Identification of volatile and extractable chloroorganics in rain and snow

Article abstract of DOI:10.1021/es980480i

Organics enriched from samples of rain, snow, and glacier ice were analyzed to determine the chemical structure of the chloroorganic compounds that were most abundant in such media. Gas chromatography with atomic emission detection (GC-AED) was used to provide an overview of the presence of volatile and extractable chloroorganics at four sites in Europe and one site in Antarctica. Real samples and isomer-specific standards were then analyzed by gas chromatography with mass-spectrometric detection (GC-MS) to identify and confirm the structure of the chloroorganics that had been detected in the GC-AED analysis. The results revealed that of the volatile chloroorganics found in the samples, dichloronitromethane, a compound not previously reported to occur in the ambient environment, was often present in the highest concentration (up to 130 ng L^-1). Chlorobenzenes were detected mainly as 1,4-dichlorobenzene and minor amounts of 1,3- and 1,2- dichlorobenzene and one isomer of tetrachlorobenzene (1,2,4,5- or 1,2,3,5- tetrachlorobenzene). Chlorinated alkyl phosphates, which were normally responsible for the largest peaks in the chlorine-specific chromatograms of hexane- or ether-extractable compounds, were present as tris(2- chloroethyl)phosphate, tris(1-chloro-2-propyl)phosphate, and one of the isomers bis(1-chloro-2-propyl)(3-chloro-1-propyl)phosphate er bis(1-chloro- 2-propyl)(2-chloro-1-propyl)phosphate. All of the chloroorganic compounds mentioned, i.e., dichloronitromethane, chlorobenzenes, and chlorinated alkyl phosphates, were detected at relatively remote sites in the northern hemisphere, whereas only chlorobenzenes were found in a reference sample of snow from Antarctica. Organics enriched from samples of rain, snow, and glacier ice were analyzed to determine the chemical structure of the chloroorganic compounds that were most abundant in such media. Gas chromatography with atomic emission detection (GC-AED) was used to provide an overview of the presence of volatile and extractable chloroorganics at four sites in Europe and one site in Antarctica. Real samples and isomer-specific standards were then analyzed by gas chromatography with mass-spectrometric detection (GC-MS)to identify and confirm the structure of the chloroorganics that had been detected in the GC-AED analysis. The results revealed that of the volatile chloroorganics found in the samples, dichloronitromethane, a compound not previously reported to occur in the ambient environment, was often present in the highest concentration (up to 130 ng L^-1). Chlorobenzenes were detected mainly as 1,4-dichlorobenzene and minor amounts of 1,3- and 1,2-dichlorobenzene and one isomer of tetrachlorobenzene (1,2,4,5- or 1,2,3,5-tetrachlorobenzene). Chlorinated alkyl phosphates, which were normally responsible for the largest peaks in the chlorine-specific chromatograms of hexane- or ether-extractable compounds, were present as tris(2-chloroethyl)phosphate, tris(1-chloro-2-propyl)phosphate, and one of the isomers bis(1-chloro-2-propyl)(3-chloro-1-propyl)phosphate or bis(1-chloro-2-propyl)(2-chloro-1-propyl)phosphate. All of the chloroorganic compounds mentioned, i.e., dichloronitromethane, chlorobenzenes, and chlorinated alkyl phosphates, were detected at relatively remote sites in the northern hemisphere, whereas only chlorobenzenes were found in a reference sample of snow from Antarctica.

Full text of DOI:10.1021/es980480i

Environ. Sci. Technol. 1998, 32, 3935-3940
In an attempt to detect chloroorganic compounds that
Identification of Volatile and
Extractable Chloroorganics in Rain
and Snow
may have been overlooked in previous studies of such
substances in rain and snow, Laniewski and co-workers (25)
determined the group parameters AOX (adsorbable organic
halogens) and EOX (extractable organic halogens) and applied
GC-AED (gas chromatography-atomic emission detection)
and GC-MS (gas chromatography-mass spectrometry) to
analyze specific compounds. The cited study revealed that
chloroorganics that are not normally analyzed were often
responsible for the largest peaks in the chromatograms
recorded in the chlorine channel of the atomic emission
detector. In particular, chlorinated alkyl phosphates, chlo-
robenzenes, and an unknown volatile compound were found
to be among the most abundant chloroorganics in the
fractions of volatiles and extractables.
