional scale.
Similar to the deposition flux of PFASs calculated for Tsukuba (and
as reported for Kawaguchi), deposition flux of PFASs in the Pacific
Ocean, off the coast of Japan, was the highest in the first 1-mm deposition, and gradually decreased with an increment in rainfall amount.
This reiterates significant scavenging of PFASs from the atmosphere at
the beginning of the rainfall event, both in continental and off-shore
locations.
3.2. Snow
The concentrations of PFASs in snow collected from five locations in
Tsukuba ranged from 3.04 to 40.5 ng/L. The concentrations and compositions of PFASs were compared between fresh snow (day 0) and “deposited” (i.e., aged) snow collected 4 to 7 days after the snowfall
(Fig. 4). The concentrations of PFASs in fresh snow samples (day 0)
were slightly higher in Tsukuba city (urban area) than the samples collected at Mt. Tsukuba (remote area, 10 km away from urban center),
suggesting fresh snow (day 0) may reflect local pollution by PFASs.
The overall concentrations of PFASs in “deposited” snow samples
(4–7 days after snowfall) were higher than those found in fresh snow
samples (day 0) in all of the five sampling locations. The concentrations
of several PFCAs in “deposited” snow increased remarkably after
4–7 days of initial deposition, and this increase was more pronounced
for PFOA and PFNA. Long chain PFCAs (C10–C12) and PFOS were not
detected in fresh snow (day 0), but were detected in “aged” snow samples (days 4, 6 and 7). The concentrations of PFASs in “aged” snow samples may reflect sequestration of contaminants from the atmosphere
over time and ice surface chemical reactions that lead to the transformation of some PFASs (Kim and Kannan, 2007). Photochemical reactions on snow or ice surface can lead to the formation of PFCAs from
precursor molecules (Webster and Ellis, 2010, 2012; Young et al.,
2007). Remarkable increase in the concentrations of PFOA and PFNA
on “aged” snow suggests the formation of these PFCAs from the degradation of fluorotelomer alcohols by surface reactions on snow/ice. We
did not analyze fluorotelomer alcohols in snow, but these compounds
are present in very high concentrations in air (Kim and Kannan,
2007). Overall, these findings suggest that fresh snow can reflect contamination levels of PFASs in the atmosphere, but the collection of
snow deposited on the ground for several days may not provide accurate reflection of atmospheric profiles of PFASs.
The changes in concentrations of individual PFASs with aging of
snow samples deposited on the ground were calculated as follows;
Increment in PFAS concentrations ðng=LÞ
¼ Concentration on day X ðng=LÞ–day 0 concentration ðng=LÞ
Concentrations of PFOA, PFNA, longer chain of PFCAs, PFOS, and
PFOSA increased in all of the “aged” snow samples (collected at
4–7 days after initial deposition) (Fig. S3). However, the rate of increase
in concentrations of PFBA, PFPeA, PFHxA, and PFHpA in “aged” snow samples varied depending on the sampling locations. The “aged” snow samples collected at the center of Tsukuba city (Stations A and B) showed
Fig. 4. Time trends of PFAS concentrations in fresh (day 0) and aged snow samples (collected
4, 6, or 7 days after snowfall) collected from five locations in Tsukuba city, Japan between
21and 28 January 2006. A: center of ground B: on the grass C: rooftop D: residential area
and E: Mt. Tsukuba. 8:2FTUCA was not analyzed. PFEtS, PFPrS, PFBS and PFHxS were below
the LOD in all samples.
comparable or slight reduction in concentrations of PFBA, PFPeA, PFHxA
and PFHpA with time, whereas the samples at the rooftop, residential
ground, and the mountain (Stations C, D and E) showed a clear increment
in concentrations (Fig. 4). It is probable that snow deposited on the
ground sequester contaminants in the atmosphere by cold trapping
mechanism, apart from photochemical reactions that could occur on
ice/snow surface. The differences in the rates of sequestration of PFASs
with time appear to be related to the existence of local sources of PFASs
and prevailing weather conditions.
3.3. Snow and ice core
Ice core and snow core samples have been used to elucidate the
temporal changes in the deposition of PFASs. However, significant
30
S. Taniyasu et al. / Environment International 55 (2013) 25–32
perturbations from thawing and freezing cycles can alter the depositional profiles of chemicals. The vertical profiles of individual PFASs collected at six depths of a snow core collected from Mt. Tateyama are
shown in Fig. 5. The total concentrations of the sum of PFASs in snow
core samples ranged from 0.884 to 4.51 ng/L. PFOA and PFNA were
the predominant compounds in samples collected at various depths.
