RESEARCH T R 1 A N G L E IN S T I 'T U T:E
DEVELOPMENT OF ANALYTICAL TECHNIQUES FOR
MEASURING AMBIENT ATMOSPHERIC CARCINOGENIC VAPORS
EPA Contract No. 68-02-1228
RTI Project No. 31U-885
Monthly Technical Progress Report
No. 6
Period: December 1 - December 31, 1975
E. D. Pellizzari, Project^JLgader
Date: January 23, 1976^
Prepared for the Environmental Protection Agency, Research Triangle Park,
North Carolina 27711
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27709
-------
TABLE OF CONTENTS
!£-£
Introduction and Rationale 1
Methods of Procedure 2
Sampling Methods 2
Gas-Liquid Chromatography/Mass Spectrcinetry Computer
Analysis of Ambient Air Samples 6
Artifact Experiments . 10
Results and Discussion 12
References 31
-------
INTRODUCTION AND RATIONALE
On October 3, 1975, personnel from the Chemistry and Life Sciences
group at RTI attended a meeting held by Dr. A. P. Altschuller at ESRL,
Environmental Protection Agency, Research Triangle Park. In response
to an OAQPS request, RTI was asked to participate in assessing the
nitrosamine' problem in the Baltimore, MD area. The effort consisted of
obtaining and comparing data on any ambient nitrosamines as measured by
gc/ms methods (RTI) and by thermal energy analysis (Thermal Electron
Corp.). Samples were to be collected primarily in parallel between
these two contractors. The Research Triangle Institute proceeded to
undertake a sampling program during the week of October 13-17 and during
the period of November 19-25, 1975 in the Baltimore area, specifically
near and on the FMC property.
During this sampling period, the Environmental Protection Agency pro-
vided a meterologist to assist the RTI sampling team in obtaining favorable
locations with respect to downwind locations from the FMC property.
The primary objectives were to identify unequivocally the presence of
N-nitrosodimethylamine in ambient air and estimate the levels in this area.
-------
METHODS OF PROCEDURE
Sampling Methods
The sampling procedure employed for this study has been previously
described (1,2) which consisted of concentrating nitrosamines and other
organic vapors on a 1.5 x 6.0 cm bed of Tenax GC (35/60) in a glass
cartridge. All sampling cartridges were preconditioned by heating to
275°C for a period of 20 min under a helium purge of 20-30 ml/min.
After cooling in precleaned Corex centrifuge tubes, the containers were
sealed to prevent contamination of the cartridge. Sampling cartridges
I
prepared in this .manner were carried by automobile or air freight to
the sampling site; 2-3 cartridges were designated as blanks to determine
whether any of the cartridges might have been contaminated by the pack-
I/
ing and transportation procedure. Cartridges containing known quantities
/
(100 and 300 ng) of N-nitrosodimethylamine (DMN) were prepared and carried
to and from the field, stored and the percent recoveries were determined.
Ambient air samples were collected with Nutech Model 221-A AC/DC
portable samplers (2). In general, a sampling rate of 1 £/min/cartridge
was used throughout this study.
The sampling locations are shown in Figures 1 and 2 and the sampling
protocol used in this study is given in Table 1.
Breakthrough volumes for DMN were determined as previously described
(2). This consisted of injecting DMN vapor onto a gas chromatographic
column packed with the sorbent Tenax GC and determining the elution volume
at a series of decreasing temperatures. By plotting the log of the elu-
tion volume vs temperature, and extrapolating to ambient temperatures,
(50-90°F) the breakthrough volumes for DMN were determined. These values
-------
Figure l. EAST BROOKLYN,
BALTIMORE, MARYLAND
CHESSIE
COAL PIERS
CURTIS BAY
SCALE: ONE INCH =0.5 miles
-------
Figure 2. Map of FMC plant.
1.
; 2.
3.
4.
' 6.
7.
9.
10.
11.
11A
12.