K R Z YS Z T O F L A N I E W S K I , †
H A N S B O R EÄ N , † A N D
A N D E R S G R I M V A L L * ,
‡
Department of Water and Environmental Studies and
Department of Mathematics, Link o¨ ping University,
SE-581 83 Link o¨ ping, Sweden
Organics enriched from samples of rain, snow, and
glacier ice were analyzed to determine the chemical
structure of the chloroorganic compounds that were most
abundant in such media. Gas chromatography with
atomic emission detection (GC-AED) was used to provide
an overview of the presence of volatile and extractable
chloroorganics at four sites in Europe and one site in
Antarctica. Real samples and isomer-specific standards
were then analyzed by gas chromatography with mass-
spectrometric detection (GC-MS) to identify and confirm the
structure of the chloroorganics that had been detected
in the GC-AED analysis. The results revealed that of the
volatile chloroorganics found in the samples, dichloroni-
tromethane, a compound not previously reported to occur
in the ambient environment, was often present in the
highest concentration (up to 130 ng L ). Chlorobenzenes
were detected mainly as 1,4-dichlorobenzene and minor
amounts of 1,3- and 1,2-dichlorobenzene and one isomer of
tetrachlorobenzene (1,2,4,5- or 1,2,3,5-tetrachlorobenzene).
Chlorinated alkyl phosphates, which were normally
responsible for the largest peaks in the chlorine-specific
chromatograms of hexane- or ether-extractable compounds,
were present as tris(2-chloroethyl)phosphate, tris(1-chloro-
The present study focused on chloroorganics that can be
enriched by using purge-and-trap techniques or liquid-liquid
extraction with hexane or diethyl ether. Organic extracts of
such compounds were derived from samples of rain, snow,
and glacier ice, and the presence of chloroorganic compounds
was examined by using a combination of GC-AED and GC-
MS. The main objectives of our investigation were as follows:
(
i) to determine the structure of the volatile and extractable
precipitation-borne compounds that produce the largest
peaks in chromatograms recorded in the chlorine channel
of an AED system;
(ii) to undertake a special study of the presence of
chlorobenzenes and chlorinated alkyl phosphates in pre-
cipitation.
To facilitate general conclusions regarding the presence
of chloroorganics in rain and snow, the analyzed samples
were selected to represent both urban and remote areas.
-
1
Experimental Section
Sam pling. Snow samples from Sweden and Poland were
collected after individual snowfalls by pushing 5-L polypro-
pylene jars into fresh loose snow packs of sufficient depth,
i.e., 20-50 cm deep, avoiding the snow that was close to the
ice on lakes or the ground. The sample from Antarctica was
collected by pushing 2-L polyethylene bottles into the
snowpack just below the surface. The ambient temperature
during snow sampling was below 0 °C. A sample from a
glacier in Sweden was obtained by sawing a block of ice out
of a layer formed several hundred years ago. Dating was
based on annual growth of the glacier (26).
2-propyl)phosphate, and one of the isomers bis(1-chloro-
2-propyl)(3-chloro-1-propyl)phosphate or bis(1-chloro-2-
propyl)(2-chloro-1-propyl)phosphate. All of the chloroorganic
compounds mentioned, i.e., dichloronitromethane,
chlorobenzenes, and chlorinated alkyl phosphates, were
detected at relatively remote sites in the northern hemisphere,
whereas only chlorobenzenes were found in a reference
sample of snow from Antarctica.
Bulk precipitation (rainfall) from Ireland was sampled by
using a funnel-shaped, open collector made of stainless steel
2
(
surface area of 3 m , height of 1.5 m). The sampled
precipitation was accumulated in 10-L glass bottles.
The samples of snow from Poland were brought directly
to Link o¨ ping University in Sweden for prompt analysis. All
other samples were transported frozen to the laboratory and
then kept in a freezer until analyzed. The outer parts of the
ice block from the glacier were removed by sawing out
immediately before thawing. Information about sample
types and sampling sites is summarized in Table 1.
Introduction
Most of the studies of chloroorganics in rain and snow have
dealt primarily with compounds that have long been
recognized as harmful environmental pollutants. Some
investigators have made relatively broad analyses of priority
pollutants (1-5), whereas others have undertaken more
specific studies of PCBs (6-10), chlorinated dioxins (11, 12),
chlorophenols (13), or pesticides, such as DDT, lindane, and
dieldrin (14, 15). A number of scientists have studied the
Stripping Enrichm ent of Volatiles. Volatile chloroor-
ganics with boiling points ranging from 80 to 350 °C were
enriched by using a dynamic headspace technique (27).
Samples of rain or melted snow or ice were purged with N2
for 2 h at 60 °C. The compounds released were adsorbed on
an activated carbon filter at 80 °C and then desorbed by
extracting the filter with three portions of CH Cl (10 + 5 +
occurrence of chlorinated C
2
4
-C hydrocarbons (3, 16, 17)
and haloacetic acids (3, 18-22) in precipitation. Recently,
the first results regarding chlorinated alkyl phosphates in
rain and snow were published (23-25).