The highest concentrations of total PFASs were found at the surface
layer of the snow core. Although the vertical profiles of individual
PFASs were different, the overall vertical profiles of PFCAs (PFPeA,
PFOA, PFDoDA, and PFDA) were similar. Similar results were found for
snow core samples collected from Mt. Zao (at 1736 m) which is not
influenced by anthropogenic activity except for winter sport activities
such as skiing (Fig. S4). Samples were collected in two stations (ZA
and ZB) at Mt. Zao, which were located approximately 5 km from
each other, but showed a remarkable difference in the composition of
PFASs. In both stations, surface snow contained higher concentrations
of PFASs than sub-surface snow (depth of 0.5 to 1.3 m). PFAS profiles
in the two stations were different, although the samples were collected
at different years and at different sides of the mountain. PFOA was the
predominant compound at the location ZA. In contrast, snow samples
from location ZB, near a ski resort, contained PFNA as the predominant
compound followed by PFOA, PFUnDA and PFHpA, but no other PFASs
were found. Surface snow was also collected from station ZC, near the
ski resort on the same side of the mountain near station ZB, but approximately 1 km away from each other. PFNA was found at more than
twenty times higher in station ZC than in station ZB. PFAS profiles
found at station ZB were similar to those found at station ZC suggesting
the influence of skiing activity.
The vertical profile of PFBS was different from those found for
other PFASs in snow core samples from Mt. Tateyama. Although
PFBS was not found in the surface, highest concentrations of this compound were found in the deepest layers (6–8 m deep). This pattern
could be explained by snow chemistry and micro-meteorology.
Even at temperatures below the melting point of ice, solarized snow
surface can melt. The melt-water can percolate and reach deeper
layers. PFBS is a water soluble compound and can percolate by the
movement of melt-water. Although the production and usage of
PFBS began only in the past 10 years, occurrence of this compound
in deeper layers of snow core supports movement by percolation.
Therefore, snow/ice core does not seem to conserve the chronology
of pollutants deposition. It is worth to mention that PFSAs and
PFCAs showed different vertical profiles even within the same snow
core sample.
3.4. Sea water
Snow melt-water can be a source of PFASs in surface waters which
can eventually enter into oceans through riverine discharges (Zhao et
al., 2012; Zushi et al., 2008). To investigate the transportation of PFASs
originating from snow samples at Mt. Tateyama (2450 m; Toyama prefecture) to coastal oceans, snow collected in Toyama city (23 m) and
seawater collected from the Toyama Bay (0 m) were analyzed (Fig. 6).
The concentrations and compositions of PFASs were similar between
snow sample collected at an altitude of 2450 m and at 23 m in Toyama,
indicating no altitudinal change in PFAS profiles, except for PFBA, PFPeA
and PFOS. However, LODs for PFBA (b 200 pg/L) and PFPeA (50 pg/L) in
snow sample from Toyama city were comparable to those found in
snow from Mt. Tateyama. In the snow samples collected at two different
altitudes, the concentrations of PFNA were comparable to those of
PFOA. No altitudinal difference in the concentrations of PFASs reflects
similarity/homogeneity in atmospheric vertical profiles of PFASs. For
seawater samples collected between coastal and open ocean environments, the concentrations of PFNA were much lower than that of
PFOA (Yamashita et al., 2008). This indicates that PFASs can be
transported vertically from the ground level to 2.5 km above the
ground. However, the concentrations of PFASs decreased dramatically
across 25 km horizontal distance (coastal to open oceans) (Fig. 6). Although the concentration of PFOS in water samples was approximately
half of PFOA concentration in inland waters in Japan (Senthilkumar et
al., 2007; Taniyasu et al., 2003; Yamashita et al., 2008), PFOS was
found seventy times lower in open ocean waters 285 km away from
the coast. Interestingly, only PFOA was found in the open ocean surface
waters. In addition to transport by oceanic currents, volatile precursors
of PFOA and PFOS may be the sources of this compound in remote areas.
