Environ Sci Technol 39, 2005, 1995 2003 .pdf
Original filename: Environ Sci Technol 39, 2005, 1995-2003.pdf
Title: No Job Name
This PDF 1.2 document has been generated by Parlance Publisher 5.0/(Xyvision Postscript Formatter) 3.0 3 / Acrobat Distiller Command 3.01 for Solaris 2.3 and later (SPARC), and has been sent on pdf-archive.com on 01/11/2015 at 18:03, from IP address 71.17.x.x.
The current document download page has been viewed 598 times.
File size: 666 KB (9 pages).
Privacy: public file
Download original PDF file
Environ. Sci. Technol. 2005, 39, 1995-2003
Polychlorinated Dioxins and Furans
from the World Trade Center
Attacks in Exterior Window Films
from Lower Manhattan in
New York City
Department of Chemistry, University of Victoria, P.O. Box
3065, Victoria, BC, Canada, V8W 3V6
MICHAEL G. IKONOMOU*
Institute of Ocean Sciences, Fisheries and Oceans Canada,
9860 West Saanich Road, Sidney, BC, Canada V8L 4B2
CRAIG M. BUTT,
MIRIAM L. DIAMOND, AND
Department of Geography, University of Toronto, 45 St.
George Street, Toronto, ON, Canada, M5S 3G3
Samples of ambient organic films deposited on exterior
window surfaces from lower Manhattan and Brooklyn in
New York City were collected six weeks after the terrorist
attacks at the World Trade Center (WTC) on September
11, 2001 and analyzed for polychlorinated dibenzo-p-dioxins
and dibenzofurans (PCDD/Fs). Total tetra- through octaCDD/F concentrations in window films within 1 km of the
WTC site in lower Manhattan ranged up to 630 000 pg/m2
(estimated as a mass concentration of ca. 1 300 000 pg/
g) and a maximum toxic equivalent (TEQ) concentration of
4700 TEQ/m2 (ca. 10 000 pg TEQ/g). Measurements at a
background site 3.5 km away in Brooklyn showed lower
concentrations at 130 pg TEQ/m2 (260 pg TEQ/g). Ambient gasphase PCDD/F concentrations estimated for each site
using an equilibrium partitioning model suggested
concentrations ranging from ca. 2700 fg-TEQ/m3 near the
WTC site to the more typical urban concentration of 20 fgTEQ/m3 at the Brooklyn site. Multivariate analyses of 2,3,7,8substitued congeners and homologue group profiles
suggested unique patterns in films near the WTC site
compared to that observed at background sites in the study
area and in other literature-derived combustion source
profiles. Homologue profiles near the WTC site were
dominated by tetra-, penta-, and Hexa-CDD/Fs, and 2,3,7,8substituted profiles contained mostly octa- and hexachlorinated congeners. In comparison, profiles in Brooklyn and
near mid-Manhattan exhibited congener and homologue
patterns comprised mainly of hepta- and octa-CDDs, similar
to that commonly reported in background air and soil.
The terrorist attacks at the World Trade Center (WTC)
complex in New York City on September 11, 2001 resulted
* Corresponding author phone: (250)363-6804; fax: (250)363-6807
10.1021/es049211k CCC: $30.25
Published on Web 02/15/2005
2005 American Chemical Society
in the immediate loss of life of nearly 3000 persons. Each of
the twin towers at the WTC site was attacked by a hijacked
commercial airplane and destroyed by the combined effects
of the plane impacts and the fires ignited by the jet fuel
carried by the planes. These events marked the first time
high-rise buildings have been destroyed by such tactics (1).
Before the attacks, the north and south towers of the WTC
were the tallest buildings in New York City (NYC) at 110
stories each, with respective heights of 417 and 415 m. The
hijacked planes collided with the north and south towers at
8:46 AM EST and 9:03 AM EST, each carrying an estimated
34 000 L and 31 000 L of jet fuel, respectively (2). The initial
building fires weakened the steel floor joists, causing the
eventual collapse of the north and south towers within 104
and 62 min after being hit, respectively. WTC-7, a 47-story
office building, was also damaged by the collapsing towers,
subsequently caught fire, and collapsed later in the afternoon
of September 11, 2001.
The terrorist attacks on the WTC resulted in a single-day
death toll greater than at any other time in American history
except the Civil War battle of Antietam (3). When the WTC
towers collapsed, more than 1.2 million tons of building
material came down in the core of lower Manhattan (4). The
resulting fires caused by the nearly full airline fuel tanks at
the times of the impacts were estimated to range from 750
°C (5-7) up to 2500 °C (8). Following the collapse of the
towers, fires continued to burn at the site for several months
at lower temperatures. The large quantities of plastics in
building materials (e.g., PVC piping and coated copper wire),
consumer and office goods present in each tower, and
∼130 000 gallons of PCB contaminated transformer oil (4)
stored below ground level represented a potential store of
halogenated organic contaminants arising from subsequent
processes of combustion and pyrolysis.
In the present study, we investigated concentrations and
patterns of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) in exterior surface films obtained from
windows in Lower Manhattan and Brooklyn six weeks after
the WTC attacks. Surface films such as these accumulate on
impervious surfaces that are characteristic of urban areas
(e.g., windows, building exteriors, asphalt). The films are
derived from the deposition of numerous organic and
inorganic compounds present in ambient urban air (9, 10).