2
2
5
µL). An internal standard, 1-chlorooctane, was added to
*
Corresponding author phone: +46 13 281482; fax +46 13 100746;
the final extract.
Liquid-Liquid Extraction of Neutrals and Bases. Ex-
tractable neutrals and bases in rain and melted snow or ice
e-mail: angri@mai.liu.se.
†
Department of Water and Environmental Studies.
Department of Mathematics.
‡
1
0.1021/es980480i CCC: $15.00

1998 Am erican Chem ical Society
VOL. 32, NO. 24, 1998 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 9 3 5
Published on Web 10/31/1998

TABLE 1. Sample Types and Sampling Sites
medium
sampling site
site environment
position
sampling occasion
sample size
rain
snow
Mace Head, Republic of Ireland
city of Gdansk, Poland
seashore in rem ote area
densely built area
53°20′N 09°54′W March 1996
10 La
10 L
54°22′N 18°35′E
J anuary 1996
J anuary 1997
J anuary 1997
a
5 L
snow
Lakes: J a¨ rnlunden,
Stora R a¨ ngen,
O¨ vre F o¨ llingen,
southern Sweden
Queen Maud’s Land, Antarctica
ice-covered lake in rural area 58°12′N 15°37′E
58°18′N 15°40′E
10 La
10 La
10 La
58°00′N 15°31′E
snow
snow and ice
72°51′S 04°21′E
68°10′N 18°40′E
Decem ber 1996-
February 1997
J anuary 1996
4 La
ice
the M a˚ rm a Glacier, northern Sweden ice and rocks in rem ote area
2.5 kgb
a
Snow volum e before thawing. b Before rem oval of the outer parts.
were enriched by liquid-liquid extraction. Prior to the
extraction, the original samples were concentrated 50-100
times by rotary evaporation at 40 °C, and the pH was adjusted
to 9.9 by addition of NaOH. The concentrates obtained were
then extracted with hexane and thereafter with diethyl ether.
The hexane and ether phases were evaporated to a small
ronitromethane was the main chlorination product in all
three of the analyzed extracts. A small peak representing
dichloronitromethane appeared when analyzing the last
extract.
Results
2
volume (100 µL) under a gentle stream of N . An internal
The chromatograms recorded in the chlorine channel of the
GC-AED system showed that only a few compounds in the
standard, 1-chlorooctane, was added to the final extracts.
Gas Chrom atographic Analyses. The GC-AED analyses
were carried out on a Hewlett-Packard 5890 GC system
equipped with an HP 5921A atomic emission detector.
Carbon, chlorine, bromine, nitrogen, and oxygen were
monitored at wavelengths of 496, 479, 478, 174, and 777 nm,
respectively. Gas chromatographic conditions: SGE BPX5
column (5% phenyl equivalent modified siloxane, 24 m ×
analyzed precipitation samples produced distinct peaks,
-1
measured as chlorine, at a detection level of 0.5-1 ng L
.
It had previously been reported that the chloroorganics
normally present at ppt levels are chlorinated nitrogen
compounds, chlorobenzenes, or chlorinated alkyl phosphates
(
25), hence we focused our attention on these three groups
of compounds. However, it should be also mentioned that
other chloroorganics, such as lindane and chlorofluorocar-
bons that have at least six carbon atoms, occasionally
occurred in the lower ppt range. In addition, a relatively
large amount of a chlorinated organometallic compound was
detected in one of the analyzed samples.
Chlorinated Nitrogen Com pounds. Identification. We
have previously reported that extracts obtained by stripping
enrichment of organics in rain and snow can contain
comparatively large amounts of a compound that is active
in the chlorine channel of a GC-AED system and produces
0
.32 mm, 0.5 µm phase thickness); carrier gas helium, at a
flow rate of 40 cm/ s; splitless injection, splitless time 1 min;
injector temperature 250 °C; temperature program, 40 °C for
5
min, 5 °C/ min, and 250 °C for 10 min.
The GC-MS analyses in full scan or selected ion monitoring
(
SIM) mode were performed on a Hewlett-Packard 6890 GC
system combined with an HP 5973 mass selective detector
and equipped with an HP-5MS column (5% phenyl methyl
silicone, 30 m × 0.25 mm, 0.25 µm phase thickness). Helium
was used as a carrier gas, at a flow rate of 40 cm/ s. Injection
conditions: split or pulsed splitless injection. All other
conditions were identical to those of the GC-AED analysis.