In summary, our results suggest that data for PFASs determined
from precipitation samples should be interpreted with caution, and
adequate description of samples (e.g., time of sampling), meteorological conditions, and geographical features should be described along
with results. In addition, it is more appropriate to present PFAS data
for precipitation as flux (i.e., amount per square area, with total
amount of rainfall) rather than concentration. The origin and pathway of cloud is also an important parameter for the tracking of source
of contaminants in precipitation (Franco et al., 2011). As clouds move,
they can carry contaminants over long distances. Our previous studies
indicated that PFASs in surface and deep-sea water can provide information on the transport of PFASs on a global scale by oceanic currents
(Yamashita et al., 2008). Precipitation samples such as rainfall and
Fig. 5. Vertical profiles of PFAS concentrations in snow core collected from Mt. Tateyama (at 2450 m) at six different depths in 15 May 2005.
S. Taniyasu et al. / Environment International 55 (2013) 25–32
31
Fig. 6. Comparison of horizontal and vertical differences in the concentrations (pg/L) of PFASs in precipitation and surface seawater samples. Inland snow samples were collected at
Toyama city in February 2006 and at Mt. Tateyama in May 2005. Surface seawater samples were collected from Toyama Bay and Japan Sea in May 2005. PFTeDA, PFHxDA, PFOcDA,
8:2 FTUCA, PFEtS, and PFPrS were below the LODs in all the samples.
snow can be useful in tracking the transport of PFASs on a local and/or
regional scale. Nevertheless, careful planning and designing of study
as well as data analysis are necessary for a better understanding of
transport of PFASs in the environment.
Acknowledgments
Part of this study was funded by the Ministry of the Environment
(project no. B-1106) and the Japan Society for the Promotion of
Science (project no. 23710032).
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.envint.2013.02.005.
References
Ahrens L, Felizeter S, Ebinghaus R. Spatial distribution of polyfluoroalkyl compounds in
seawater of the German Bight. Chemosphere 2009;76:179–84.
Barton CA, Kaiser MA, Russell MH. Partitioning and removal of perfluorooctanoate during rain events: the importance of physical-chemical properties. J Environ Monit
2007;9:839–46.
Butt CM, Berger U, Bossi R, Tomy GT. Levels and trends of poly- and perfluorinated
compounds in the arctic environment. Sci Total Environ 2010;408:2936–65.
Cai M, Yang H, Xie Z, Zhao Z, Wang F, Lu Z, et al. Per- and polyfluoroalkyl substances in
snow, lake, surface runoff water and coastal seawater in Fildes Peninsula, King
George Island, Antarctica. J Hazard Mater 2012;209–210:335–42.
Dreyer A, Matthias V, Weinberg I, Ebinghaus R. Wet deposition of poly- and perfluorinated
compounds in Northern Germany. Environ Pollut 2010;158:1221–7.
Franco A, Hauschild M, Jolliet O, Trapp S. Atmospheric fate of non-volatile and ionizable
compounds. Chemosphere 2011;85:1353–9.
Giesy JP, Kannan K. Global distribution of perflurooctanesulfonate in wildlife. Environ
Sci Technol 2001;35:1339–42.
Kannan K. Perfluoroalkyl and polyfluoroalkyl substances: current and future perspectives. Environ Chem 2011;8:333–8.
Kim S-K, Kannan K. Pefluorinated acids in air, rain, snow, surface runoff, and lakes: relative importance of pathways to contamination to urban lakes. Environ Sci Technol
2007;41:8328–34.
Koester CJ, Hites RA. Wet and dry deposition of chlorinated dioxins and dibenzofurans.
Environ Sci Technol 1992;26:1375–82.
Kwok KY, Taniyasu S, Yeung LWY, Murphy MB, Lam PKS, Horii Y, et al. Flux of
perfluorinated chemicals through wet deposition in Japan, the United States, and
several other countries. Environ Sci Technol 2010;44:7043–9.
Liu W, Jin YH, Quan X, Sasaki K, Saito N, Nakayama SF, et al. Perfluorosulfonates and
perfluorocarboxylates in snow and rain in Dalian, China. Environ Int 2009;35:737–42.
Rayne S, Forest K. Perfluoroalkyl sulfonic and carboxylic acids: a critical review of physicochemical properties, levels and patterns in waters and wastewaters, and treatment methods. J Environ Sci Health A 2009;44:1145–99.
Scott BF, Spencer C, Mabury SA, Muir DCG. Poly and perfluorinated carboxylates in
North American precipitation. Environ Sci Technol 2006;40:7167–74.