As such, the films provide a sample of the complex mixture
of semivolatile organic compounds (SVOCs) to which humans
and biota are exposed through inhalation of urban air and
ingestion of, and contact with, urban soils and vegetation.
Furthermore, exterior surfaces of windows were chosen as
convenient impervious surfaces because of their ubiquity,
ease of sampling, relatively inert nature, and absence of
inherent contamination (10). Material accumulated on the
exterior of windows has also proven to be a convenient and
reproducible means of obtaining an integrated sample of
atmospherically deposited chemicals. Particles and particlesorbed chemicals are efficiently captured on such surface
films (9) that are presumed to initially accumulate because
of the condensation of gas-phase compounds, followed by
the deposition of particulate matter. As the film continues
to grow and develop, gas-phase compounds may partition
between the air and the organic phase of the film (10, 11).
The primary objectives of this study were to use window
films as (a) indicators of the potential degree of contamination
by PCDD/Fs in lower Manhattan and surrounding regions
from the WTC attacks on September 11, 2001 and subsequent
fires and (b) to explore previously unreported unique PCDD/
Fs congener patterns that resulted from the WTC fires.
VOL. 39, NO. 7, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
FIGURE 1. Sampling locations for window films in New York City. Sampling site labels and superscripts correspond with site descriptions
in other figures and tables. The height of the vertical bars in (a) and (b) correspond with relative ΣPCDD/F and ΣPCDD/F-TEQ concentrations
between sites as given in Table 1, respectively. The base of each bar indicates the geographical location of the sampling site.
TABLE 1. Concentrations (in pg/m2) of 2,3,7,8-Substituted PCDD/Fs, ΣP4-8CDD/F, and TEQs in Window Films from New York City
date sampled (dd/mmm/yy)
date of last cleaning
window area (m2)
Further, window films have been used to examine the extent
and magnitude of September 11th contamination with
respect to polycyclic aromatic hydrocarbons (PAH), polychlorinated biphenyls (PCB), polybrominated diphenyl ethers
(PBDE), polychlorinated naphthalenes (PCN), and organochlorinated pesticides (12).
Sample Collection. Organic film samples were collected from
the outside of windows at eight sites in lower Manhattan
and Brooklyn, New York City (Figure 1) by scrubbing the
surfaces with precleaned laboratory Kimwipes soaked in
HPLC grade 2-propanol. All sampling locations were on
window surfaces facing the WTC site. Between 0.6 and 5.6
m2 of window surface area was cleaned at each site depending
on the apparent quantity of organic film (Table 1). Kimwipes
were precleaned by extracting with CH2Cl2, air-drying, and
storing in precleaned clear glass jars. Field blanks were
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 7, 2005
(Broadway & (Park Row/
prepared at three sites by soaking 10 precleaned Kimwipes
with 2-propanol and wafting in the air until dry (Supporting
Information Table S1). Field blank preparation was identical
to sample preparation, except that field blank Kimwipes did
not contact window surfaces. Windows were wiped until the
window surface appeared to have no residual surface film
left. As has been demonstrated elsewhere (10), it was assumed
that all organic contaminants present on the window surface
were captured by the 2-propanol soaked Kimwipes. Sampling
was conducted between October 27 and October 29, 2001.
Average air temperatures over the period from September
11 to October 27, 2001 were calculated using meteorological
data provided by the National Weather Service and available
at www.erh.noaa.gov. The average temperature of 17.1 °C is
the geometric mean of daily average temperatures over this
period. Samples were collected from either ground level or
second story windows and a 10-cm border was left on all
windows to prevent direct contamination from building
materials. Following collection, samples were stored in
precleaned glass jars, kept in the dark at all times to prevent
analyte photodegradation, and frozen at -20 °C until analysis.
Further details on the collection of urban organic films using
this method, along with quality control/quality assurance
(QA/QC) protocols and validation for SVOCs, are provided
Sample Analysis. Sampled Kimwipes were Soxhlet extracted overnight with a 80:20 toluene:acetone mixture. A
suite of nine 13C-labeled PCDD/F surrogate method internal
standards (2,3,7,8-DF, 1,2,3,7,8-DF, 1,2,3,4,7,8-DF, 1,2,3,4,6,7,8DF, 2,3,7,8-DD, 1,2,3,7,8-DD, 1,2,3,6,7,8-DD, 1,2,3,4,6,7,8DD, and OCDD) were added prior to extraction. Extracts
were acid/base washed with H2SO4 and KOH, respectively,
and then returned to neutral pH by rinsing with HPLC grade
water. The extract was then reduced to dryness and reconstituted in 1:1 CH2Cl2:hexane. Sample cleanup took place
using three stages. First, the sample was passed through a
multilayered acidic/basic silica column, eluted with 1:1 CH2Cl2:hexane, reduced to dryness using UHP-N2, and subsequently reconstituted in hexane. To remove sulfur impurities,
the sample was then passed through a copper column with
hexane as the eluant. Next, the sample was again reduced
in volume and applied directly to an alumina column for
fractionation. The fraction containing PCDD/Fs was eluted
with 1:1 CH2Cl2:hexane, reduced to dryness using UHP-N2,
and reconstituted in toluene. 13C-labeled PCDD recovery
standards (1,2,3,4-DD and 1,2,3,7,8,9-DD) were added to each
sample extract immediately prior to analysis. Samples were
analyzed using high-resolution gas chromatography with
high-resolution mass spectrometry (HRGC-HRMS) in the
selected ion monitoring (SIM) mode operating at 10 000
resolution. Recoveries of internal standards ranged from 45
to 98%. Details on the sample cleanup steps, the instrumental
analysis protocol used for full congener PCDD/Fs determination, the criteria used for identification, the QA/QC
measures applied, and the procedures used for quantification
are provided elsewhere (13). All 135 possible PCDD/F
congeners from tetra- through octachlorinated were quantitated using these methods.