13A
13B
13C
14A
14 B
^ E"T!-!fP% "£ - ••'•( Ir11^,. A-rr-<-<
L^^^Lnr^l?^5^ HnSfe^ /
=^D a t
*-""^"^
. 5T5.# -Vr=^^//,.3
I"0 LW e 'S'«/^7O
ZwkVL !' A ///_//
X= INCINERATOR STACK
SS= SAMPLING SITE
Power House
Stack
Dcpon Bldg.
Refined Chemicals Bldg.
Ethion & Butoxide Bldg.
P2S8 Conveyor & Storage Bldg
Diallyl Phthalate Bldg.
Packaging Bldg.
General Shop & Storeroom
Maintenance Office Trailer
Fire Pump House
Warehouse
Shop
& Shipping Dept.
Carpenter
Warehouse
Storage
Storage
.14C. Drum Recovery
Dapon Manufacturing
Warehouse
16. Warehouse
18. Lab
19. Main Office Bldg.
20. Gate House
25. Booster Pump House
26. Anhydrous Ammonia
28. Dinazinc Control Rm
29. Lab
30. CO Compressor Shed
31. Production Office
34. Pyrethrum Refining
38. Hydrcgenators
39. Nitroso
40. Purification & Reco
very
41.
43.
A4.
44.
47.
48.
49.
51.
52.
53.
54.
55.
59.
61.
62.
63.
Valve House
PuT.p House
Storage
Storage
Pump House
Chloroformates
Scdiuir. Condensa-
tions
Sodium Storage
Tedion Bldg.
Lockers
Lockers
Cafeteria
Paint Storage
Maint. & Prod.
Central Eng.
J. F. Wilson
Trailer
69.
70.
71.
72.
73.
74,
75.
76,
77,
73.
79,
30.
31.
Hydrogenation
7-OH Production
Control Lab
Incinerator Area
Elec . Control Rin.
7-OK Warehouse
Effluent Control
Rr.;.
>'otor Control &
Locker Km.
7-OH Shop
Cl2 Control Rm.
Fire Pur.-.p House
Locker Km.
Compressor Bldg.
82.
86.
37.
83.
89.
91.
92.
93.
95.
98.
CT.
TF.
£3.
Locker P.TJ.
Inert Gas Generator
Locker Rn.
Calgon Carbon Treat-
ment Bldg.
Shop
Resin Seai-Works
Plant
Pilot Plant
Elec. Control Ra.
Orvis Pilot Plant
Compressor Eldg.
Cooling Tower
Tank Farm
Sub Station
-------
Table 1. SAMPLING PROTOCOL FOR BALTIMORE STUDY
Site Cartridge Replicates
FMCS
FMC
FMC
FMC
FMC
FMC
FMC
PS
PS
CP
CP
1
5
9
13
17
1
5
1
5
1
5
_
-
-
-
-
_
-
-
-
T
a
i.
6
10
14
IS
2
6
2
6
2
6
- 3 -
- 7 —
- 11
- 15
- 19
- 3 -
— 7 —
- 3 -
- 7 -
- 3 -
- 7 -
-m/e
4
8
- 12
- 16
- 20
4
8
4
8
4
8
74
11
3
7
11
3
11
3
10
2
9
p
Time Period
:00
:00
:00
:00
:00
:00
:00
:00
:00
:56
:10
AM
PM
PM
PM
AM
AM
PM
AM
PM
AM
PM
- 2:50
- 6:50
- 10:50
-2:50
- 6:50
- 2:50
- 6:50
- 1:50
- 5:50
- 1:46
- 6:00
PM
PM
PM
AM
AM
PM
PM
PM
PM
PM
PM
Date
10/14/75
10/14/75
10/14/75
10/14 - 15/75
10/15/75
10/15/75
10/15/75
10/16/75
10/16/75
10/17/75
10/17/75
(Quant.)
GC-MS-COMP (Qual
FMC (W of Dimazine Destructor)
Town of Fairfield
Near Memirac Corp .