Senthilkumar K, Ohi E, Sajwan K, Takasuga T, Kannan K. Perfluorinated compounds in
river water, river sediment, market fish, and wildlife samples from Japan. Bull
Environ Contam Toxicol 2007;79:427–31.
Shoeib M, Harner T, Vlahos P. Perfluorinated chemicals in the Arctic atmosphere. Environ Sci Technol 2006;40:7577–83.
Simcik MF, Dorweiler KJ. Ratio of perfluorochemical concentrations as a tracer of atmospheric deposition to surface waters. Environ Sci Technol 2005;39:8678–83.
Stemmler I, Lammel G. Pathways of PFOA to the Arctic: variabilities and contributions
of oceanic currents and atmospheric transport and chemistry sources. Atmos Chem
Phys 2010;10:9965–80.
Taniyasu S, Kannan K, Horii Y, Hanari N, Yamashita N. A survey of perfluorooctane sulfonate (PFOS) and related perfluorinated organic compounds in water, fish, birds
and humans from Japan. Environ Sci Technol 2003;37:2634–9.
Taniyasu S, Kannan K, So MK, Gulkowska A, Sinclair E, Okazawa T, et al. Analysis of
fluorotelomer alcohols, fluorotelorner acids, and short- and long-chain perfluorinated
acids in water and biota. J Chromatogr A 2005;1093:89–97.
Taniyasu S, Rostkowski P, So MK, Miyake Y, Lam PKS, Falandysz J, et al. Ultra trace analysis of persistent perfluorinated chemicals in the wet deposition matrixes. 232nd
ACS National Meeting, 2006, San Francisco, USA; 2006.
Taniyasu S, Kannan K, Yeung LWY, Kwok KY, Lam PKS, Yamashita N. Analysis of
trifluoroacetic acid and other short-chain perfluorinated acids (C2–C4) in precipitation by liquid chromatography-tandem mass spectrometry: comparison to patterns
of long-chain perfluorinated acids (C5–C18). Anal Chim Acta 2008;619:221–30.
Tao L, Kannan K, Kajiwara N, Fillmann G, Takahashi S, Tanabe S. Perfluorooctanesulfonate
and related fluorochemicals in albatrosses, elephant seals, penguins and polar skuas
from the Southern Ocean. Environ Sci Technol 2006;40:7642–8.
Webster E, Ellis DA. Potential role of spray generation in the atmospheric transport of
perfluorocarboxylic acids. Environ Toxicol Chem 2010;29:1703–8.
32
S. Taniyasu et al. / Environment International 55 (2013) 25–32
Webster EM, Ellis DA. Understanding the atmospheric measurement and behavior of
perfluorooctanoic acid. Environ Toxicol Chem 2012;31:2041–6.
Yamashita N. Global monitoring and trace analysis of perfluorinated chemicals in environment. An International Symposium on Fluorinated Alkyl Organics in the Environment (FLUOROS), Toronto, ON, Canada; 2005.
Yamashita N, Kannan K, Taniyasu S, Horii Y, Hanari N, Okazawa T, et al. Environmental
contamination by perfluorinated carboxylates and sulfonates following the use of
fire-fighting foam in Tomakomai, Japan. Organohalogen Compd 2004;66:4063–8.
Yamashita N, Kannan K, Taniyasu S, Horii Y, Petrick G, Gamo T. A global survey of
perfluorinated acids in oceans. Mar Pollut Bull 2005;51:658–68.
Yamashita N, Taniyasu S, Petrick G, Si W, Gamo T, Lam PKS, et al. Perfluorinated acids as
novel chemical tracers of global circulation of ocean waters. Chemosphere 2008;70:
1247–55.
Young CJ, Furdui VI, Franklin J, Koerner RM, Muir DCG, Mabury SA. Perfluorinated acids
in arctic snow: new evidence for atmospheric formation. Environ Sci Technol
2007;41:3455–61.
Zhao Z, Xie Z, Möller A, Sturm R, Tang J, Zhang G, et al. Distribution and long-range
transport of polyfluoroalkyl substances in the Arctic, Atlantic Ocean and Antarctic
coast. Environ Pollut 2012;170:71–7.
Zushi Y, Takeda T, Masunaga S. Existence of nonpoint source of perfluorinated compounds and their loads in the Tsurumi River basin, Japan. Chemosphere 2008;71:
1566–73.