The GC column we routinely use in our laboratory for
PCDD/F analyses is a 60-m DB5 fused silica capillary column
DB5 (0.25-mm i.d. with 0.1-µm film thickness) from J&W
Scientific (Folsom, CA). The temperature program is as
follows: 100 °C for 2 min following injection; 20 °C/min to
200 °C; 1 °C/min to 215 °C; 7 min at 215 °C; 4 °C/min to 300
°C; and held for 3 min at this temperature. UHP-He is the
carrier gas and is held at a constant head pressure of 25 psi
to maintain a linear velocity of 35 cm/s. The sample volume
injected into the splitless injector is 1 µL of sample extract
plus 0.5 µL of air with the purge valve activated 2 min after
injection. The splitless injector port, direct GC-MS interface,
and the MS ion source are maintained at 260 °C, 290 °C, and
300 °C, respectively. Using the DB5 column under such
conditions, 2,3,7,8-TeCDF coelutes with five other tetraCDFs: 2,3,4,8-, 2,3,4,7-, 2,3,4,6-, 1,2,4,6-, and 1,2,4,9-TeCDF.
Because the amount of 2,3,7,8-TeCDF detected in the window
film samples nearest the WTC site was substantial, and this
congener is of toxicological relevance, all samples were
reanalyzed using a DB225 GC column that has the capacity
to separate 2,3,7,8-TeCDF from these coeluting congeners.
An aliquot of the same extract analyzed on the DB5 column
was subsequently analyzed on the 30-m DB225 (50% cyanopropylphenyl methylpolysiloxane stationary phase) fused
silica capillary column (0.25-mm i.d. with 0.15-µm film
thickness) from J&W Scientific (Folsom, CA). The GC temperature program for the DB225 column was as follows: 140
°C for 2 min following injection; 20 °C/min to 200 °C; 1 °C/
min to 220 °C; and 10 °C/min to 240 °C. The splitless injector
port, direct GC-MS interface, and the MS ion source were
maintained at 220 °C, 260 °C, and 270 °C, respectively. UHPHe was the carrier gas at a constant head pressure of 10 psi.
All other HRGC-HRMS conditions were identical to those
used with the DB5 column.
Data Analysis. Cluster analysis (with the standardized
Euclidean measure and Ward clustering method) and
principal components analysis were performed using KyPlot
(v.2.0 b.9). 2,3,7,8-Substituted PCDD/F congener patterns
and references for the literature samples used in constructing
Figures 4 and 5 are provided in the Supplementary Information. Homologue totals (e.g., tetra-CDDs) are the sum of all
possible congeners having the particular level of chlorination.
The sum of homologue totals for each degree of chlorination
from tetra- through octasubstituted for both the CDDs and
CDFs is equal to ΣP4-8CDD/F. Homologue contributions
toward ΣP4-8CDD/F were calculated by dividing the homologue total by ΣP4-8CDD/F. In addition, 2,3,7,8-substituted
PCDD/F contributions were calculated toward the sum of
P4-8CDD/F. Using this approach, the sum of homologue
contributions will equal 100%, whereas the sum of 2,3,7,8substituted PCDD/F contributions will not equal 100% (as
there are significant contributions toward ΣP4-8CDD/F from
non-2,3,7,8 substituted congeners). TEQ concentrations were
calculated using the World Health Organization’s 1998 TEF
Results and Discussion
PCDD/F Concentrations in Window Films from Manhattan
and Brooklyn after September 11, 2001. Samples of window
films from seven sites in lower Manhattan and one site in
Brooklyn (Figure 1) in New York City were obtained October
27-29, 2001, approximately six weeks after the terrorist
attacks at the WTC site on September 11, 2001. Window film
samples contained ΣP4-8CDD/F concentrations at sites within
1 km of the WTC up to 630 000 pg/m2 (Table 1 and Figure
1), in comparison to concentrations from 1440 to 5470 pg/
m2 at sites >3.5-4.0 km from the WTC site that are
representative of background conditions in Brooklyn and
northern lower Manhattan. ΣP4-8CDD/F concentrations in
window films decreased rapidly as a function of distance
from the WTC site (Figure 1). Concentrations of all 135
congeners in the three blank samples taken in Brooklyn and
in lower Manhattan (at the Church & Warren and NYU) were
below the method detection limits for P4-8CDDs (3.3-14.2
pg per congener) and P4-8CDFs (1.8-12.1 pg per congener),
respectively. These results suggest negligible PCDD/F contamination of Kimwipe window film samples during exposure
to the ambient atmosphere at each site, even those closest
to the WTC (e.g., Church & Warren). In lower Manhattan,
window films appeared to have greater mass than those found
in other urban areas (∼100-200 mg/m2) (10), possibly up to
400-500 mg/m2. Using an upper estimate of 500 mg/m2
window film in lower Manhattan, P4-8CDD/F concentrations
are conservatively estimated at 1 300 000 pg/g (pg analyte
per gram of surface film) in window films near the WTC.