Blank Tests. All solvents were of analytical grade (Merck)
and distilled before use. The purity of the distilled solvents
was ascertained by GC-AED and GC-MS analyses of extracts
of solvents evaporated to a small volume. The risk of
contaminating the precipitation samples during preparation
and analysis was examined by analyzing ultrapure water
2
CHCl fragments upon GC-MS analysis (25). The present
study confirmed the occurrence of such a compound. The
GC-AED chromatogram in Figure 1A shows that, when a
wavelength (479 nm) characteristic of chlorine was moni-
tored, a significant peak appeared shortly after the solvent
peak (retention time 5.475 min). Examination of so-called
snapshots of emission spectra provided further evidence that
this peak was caused by a chlorine-containing compound.
The mass spectrum of the same compound (Figure 2) displays
a cluster of three peaks (m/ z 83, 85, and 87) that probably
(
Millipore Milli-Q) treated in the same manner as the real
samples.
Detection and Quantification. An internal standard (1-
represent the formation of CHCl
at m/ z 48 and 50 might be attributed to CHCl fragments, and
a peak at m/ z 30 suggests that NO or CH NH fragments
2
fragments. A pair of peaks
chlorooctane) was added to the final extracts to monitor both
retention times and the chromatographic response in each
run and to estimate concentrations of detected compounds.
Quantification was based on the response to chlorine
2
2
were formed. GC-AED chromatograms recorded at wave-
lengths characteristic of nitrogen (Figure 1B) and oxygen
(not shown) indicated that the investigated compound did
contain the former element and probably also the latter.
Together, the cited observations raised the question of
whether the unknown chloroorganic compound could be
(
wavelength 479 nm) in the AED system, and a detection
-
1
limit of 0.5-1 ng L was achieved for all the reference
substances. Tris(2-chloroethyl)phosphate was found in both
volatile and extractable fractions.
Synthesis of Di- and Trichloronitrom ethane. Chlori-
nated nitrom ethanes were synthesized by m ixing ni-
2 2
dichloronitromethane (CHCl NO ).
GC-MS analysis of reaction products of aqueous chlori-
nation of nitromethane showed that trichloronitromethane
was the main product and that dichloronitromethane was
also formed. Moreover, the unknown chloroorganic had
practically the same mass spectrum and retention time as
the synthesized dichloronitromethane (Figures 2-4).
Occurrence. Dichloronitromethane was the most abun-
dant volatile chloroorganic in all of the samples from Poland,
Sweden, and Ireland. The highest concentration (130 ng
-3
tromethane (CH
3
3
NO
2
;4 ×10 mol) and sodium hypochlorite
-
(
NaClO; 4 × 10 mol) in 50 mL of water. Immediately after
mixing, 10 mL of the solution was extracted with 2 × 2 mL
of CH Cl . After 30 min, an additional 10 mL of the reaction
2
2
-
3
mixture was extracted. Another portion of NaClO (2 × 10
mol) was then added to the remaining 30 mL of the reaction
mixture, and a third 10-mL portion of this mixture was
extracted with CH
GC-MS analysis of the CH
2
Cl
2
. The chromatograms obtained upon
Cl extracts showed that trichlo-
-
1
2
2
L ) was observed in a rainwater sample from Ireland,
3
9 3 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 32, NO. 24, 1998

FIGURE 1. Chromatograms recorded during GC-AED analysis of a stripping extract derived from snow collected in the city of Gdansk,
Poland. The two chromatograms show the response to chlorine (A) and nitrogen (B) measured at wavelengths of 479 and 174 nm, respectively.
FIGURE 2. Mass spectrum of the chloroorganic compound that was
most abundant in the stripping extract of a snow sample collected
in the city of Gdansk, Poland.
FIGURE 3. Mass spectrum of dichloronitromethane synthesized
through aqueous chlorination of nitromethane with sodium hy-
pochlorite.
whereas the Swedish snow samples contained an average of
-
1
1
0 ng L . The amount of dichloronitromethane was below
-
1
the detection level (<0.5 ng L ) in glacier ice from northern
Sweden and in the reference sample of snow from Antarctica,
and this compound was not detected in any of the blank
samples. Trichloronitromethane was not found in any of
the analyzed samples.
Chlorobenzenes. Identification. We have previously
reported that dichlorobenzene can occur in extracts obtained
by stripping of precipitation and that the isomer that was
most abundant in such extracts had the same retention time
as 1,4-dichlorobenzene (23, 25). In the present study, it was
noted that the chromatograms recorded in the chlorine
channel of the AED system often contained small peaks that
might represent other isomers of dichlorobenzene (Figure
1
A). GC-MS analysis in the full scan mode confirmed this
conjecture, and analysis of model compounds showed that
all three isomers, i.e. 1,2-, 1,3-, and 1,4-dichlorobenzene,
were present in stripping extracts of precipitation. The
presence of one isomer of tetrachlorobenzene was demon-
strated by GC-MS analysis in the full scan mode as well as
by monitoring of selected ions (m/ z 214, 216, and 218).