FMC (SW of Dimazine Destructor)
ir
II
it
Northbridge and Convery St.
"
1-2
1-2
1 - 2
1~* - 2
1° - 2
1-2
1^-2
1. - 2
1-2
m/e 74 (Quant
•)
•)
2
3
8
11
1
6
6
1
3
:00
:45
:20
:50
:55
:35
:35
:4S
:59
PM
PM
PM
AM
PM
PM
PM
PM
PM
- 4:00
- 5:45
PM
PM
- 10:20 PM
- 1:50
- 3:55
- 8:35
- 8:35
- 3:48
- 5:59
PM
PM
PM
PM
PM
PM
11/19/75
11/20/75
11/20/75
11/24/75
11/24/75
11/24/75
11/24/75
11/25/75
11/25/75
Parking lot
Part of Artifact Experiment
-------
wer-: used for quantitative collection of DMN, if present, during field
sampling. The breakthrough volumes are shown in Table 2.
Gas-L-iquid Chromatography/Mass Spectrometry Computer Analysis of Ambient
Air Samples
The instrumental system (glc-ms-comp) used for the qualitative and
quantitative analyses of nitrosamines and other ambient air pollutants
and the inlet manifold used for recovering vapors trapped on Tenax GC
cartridge samplers were as described elsewhere (1,2). The descrbed
vapors were resolved by capillar}7 gas-liquid chromatography and mass
cracking patterns were automatically and continuously obtained through-
out the glc run with a Varian CH-7 gas chromatograph/irsass spectrometer.
The cracking patterns and retention time data were accumulated on a
magnetic tape deck and further processed by an on-line Varian 620i com-
puter. Computer programs (KOSB) were employed which converted the
acquired spectra into a sequential series of mass spectra and uere cor-
related to peak retention time on a total ion current plot. Data output
from the 6201 computer was provided in two forms: (1) a teletype listing
which contained the mass spectrum number, number of peaks in the cracking
pattern, total maximum and minimum m/e peak intensity and standard devia-
tion from calibration m/e and (2) an electrostatic plot of total ion
current plots and/or normalized mass spectral. Single ion plots (e^.g_.
m/e 74) were obtained as an analog output.
The operating parameters for the glc-ms-comp system for analysis
of samples collected on glass cartridges from the Baltimore, MD area are
shown in Table 3. Ambient air samples were analyzed en 55, 80 or 100
meter glass SCOT columns coated with either OV-101, OV-225 or DECS sta-
tionary phase, respectively. The desorption of ambient air pollutants
-------
Table 2. BREAKTHROUGH VOL FOR DMN
?F
50
55
60
65
70
75
80
85
90
95
100
105
110
2.2 g /cartridge
(1.5 x 6.0)
liters
385
332
280
242
204
224
163
156
148
127
107
93
79
2.93 g/cartridge
(1.5 x 8.0)
liters
513
493
373
322
272
244
217
207
197
170
143
123
103
-------
Table 3. OPERATING PARAMETERS FOR GLC-MS-COMP SYSTEM
Parameter
Setting
Inlet-manifold
desorption chamber
valve
capillary trap - minimum
maximum
thermal desorption time ""
GLC
OV-101 glass SCOT (100M)
OV-225 glass (80M)
DECS glass SCOT (55M)
carrier (He) flow
MS
single stage glass jet separator
ion source vacuum
filament current
multiplier
scan rate, automatic-cyclic
scan range
265°-270°
175°
-195°C
4175°C
~4 min
30-225°C, 4°C/min
80-2106C, 4°C/min
70-205°C, 4°C/min
1.5 ml/min
200° C
~2 x 10~ torr
300 yA
5.5
1 sec/decade
m/e 20 -»• 300
-------
including nitrosamines from the Tenax cartridge samplers x^as achieved
at 265-270°C. A single stage glass jet separator interfaced the SCOT
capillary columns to the mass spectrometer and was maintained at 200°C.