Similar spatial patterns were also observed for the 17
individual 2,3,7,8-substituted PCDD/Fs. For all 2,3,7,8substituted PCDD/F congeners except 2,3,4,7,8-PeCDF, the
highest concentrations were observed nearest the WTC site
and decreasing levels were found moving north into midManhattan and southeast into Brooklyn. The anomalous
spatial pattern for 2,3,4,7,8-PeCDF (exhibiting a maximum
of 330 pg/m2 at the Union Square site, with all other
concentrations in the range 0.9-24 pg/m2) may be a result
of its coelution with 1,2,4,8,9-, 1,3,4,8,9-, and 1,2,3,6,9-PeCDF,
making the reported values subject to background regional
PCDD/F patterns. In other words, there may be a regional
background PCDD/F pattern with localized PCDD/F “hot
spots” for certain congeners in Manhattan superimposed on
VOL. 39, NO. 7, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
TABLE 2. Predicted Gas-Phase Air Concentrations (in fg/m3) of 2,3,7,8-Substituted PCDD/Fs and TEQs in New York City
Back-Calculated from Window Films, on the Basis of an Equilibrium Partitioning Approach
the PCDD/F signature produced by the WTC attacks. Full
congener concentrations of the 135 individual P4-8CDD/Fs
analyzed in this study at each site are provided in Supporting
Information Table S2. Concentrations of the 118 non-2,3,7,8substituted P4-8CDD/Fs followed a similar spatial pattern to
the 2,3,7,8-subsituted congeners discussed above. The highest
concentrations for 134 of 135 congeners (except for 2,3,4,7,8pentaCDF as discussed above) were observed nearest the
WTC site. This attests to the WTC attacks as the dominant
source of all tetra- through octa-CDD/F congeners in lower
Calculated Equilibrium Air Concentrations of PCDD/
Fs. Surface films also play an important role in mediating
chemical transport among environmental compartments (15,
16). These films act as transient sinks for chemicals whereby
low vapor pressure compounds are transferred to surface
waters via precipitation and subsequent runoff, and higher
vapor pressure compounds may volatilize into the atmosphere (16). The rapid response time of the mass of chemicals
in the film (days to weeks) contrasts sharply with that of soil
(years) (15). Within the atmosphere, SVOCs (such as PCDD/
Fs) sorbed to particulate matter are assumed to be at
equilibrium with the gas phase (17). This equilibrium
partitioning can also be extended to surface films (18), and
gas-phase concentrations can thus be calculated as Cg-air )
Cfilm/(foc × KOA), where Cg-air is the concentration of a
particular analyte in the atmospheric gas phase, Cfilm is the
analyte concentration in the surface film, foc is organic carbon
content of the surface film (on a mass basis; typically 0.2 for
exterior surface films (10)), and KOA is the octanol-air partition
coefficient (15). This technique has been successfully used
to estimate gas-phase air concentrations of polybrominated
diphenyl ethers in Southern Ontario, Canada (19). For PCDD/
Fs at approximately 17 °C, values of log KOA range from 10.90
for 2,3,7,8-TeCDF to 13.00 for octa-CDD (20).
Previous work has reinforced the validity of assuming
equilibrium between the atmospheric gas phase and surface
films for persistent SVOCs (i.e., those with minimal degradation within the film such as PCBs) (18). The potential to
reconstruct gas-phase PCDD/F air concentrations using this
equilibrium partitioning approach in NYC is complicated by
the reported rapid decrease in total atmospheric (i.e., gas +
particle phase) PCDD/F concentrations following the WTC
attacks (21). On the basis of previous studies of air particles,
the kinetics of sorption-desorption are not assumed to be
rate-limiting, and that diffusion, which is presumed to be
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 7, 2005
the rate-limiting step, is rapid and occurs within a time scale
of hours to days. Given this potential for rapid air-film
exchange, chemical concentrations in surface films are likely
representative of atmospheric conditions at the time of
sampling, although it must be remembered that the surface
film material itself, and hence the mass of contaminants
within the film, accumulates over time. Using this equilibrium
partitioning approach under changing atmospheric contaminant levels (conditions were observed to occur after the
WTC attacks), back-calculated gas-phase air concentrations
of SVOCs can provide order-of-magnitude estimates.
Predicted PCDD/F-TEQ gas-phase air concentrations
(Table 2) ranged from 6 to 20 fg-TEQ/m3 at sites most remote
from the WTC (NYU and Brooklyn, respectively) and up to
2730 fg-TEQ/m3 near the former site of the WTC (Church &
Warren) in a pattern analogous to ΣPCDD/F concentrations
in window films. These predicted PCDD/F-TEQ concentrations are within the range reported for air samples taken at
the same time by the U.S. EPA in lower Manhattan (100010 000 fg-TEQ/m3 throughout most of lower Manhattan and
increasing up to 175 pg-TEQ/m3 at the WTC site) (21). The
U.S. EPA data are gas + particle-phase PCDD/Fs and are
expected to exceed our estimated gas-phase only concentrations by severalfold since most PCDD/F are found in the
particle phase. The estimated gas-phase PCDD/F-TEQs near
the former WTC site are up to 2.5 orders of magnitude higher
than at background sites in Brooklyn and closer to midManhattan. By comparison, atmospheric gas + particle-phase
PCDD/F-TEQ concentrations reported in the literature can
typically be divided into the following three groups on the
basis of proximity to human activities: remote regions (<10
fg-TEQ/m3), rural areas (20-50 fg-TEQ/m3), and urban/
industrial centers (100-400 fg-TEQ/m3) (22). Thus, estimated
equilibrium gas-phase PCDD/F-TEQs near the WTC site in
lower Manhattan are within and some well above the high
end of the range reported for gas + particle phases in urban
air. Again, the comparisons above are between estimated
gas-phase concentrations in lower Manhattan and Brooklyn
and total air concentrations (both gas- and particle-phase
PCDD/Fs) reported elsewhere. The WTC samples examined
in this study were also analyzed for PCBs and PCNs and have
been presented elsewhere (12). The total TEQ that resulted
from the PCB and PCN measurements was very small and
accounted for only 0.5-5% of the combined total TEQ where
the PCCD/Fs TEQ was incorporated (12).