Analysis of model compounds showed that the detected
isomer was 1,2,4,5- or 1,2,3,5-tetrachlorobenzene. Pen-
tachlorobenzene was found in the analyzed extracts. How-
ever, this compound was also detected in the blank samples,
thus it was uncertain whether it had actually been present
in the original precipitation samples.
FIGURE 4. GC-MS-TIC (total ion current) chromatograms of extracts
obtained from aqueous chlorination of nitromethane (A) and from
Mace Head rainwater, Ireland (B).
in all analyzed samples. None of the chlorobenzenes were
found in any of the blank samples.
Chlorinated Alkyl Phosphates. Identification. It has
already been reported elsewhere (23-25) that tris(2-chlo-
roethyl)phosphate can occur in rain and snow. The GC-
AED runs in the present study showed that extracts obtained
by liquid-liquid extraction of precipitation normally con-
tained a chlorinated compound with a retention time that
was identical to that of tris(2-chloroethyl)phosphate (Figure
5). Moreover, the model compound and the organochlorine
detected in precipitation had almost identical mass spectra
in the GC-MS analysis in the full scan mode.
Occurrence. Chlorobenzenes were found to be ubiquitous
in precipitation. All of the analyzed samples contained the
three possible isomers of dichlorobenzene, and 1,4-dichlo-
robenzene was invariably present in higher concentrations
-
1
(
up to 5 ng L ) than 1,2- and 1,3-dichlorobenzene. The
Closer examination of the chromatogram A in Figure 5
and other chromatograms recorded in the chlorine channel
detected isomer of tetrachlorobenzene, was also present
VOL. 32, NO. 24, 1998 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 9 3 7

FIGURE 5. GC-AED chromatograms of organics enriched from snow by hexane extraction (A) followed by diethyl ether extraction (B).
The sample was collected on the ice of Lake J a1 rnlunden in southern Sweden. The response to chlorine in the AED system was monitored
at 479 nm.
2
-propyl)phosphate (CAS No. 13674-84-5). The mass spec-
trum of isomer II differed from that of isomer I in two
important respects: the ion cluster characteristic of CH Cl
losses at branching points was weaker, and a clearly visible
peak at m/ z 263 indicated R-cleavage and loss of CH CH Cl
or CH CHCl fragments from 3-chloro-1-propyl or 2-chloro-
-propyl structures, respectively. Comparison of major ions
2
2
2
3
1
and retention times with those reported in a study of
chloroalkyl phosphates in wastewater (28) indicated that
isomer II was the nonregistered compound bis(1-chloro-2-
propyl)(3-chloro-1-propyl)phosphate. However, the regis-
tered compound bis(1-chloro-2-propyl)(2-chloro-1-propyl)-
phosphate (CAS No. 76025-08-6) may also produce a similar
mass spectrum.
Occurrence. Tris(2-chloroethyl)phosphate was present in
all of the analyzed samples from the northern hemisphere.
Normally, it was also the most abundant chloroorganic
compound in the fraction of extractable organics. Interest-
-
1
ingly, the highest concentration (21 ng L ) was observed in
the sample of glacier ice from northern Sweden; in com-
parison, concentrations in the other samples ranged from 1
-
1
to 17 ng L . GC-AED analysis of hexane and ether extracts
of snow from Antarctica did not indicate the presence of
tris(2-chloroethyl)phosphate at that site. Furthermore, that
compound was not found in Milli-Q water that had been
evaporated and extracted in the same manner as the
precipitation samples. GC-MS-SIM analysis of the diethyl
ether extract of ice from the M a˚ rma Glacier confirmed the
results of the GC-AED analysis, whereas the GC-MS-SIM
response to the extract of Antarctica snow was below that of
Milli-Q water, which was used as a blank.
FIGURE 6. Mass spectra of two isomers of tris(chloropropyl)-
phosphate found in an ice sample from the Må rma Glacier in northern
Sweden.