Identification of resolved components was achieved by comparing
the mass cracking pattern of the unknown mass spectra to an 8 major peak
index of mass spectra (1,2). In several cases the identification was
confirmed by comparison with authentic compounds of the mass spectrum
and the elution temperature on two different columns. Particular note
was made of the relationship of the boiling point of the identified
compound to its elution temperature and to its order of elution of cons-
tituent in homologous series since the OV-101 SCOT capillary column
separates primarily on the basis of boiling point.
For quantitative analysis, the mass spectrometer was set in the
single ion plotting mode whereby the selected master charge ratio was
recorded on analog output. Quantitation of K-nitrosodimethylamine (by
monitoring only m/e 74) was conducted on a 55 meter glass SCOT capillary
coated with DECS.
A standard curve for the response of the mass spectrometer set at
m/e 74 vs the concentration of DMN was prepared by introducing known
quantities of DMN vapor. Synthetic air-DMN vapor was synthesized and
specified quantities collected on Tenax GC cartridges. By thermally
desorbing the cartridge and monitoring the m/e 74, responses vs concen-
tration were obtained. Cartridge samplers containing unknown concentra-
tions of DMN were analyzed by monitoring the m/e 74 ion and obtaining the
quantity/cartridge from the standard curve. Based on the volume of air
sampled and the breakthrough volume for DMN, the concentration of N-
nitrosodimethylamine was calculated in yg/m of ambient air.
-------
10
Artifact Experiments
The possibility of formation of DMN on the Tenax sampling cartridge
was investigated. Since the Tenax cartridge sampler would concentrate
the dimethylamine if present in ambient air (breakthrough volume approxi-
mately 35 H at 25°C), it is conceivable that in the presence of NO > the
dimethylamine could become nitrosated to form DMN. Two types of experi-
ments were conducted to either support or refute this possibility. The
first set of experiments were conducted under laboratory conditions.
Synthetic mixtures of NO, water vapor, and air were prepared at known con-
centrations of 1, 10 and 250 parts-per-billion (in NO) in 7 £ glass bulbs,
Tenax cartridges were loaded with approximately 5 parts-per-million of
dimethylamine and the NO __ air vapor mixture was drax-m through the cart-
A
ridge. Cartridges were subsequently analyzed by single ion monitoring
of the in/e 74. In another set of experiments, NO (~2 x 10 mole/min,
1 ppm) was introduced into a stream of laboratory air which was pulled
across a permeation tube containing dimethylamine at 1 £/min and into
the Tenax cartridge. The sampling period was approximately 30 min. The
dimethylamine permeation tube permeated at a rate of ~9 x 1C mole/min
(~4.5 ppm). Thus, the amine to NO ratio was ~45:1.
Experiments were also conducted during field sampling to demonstrate
whether an enhancement in the collection of N-nitrosodimethylamine could
occur above background when Tenax cartridges were either pre-loaded with
dimethylamine or when field air was drawn across a permeation tube con-
taining dimethylamine and into the Tenax cartridge. The permeation tube
containing dimethylamine (~2.4 x 10 g/min) was placed in front of the
glass fiber filter on the Tenax cartridge. Therefore in these experi-
ments, the dimethylamine was introduced prior and after the glass fiber
-------
11
filter. All experiments were conducted in at least triplicate. A total
of ~120 £ of ambient air was sampled in all cases.
Further experiments were conducted to also demonstrate whether sto-
rage of cartridges which had been exposed to dimethylamine and NO might
J^
gradually produce DMN. Replicate cartridges were stored at room tempera-
ture and at -15°C for periods of 1, 2 and 3 weeks.
-------
12
RESULTS AND DISCUSSION
As shown in Figure 1, the sampling sites selected were (1) on the
Fl-IC property, (2) at the Patapsco Sewage Treatment Plant, and (3) the
Chessie Coal Piers. During the ambient air sampling on the F>!C pro-
perty, the samplers were located either in the parking lot area or near
the dimazine thermal destructor as shown in Figure 2.
The sampling protocol for the Baltimore study was depicted in
Table 1. During the week of October 14, the sampling protocol included
the collection of quadruplicate cartridges during each time period.
One cartridge from the series of replicates were committed to gas chroraa-
tography/mass spectrometer computer for qualitative analyses. Another
series of cartridges were designated for quantitation. The remaining
cartridges served either as duplicates for qualitative or quantitative
analysis. During the period from November 19 to November 25, the samp-
ling protocol was slightly modified to allow only the collection of
duplicate cartridges as well as inclusion of the field artifact studies
(see Table 1). For quantitative analysis in all cases the calculation
of the amount of N-nitrosodimethylamine in ambient air was based on the
breakthrough volume and the ambient air temperature at the time of field
sampling.
Figure 3 depicts the profile of ambient air pollutants in a sample
taken during the day in the FMC parking lot area. The volume of air col-
lected represents 75 I. A 100 meter glass SCOT column containing OV-101
stationary phase was used to effect this separation. The remaining
conditions for glc-ms-comp analysis is shown in Table 3. In contrast to
Figure 3, Figure 4 represents a profile of ambient air pollutants in a
sample taken during the subsequent evening at the same location in the
-------
20
68
I »\" I
l\ tz\\ '420
4jyd
116
TEMPERATURE (°c)
164
212
12
15
18
21
24
27
30 33
TIME (min)
36
39
43
51
54
57
60
Figure 3. Profile of ambient air pollutants in a sample ts.ken during the day on FtfC parking lot.
A 100M glass SCOT column containing OV-101 was used, see Table 3 for GLC/MS/COMP conditions.
u>
-------
23
20
0~
28 32
68
16
20
36
n
54
i
50
49 33 64 69
48
'.2
{43
f. ^
«45
US
;i
%°stt
85
••':! v\n
\ r-ir i w
115
TEMPERATURE.(°c)
164
_i_
'S5
3-;: 102
'^•';i-!C%.^
10'
212
L_
IOS
2:0
111 112
12
15 18
21
•& 2T
TO 33
TIME (min)
3S 39
45
57
SO 63
Figure 4. Profile of ambient air pollutants in a sample-taken during the night on FMC parking lot,
A 100M OV-101 glass SCOT column was used, see Table 3 for GLC/MS/COMP conditions.
-------
15
FMC parking lot. Of particular interest here is the comparison of the
overall quantity of ambient air pollutants which are significantly
greater during the night sampling period as compared to the day. This
phenomenon is frequently observed in our other studies where day and
night sampling have been compared in geographical areac such as Los
Angeles, CA, Houston, TX, St. Louis, MO, Denver, CO and the Kanawha
Valley in W. VA.
Figure 5 depicts a profile of ambient air pollutants in a sample
taken during the daylight hours near the Patapsco Sewage Treatment Plant.
This sampling location is upwind from the FMC Corporation plant but down-
wind from the Sewage Treatment plant. The profile depicted in Figures 3
and 4 were taken downwind from the FMC Corporation.
Many of the pollutants resolved in Figures 3-5 have been identified,
however only a partial listing will be discussed later.
A profile of ambient air pollutants in a sample taken during the
daylight hours on the FMC parking lot which is identical to the samples.
taken as shown in Figure 3, is given in Figure 6. The chromatographic
separation of constituents was achieved in this case on a semi-polar
stationary phase, OV-225, coated in a 80M glass SCOT column. Semi-polar
and polar phases are not well suited for resolution of non-polar pollu-
tants. However, the resolution of semi-polar and polar constituents are
significantly better. Figure. 7 depicts the profile of ambient air pol-
lutants in a sample taken again during the daylight in the FMC parking
lot. In this case, the chromatographic column consisted of a 55M DECS
glass SCOT column.
The mass cracking pattern for N-nitrosodimethylamine as reported
in the literature is shown in Figure 8. Upon electron impact, the
-------
20
6-
26 30
68
JL
40
11
39,.
• n
58
74
44
43-
V
K
r.
70
G3
a 68
•9 67
63
75
70
77
70
U 62\'S TO 6C
-,fe S;i*j'& ,V'
4?f Jiffli'1'™/^
.giMyuy ^^".v
JJ1
TEMPERATURE (°c)
116
164
212
i
98
73
100
ICI
i02
230
103
106
12 15 18 21 24 27
30 33
TIME (min)
36
42 45
5!
54 57
6O
l
63
Figure 5. Profile of ambient air pollutants in a sample taken during the day near Patapsco Sewage Plant.
A 100M OV-101 glass SCOT column was used, see Table 3 for GLC/MS/COMP conditions.
-------
lOOp
90-
80-
UJ
g TO
0.
J
o
_)
-------
UJ
lOOp
90-
80-
o:
5 TO
H 60
ui
!r 50
O
z 40h
O
_l 30"
P 20
10
air
DMN
Figure 7. Profile of ambient air pollutants in a sample taken during the day on FMC
parking lot. A 55M DEGS glass SCOT column was used, see Table 3 for CLC/MS/COMP
conditions.
00
-------
19
Figure 8. Mass cracking pattern for N-nitrosodimethylamine
CH
CH
3X b
xK-NO
/
4
'N-N=0
m/e 74
-H, UNO
CH2 = N = CH2
m/e 42
-------
20
N-nitroscdimethylamine forms a radical ion with an m/e of 74. This ion
represents the most intense ion in the mass spectrum. Furthermore, the
formation of the ion m/e of 42 is observed which represents ~68% of the
base peak.
Figure 9 depicts the mass spectra of an authentic sample of N-nitroso-
dimethylaniine which was chromatographed on the 100M glass SCOT coated with
OV-225. Also shown in. Figure 9 is an ambient air sample taken at the FMC
parking lot which represents a mass spectrum from Figure 6. Comparison
of this mass spectrum with the one for the. authentic N-nitrosodirnethyl-
ainine indicates that the unknown peak in Figure 6 is indeed N-nitro.^o-
dimethylarair.e. The mass spectra are essentially superimposablc. The
retention time of the authentic N-nitrosodimethylamine was identical to
that of the unknown peak in Figure 6. ^
/
Likewise, Figure 10 depicts the mass spectrum of a peak in Figure 7
which subsequently has been identified as K-nitrosodimethylamine. This
mass spectra was obtained on the 55M glass SCOT DECS column. Furthermore,
the retention times of the authentic N-nitrosodimethylanine standard was
identical to that of the unknown peak in Figure 7.
A partial listing of the ambient air pollutants which have been ten-
tatively identified in samples of ambient air taken in the FMC parking lot
is given in Table 4. As it can be readily seen, many chlorinated hydro-
carbons have been detected. The compounds which have been marked with an
asterick represents air pollutants which have been identified for the
first time. The remaining constituents have been observed fequently in
other ambient air samples taken in areas such as Los Angeles, CA, Hous-
ton, TX, St. Louis, MO, Denver, CO and the Kanawha Valley, W VA. Of
particular interest in this listing are the dimethylvinylchloride. The
-------
JDMN
IOOMGSCOT OV225
70-225 ec
AFMC4
IOOMGSCOT OV225
70-225°c
too
90
(PERCENT)
-j co
0 0
£ 6°
i»
UJ
p 30
_i
CC 2°
10
-
-
-
•
-
-
-
.
1
10 2
0
|
0
J
3
,|
"J'"t
0 40
42
i illll
J
m
Hhjlli;
1
6
0
/n —
Ililji1"
0 7
ill
J
j*
m
in
8
«•
/e '
iiiiiiii:
0 9
r^
IlijjliU
0
1C
1
X)
o i:
io i:
"'1"'j""l'"'
JO 140
!
^o (!
K
>
3 2
I
A
0 c
!
Ji i
i
i *
-0 4C
42
1
5
• M.' . M .
60 70
0
/Li.
n/e 74
i
i iji I i 'Mi
SO SO CO 120 IZO 140
100
J*5. '
Figure 9. Mass spectra of N-nitrosodimethylamine. AFKC4 represents mass spectrs. from Figure 6.
-------
100
90
80
co 70
LU
t 60
AFMC5
LU
u
cc
50
30
20
10
m/e 74
60MGSCOT DECS
I .P.. I.;.. I,
Illijilll 111111111 Hil IIII Illllllii lllli.il!
II II
10 20 30 40
D 5
m
'"T"|""i""|""l""|""!""|""!""
GO 70 80 90
0 l(
/o . , It*
X!
!
0
1
!
£> \:
"T
50 1^
K)
i;
1 |
ISO 170
30
Figure 10. Mass spectrum of N--nitrcsodimethylamine, see Figure 7
for chromatographic profile.
-------
23
Table 4. PARTIAL LISTING OF AMBIENT AIR POLLUTANTS
TENTATIVELY IDENTIFIED IN SAMPLES
FROM FMC PARKING LOT
Compound Chromatographic Peak No.a
*methylene chloride 2
*chloroform 5
l-chloro-2-methylpropene 6
3-chloro-2-methylpropene 7
*l,l,l-trichloroethane 9
*carbon tetrachloride 11
methylene dibromide 15
*trichloroethyler.e 15B
l-chloro-2-brorr.opropane 19
2,3-dichlorobutane 21
chlorobutens isomer 2,2
*tetrachloroethylene 27
dichlorobutene isomer 28
2-chloro-l,3-butadiene 29
*chlorobenzene 30
dibromochlorofluoromethane 33
1,1-dichloropropene + trichlorobutane 40
*in-dichlorobenzene 46
*^-dichlorobenzene 52
a
See Fig. 3 for corresponding profile.
These pollutants have been also found in many other areas of the country,
-------
24
unequivocal identification of many of these compounds awaits the com-
parison of the mass spectra and the retention time of authentic com-
pounds to those listed in Table 4.
The standard curve for N-nitrosocliinethylamine generated on the gas
chromatograph/mass spectrometer during single ion monitoring of m/e Ik
is given in Figure 11. Because of the concentrations anticipated as a
result of the qualitative analysis, the standard curve was constructed
from 10-3,000 ng. This would represent the linear concentration range
which is present on a cartridge sampler. The detection limit observed
for the m/e 74 ion is ~.5 parts-per-trillion when the ambient air
temperature during field sampling is 85°F. Thus, the minimum amount
which can be quantitated is 5 parts-per-trillion.
Single ion plots for N-nitrosodimethylamine of replicate cartridges
are shown in Figure 12 and 13. It is evident from these two figures that
the replicate cartridge sampling technique is highly reproducible. All
single ion monitoring for quantitative analysis was conducted on the
55M glass SCOT capillary coated with DECS.
Using the described procedures for quantiting N-nitrosodircethylamine,
the concentration of the nitrosodimethylamine in ambient air was deter-
mined in the immediate area of the FMC plant in Baltimore, MD. Table 5
depicts these results. During the week of October 14, the highest values
observed were ~13.4 ug/m in the late afternoon. Only trace quantities
were observed upwind from the FMC plant at the Patapsco Sewage Treatment
Plant. Significant quantities were also measured at ~l/2 mile downwind
from the FMC Corporation at the Chessie Coal Piers.
-------
25
3OOr-
200-
20O 400 600 600 IOOO
"9
2000 3OOO
Figure 11. Standard cruve for N-nitrosodimethylamine
-------
10.0
9.0
ao
7.0
o
cc
LJ
5.0
4.0
3.0
2.C
xl
x3
2S min
Figure 12. Single ion plot (m/e 74) for N-nitrosodimethylamine, see Table 3 for GLC/MS/CCMP
conditions.
-------
IQOr
9.0-
ao-
7.0
CO
cc
UJ
p 5.0
<
a;
2 4.0
3.C
2.0
1.0
xl
x3
26m in.
Figure 13. Single ion plot (m/e 74) for N-nitrosodimethylamine (replicate cartridge of Fig. 12),
see Table 3 for GLC/MS/COMP conditions.
Kl
—I
-------
Table 5. SAMPLING CONDITIONS AND CONCENTRATION OF N-NITROSODIMETHYLAMINE IN AMBIENT AIR
Wind Molecular
Date
10/14
10/15
10/15
10/16
10/17
Time
(EOT)
11-2
3-6:
7-10
11-2
3-6:
11-2
3-6:
10-1
2-5:
:50
50
:50
:50
50
:50
50
:50
50
9:56-1:
2:10-6:
PM
PM
PM
AM
AM
PM
PM
PM
PM
46 PM
00 PM
Temperature RH
Location (°F) (%)
FMC (P.
FMC (P.
FMC (P.
FMC (P.
FMC (P.
FMC (P.
FMC (P.
Sewage
Sewage
Chessie
Chessie
Lot)
Lot)
Lot)
Lot)
Lot)
Lot)
Lot)
Plant
Plant
Pier
Pier
83
85
74
65
61
85
83
72
72
55
55
40-50
45-50
65-85
90-97
97
40-50
40-50
45-57
45-50
88-94
94-100
Direc-
tion
WNW
WNW
WSW
Calm .
S
WSW
WSW
NNW
NNW
ENE
E
Speed
(KTS) DECS
10 x
7 x
7 x
_
8-10 x
5-8 x
9-11
5-12
7-9 x
10-14 x
Verification DMN Concentration
OV-225 (ng/m3)
2,133
10,500 ± 1,
(±11%)
167
x 1,375 ± 125
(±9%)
416
571
3,200
x 13,437 i 937
(±7%)
trace
trace
909'
x 84
NJ
00
-------
Table 6. SAMPLING CONDITIONS AND CONCENTRATIONS OF N-NITROSODIMETHYLAMT.NE
Date
Time
Wind/Temperature
Location
JJg/m"
11/19/75 2:00 PM-4:00 PM
11/20/75 3:45 PM-5:45 PM
11/20/75 8:20 PM-10:20 PM
11/24/75
11/24/75
11/24/75
11/25/75
11:50 PM-1:50 PM
1:55 PM-3:55 PM
6:35 PM-8:35 PM
1:48 PM-3:48 PM
E-SE
6KN-SW,
3KN-SE, 63°
6KN-S, 57'
West-Northwest of dimazine
thermal destructor on FMC
property.
Town of Fairfield, 200 yd
from residential area,
Conoco parking lot.
Memirac Corp., north of FMC
plant.
FMC lot, SV7 of dimazine plant
Northbridge & Cannery St,
downwind of FMC
3"2 ± 1.5
1.95 -
1.36 ± 0.51
20 ± 4
14 i 0.2
26-'± 0.5
7.6
to
-------
30
Approximately one month later during the period of November 19-25,
additional values for N-nitrosodimethylamine were obtained. Again on
the FMC property, high values of this nitrosamine were detected. How-
ever, the sampling locations were much closer to the general area of
the dimazine thermal destructor. The concentrations ranged from 14-32
Ug/m •
Experiments on artifact formation, i^.e^. production of DMN on the
cartridge sampler during field sampling could not be detected when the
air stream was deliberately spiked with dinethylamine from a permeation
tube. Laboratory experiments indicated that 10 or more ppm of NO/NO,,
spiked into an air stream containing DMA can produce DMN; 1 ppm or less
yielded undetectable quantities. We found it necessary to purify our
DMA by subambient vacuum distillation prior to any field experiments
since it contained 1 part in 10 DMN.
------- |