FIGURE 2. Tetra- through octa-CDD/F homologue profiles for window films in New York City.
PCDD/F Congener Patterns Resulting from the WTC
Attacks. On a homologue basis, PCDD/F patterns in window
films from lower Manhattan near the WTC have a distinct
profile compared to the background site in Brooklyn and
typical urban areas (Figure 2 and Supporting Information
Table S3). Homologue patterns in lower Manhattan are
dominated on a mass basis by tetra-, penta-, and hexa-CDDs
(46-67% of ΣPCDD/F), with lower contributions of heptaand octa-CDDs (3-20%) compared to the Brooklyn background site (55%). The patterns observed at sites near the
WTC are consistent with atmospheric particulate samples
exposed to combustion sources of PCDD/Fs where the
homologue contribution decreases with increasing chlorination (23-27). Furthermore, the Brooklyn homologue pattern
is consistent with typical urban and rural air samples that
are not directly influenced by combustion sources. In typical
urban and rural air samples, the homologue group contribution generally increases with increasing chlorination (i.e.,
more chlorinated congener groups comprise more of total
PCDD/Fs) (22, 28-31). The concordance of the Brooklyn
profile with typical urban profiles reaffirms this site as a
background reference. However, PCDD/F homologue patterns vary widely in combustion sources, perhaps because
of differing combustion conditions and feedstocks. Thus,
increasing PCDD homologue contributions with increasing
chlorination for incinerator effluents and diesel engine
exhaust have also been reported (28, 32-34). The higher
contribution of penta- and hexa-CDDs in lower Manhattan
relative to conventional solid-waste combustion may have
resulted from the large proportion of office paper burned at
the WTC site, as previous studies have shown paper
combustion to form high proportions of these homologue
groups (26, 27).
For PCDFs in particular, lower Manhattan films impacted
by the WTC fires displayed a general decrease in homologue
contribution with increasing chlorination, whereas in Brooklyn all PCDF homologue groups had approximately equal
contribution with no discernible pattern. A decreasing PCDF
homologue contribution with increasing chlorination is
consistent with ambient urban and remote air patterns (22,
24, 30) but also with combustion sources (17, 23, 26-28, 32,
35, 36). Increasing PCDF homologue contributions with
increasing chlorination have been reported for incinerator
effluents (34), diesel engine exhaust (33), and urban background air (28). Thus, PCDF homologue patterns cannot be
used to unambiguously diagnose sources. At all sites in lower
Manhattan, except Church & Warren nearest the WTC site
and the Museum at the southern end of the island of
Manhattan, homologue profiles followed the trend P4CDF ≈
P5CDF > P6CDF > P7CDF ≈ P8CDF. At these other two sites,
P4CDF was the dominant PCDF homologue group with a
consistent pattern of decreasing contribution with increasing
chlorination. The slightly higher contribution of P5CDF than
P4CDF at some sites in lower Manhattan is also similar to
other reported PCDF combustion signatures (24, 31).
The fires resulting from the WTC attacks also produced
unusual 2,3,7,8-substituted PCDD/F congener patterns (Figure 3). Here, we use the congener data from the DB225
column that separates 2,3,7,8-TeCDF from the other TeCDF
congeners. The seven sites in lower Manhattan exhibit a
VOL. 39, NO. 7, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
FIGURE 3. 2,3,7,8-Substituted PCDD/F congener profiles for window films in New York City. Contributions of 2,3,7,8-DF at each sampling
site shown above are those obtained from use of the DB225 analytical column which resolves 2,3,7,8-DF from its coeluting tetra-CDF
significantly lower proportion of octa-CDD (0.7-12% of ΣP4-8CDD/F) than in Brooklyn (37%). The second most abundant
2,3,7,8-substituted congener was 1,2,3,4,6,7,8-HpCDF which
comprised <1% of ΣP4-8CDD/F. Of the 17 2,3,7,8-substituted
congeners measured, 2,3,7,8-TeCDF contributed the most
toward PCDD/F-TEQs at each site (14.6-35.2%) when
comparing congener contributions on a TEQ basis for the
estimated gas-phase air concentrations. The contribution of
2,3,7,8-TeCDF toward PCDD/F-TEQs decreased with distance
from the WTC site. For example, the highest 2,3,7,8-TeCDF
contributions were found at Church & Warren (35%), the
Museum (32%), and Park Row/Spruce (28%), with lower
contributions at the most distant Union Square (15%), NYU
(18%), and Brooklyn (15%) sites. 1,2,3,7,8-PeCDD and 2,3,7,8TeCDD were the next most significant contributors toward
predicted gas-phase PCDD/F-TEQs, ranging from 9 to 20%
and 0.8 to 10%, respectively, with no apparent spatial patterns.
