N. VARDAR ET AL.
Copyright © 2013 SciRes. CWE EE
over the past few years. Passive samplers can be dep-
loyed at loca- tions where it is difficult or impractical to
install and maintain Hi-Vol samplers. These devices are
simple and inexpensive and do not need field calibration,
electricity nor technica l personnel at t he sampli ng site. In
addition, they can be deployed in many locations con-
currently due to their low cost.
Puerto Rico is located in the Caribbean with mainly
north-easterl y trade winds. Due to its location, the Is land
enjoys a tropical climate and a lso experiences the Atlantic
hurricane season. In this study, effect of ambient tem-
perature on passive samplers and seasonal distribution of
PAHs were evaluated Characterization of the PAHs in
the ambient air has been studied for a long time in other
parts of the world. A vast number of publications are
available in the literature for both urban and rural areas
from developed and developing countries [1-3,5-7].
How-ever, only limited data on atmospheric PAHs had
been acquired for Puerto Rico, and to our knowledge,
this is the first study to compare PAH concentrations
spatially distr ibuted across the island.
2. Materials and Method
2.1. Sampling Locations
Puerto Rico is a tropical island located in the Caribbean
Sea with a population of around 4 million. It has 9100
km2 total area and is one of the most densely populated
islands in the world. PUF pa ssive air samplers consisting
of polyurethane foam disks of 14 cm diameter and 1.2
cm thickness housed in two stainless steel bowls were
dep lo yed i n Ba yamo n, C aye y, Caroli na, Gua ya ma, Manati,
San Sebastian, and Naguabo (Figure 1).
Two sampling campaigns were achieved in 2010 for
three-month periods except in Naguabo where sampling
took place for six motnhs. Sampling parameters for the
samples are summarized in Table 1. Sampling sites of
Cayey and Manati are considered as urban, medium den-
sity, and residential. Bayamon and Carolina is described
as urban, high-density residential/industrial area. Naguabo
site is characterized as low-density coastal/resident ial.
San Sebastian is located in west part of the Island and
Figure 1. Map of Puerto Rico showing sampling locations.
Table 1. Passive Sampling Campaign Parameters.
Sampling
Campaign Start
Date End
Date Temp .
(
C) Rain
(cm)
Non-Hurricane
Seas on March 2010 June 2010 27.7 17.9
Hurric ane S eason June 2010 October 2010 28 .6 21.8
considered as urban, medium density, and residential.
Last, Guaya ma sampling site is situated in a coasta l/rural
region.
The temperature in the south of the Island is usually a
few degrees higher than the north. Between winter and
summe r, t here is onl y a te mpe ra ture s wing o f aro und 3˚C
Rainfall tends to be evenly distributed throughout the
year, but doubles during the months from May to Octo-
ber, as falls fro m November to April, with a driest period
from January to April. The wind patterns across the is-
land are basically zonal, from east to west.
2.2. Analy tical Proc edu re s
Information for sample preparation, extraction and analysis
is given in detail elsewhere [8]. Briefly, each PUF was
Soxhlet extracted with a 20:80 dichloromethane (DCM):
petroleu m ether (PE) solution for 24 h. All samples were
spiked with PAHs surrogate standards prior to extraction.
Four deuterated PAHs were used as surrogate standard
and P yre ne -d10 for volumetric corrections. The recoveries
of the following surrogate standards were used to correct
the amounts of specific PAHs found in the samples:
Ace na p h t hen e -d10 for acenaphthene (ACE), acenaphthylene
(ACT) and fluorine (FLN), phenanthrene-d10 for phe-
nanthrene (PHE), anthracene (ANT), and fluoranthene
(FL), chrysene-d12 for pyrene (PY), benz(a) anthracene
(BaA) and chrysene (CHR), and perylene-d12 for benzo
(b) fluoranthe (B bF), benzo(k)fluoranthe (BkF), benzo (a)
pyrene (BaP), indeno (1,2cd) pyrene (IcdP), dibenz (a,h)
anthracene ( DahA), and benzo(ghi)perylene (BghiP). T he
average recoveries for surrogates in field samples were
79% ± 31% for acenaphthene-d10, 86% ± 13% for phe-
nanthrene-d10, 84% ± 28% for chrysene-d12, 80% ± 15%
for perylene-d12.
The analysis of the samples was performed using a
Varian 450-GC coupled to an ion trap mass spectrometer
Varian 240 MS. A Varian factor 4 capillary column (30
m, 0.25 mm, 0.25 µm) was used. The GC oven tempera-
ture was programmed from 60℃(held one minute) to
130℃ at 7℃ min-1, then raised to 200℃ at 5℃ min-1,
and finally increased from 260 ℃ to 320℃ at 6℃ min-1.
The injector temperature was maintained at 295℃. The
linearity in the response of the GC/MS system was eva-
luated with calibration standards, at five different levels
of concentration (0.012, 0.06, 0.3, 0.6 and 1.2 ng/mL).
The instrument was Auto tuned at the start of runs with