of the AED system revealed the presence of two more
chlorinated compounds eluting just after tris(2-chloroethyl)-
phosphate. Scott and co-workers (24) have reported that
tris(chloropropyl)phosphates can occur in precipitation, and
GC-MS analysis in the full scan mode indicated that the
organochlorines detected in our samples were two isomers
of that compound (Figure 6). Standards of different isomers
of tris(chloropropyl)phosphate were not commercially avail-
able, and the GC-MS library did not provide any isomer-
specific information, hence identification of the two isomers
was based on general theories of mass spectrometry. As can
be seen in Figure 6, isomer I produced a relatively intense
ion cluster (m/ z 277, 279, and 281) that is characteristic of
GC-AED analysis showed that the hexane and ether
extracts of samples from Sweden contained both of the
investigated isomers of tris(chloropropyl)phosphate, whereas
rainwater from Ireland and snow from Gdansk, Poland,
contained only tris(1-chloro-2-propyl)phosphate. The high-
est concentrations of tris(1-chloro-2-propyl)phosphate and
bis(1-chloro-2-propyl)(3-chloro-1-propyl)phosphate or bis(1-
chloro-2-propyl)(2-chloro-1-propyl)phosphate were 3.0 and
1.5 ng L , respectively. The amount of tris(chloropropyl)-
phosphate in the reference sample of snow from Antarctica
and in the blank sample of Milli-Q water was below the
detection level of the GC-AED analysis. To further investigate
the presence of that compound at remote sites, hexane
extracts of Milli-Q water, snow from Antarctica and ice from
the Mårma Glacier were also analyzed by GC-MS-SIM. The
diagrams shown in Figure 7 confirmed that two isomers of
tris(chloropropyl)phosphate were present in the ice sample,
whereas the response to the extract of Antarctica snow was
almost identical to that of Milli-Q water.
-
1
losses of CH
structures. The absence of the ion m/ z 263, which is
characteristic of R-cleavage and loss of CH CH Cl or CH
2
Cl fragments at branching points in isopropyl
2
2
3
-
CHCl fragments, indicated strongly that all three of the side
carbon chains of isomer I were 1-chloro-2-propyl. Thus,
isomer I was identified as the flame retardant tris(1-chloro-
3
9 3 8
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 32, NO. 24, 1998

FIGURE 7. Formation of ions characteristic of tris(1-chloro-2-propyl)phosphate (A) and bis(1-chloro-2-propyl)(3-chloro-1-propyl)phosphate
or bis(1-chloro-2-propyl)(2-chloro-1-propyl) phosphate (B) when hexane extracts of Milli-Q water, ice from the M a˚ rma Glacier in Sweden
and surface snow from Antarctica were subjected to GC-MS-SIM analysis. The bars represent the area ratios of the ions m/z 277 and
279 to m/z 105, which was characteristic of the internal standard 1-chlorooctane.
FIGURE 8. GC-AED analysis of a diethyl ether extract of a snow sample collected in the city of Gdansk, Poland. The chromatograms show
the response in the tin channel (A) and the chlorine channel (B).
TABLE 2. Observations of Dichloronitromethane, Chlorobenzenes, and Chlorinated Alkyl Phosphates in Different Types of
a
Precipitation Samples
type of sample
snow
snow
rainwater
glacier ice
surficial snow
compound
southern Sweden Gdansk, Poland Mace Head, Ireland northern Sweden Antarctica
dichloronitrom ethane
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
-
+
+
+
+
+
+
+
-
-
+
+
+
+
+
+
+
-
+
+
+
+
-
-
-
1
1
1
,4-dichlorobenzene
,3-dichlorobenzene
,2-dichlorobenzene
tetrachlorobenzene
tris(2-chloroethyl)phosphate
tris(1-chloro-2-propyl)phosphate
bis(1-chloro-2-propyl)(3-chloro-1-propyl)phosphate or
bis(1-chloro-2-propyl)(2-chloro-1-propyl)phosphate
tributyltin chloride
-
+
-
-
-
a
Confirm ed presence in at least one sam ple is indicated by “+”, and “-” stands for not detected or inconclusive results.
Organom etallic Com pounds. The hexane and ether
extracts of a snow sample collected in the city of Gdansk
contained fairly large amounts of a chlorinated compound
that was absent in all of the other extracts. GC-AED runs in
the chlorine and tin channels, respectively, revealed the
presence of an organotin compound (Figure 8). GC-MS
analysis of the two extracts indicated that the detected
compound was tributyltin chloride, and this was subse-
quently confirmed by GC-MS and GC-AED analyses of a
standard solution of that compound.
rines that were most abundant in the investigated samples
of rain, snow, and ice. Almost all chloroorganics that were
regularly present in concentrations of at least 1 ng L were
identified.
-
1
The discovery that dichloronitromethane is very wide-
spread in precipitation is by far the most remarkable finding
of the present study. To the best of our knowledge, there are
no reports suggesting that this compound has been detected
in the ambient environment. Nevertheless, it is an indisput-
able fact that dichloronitromethane was present in all of our
precipitation samples collected in southern Sweden, Poland,
and Ireland. Furthermore, of all the volatile organochlorines
detected, dichloronitromethane was usually found in the
highest concentration. The comprehensive and unequivocal
information regarding elemental composition, the mass
spectrum, and retention time allowed the identification of
Sum m arizing Table. Table 2 illustrates the presence of
chloroorganics in different types of precipitation samples.
Discussion
Combined GC-AED and GC-MS proved to be an efficient
method to reveal the chemical structure of the organochlo-
VOL. 32, NO. 24, 1998 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 9 3 9

the chemical structure, and there were no indications
whatsoever of blank problems or contamination of samples.
The absence of dichloronitromethane at the two most
remote sampling sites indicates that anthropogenic sources
predominate. Apart from that, there are few clues as to
sources or precursors of the com pound. Dichloroni-
tromethane is not intentionally produced or manufactured
on a large scale. However, it was recently shown that both
mono- and dichloronitromethane can be formed during
pyrolysis of organic matter in the presence of reactive chlorine
Literature Cited
(
(
(
1) Strachan, W. M. J. Environ. Sci. Technol. 1985, 19, 677.
2) Strachan, W. M. J. Environ. Toxicol. Chem. 1988, 22, 871.
3) Schleyer, R.; Renner, I.; M u¨ hlhausen, D. WaBoLu-Hefte 1991,
5, 31.
(4) Hart, K. M.; Tremp, J.; Molnar, E.; Geiger, W. Water Air Soil
Pollut. 1993, 68, 91.
(
5) Hoff, R. M.; Strachan, W. M. J.; Sweet, C. W.; Chan, C. H.;
Shackleton, M.; Bidleman, T. F.; Brice, K. A.; Burniston, D. A.;
Cussion, S.; Gatz, D. F.; Harlin, K.; Schroeder, W. H. Atmos.
Environ. 1996, 30, 3505.
(
29). Hence, it cannot be excluded that uncontrolled
(
6) Chevreuil, M.; Chesterikoff, A.; L e´ tolle, R.; Granier, L. Water Air
Soil Pollut. 1989, 43, 73.
formation of dichloronitromethane can take place, for
example during incineration processes. Formation of dichlo-
ronitromethane during chlorine disinfection of drinking water
or wastewater is possible but seems to be negligible (30).
The trichlorinated analogue of nitromethane (chloropi-
crin) is used as a fumigant in agriculture (31), and the same
compound can also be a byproduct of aqueous chlorination
of organic matter (32, 33). However, there is no obvious
connection between the occurrence of di- and trichloroni-
tromethane in the environment. It is not very plausible that
trichloronitromethane is transformed to dichloronitromethane
in the atmosphere (34, 35), and it is difficult to imagine that
microbial dechlorination of trichloronitromethane under
anaerobic conditions (36) could render dichloronitromethane
the most abundant volatile chloroorganic in precipitation.
Haloalkyl phosphates are used as flame retardants in
polymers (37, 38), and the occurrence of such compounds
in rainwater samples has previously been noted by our study
team (23, 25) and by Scott and co-workers (24), who analyzed
samples collected in urban areas close to Lake Ontario in
Canada. In addition, Aston and co-workers (39) found such
compounds in pine needles in the Sierra Nevada Mountains
in the United States. The results of the present study are
novel in two respects: the structures of the two dominating
isomers of tris(chloropropyl)phosphate are suggested; the
presence of haloalkyl phosphates in the sample of glacier ice
from northern Sweden strongly indicates that such com-
pounds are subjected to long-range airborne transport. When
planning this investigation, we assumed that any chloroor-
ganics present in old glacier must have natural sources,
because of the age of the ice sample (26). However, detection
of lindane in the analyzed ice sample suggested that recent
precipitation had contaminated older layers of ice. The
concentrations of haloalkyl phosphates in surface snow from
Antarctica were below or slightly above the limit of detection.
Dichlorobenzenes are used as fumigants and sanitizers
and as intermediates in the manufacture of dyestuffs,
insecticides, and pharmaceuticals (30). The presence of 1,2-,
(7) Murray, M. W.; Andren, A. W. Atmos. Environ. 1992, 26A, 883.
(8) Galassi, S.; Gosso, E.; Tartari, G. Chemosphere 1993, 27, 2287.
(9) Chan, C. H.; Bruce, G.; Harrison, B J. Great Lakes Res. 1994, 20,
5
46.
(
(
10) Poster, D. L.; Baker, J. E. Environ. Sci. Technol. 1996, 30, 341.
11) Tashiro, C.; Clement, R. E.; Lusis, M.; Orr, D.; Reid, N.
Chemosphere 1989, 18, 777.
12) Koester, C. J.; Hites, R. A. Environ. Sci. Technol. 1992, 26, 1375.
13) Paasivirta, J.; Knuutila, M.; Paukku, R.; Herve, S. Chemosphere
1985, 14, 1741.
(
(
(14) Atlas, E.; Giam, C. S. Science 1981, 211, 163.
(15) Brun, G. L.; Howell, G. D.; O’Neill, H. J. Environ. Sci. Technol.
1
991, 25, 1249.
(
(
16) Class, T.; Ballschmiter, K. Fresenius’ Z. Anal. Chem. 1986, 325,
1
.
17) Jung, W. J.; Fujita, M.; Sohn, D. H. Jpn. J. Toxicol. Environ. Health
992, 38, 490.
1
(
(
18) Frank, H. Ambio 1991, 20, 13.
19) Frank H.; Renschen D.; Klein A.; Scholl H. J. High Resol.
Chromatogr. 1995, 18, 83.
(20) Reimann, S.; Grob, K.; Frank, H. Environ. Sci. Technol. 1996, 30,
340.
2
(
(
(
21) Zehavi, D.; Seiber, J. N. Anal. Chem. 1996, 68, 3450.
22) Juuti, S. Ph.D. Dissertation, University of Kuopio, Kuopio, 1997.
23) Laniewski, K.; Bor e´ n, H.; Grimvall, A.; Jonsson, S.; von Sydow,
L. In Naturally-Produced Organohalogens; Grimvall, A., de Leer,
E. W. B., Eds.; Kluwer Academic Publishers: Dordrecht, 1995;
Environment & Chemistry Vol. 1, pp 113-129.
(24) Scott, B. F.; Sverko, E.; Maguire, R. J. Water Qual. Res. J. Can.
1996, 31, 341.
(25) Laniewski, K.; Bor e´ n, H.; Grimvall, A. Accepted for publication
in Chemosphere.
(
(
(
26) Holmlund, P.; N a¨ slund, J.-O.; Richardson, C. Geogr. Ann. 1996,
8A (2-3), 147.
7
27) Bor e´ n, H.; Grimvall, A.; Palmborg, J.; S a¨ venhed, R.; Wigilius, B.
J. Chromatogr. 1985, 348, 67.
28) Thruston, Jr., A. D.; Richardson, S. D.; McGuire, J. M.; Collette
T. W. J. Am. Soc. Mass Spectrom. 1991, 2, 419.
(29) Laniewski, K.; Bor e´ n, H.; Grimvall, A. J. Chromatogr. Accepted
for publication.
(
(
30) Fawell, J. K.; Hunt, S. Environmental Toxicology: Organic
Pollutants; Ellis Horwood Limited: Chichester, 1988.
31) Sassaman, J. F.; Jacobs, M. M.; Kelley, J. M. Pesticide background
statements. Volume 2. Fungicides and fumigants; Agriculture
Handbook-661; United States Department of Agriculture: Wash-
ington, DC, 1986; 620 pp.
1
,3-, and 1,4-dichlorobenzene in rain and atmosphere from
urban and suburban areas was documented more than a
decade ago (40). The results of the present study showed
that all three isomers of dichlorobenzene can also be found
in precipitation collected at remote sites. Tetrachloroben-
zene is an intermediate in the synthesis of 2,4,5-trichlo-
rophenol and a metabolite of hexachlorobenzene and lindane
(32) Thibaud, H.; De Laat, J.; Dore, M. Water Supply 1988, 6, 243.
(33) Williams, D. T.; LeBel, G. L.; Benoit, F. M. Chemosphere 1997,
3
4, 299.
(
(
34) Fazekas, G. B.; Takacs, G. A. J. Photochem. 1983, 21, 9.
35) Carter, W. P. L.; Luo, D.; Malkina, I. L. Atmos. Environ. 1997,
(
30). We are apparently the first to report that one of the
isomers of tetrachlorobenzene, i.e., 1,2,4,5- or 1,2,3,5-
tetrachlorobenzene, is widespread in rain and snow. It is
most probably 1,2,4,5-tetrachlorobenzne, but at present stage
the exact identity is not known.
3
1, 1425.
(
36) Wilhelm, S. N.; Shepler, K.; Lawrence, L. J.; Lee, H. ASC Symp.
Ser. 1997, 652, 79-93.
37) Hutzinger, O.; Sundstr o¨ m, G. Chemosphere 1976, 1, 3.
(
Organotin compounds are used often as antifouling agents
for ships, but we did not find any evidence that such
substances are prevalent in precipitation. Tributyltin chloride
was detected in only one of the samples, and that particular
snow sample had been collected in the vicinity of a shipyard.
(38) Green, J. J. Fire Sci. 1992, 10, 470.
(39) Aston, L. S.; Noda, J.; Seiber, J. N.; Reece, C. A. Bull. Environ.
Contam. Toxicol. 1996, 57, 859.
(
40) Pankow, J. F.; Isabelle, L. M.; Asher, W. E. Environ. Sci. Technol.
984, 18, 310.
1
Received for review May 12, 1998. Accepted September 11,
99819981998.
Acknowledgments
The authors are grateful for financial support from EuroChlor
1
and the Swedish Environmental Protection Agency.
ES980480I
3
9 4 0
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 32, NO. 24, 1998