These results are unique because for almost all atmospheric
data reported, 2,3,4,7,8-PeCDF contributes most to PCDD/
F-TEQs, typically in the range from 20 to 40% (22).
With the exception of octa-CDF, which was less abundant
in lower Manhattan (1-4%) than in Brooklyn (40%), the
remaining 2,3,7,8-substituted congeners did not change
substantially in their contribution toward ΣP4-8CDD/F relative to the background site in Brooklyn. These results are in
contrast to the analysis of the dust that settled immediately
after the WTC attacks both at and near the WTC site, in which
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 7, 2005
the dust samples showed increasing contributions of 2,3,7,8substituted PCDD/Fs congeners with increasing chlorination
(37). The difference between these results likely resides in
both the different dates of sampling and the composition of
material sampled. Window films sampled six weeks after the
WTC attacks were most likely subject to precipitation during
the intervening time. Climatic data indicates that rain events
occurred on 13 days between September 11th and the time
of sampling (www.erh.noaa.gov). As well, the films sampled
here would have accumulated by wet and dry deposition of
coarse and fine particles as well as gas-film partitioning (15).
By contrast, surface samples taken immediately after the
attacks were comprised of g10-µm dust particles (>98% by
mass) that would have preferentially captured particle
associated PCDD/Fs, such as penta- through octa-CDD/Fs,
which have decreasing proportions in the gas phase with
increasing chlorination (e.g., minimum gas-phase fractions
for octa-CDD/F of 1.3% and 1.9%, respectively (20)). Furthermore, the high organic content of window films (38)
versus the largely inorganic and ionic content of the dust
further favors partitioning of PCDD/Fs into these surface
films. Finally, the differences could be due to different
combustion conditions that contributed to the PCDD/F
profiles sampled immediately after the WTC attacks versus
the smoldering conditions that prevailed in the ensuing
FIGURE 4. Cluster analysis plot of 2,3,7,8-substituted PCDD/F congener patterns in window films from lower Manhattan and Brooklyn in
New York City and samples from other combustion processes known to produce PCDD/Fs. The cluster analysis was performed using the
contributions of 2,3,7,8-DF, including those reported both in the literature references and in the current study, determined on DB5 or SP2331
columns which are unable to resolve 2,3,7,8-DF from its other coeluting tetra-CDF congeners.
Cluster analysis (CA; Figure 4) and principal components
analysis (PCA; Figure 5) were used to further interpret 2,3,7,8substituted PCDD/F film profiles relative to source profiles
(as percent in total 2,3,7,8-substituted PCDD/Fs, see Supporting Information Tables S4 and S5 for published congener
patterns corresponding references). These multivariate analyses were performed using the contributions of 2,3,7,8DF,
including those reported both in the literature references
and in the current study, determined on DB5 or SP2331
columns which are unable to resolve 2,3,7,8DF from its other
coeluting tetra-CDF congeners. The analyses show that the
samples from lower Manhattan group together (suggesting
similar congener patterns and a common source) and apart
from the Brooklyn sample, which is more remote from the
WTC site. In the CA plot, a close correlation is observed
between the Brooklyn window sample and other normal
urban and rural air samples, suggesting that the Brooklyn
site was largely unaffected by the WTC PCDD/F plume. The
lower Manhattan samples are also distinct from, but most
closely related to, other combustion and pyrolysis samples,
including those from various types of incinerators, hydrocarbon combustion (e.g., heavy oil, diesel fuel, paper), PVC
combustion, and combinations of these processes and
materials (e.g., PVC + paper combustion). Similar results
were observed for the PCA plot, whereby the lower Manhattan
FIGURE 5. Principal component analysis plot of 2,3,7,8-substituted
PCDD/F congener patterns in window films from lower Manhattan
and Brooklyn in New York City and samples from other combustion
processes known to produce PCDD/Fs. This was performed using
the contributions of 2,3,7,8-DF, including those reported both in the
literature references and in the current study, determined on DB5
or SP2331 columns which are unable to resolve 2,3,7,8-DF from its
other coeluting tetra-CDF congeners.
VOL. 39, NO. 7, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
window film samples grouped separately from other known
combustion sources and the Brooklyn window film, which
grouped with other normal urban and rural air samples.
We thank the building owners and managers of New York
City who granted permission for sample collection. J.
Archbold, K. Tsoi, and H. Jones of University of Toronto
assisted with sampling. Funding was provided by Environment Canada. We thank the analysts of the Regional Dioxin
Laboratory at the Institute of Ocean Sciences for their
assistance with sample analyses. We are also grateful to the
Department of Fisheries and Oceans for supporting us with
our collaborative research.
Supporting Information Available
Additional tables as noted in the text. This material is available
free of charge via the Internet at http://pubs.acs.org.
(1) Rehm, R. G.; Pitts, W. M.; Baum, H. R.; Evans, D. D.; Prasad, K.;
McGrattan, K. B.; Forney, G. P. Initial Model for Fires in the
World Trade Center Towers; Building and Fire Research Laboratory, National Institute of Standards and Technology: Gaithersburg, MD, 2002.
(2) World Trade Center Building Performance Study: Data Collection,
Preliminary Observations, and Recommendations; Federal Emergency Management Agency, Federal Insurance and Mitigation
Administration: Washington, DC, 2002.
(3) Lipton, E. A New Count of the Dead, but Little Sense of Relief.
New York Times, Dec 2, 2001, 41.
(4) Nordgren, M. D.; Goldstein, E. A.; Izeman, M. A. The Environmental Impacts of the World Trade Center Attacks: A Preliminary
Assessment; Natural Resources Defense Council: Washington,
(5) Eagar, T. W.; Musso, C. Why did the World Trade Center Collapse?
Science, engineering, and speculation. JOM-J. Min. Met. Mater.
Soc. 2001, 53, 8-11.
(6) Bazant, Z. P.; Zhou, Y. Why did the World Trade Center collapse?
Simple analysis. J. Engr. Mech.-ASCE 2002, 128, 2-6.
(7) Ashley, S. When the Twin Towers Fell. Sci. Am. [Online] http://
www.sciam.com/explorations/2001/100901wtc/ (accessed Dec
(8) Investigative Reports. Anatomy of September 11th, Broadcast
September 9, 2002 (Arts & Entertainment Network).
(9) Liu, Q. T.; Diamond, M. L.; Gingrich, S. E.; Ondov, J. M.;
Maciejczyk, P.; Stern, G. A. Accumulation of metals, trace
elements and semi-volatile organic compounds on exterior
window surfaces in Baltimore. Environ. Pollut. 2003, 122, 5161.
(10) Diamond, M. L.; Gingrich, S. E.; Fertuck, K.; McCarry, B. E.;
Stern, G. A.; Billeck, B.; Grift, B.; Brooker, D.; Yager, T. D. Evidence
for organic film on an impervious urban surface: characterization and potential teratogenic effects. Environ. Sci. Technol.
2000, 34, 2900-2908.
(11) Gingrich, S. E.; Diamond, M. L. Atmospherically derived organic
surface films along an urban-rural gradient. Environ. Sci.
Technol. 2001, 35, 4031-4037.
(12) Butt, C. M.; Diamond, M. L.; Truong, J.; Ikonomou, M. G.; Helm,
P. A.; Stern, G. A. Semivolatile organic compounds in window
films from lower Manhattan after the September 11th World
Trade Center attacks. Environ. Sci. Technol. 2004, 38, 35143524.
(13) Ikonomou, M. G.; Fraser, T. L.; Crewe, N. F.; Fischer, M. B.;
Rogers, I. H.; He, T.; Sather, P. J.; Lamb, R. F. A comprehensive
multi-residue ultra-trace analytical method, based on HRGC/
HRMS, for the determination of PCDDs, PCDFs, PCBs, PBDEs,
PCDEs, and organohalogen pesticides in six different environmental matrices. Can. Technol. Rep. Fish. Aquat. Sci. 2001, 2389,
(14) Van den Berg, M.; Birnbaum, L.; Bosveld, A. T. C.; Brunstrom,
B.; Cook, P.; Feeley, M.; Giesy, J. P.; Hanberg, A.; Hasegawa, R.;
Kennedy, S. W.; Kubiak, T.; Larsen, J. C.; van Leeuwen, F. X. R.;
Liem, A. K. D.; Nolt, C.; Peterson, R. E.; Poellinger, L.; Safe, S.;
Schrenk, D.; Tillitt, D.; Tysklind, M.; Younes, M.; Waern, F.;
Zacharewski, T. Toxic equivalency factors (TEFs) for PCBs,
PCDDs, PCDFs for humans and wildlife. Environ. Health
Perspect. 1998, 106, 775-792.
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 7, 2005
(15) Diamond, M. L.; Priemer, D. A.; Law, N. L. Developing a
multimedia model of chemical dynamics in an urban area.
Chemosphere 2001, 44, 1655-1667.
(16) Priemer, D. A.; Diamond, M. L. Application of the multimedia
urban model to compare the fate of SOCs in an urban forested
watershed. Environ. Sci. Technol. 2002, 36, 1004-1013.
(17) Harnly, M.; Stephens, R.; Mclaughlin, C.; Marcotte, J.; Petreas,
M.; Goldman, L. Polychlorinated dibenzo-p-dioxin and dibenzofuran contamination at metal recovery facilities, open burn
sites, and a railroad car incineration facility. Environ. Sci.
Technol. 1995, 29, 677-684.
(18) Butt, C. M. Chemical and Physical Characterization of Organic
Films on an Impervious Surface. M. Sc. Thesis, Department of
Geography, University of Toronto, Toronto, ON, Canada, 2003.
(19) Butt, C. M.; Diamond, M. L.; Truong, J.; Ikonomou, M. G.; ter
Shure, A. F. H. Spatial distribution of polybrominated diphenyl
ethers in southern Ontario as measured in indoor and outdoor
window organic films. Environ. Sci. Technol. 2004, 38, 724731.
(20) Harner, T.; Green, N. J. L.; Jones, K. C. Measurements of octanolair partition coefficients for PCDD/Fs: a tool in assessing airsoil equilibrium status. Environ. Sci. Technol. 2000, 34, 31093114.
(21) U.S. EPA. Exposure and Human Health Evaluation of Airborne
Pollution from the World Trade Center Disaster; EPA/600/P2/002A(External Review Draft); United States Environmental
Protection Agency, Office of Research and Development,
National Center for Environmental Assessment: Washington,
(22) Lohmann, R.; Jones, K. C. Dioxins and furans in air and
deposition: a review of levels, behaviour and processes. Sci.
Total Environ. 1998, 219, 53-81.
(23) Anthony, E. J.; Jia, L.; Granatstein, D. L. Dioxin and furan
formation in FBC boilers. Environ. Sci. Technol. 2001, 35, 30023007.
(24) Lugar, R. M.; Harless, R. L.; Dupuy, A. E.; McDaniel, D. D. Results
of monitoring for polychlorinated dibenzo-p-dioxins and
dibenzofurans in ambient air at McMurdo Station, Antarctica.
Environ. Sci. Technol. 1996, 30, 555-561.
(25) Pandompatam, B.; Kumar, Y.; Guo, I.; Liem, A. J. Comparison
of PCDD and PCDF emissions from hog fuel boilers and hospital
waste incinerators. Chemosphere 1997, 34, 1065-1073.
(26) Katami, T.; Ohno, N.; Yasuhara, A.; Shibamoto, T. Formation of
dioxins from sodium chloride-impregnated newspapers by
combustion. Bull. Environ. Contam. Toxicol. 2000, 64, 372376.
(27) Yasuhara, A.; Katami, T.; Okuda, T.; Ohno, N.; Shibamoto, T.
Formation of dioxins during the combustion of newspapers in
the presence of sodium chloride and poly(vinyl chloride).
Environ. Sci. Technol. 2001, 35, 1373-1378.
(28) Lorber, M.; Pinsky, P.; Gehring, P.; Braverman, C.; Winters, D.;
Sovocool, W. Relationships between dioxins in soil, air, ash,
and emissions from a municipal solid waste incinerator emitting
large amounts and dioxins. Chemosphere 1998, 37, 2173-2197.
(29) Duarte-Davidson, R.; Sewart, A.; Alcock, R. E.; Cousins, I. T.;
Jones, K. C. Exploring the balance between sources, deposition,
and the environmental burden of PCDD/Fs in the UK terrestrial
environment: an aid to identifying uncertainties and research
needs. Environ. Sci. Technol. 1997, 31, 1-11.
(30) Lee, R. G. M.; Green, N. J. L.; Lohmann, R.; Jones, K. C. Seasonal,
anthropogenic air mass, and meteorological influences on the
atmospheric concentrations of polychlorinated dibenzo-pdioxins and dibenzofurans (PCDD/Fs): evidence for the
importance of diffuse combustion sources. Environ. Sci. Technol.
1999, 33, 2864-2871.
(31) Brzuzy, L. P.; Hites, R. A. Global mass balance for polychlorinated
dibenzo-p-dioxins and dibenzofurans. Environ. Sci. Technol.
1996, 30, 1797-1804.
(32) Sakai, S. I.; Hayakawa, K.; Takatsuki, H.; Kawakami, I. Dioxinlike PCBs released from waste incineration and their deposition
flux. Environ. Sci. Technol. 2001, 35, 3601-3607.
(33) Ryan, J. V.; Gullett, B. K. On-road emission sampling of a heavyduty diesel vehicle for polychlorinated dibenzo-p-dioxins and
polychlorinated dibenzofurans. Environ. Sci. Technol. 2000, 34,
(34) Berlincioni, M.; Didomenico, A. Polychlorodibenzo-para-dioxins
and polychlorodibenzofurans in the soil near the municipal
incinerator of Florence, Italy. Environ. Sci. Technol. 1987, 21,
(35) Safe, S.; Brown, K. W.; Donnelly, K. C.; Anderson, C. S.;
Markiewicz, K. V.; Mclachlan, M. S.; Reischl, A.; Hutzinger, O.
Polychlorinated dibenzo-para-dioxins and dibenzofurans as-
sociated with wood-preserving chemical sites - biomonitoring
with pine needles. Environ. Sci. Technol. 1992, 26, 394-396.
(36) Eitzer, B. D.; Hites, R. A. Polychlorinated dibenzo-para-dioxins
and dibenzofurans in the ambient atmosphere of Bloomington,
Indiana. Environ. Sci. Technol. 1989, 23, 1396-1401.
(37) Lioy, P. J.; Weisel, C. P.; Millette, J. R.; Eisenreich, S.; Vallero, D.;
Offenberg, J.; Buckley, B.; Turpin, B.; Zhong, M. H.; Cohen, M.
D.; Prophete, C.; Yang, I.; Stiles, R.; Chee, G.; Johnson, W.; Porcja,
R.; Alimokhtari, S.; Hale, R. C.; Weschler, C.; Chen, L. C.
Characterization of the dust/smoke aerosol that settled east of
the World Trade Center (WTC) in lower Manhattan after the
collapse of the WTC 11 September 2001. Environ. Health
Perspect. 2002, 110, 703-714.
(38) Liu, Q. T.; Chen, R.; McCarry, B. E.; Diamond, M. L.; Bahavar,
B. Characterization of polar organic compounds in the organic
film on indoor and outdoor glass windows. Environ. Sci. Technol.
2003, 37, 2340-2349.
Received for review May 27, 2004. Revised manuscript received December 16, 2004. Accepted December 20, 2004.
VOL. 39, NO. 7, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY