Pre-Publication Copy
THE ANALYSIS OF PURGEABLE
ORGANICS IN THE DRINKING WATER
OF FIVE U.S. CITIES
USING GAS CHROMATOGRAPHY/
MASS SPECTROMETRY (GC/MS)
HEALTH EFFECTS RESEARCH LABORATORY
WATER QUALITY DIVISION
ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO
JUNE, 1975
-------
Pre-Publlcation Copy
GC/MS DETERMINATION OF VOLATILES FOR THE
NATIONAL ORGANICS RECONNAISSANCE SURVEY (NORS)
ON DRINKING WATER
by
Frederick C. Kopfler
Robert G. Melton
Robert D. Lingg
W. Emile Coleman
THE OCCURRENCE OF VOLATILE ORGANICS IN
FIVE DRINKING WATER SUPPLIES USING
GAS CHROMATOGRAPHY/MASS SPECfROMETER
(GC/MS)
by
W. Emile Coleman
Robert D. Lingg
Robert G. Melton
Frederick C. Kopfler
HEALTH EFFECTS RESEARCH LABORATORY
WATER QUALITY DIVISION
ENVIRONMENTAL RESEARCH CENTER
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO
JUNE, 1975
-------
LITERATURE REFERENCES
To Attached Papers Which Were Presented at
The First Chemical Congress of the North American Continent
Mexico City, December 1975
1. F.C. Kopfler, R.G. Melton, R.D. Lingg, and W.E. Coleman,
GC/MS Determination of Volatiles for the National
Organics Reconnaissance Survey (NORS) of Drinking
Water in "Identification and Analysis of Organic
Pollutants in Water," 1st ed, L.H. Keith, Ed.,
Ann Arbor Science Publishers Inc., Ann Arbor, MI.,
1976, Chapter 6.
2. W.E. Coleman, R.D. Lingg, R.G. Melton, and F.C. Kopfler,
The Occurrence of Volatile Organics In Five Drinking
Water Supplies Using GC/MS in "Identification and
Analysis of Organic Pollutants in Water," 1st ed,
L.H. Keith, Ed., Ann Arbor Science Publishers Inc.,
Ann Arbor, MI., 1976, Chapter 21.
-------
GC/MS DETERMINATION OF VOLATILES FOR THE
NATIONAL ORGANICS RECONNAISSANCE SURVEY (NORS) DRINKING WATER
by
Frederick C. Kopfler, Robert G. Melton
Robert D. Lingg, and W. Emile Coleman
U.S. Environmental Protection Agency (EPA)
Health Effects Research Laboratory
Water Quality Division
Cincinnati, Ohio 45268
INTRODUCTION
During the past several years the technique of partitioning trace
organics from aqueous samples into a gas phase for analysis by chroma-
tography has become popular. This technique is advantageous because
it is suitable for the analysis of more volatile compounds which would
be lost in routine extraction procedures. A decision was made to use
this technique to complement methods for isolation of higher molecular
weight organics from drinking water(1).
At the time this decision was made several methods had been
developed. McAuliffe(2) repeatedly equilibrated the aqueous sample
with helium and analyzed the gas phase directly. Distribution coeffi-
cients obtained by the repeated equilibrations supplemented the reten-
tion times in identifying unknown organic compounds.
Rook(3) developed a static head-gas method in which 10 1 of water
are heated at 60°C for 12 hours after which the head-gas is forced
through a small amount of activated silica which traps the organics
for subsequent gas chromatographic analysis. Mieure and Dietrich(A)
-------
developed a method in which the head space over an aqueous sample is
continuously swept through a porous polymer trap. Zlatkis, et al.(5)
developed a similar method, but the aqueous sample is heated almost to
boiling to thermally extract the volatile organics into the head
space. This latter method has been applied to drinking water by
Dowty, et al.(6,7).
Grob(8) reported an elegant system in which a small volume of air
is recycled through a water sample and a small charcoal filter. The
organics are eluted from the charcoal with CS2 for analysis.
While these existing methods were being considered for use in
this laboratory, another method for isolating volatile organics from
water was developed in another EPA Laboratory by Bellar and Lichten-
berg(9) and its validity for the analysis of drinking water demon-
strated (10). Consequently, this latter method was adopted in our
laboratory because this group of workers was available to give assist-
ance in overcoming the technical difficulties encountered when an
unfamiliar technique is implemented.
In this technique, five ml of sample were introduced by syringe
into the sample compartment of a specially designed purging device.
The sample was purged for 11 minutes with 20 ml/min of a purified
gas such as helium or nitrogen. The purge gas exited through a 1/8"
stainless steel tube filled with porous polymer or silica gel to trap
the organics. The organics could then be thermally desorbed and back-
flushed onto a gas chromatographic column.
-------
In August of 1975 this laboratory was requested to assist in the
analysis of the drinking water of New Orleans, Louisiana. This purging
method was used in conjunction with a combined gas chromatograph-mass
spectrometer (GC/MS) and a gas chromatograph with flame ionization
detectors (GC/FID). Only 15 compounds were identified(ll); many
other compounds were present in concentrations too low to produce
conclusive spectra.
Soon after the completion of the New Orleans study the Adminis-
trator of the U.S. Environmental Protection Agency announced that a
nationwide study, NORS, would be undertaken to determine the concentration
and potential effects of certain organic chemicals in drinking water.
The objective of this laboratory was to identify as comprehensively
as possible the volatile organic compounds present in the tap water of
five U.S. cities. Ten of the identified compounds were then to be
selected for quantitative analysis.
Results of the New Orleans survey and other field studies demon-
strated that achievement of these objectives required both a reliable
method of sample collection and a method for collecting an increased
amount of organics from the sample. The purpose of this presentation is
to describe the development and use of these techniques for collect-
ing and preparing the NORS samples for GC/MS analysis. The results of
these analyses will be presented by Coleman, et al.(12)
-------
SAMPLE COLLECTION AND TRANSPORTATION
From previous experience In collecting and transporting samples
from the field for volatile organic analysis (VGA) it had been found
that serum vials closed with Teflon-lined rubber septa held securely
in place with crimped-on aluminum seals were effective in preventing
leakage of the samples during transportation. Vials filled carefully
to the top with water and capped so that no air space remained (Fig. 1)
developed no significant head space during storage.
At the time the survey was announced, there was evidence indicating
that chlorination of drinking water resulted in the formation
of halogenated organic compounds in the water(3,9). Since drinking water
contains a chlorine residual, it was necessary to determine if the con-
centration of organics could change with time. The effect of storing
tap water samples was investigated by filling a group of 50 ml vials
with tap water. These samples were then divided into two groups, one
maintained at 3°C and the other at 20°C. At selected intervals, 2
vials from each group were removed, placed in a 25°C water bath for 30
minutes and analyzed for chloroform and bromodichloromethane by the
method of Bellar and Lichtenberg(9). The results are illus-
trated in Figures 2 and 3. The concentration of these compounds clearly
increases even when the samples are kept cold.
Therefore samples for this survey were collected in 50 or 140 ml
serum vials which had previously been heated at 550°C for 4 hours.
The samples were packed in an ice chest with an ice substitute which
maintained the temperature at 0 to 48C. Because samples shipped by
air mail or air freight could require more than a week to reach the
laboratory, they were then hand carried by air transportation to the
A
-------
laboratory and stored in a refrigerator free of organic solvents.
The samples from each city were analyzed between 24 to 168 hours after
collection.
QUALITATIVE ANALYSIS
The Effect of Temperature
It had been demonstrated that recovery of volatile organics
could be enhanced by elevating the temperature during purging(5,8).
Because formation of haloforms in the water samples was temperature-
dependent, the effect of elevated temperature on some volatile organics in
drinking water samples were incorporated in a broader study. The water
was heated to boiling in an open container and an aliquot transferred
to a serum vial and capped as previously described. After 30 minutes
additional samples of the boiling water were obtained. These samples
were allowed to cool and were analyzed at room temperature as were
samples of the original water. The results are illustrated in columns
D and E of Table 1 as percent change in concentration from the
original tap water. It can be seen from the data in Column E that
boiling the water containing a chlorine residual increased the
concentration of haloforms. When continued for 30 minutes (Column
D), an efficient thermal extraction of the alkyl halides was achieved.
This increase in haloform concentration upon heating made the direct
incorporation of thermal extraction into the analysis of drinking
water impossible. The reason in quantitative analysis is obvious, but
the increase in chloroform content can also cause a problem in qualita-
tive analysis: chromatograms from a computerized GC/MS are
-------
normalized on the largest peak, and since in many samples chloroform
is the volatile compound present in highest concentration, an increase
in chloroform concentration tends to obscure peaks of compounds present
in low concentration.
The effect of chemically reducing the residual available chlorine
in a tap water sample before increasing its temperature was investigated.
Sodium thiosulfate was not used since thiosulfate can react with
alkyl halides to produce Bunte salts (S-alkyl thiosulfates)(13). Sodium
sulfite can also react with alkyl halides to produce sulfonic acids and
therefore it could not be used in these types of analyses.
The reducing agent chosen for this work was potassium ferrocyanide
because of its reduction potential and because there were no known
reactions with organic compounds. The ferricyanide ion produced upon oxidation of
ferrocyanide can react with phenols(14), but since neither reactants
nor products are volatile under the purging conditions, no interfer-
ences were observed. A small increase in cyanogen chloride concentration
may have occurred in drinking water samples to which ferrocyanide was
added. At low pH however, this problem can become severe, so if
ferrocyanide is to be added to the sample, it should be made alkaline
before addition of this reducing agent.
It is important to point out that even though residual chlorine
was totally eliminated an increase of chloroform occurred when the sample
was heated to 95°C. This same increase in chloroform content was
observed in samples containing ferrocyanide after standing at 25°C
for 24 hours. No further increase was observed in these samples after
standing for an additional 24 hours. The same increase was observed
-------
after 24 hours in samples containing a 10-fold excess of ferrocyanide.
Thus, it appears that at the time of taking the samples some chlorine
is already combined with an organic molecule that will ultimately under-
go a conversion resulting in production of chloroform.
The increased yield of organics observed when the sample is purged
at 95°C outweighed the disadvantages and a thermal extraction step was
incorporated into the final method.
Design of Larger Purging Devices
A direct 100-fold scale-up of the 5 ml purging device of Bellar
and Lichtenberg was not satisfactory. The next apparatus constructed
was modeled after the purging device used by Saunders, et al.(15)
but the sample chamber was surrounded by a jacket which permitted
rapid heating of the sample by circulating water from a bath main-
tained at 95°C. For qualitative analysis a device with a 500 ml
sample chamber was constructed.
For several reasons a similar device, illustrated in Figure
4, containing 140 ml of sample was constructed for quantitative analysis.
There was skepticism about transferring the contents of more than one
serum vial quantitatively. Complete removal of alkyl halides was not
completed in a reasonable time at room temperature in the larger device.
It was impossible to secure enough sample for several replicate quantita-
tive determinations if the larger device was used.
The smaller purging device is similar to a gas washing bottle
equipped with a 29/42 ground glass joint on the top, a 20 mm medium
fritted disc on the end of the gas tube to dispense the purge gas, a
-------
5 mm ID sample introduction port fitted with a silicone rubber septum,
an 11-gauge by 304.8 mm (12 inch) stainless steel hypodermic needle
through the septum, a stainless steel stopcock with Leur-Lock fittings
on the needle, and a water jacket surrounding the sample reservoir.
The design of the larger device is the same, but a 457 mm (18 inch)
stainless steel needle was required for the sample introduction port.
Both of these purging devices, as well as that of Bellar and Lich-
tenberg are available from Paxton Woods Glass Shop, 7500 Brill Road,
Cincinnati, Ohio 45243.1
Traps
The traps used with the smaller device were identical with those
described by Bellar and Lichtenberg(S). The trap used with the larger
purging device was similar but was constructed of 1/4 inch stainless
steel tubing (4.2 mm ID) instead of 1/8 inch tubing. All traps were
filled with 60/80 mesh Tenax GC and were conditioned at approximately
200°C with a helium flow of 20 ml/min for 16 to 24 hours. The traps
were reconditioned for 20 minutes before use each day. The traps
are available from Cincinnati Valve and Fitting, 3710 Southern Avenue,
Cincinnati, Ohio 45227.1
Low Organic Water
Another problem encountered during the New Orleans survey was ob-
taining a source of water free of volatile organics to be used in develop-
ment of a procedural blank. Distilled water from a number of sources
•^•Mention of commercial sources does not necessarily imply endorsement
by the U.S. Government.
-------
was analyzed as were distilled water further treated with activated
carbon and purified water produced by a Millipore Corp. Super-Q System.
All were contaminated with several volatile compounds.
An acceptable "low organic" water was produced from distilled water
by passing it through a Millipore Corp. Super-Q system followed by
purging with ultra pure helium for 48 hours at 95°C. This water
still contained traces of polar organics such as ethanol and acetalde-
hyde. It was then transferred to serum vials, sealed and stored in
the refrigerator with the samples until analyzed.
Fractional Purging
The initial analyses of samples conducted with the largest purging
device resulted in chromatograms dominated by haloforms to such an
extent as to obscure many compounds present in lower concentration.
The slowness with which polar organics were purged in the preparation
of the "low organic water" suggested that a fractional purging scheme
could be developed to aid in the qualitative analyses of the samples.
Figure 5 illustrates the relative purging efficiencies of five
compounds. These data were obtained by performing six successive
analyses of a solution containing approximately 50 yg/liter of each
component except ethanol. To obtain a similar response for ethanol
with the GC/FID a concentration of 2 mg/liter was required. Clearly
all organics do not purge at the same rate, nor do they all purge
quantitatively. It was found that purging at 95°C for 30 minutes
removed most of the alkyl halides and other non-polar compounds.
A sufficient quantity of polar compounds remained so that a subsequent
-------
purging of the sample allowed these organics to be trapped and analyzed
separately. This aided in obtaining conclusive spectra of the polar
compounds.
Qualitative Procedure used in NORS.
Organics were isolated for qualitative analysis from 500-600
ml aliquots of each sample. The organics were fractionated by conducting
three consecutive 30-minute purges of the same sample. The first purge
was conducted at 6°C to remove the most volatile, non-water soluble
organics. A stoichiometric amount of potassium ferrocyanide was then
added to reduce chlorine, and the sample was then purged at 95°C to
remove most of the remaining volatile, non-water soluble organics.
Finally, a third purge was conducted at 95°C to detect the presence
of volatile, water soluble organics.
Details of the qualitative procedure are as follows: (1) the
level of total chlorine residual in an aliquot of the drinking water
sample was determined by amperometric titration; (2) The trap was
attached to the purging device; (3) the content of five 125 ml sample
bottles were forced with helium into the reservoir through a teflon
transfer line as illustrated in Figure 6; (4) The sample was purged for
30 minutes at 6°C with the slow stirring of a magnetic stirring bar;
(5) The sample purging and stirring was stopped, the trap was removed and
capped and the helium exit of the purging device was capped; (6) After
the organics had been desorbed from the trap into the GC/MS, the cooled
trap was replaced onto the purging device; (7) The stoichiometric amount
of potassium ferrocyanide necessary to reduce the chlorine was added
through the sample inlet, and the solution was stirred for 10 minutes
10
-------
at 6°C; (8) The temperature was raised to 95°C and the water was again
purged for 30 minutes; (9) The trap was removed and the adsorbed organics
analyzed while the purging of the water continued; (10) the trap was
again cooled and attached; (11) The organics were collected for another
30 minutes and analyzed.
The "low organic water" was analyzed immediately prior to the analysis
of each sample. A compound was deemed present in a sample if its concen-
tration was greater in the sample than in the corresponding procedural
blank(12).
Figures 7, 8, and 9 are the reconstructed chromatograms obtained
during the qualitative analysis of the Miami, Florida, sample. The
benefit of the fractional purging is best illustrated in the following
observations. Figure 7 shows the detection of extremely volatile
compounds such as vinyl chloride. The chromatogram in Figure 8 is
dominated by alkyl halides, especially the haloforms. The presence of
many more water soluble alcohols, aldehydes and ketones can easily be
detected in the third analysis of the sequence, Figure 9.
QUANTITATIVE ANALYSIS
The quantitative analyses of these samples was the most complex
task to be accomplished in the NORS. Before the survey began a decision
was made to determine the concentration of ten volatile organics
deemed significant because of their potential health effects. These
compounds were chosen only after the qualitative analysis of all five
samples was completed. The increase in haloform concentration in the
samples precluded delaying quantitative analysis until after all
qualitative analyses were completed and selection of compounds was
11
-------
made. Thus, it was necessary to obtain data on each sample that
could later be used to quantitate any compound identified in the
samples.
Given these conditions, the only approach to quantitation was
to use an internal standard (IS) in all samples and blanks, to analyze
them under identical conditions with computerized GC/MS and to store the
data on magnetic tapes. After the compounds to be quantified were
selected, solutions containing known amounts of these compounds and the
IS were anlyzed in the same manner. The IS consisted of a solution of
two compounds: 2-bromobutane and 2-iodopropane. These compounds were
chosen because they are easily purged from water and together have a good
distribution of mass peaks that are 15% or greater of the base peaks.
The internal standards were prepared from reagents supplied by
Chem-Services, Inc. A 10-yl syringe was used to carefully measure
2 yl of each of the two compounds. This volume of each compound was
injected into a 1 liter volumetric flask nearly filled to the mark with
water. After dilution to the mark and gentle agitation to insure complete
mixing, the IS solution was bottled in 50-ml serum vials and capped as
described above. Fresh internal standard was prepared each day when
quantitation runs were made because both compounds slowly decompose
in aqueous solution.
Purging Procedure for Quantitative Analysis
To insure that a constant amount was added at each run, the IS
from these capped serum bottles was added directly to the 140 ml purg-
ing apparatus. Before sample water was introduced, the sample trans-
12
-------
fer line was disconnected from the Luer-lock valve and a 5 mm NMR
sample tube septum was placed over the opening of the valve (Figure 1).
With the trap in place, a 500 yl syringe was used to withdraw 300 yl
of IS from one of the prepared vials. The NMR tube septum was then
pierced with the syringe needle and the Luer-lock valve was opened.
After passing the syringe needle through the valve opening, the 300 yl
of IS was discharged directly into the purging device. Upon withdrawing
the syringe needle beyond the valve opening, the valve was closed,
and the syringe withdrawn completely. The septum was removed, the
sample transfer tube was reconnected, and the sample was then transferred
from a serum vial into the apparatus via the closed transfer loop
previously described for qualitative analysis.
Samples were purged for 15 minutes at 25°C with a helium flow
rate of 20 ml/min. Four replicate analyses were performed on each
sample using a 1.52 m (5 ft.) column of 60/80 mesh Chromasorb 101.
Figure 10 illustrates the reproducibility of the four quantitative
chromatograms obtained on one of the samples.
Analytical Standards
The compounds selected for quantitation were liquids at room temperature.
Aqueous standards of these compounds were prepared by adding 2 yl of each
to a 1 liter volumetric flask almost filled with distilled water. The
flask was then diluted to volume with distilled water and agitated to
achieve a complete solution. Dilutions of this stock solution produced
a series of standards containing approximately 5, 2.5, 1, 0.5, 0.3, 0.1
and 0.05 yg/1 of each compound. The exact concentration of each compound
13
-------
was calculated from its density. These solutions were transferred to
140 ml serum vials and capped until analyzed.
Four replicates of this set of standards were preapred. Each standard
solution was fortified with the IS and analyzed as described for the
samples. The intensity of the IS response in each chromatogram was used
to compensate for instrument variations which existed among analyses,
allowing comparison of the samples to the standards. This technique
has recently been discussed by Young(16).
Many problems were encountered in obtaining quantitative data on these
samples prior to obtaining the qualitative data. The application of
computerized GC/MS to this problem is to be discussed in a forthcoming
paper by Lingg, et al.(17).
14
-------
REFERENCES
1. Kopfler, F.C., Melton, R.G., Mullaney, J.L. and Tardiff, R.G.,
"Human Exposure to Water Pollutants," presented before the
Division of Environmental Chemistry, American Chemical Society,
Philadelphia, PA., April (1975).
2. McAuliffe, C., Chem Tech 1, 46 (1971).
3. Rook, J.J., Water Treatment Exam., 21, 259 (1972).
4. Mieure, J.P. and Dietrich, M.W., J_. Chromatogr. Sci., 11, 559
(1973).
5. Zlatkis, A., Lichtenstein, H.A., and Tishbee, A., Chromatographia
£, 67 (1973).
6. Dowty, B., Carlisle, D., and Laseter, J., Science, 187, 75 (1975).
7. Dowty, B., Carlisle, D. , and Laseter, J., Envir. Sci. Tech. 9^,
762 (1975).
8. Grob, K., J_. Chromatogr. 84, 255 (1973).
9. Bellar, T.A. and Lichtenberg, J.J., _J. Amer. Water Works Assn.,
66, 739 (1974).
10. Bellar, T.A., Lichtenberg, J.J. and Kroner, R.C., J^. Amer. Works
Assn., _66, 703 (1974).
11. "Analytical Report - New Orleans Area Water Supply Study." U.S.
Environmental Protection Agency Report, EPA-906/10-74-002 Region
VI (Dallas, Texas) Surveillance and Analysis Division (1975).
12. Coleman, W.E., Lingg, R.D., Melton, R.G., and Kopfler, F.C.
"The Occurrence of Volatile Orgaiiics in Five Drinking Water
Supplies Using Gas Chromatography/Mass Spectrometry (GC/MS),"
To be presented at the First Chemical Congress of the North
American Continent, Mexico City, December (1975).
13. March, J. "Advanced Organic Chemistry: Reactions, Mechanisms,
and Structure," McGraw-Hill Book Co., New York, N.Y., 1968,
p 330.
14. Fieser, L.F. and Fieser, M.F., "Reagents for Organic Synthesis,"
John Wiley and Sons, New York, N.Y., 1974, p. 929.
-------
15. Saunders, R.A., Griffith, J.R., and Saalfeld, F.E., Biomed.
Mass Spectrom. 1^, 192 (1974).
16. Young, I.G. American Laboratory, 1_ 11, (1975).
17. Lingg, R.D., Melton, R.G., Kopfler, F.C., and Coleman, W.6.,
GC/MS techniques for the quantification of volatile organics
in tap water. In Preparation (1975).
-------
ALUMINUM SEAL
TEFLON - FACED SEPTUM
- TEFLON
CONVEX MENISCUS (SAMPLE)
MUFFLED SERUM VIAL
COLLECTION OF SAMPLE WITHOUT HEADSPACE
-------
1801 1 1 1 \ r—i 1 1 1 1
6
8 10 12 14 16 18 20
days stored
Chloroform Concentrations in Stored Drinking Water
-------
I 1 I —I 1 1 I
Ill II 1 1 1 \ 1 1 1 1 1 1
days stored
Bromodichloromethane Concentrations
in Stored Drinking Water Samples
-------
PERCENTAGE INCREASE OR DECREASE OF FIVE VOLATILE ORGANICS ABOVE
OR BELOW LEVELS PRESENT IN CINCINNATI COLD TAP WATER.
SAMPLE TYPE
COMPOUND
TRICHLOROMETHANE
1,1,1 - TRICHLOROETHANE
TETRACHLOROMETHANE
BROMODICHLOROMETHANE
CHLORODIBROMOMETHANE
(A)
COLD
TAP
0
0
0
0
0
(B)
CARBON
FILTERED
(-) 99.5
{-) 100.0
(-) 100.0
(-100.0)
(-) 100.0
(C)
CARBON
FILTERED
& BOILED
(-) 99.4
(-) 100.0
(-) 100.0
(-) 100.0
(-) 100.0
(D)
BOILED
30 MIN.
(-) 94.0
(-) 100.0
(-) 97.3
(-) 97.2
(-) 100.0
(E)
BOILED
5 SEC.
(+) 96.5
(-) 50.4
{-) 61.6
(+) 72.4
(+) 121.8
(F)
HOT
TAP
(+) 108.0
(-) 14.5
(-) 18.0
(+) 78.6
(+) 141.5
(A) Total chlorine residual (TCR) = .66 mg/1; Non-volatile total organic carbon (NVTOC) = 1,1 mg/1;
Conductivity = 276 juMhos/cm; pH = 9.1.
(B) TCR =0 mg/1; NVTOC<0.1 mg/1; pH = 8.7
(D) TCR = 0 mg/1; NVTOC = 1.1 mg/1; pH = 9.4
(E) TCR = .06 mg/1; NVTOC = 1.1 mg/1; pH = 9.0
(F) TCR = .04 mg/1; NVTOC = 1.1 mg/1; pH = 8.9; Contact time of cold tap water (A) in
domestic hot water tank (glass lined, 60 C) = 8 hr.
-------
B-D
26.7 cm Height
MS09
Luer-Lok Stopcock
(Stainless Steel)
Injection Septum
(6mm O.D. x 9mm Height)
5mm I.D. Sample
Introduction Port
10 Gauge Needle
-4- 1/4" O.D.
T Helium Inlet
! ! Ik- 29/42 f Joint
ir-r'/
3/8" O.D.
Water Jacket
Exit
3/8" O.D.
JL
Water Jacket [«
Inlet 35mm O.D.
20mm Medium Fritted
Filter Disc
45mm O.D.
140-ml VGA Purging Apparatus
-------
11
21 31 41 51
PURGE TIME (min)
FIVE REPETITIVE PURGES OF STANDARD SOLUTION
-------
OPEN VENT
ADSORPTION
TRAP
FLOW OF WATER
-140-ML
SERUM
BOTTLE
20-GAUGE
NEEDLE
10-GAUGE
NEEDLE
PURGE GAS INLET
ADDITION OF WATER SAMPLE TO PURGING DEVICE
-------
Q
a
j
'•
--
•
• _.
•
•••
• •
••
«M "
CHLORIDE)
^j
LOROETHEHE IVINY
*
O
A
YLIDENE CHLORIDEI
_^ %
PANONE (ACETONE
ZHLOROETHENE (VI
23
°|- '
*N **
UL_
in
-2 DICHLOROETHE
•
A
CHLOROETHANE
---
•
)
^-
<
-
'
•
1
1
*
s 7.2- DICHLOROET
)
Hi
(CHLOROFORM/
Uj
CHLOROMETHANE
ON TETRACHLORID
\\\
\\
LOROETHEHE
•
\
HANE
3ROMODICHLOROHH
(i
A
DIBROMOMETHANE
.
I
\
;
:
Q
5
0 50
SPECTRUM NUMBER
100
150
200
250
300
350
400
RGC OF 500-ml VGA SAMPLE OF MIAMI
FINISHED WATER PURGED AT 6 C, 30 MIN.
-------
ISO
200
250
300
350
400
450
500
SPECTRUM NUMBER
RGC OF 500-ml VOA SAMPLE OF MIAMI
FINISHED WATER PURGED CONSECUTIVELY AT 6C, 30 MIN; 95C, 30 MIN.
-------
0 50
SPfCTRU/M NUMBER
100
150
200
250
350
400
RGC OF 500-ml VGA SAMPLE OF MIAMI
FINISHED WATER PURGED CONSECUTIVELY AT 6C, 30 MIN; 95C, 60 MIN.
-------
REPRODUCIBILITY OF FOUR
REAL-TIME 140-ml VOA CHROMATOGRAMS
SAMPLES OF PHILADELPHIA FINISHED WATER.
-------
THE OCCURRENCE OF VOLATILE ORGANICS IN FIVE
DRINKING WATER SUPPLIES
USING GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS)
by
W. Emile Coleman, Robert D. Lingg
Robert G. Melton, and Frederick C. Kopfler
U.S. Environmental Protection Agency (EPA)
Health Effects Research Laboratory
Water Quality Division
Cincinnati, Ohio 45268
INTRODUCTION
After widespread publicity of the results of the study conducted by
U.S. Environmental Protection Agency (EPA) on New Orleans drinking water in
November of 1974(1,2,3), EPA Administrator Russell E. Train announced that
he was ordering an immediate nationwide survey to determine the concent-
ration and potential effects of certain organic chemicals in drinking
water. A National Organics Reconnaissance Survey (NORS) was undertaken
after President Ford signed "The Safe Drinking Water Act" in December of
1974. EPA was to conduct a comprehensive study of public water supplies
and drinking water sources to determine the nature, extent, sources of, and
means to control contamination by chemicals or other substances suspected
of being carcinogenic.
As part of the NORS of drinking water supplies, the then Water Supply
Research Laboratory in Cincinnati was charged with investigating the occur-
rence of volatile organics in drinking water of five selected cities:
Miami, Florida; Seattle, Washington; Ottumwa, Iowa; Philadelphia, Pennsyl-
vania, and Cincinnati, Ohio. These supplies were chosen to represent the
-------
major types of raw water sources in use in the United States today. Table 1
identifies the chemical properties of the finished water of the five cities
along with the type supply and raw water source.
The objective of this laboratory in the NORS was to characterize as
completely as possible, using existing analytical technology, the volatile
organics in finished drinking water. This paper discusses primarily the gas
chromatography/mass spectrometry (GC/MS) techniques used for obtaining the
results related to the above objective; that is, identifying volatile
organic compounds in the drinking water supplies of these five cities. Ten
selected compounds were also quantified as a part of this study. A more
comprehensive report of the five-city survey is being prepared for trans-
mi ttal to Congress(4).
As part of a storage study on the effects of residual chlorine in tap
water samples collected in the field and shipped to our laboratory for
subsequent analyses, some compounds were identified by GC/MS that might be
products of chlorination of raw water at the treatment plant. The results
of ensuing experiments also are discussed in this paper.
EXPERIMENTAL
To determine the presence of volatile organics in drinking water a
computerized GC/MS system and a modified Bellar and Lichtenberg(S) trapping
technique were employed. This modified technique is described in detail by
Kopfler et al.(6); they also describe a three-stage technique of stripping
organics from a 500-ml volume of water commonly called the "500-ml VOA
(volatile organics analysis)." The qualitative results of analyses in this
paper are all based on the 500-ml VOA. The method of quantification of the
selected compounds is described in another paper by Lingg et al.(7).
-------
Because of the wide variety (class and polarity) of compounds present
in the water, chromatographic separation of each compound on the same GC
column was not attainable. However, identification of the detected com-
pounds was enhanced by having computer capabilities which enabled searches
for specific ions (Extracted Ion Current Searches)(8) in a total ion current
(TIC) recori structed gas chromatogram (RGC) .
Apparatus
Reagents
Ultra-pure helium, at a flow rate of 20 cc/min, was used to back-flush
the organics from the trap onto the GC column and also as a carrier gas
through the column at 30 cc/min.
In-house distilled water was used to make all standards used to confirm
relative retention times of suspect compounds. The sources of standards
that we were able tc obtain are listed in Table 5. Preparation of standards
is discussed by Lingg et al.(7).
The reagents used for the chlorination experiment were all reagent
grade. The chlorine solution was prepared by bubbling gaseous chlorine
(Cl2) through nitrogen-purged distilled water to achieve a stock solution
concentration in the range of 200-1000 mg/1. Appropriate dilutions were
made from the stock solution (after the chlorine content was accurately
measured by amperometric titration) to obtain the desired chlorine con-
centration. Seattle drinking water was used to make solutions containing 50
ppb each of benzene," ethanol, nitromethane, 2-pertanone, toluene and
m-xylene; 1 ppm ferric chloride and 1.12 ppm chlorine in 140-ml sealed serum
vials(6,7).
-------
Trap and Desorber
The trap and desorber for the 500-ml VOA were essentially the same as
the ones designed by Bellar and Lichtenberg(S), except the OD of the trap is
1/4 inch and the desorber was built to accommodate the wider diameter
tubing. The desorber temperature was stabilized at 180C.
GC Columns
Two, packed, glass columns were used for the 500-ml VOA samples. One
contained Chromosorb 101, 60/80 mesh and the other Tenax-GC, 60/80 mesh. The
OD of each 3.05-m (10 ft) column was 6.35-mm; the ID, 2-mm.
Gas Chromatograph/Mass Spectrometer
A Finnigan, Model 9000, Gas Chromatograph equipped with one of the
above columns was interfaced with a Finnigan, Model 1015D, Mass Spectrometer
to obtain electron impact mass spectra.
Automated Data System
Data were acquired and analyzed with System Industries 150 computer
system. In this laboratory, all GC/MS data were acquired on a rapid access
disk memory unit. With the use of graphic software, data were displayed on
a TEKTRONIC 4010 CRT data terminal. Additional features of the automated
data system are discussed in other papers(9,10).
Mass Spectrometer Operating Parameters
Typical operating parameters for a 500-ml VOA are listed in Table 2.
Note that the medium mass range (10-250 amu) was used because it offered
additional sensitivity over the high mass range. The MS was calibrated each
day on the medium range using perfluorotributylamine (FC 43)(9).
-------
Data Acquisition System Operating Parameters
Computer software program: IFSS(ll)
Mass range scanned: 14-16; 19-27; 29-31; 33-250
Maximum Repeat count: 4
Integration time: 3 m sec
Repeat count before checking lower threshold: 4
Lower threshold: 4
Upper threshold: 4
The IFSS mode is a signal optimization mode used to adjust the Inte-
gration time as a Function of the Signal Strength. Using this option
results in a nearly constant signal-to-noise ratio and minimizes saturation
of the analog to digital converter.
The above parameters were selected so that the time required to scan
a complete ma-ss spectrum was approximately 3 seconds. In scanning the mass
range of 14-250 amu, m/e 17, 18, 28, and 32 were omitted to eliminate
normalization of the RGC on the water or air present in the background and
the sample.
Manual GC/MS Operations for 500-ml VGA
The procedure used to transfer the sample from the trap to the GC/MS
system after the trap was inserted into the desorber at 180C is described
in Table 3. Time zero CO) equals the time that the needle of the desorp-
tion device was injected through the GC septum and the start of the back-
flushing of the trap with helium. The computer scan could be started
earlier than the time indicated in Table 3; however, we observed no organic
components eluting from the GC column before water. The ionizer in the MS
-------
was turned "on" immediately after the desorption unit was withdrawn from
the GC inlet to monitor the water on the oscilloscope as the water eluted
from the GC column.
Qualitative Analysis
Before this national survey, we qualitatively analyzed 5-ml samples of
drinking water using the Bellar and Lichtenberg(S) technique. The GC/MS
system seldom detected more than 10 to 15 volatile organics unless they
were present in unusually high concentrations. By increasing the volume of
the water one hundred fold and employing three consecutive 30-min helium
purges(6) of the same water, the number of volatile organics detected in
the 500-ml VOA more than quadrupled. Furthermore, to facilitate identifi-
cations, the volatile organics isolated by the 500-ml VOA were actually
partitioned by conducting the three consecutive purges of the same water.
As stated also by Kopfler(6), the first purge at 6C removed the most vola-
tile, nonwater soluble organics (many halogenated compounds). The second
purge at 95C removed most of the remaining volatile, nonwater soluble
organics plus some water soluble compounds not detected in the first purge.
The final purge at 95C showed the presence of volatile, water-soluble
organics. Blank water, as described by Kopfler et al.(6), was treated
similarly.
GC/MS data acquisition of the volatiles isolated from the three
consecutive purges of the finished water from the five cities and of the
blank water were stored temporarily on disk memory units. The data from
each GC/MS run in the form of RGC's, mass chromatograms (MC) , and spectra
were displayed on a CRT data terminal or obtained as hard copy output on a
digital plotter.
Most of the identifications were made from the data acquired on the
second purge because of the overall increase in concentration of the
6
-------
organics at the elevated temperature. Identifications of compounds iso-
lated from the third purge were easiest because the GC peaks were well
resolved and most of the compounds originally present in the water were
removed by the previous purges. A compound was considered "detected" in
the drinking water of the five cities when the same compound, as determined
by GC/MS data, either was absent in the blank water or was above the level
of that found in the blank water. The results of the second purges at 95C
were used to make these comparisons.
Identification of the volatile organics was confirmed by running
standard compounds on the same GC column and under the same analytical
conditions as the field samples were run. The GC relative retention times
and mass spectra of these standards were compared with those of the com-
pounds detected in the drinking water samples. It was also noted whether
the ir.ass spectrum of the standard and that of the detected compound in the
drinking water agreed with a mass spectrum published in the literature(12).
The expansion feature(9) of the data system allowed us to look at very
small GC peaks that were normalized into the baseline in the original RGC.
Without this feature, many compounds would have gone undetected and identi-
fication of them would have been thwarted by improper assignment of "eyeball"
spectrum numbers.
After the mass spectra of all obvious GC peaks were obtained from each
RGC, it was apparent that many spectra were contaminated by interfering
ions from adjacent or underlying peaks that had not been detected initially.
If two or more compounds overlap on the gas chromatogram, their mass spectra
will be superimposed when this GC peak is scanned. Each chemical compound,
however, gives rise to a unique combination of ions that forms its mass
-------
spectrum and it is therefore possible to select a unique mass ion or ions
(m/e value) for each compound in the overlapping spectra that is repre-
sentative of only that compound. The computerized technique of extracting
specific ions from a TIC RGC was used extensively in this study not only to
isolate one compound but also to select classes of compounds having charac-
teristic ions. The computer program(8) developed at Battelle Columbus
Laboratories under an EPA grant is used for selected-ion monitoring of
organic compounds by computer-controlled GC/MS (quadrupole) systems. It
enhances detection of specific compounds in complex mixtures and alleviates
background interferences. Current applications range from effluent moni-
toring to detecting drug residues and metabolites in overdose victims.
Figure 1 demonstrates the practicality of this technique in drinking water
analyses. Plots (A) and (B) are RGC's of Miami's finished water and of
blank water, respectively. Plots (C) through (G) are mass chromatograms
derived from extracted ion current searches of the RGC (A) . In (C) , m/e 29,
43, and 68 represent ions CHO+, C2H30+ and C4H6N+, respectively; (D) m/e 47
represents CCL+; (E) m/e 61 and 62 represent ions C2H2C1+ and C2H3C1+; (F)
m/e 77 and 91 represent ions CeH5+ and (C7Hj+ and CBr+); (G) m/e 79 and 81
represent ions 79Br+ and 81Br+.
Quantitative Analysis
The compounds selected for quantitative assessment(4,8) are listed below;
Benzene
.Chlorobenzene
Chloroethene (Vinyl chloridej
p-Dichlorobenzene
1,1-Dichloroethene (Vinylidene chloride)
cis 1,2-Dichloroethene
trans 1,2-Dichloroethene
Tetrachloroethene
Toluene
Trichloroethene
-------
Alternates:
Nitrotrichloromethane (Chloropicrin)
p-Chlorotoluene
Vinyl chloride, chlorotoluene, and p-dichlorobenzene were not quanti-
fied in this laboratory because the standards had not been received.
Chlorination Experiment
Kopfler et al. demonstrated the effects of chlorine on the increased
production of chloroform in drinking water in their storage study(6) and
also when they applied heat to drinking water having a chlorine residual.
The results of their findings and the identification of compounds such as
nitromethane, nitrotrichloromethane (chloropicrin), benzene, chlorobenzene,
dichlorobenzenes, dichloroiodomethane, toluene, and an isomer of chloro-
toluene in the five-city survey prompted this laboratory to investigate
further the occurrence of these compounds in an experimental situation.
The three solutions for this experiment were allowed to react in the
dark at room temperature (RT) for 4 days. (A) contained 50 ppb each of
nitromethane, ethanol, 2-pentanone, benzene, toluene, and m-xylene and 1
ppm ferric chloride; (B) contained all reagents in (A) plus 1.12 ppm
chlorine, and (C) contained all reagents in (B) except ferric chloride.
Each sample was subsequently analyzed by GC/MS using a 140-ml VOA appara-
tus (6) purged at RT.
These solutions were made in Seattle drinking water rather than in
distilled water because Seattle water had no measured chlorine residual; it
was a naturally buffered system; and the reagents added and the suspected
chlorination by-products were not detected in previous analyses of Seattle
water. Because of the use of iron pipes as water carriers and the presence
of ferric ions in most finished water supplies, ferric chloride (FeCl3) was
added as a possible catalytic agent.
9
-------
RESULTS AND DISCUSSION
The results of the volatile organic analyses and of the chlorination
experiment using GC/MS techniques are presented in Tables 4 through 9 and in
Figures 2 through 7. A total of 72 compounds were identified in drinking
water supplies of the five selected cities. Table 4 lists the compounds
according to their functional groups. It is evident from this table that
aliphatic halogenated compounds accounted for the highest percentage of all
compounds found in five-city survey. Halogenated organics (aromatic and
aliphatic) accounted for 53% of the total. Seventy percent of the total was
attributed to compounds having elements other than carbon, hydrogen, and
oxygen.
Table 5 lists the 72 identified compounds as they eluted from a 10-ft
Chromosorb 101 packed column or as noted otherwise. Their relative retention
times (RRT) were compared with the RRT of chloroform (RRT=1.00). Chloroform
eluted from the column approximately 9 min after the start of the data
acquisition.
The relative distribution of compounds per city is shown in Figure 2.
The number of compounds found in each city is as follows: Cincinnati, 52;
Miami, 60; Ottumwa, 21; Philadelphia, 50; and Seattle, 23. Figure 2 also
indicates that the distribution of halogenated compounds in all cities except
Seattle was approximately 50% of the number of compounds found in each city.
The lower distribution in Seattle (=30%) could be attributed to the type of
raw water and the fact that a "zero" chlorine residual was measured in the
finished water. (The TOG content of Seattle finished water was also the
lowest of the five cities, see Table 1). The compounds found in a specific
city's supply along with their significance in drinking water is discussed by
Tardiff(4).
10
-------
Another interesting result of this study was that only 13 of the 72
compounds were common to all five cities; see Table 6. One would infer, and
evidence supports, that the type and number of compounds found in a drinking
water supply depends heavily on environmental factors such as temperature,
pollution, agricultural practices, source of water, and raw water treatment
techniques.
Table 7 summarizes the occurrence of the 72 compounds according the
carbon number, molecular weight and boiling point. After plotting the
boiling points of the 72 compounds on normal probability paper, it was
concluded from the linearity of the plot that the boiling points of the
volatile organics were normally distributed above and below the median
boiling point (77C).
The results of the quantitative analyses of the selected compounds are
presented ir. Table 8. Additional quantitative information on other identi-
fied compounds in the NORS is presented by Tardiff(4) and Symons(13) and in
an Interim Report to Congress(14).
Table 9 lists the residual chlorine and pH measurements of samples of
"spiked" Seattle drinking water. (See Chlorination Experiment under EXPERI-
MENTAL.) Note that the residual chlorine measured at RT after 4 days (1.02
ppm) indicated that 0.10 ppm C12 had reacted. Results of the 140-ml VGA by
GC/MS indicated that the same chlorination byproducts were produced whether
the ferric chloride was present or not, thus indicating that it had no
catalytic effect in the chlorination reaction.
Figure 3 demonstrates the increased production of chloroform and the
production of dichloronitromethane, nitrotrichloromethane (chloropicrin), and
an isomer of chloroxylene. Although not evident in this figure, trace amounts
of chlorotoluene were detected. It is quite obvious from comparing (A) and
11
-------
(B) in Figure 3 that ra-xylene was readily chlorinated and that chloroform was
greatly increased. It is also obvious that the purging efficiency of ethanol
and nitromethane from water is poor, as demonstrated by Kopfler et al. (6) .
The mass spectra of the chlorinated byproducts are presented in Figures 4
through 6. In Figure 5, the mass spectrum of chloropicrin is somewhat
obscured by major fragment ions (m/e 91, 92, and 65) of toluene. Identifica-
tion of compounds in this area of the RGC was very difficult because toluene,
tetrachloroethylene, chloropicrin, 1,1,2-trichloroethane, 2,6-dimethyl-4-
heptanone and bromotrichloromethane all have RRT of approximately 1.50.
Bromotrichloromethane (not found in this study) has an almost identical RRT
and mass spectrum as chloropicrin. The key spectral peak in differentiating
the latter two compounds is m/e 30 (N0+).
Figure 7 demonstrates how chloropicrin was successfully identified with
the use of extracted ion current searches.
Another identified compound that may be a byproduct of chlorination is
dichloroiodomethane. Bunn and Kleopfer(lS) found this compound in several
midwest water supplies.
The general reactions listed below summarize this brief chlorination
experiment.
CL2
Nitromethane > Nitrotrichloromethane (reacts readily)
H20 (chloropicrin)
Cl?
Benzene HO* (does not react)
Toluene Q >Chlorotoluene (reacts slowly)
C12
m-Xylene > Chloroxylene (reacts readily)
H20
12
-------
It is highly possible and probable that the compounds mentioned in this
paper have been with us for some time. Only because of powerful tools like
GC/MS and the other techniques discussed in the paper have we been able to
detect and identify low levels of volatile organics in drinking water which
heretofore had been unidentified.
13
-------
ACKNOWLEDGEMENTS
The authors wish to thank and commend the many others who
contributed to this paper: Ms. Verna E. Tilford and Ms. Diana
Routledge who typed the manuscript; Mr. Dale Dietrich and his
staff who Drovided the art work and graphics; and Mrs. Marion G.
Curry who edited the manuscript. Special commendation is due
Mr. Donald Mitchell who helped with the analyses and Dr. Richard
Schoenig who compiled an enormous amount of data and also helped
with the analyses.
-------
REFERENCES
1. "Industrial Pollution of the Lower Mississippi River in Louisiana,"
U.S. Environmental Protection Agency, Region VI, Dallas, Texas,
Surveillance and Analysis Division (1972).
2. "Analytical Results -New Orleans Area Water Supply Study," U.S.
Environmental Protection Agency Report, EPA-906/10-74-002,
Region VI (Dallas, Texas) Surveillance and Analysis Division (1975).
3. Dowty, B., Carlisle, D., and Laseter, J.L., Science 187, 75 (1975).
4. Tardiff, R.G., Budde, W.L., Coleman, W.E., DeMarco, J., Dressman, R.C.,
Eichelberger, J.W., Kaylor, W.H., Keith, L.H., Kopfler, F.C., Lingg,
R.D., McCabe, L.J., Melton, R.G., and Mullaney, J.L., "Organic
Compounds in Drinking Water: A Five-City Study," In preparation (1975).
5. Bellar, T.A., and Lichtenberg, J.J., JAWWA 66, 739 (1974).
6. Kopfler, F.C., Melton, R.G., Lingg, R.D., and Coleman, W.E., "GC/MS
Determination of Volatiles for the National Organics Reconnaissance
Survey of Drinking Water," to be presented at the First Chemical
Congress of the North American Continent, Mexico City, December 1975.
7. Lingg, R.D., Melton, R.G., Kopfler, F.C., and Coleroan, W.E... "GC/MS
Techniques for the Quantification of Volatile Organics in Tap Water,"
In preparation (1975) .
8. "Specific Ion Mass Spectrometric Detection for Gas Chromatographic
Pesticide Analysis," U.S. Environmental Protection Agency Report,
EPA-660/2-74-004, U.S. Government Printing Office, Washington, D.C.
(1974) .
9. "System/150 GC/MS Data Processor," System Industries, Inc., Sunnyvale,
California (1974).
10. Lingg, R.D., Melton, R.G., Coleman, W.E., and Kopfler, F.C., Proceedings
AWWA Water Quality Technology Conference, Dallas, Texas (1974).
11. Eichelberger, J.W. and Budde, W.L., "Expanding the Range of Con-
centrations Amenable to Gas Chromatography/Mass Spectrometry in a
Single Injection," presented at the 22nd Annual Conference on Mass
Spectrometry and Allied Topics, Philadelphia, Pa., May (1974).
12. Stenhagen, E., Abrahamsson, S., and McLafferty, F.W., "Registry of
Mass Spectral Data," John Wiley and Sons, New York (1974).
-------
13. Symons, J.M., Bellar, T.A., Carswell, J.K., DeMarco, J., Kropp, K.L.,
Robeck, G.G., Seeger, D.R., Slocum, C.J., Smith, B.L., and Stevens,
A.A., JAWWA 67_, 634 (1975) .
14. "Preliminary Assessment of Suspected Carcinogens in Drinking Water"
(with Appendices) Interim Report to Congress, U.S. Environmental
Protection Agency, Washington, D.C., 20460, June (1975).
15. Bunn, W.W., and Kleopfer, R.D., "Analyses of Drinking Water for
Organic Compounds," to be presented at the First Chemical Congress
of the North American Continent, Mexico City, December (1975).
-------
LIST OF ILLUSTRATIONS
TABLES:
1. Chemical Properties of Finished Water in Five Cities
2. Mass Spectrometer Operating Parameters for the 500-ml VOA
3. Typical Manual GC/MS Operations for a 500-ml VOA
4. Occurrence of Functional Groups in 72 Organics
5. Elution Order of 72 Organics
6. Organic Compounds Common to All Five Cities
7. Summary of 72 Volatile Organic Compounds
8. Quantitative Analysis of Selected Organics from Five-City Survey
9. Residual Chlorine and pH Measurements of Spiked Seattle Water
-------
CHEMICAL PROPERTIES OF FINISHED WATER IN FIVE CITIES
NON-VOLATILE
TYPE OF
CITY SUPPLY
CINCINNATI SURFACE
MIAMI GROUND
OTTUMWA SURFACE
PHILADELPHIA SURFACE
(TORRESDALE
PLANT)
SEATTLE SURFACE
TYPE OF TOTAL ORGANIC
RAW WATER CARBON - mg/l
INDUSTRIAL 1.3
WASTE
NATURAL 6.5
WASTE
AGRICULTURAL 2.3
WASTE
MUNICIPAL 1.9
WASTE
NATURAL 1.0
WASTE
CONDUCTIVITY CHLORINE
pMHOS/CM mg/l PH
295 2.7 8.6
350 2.3 8.7
500 1.4 9.2
260 2.0 8.3
50 0 6.6
-------
MASS SPECTROMETER OPERATING PARAMETERS
FOR THE 500-ML VOA
IONIZATION POTENTIAL 70 eV
EMISSION CURRENT 500 ^a
ION ENERGY 4V
REPELLER POTENTIAL 6V
LENS POTENTIAL 100V
ANALYZER TEMPERATURE 70C
MULTIPLIER VOLTAGE 2KV
ANALYZER PRESSURE SxlQ-^TORR
OUTPUT PREAMPLIFIER lQ-7amps/V
MASS RANGE SETTING MEDIUM (10-250)
SEPARATOR OVEN TEMPERATURE »230 C
TRANSFER LINE TEMPERATURE «225 C
-------
TYPICAL MANUAL GC/MS OPERATIONS
FOR A 500-ML VOA
OPERATION
1. PURGE TRAP
2. HEAT GC OVEN TO 120
3. START COMPUTER SCAN
4. HOLD GC OVEN
TEMPERATURE AT 120
5. START TEMPERATURE
PROGRAM AT 20C/min.
6. TERMINATE RUN
TIME(MINUTES) GC OVEN TEMPERATURE(C)
0 to 4
4 to 6
AFTER WATER ELUTES
«9 min.)
6 to 16
16
50 - 60
ROOM
ROOM to 120
120
120
120 to 200
200
N.B. - GC INJECTOR TEMPERATURE = 150C
-------
OCCURRENCE OF FUNCTIONAL GROUPS
IN 72 ORGANICS
FUNCTIONAL GROUP PRESENT IN PERCENT OF
72 ORGANICS 72 ORGANICS
HALOGENATED AROMATIC 5 7
HALOGENATED ALIPHATIC 33 46
AROMATIC RING 9 13
NITRO GROUP 3 4
SULFUR ATOM 2 3
ALCOHOL GROUP 4 6
ALDEHYDE OR KETONE GROUP 12 17
ESTER GROUP 2 3
NITRILE GROUP 6 8
ETHER GROUP 3 4
-------
Elutlon Order* of 72 Organlcs
1.
2.
3.
i».
5.
6.
7.
8.
9.
10.
11.
12.
13.
Name
Chloroethyne
(Chloroacetylene)
Chloromethane
Dlmethy lether
Methanol
Cyanogen chloride
Chloroethene
(Vinyl chloride)
Acetal dehyde
(Ethanal )
Formic acid, methyl
ester (Methyl formate)
Bromome~thane
Bromoethyne
Chloroethane
Ethanol
Dlchlorof luorome thane
Hoi ccular
Weight
60
50
U6
32
61
62
1*1*
60
91*
101*
6U
1*6
102
Number
of
Carbons
2
1
2
1
1
2
2
2
1
2
2
2
1
BoH Ing
Pojnt(C)
-32
24
-24
65
14h
-14
21
71
5
5
12
79
9
Ions Used For
Mass
Chroma tograms
m/e
60
50
1*6
31
61
27,62
29,1*1*
31,60
9U
101*
61*
31
67
Relative
Retent Ion
Time +
O.Ol*
0.05
0.06
0.10
0.15
0.16
0.18
0.29
0.31
0.37
0.38
0.39
O.UO
Source of
Standards §
**
a
a
h
I
I
b
c
1
**
a
g
j
-------
Elution Odor* of 72
11*.
15.
16.
17.
18.
19.
20.
21.
22.
23.
2i».
25.
26.
Name
Fl uorotr i chlorome thane
D ichloroethyne
(Dichloroacetylene)
Methyl cyanide
(AcetonI fi le)
Propanal
Pentane
2-Propanone (Acetone)
1,1-DichIoroethene
(Vinylidene chloride)
Diethyl ether (Ethyl ether)
2-Chloropropane
Di chl oromethane
(Methylene chloride)
Formaldehyde, dimethyl
acetal (Dimethoxymethane)
Acetic acid, methyl
ester (Methyl acetate)
lodomethane (Methyl Iodide)
Molecular
We i gh t
136
9i»
1*1
58
72
58
96
7k
78
84
76
74
11*2
Number
of
Carbons
1
2
2
3
5
3
2
t*
3
1
3
3
1
Bof 1 \nr,
Point(C)
24
82
U9
36
57
36
35
37
i»0
k6
58
1*3
Ions Uspd For
Mass
Chroma tOf;rams
m/e
101
91*
1*1,1*0
29
1*3,72
1*3
61,96
1*5,71*
1*3,78
l*9,8i*
75
1*3,71*
11*2,127
P.el at i ve
Retent ion
Time +
0.60
0.62
0.61*
0.61*
0.65
0.66
0.71
0.71
0.72
0.71*
0.75
0.76
0.77
Source of
Standa rds §
j
**
c
c
h
h
b
c
d
h
b
c
c
-------
Elution Order* of 7Z Crannies
27.
28.
29.
30.
31.
32,
33.
3i».
35.
36.
37.
38.
39.
UO.
Name
Propenolc acid, ntfMe
(Aery loni tr i 1 e)
Carbon dl sul f i de
2 -Me thy !-2-propanol
(t-3utyl alcohol)
trans 1, 2-DI chloroethcne
Nt tromethane
2 -Methyl propanal
1,1-Dlchloroethane
2-f1ethyl propenal
Propanoic acid, nltMle
(Propioni tri le)
els 1,2-Dlchloroethene
3-Hethyl -2-butanone
2-Butanone
Bromoch loromethane
Trlchloromethane
Mo I ecu lar
We i gh t
53
76
7k
96
61
72
98
70
55
96
86
72
128
118
Number
of
Carbons
3
1
k
2
1
k
2
k
3
2
5
it
1
1
Boiling
Point (C)
77
U7
83
ue
101
61
5T
68
97
59
93
80
67
61
Ions Used For
Mass
Chromatograms
m/e
53,27
76
59
61,96
30,l»6
l»3,72
63,98
70
54
61,96
U3,86
U3,72
128
83
Rel a t i ve
P.etent Ion
Time +
0.77
0.80
0.81
0.84
0.86
0.89
0.91
0.91
0.91
0.95
0.95
0.96
0.99
1.00
Source of
Standards $
C
I
C
b
c
b
b
b
c
c
b
b
b
1
(Chloroform)
-------
Elution Order* of 72 Grannies
1*1.
«*2.
U3.
1*1*.
kS.
1*6.
U7.
1*8.
1*9.
50.
51.
52.
53.
Name
2-Methy 1 propanoic acid,
nitrile ( 1 sobutyoni trlle)
1-Butanol
2-Butenal
1,1, l-THchlo roe thane
1,2-Dlchloroethane
Tetrachlorome thane
(Carbon tetrachlor Ide)
3-Methylbutanal
( 1 soval eraldehyde)
Benzene
Tr ichloroethene
(Trichloroethylene)
2-Pentanone
Bromodi ch lorome thane
Di bromomethane
3-Methylbutanolc acid,
n I +• i» I 1 *» fl«*At*-«1,M.M.M...Ifc.*f1_\
Molecular
We I gh t
69
71*
70
132
98
152
86
78
130
86
162
172
83
Number
of
Carbons
i*
i*
l*
2
2
1
5
6
2
5
1
1
5
Boil Ing
Point(C)
108
118
101
7»
8 If
77
9J
80
87
102
90
97
129
Ions Used For
Mass
Chromatograms
m/e
1*2,68
31,56
1*1,70
61,97
27,62
117,82
<*<*,86
78
95,130
1*3,86
83
171*
1*3,68
Relat ive
Retention
Time +
1.07
1.10
1.10
1.10
1.11
l.li*
1.15
1.17
1.21
1.22
1.29
1.29
1.37
Source of
Standards §
c
b
b
c
b
c
b
C
f
c
d
d
e
-------
Elutlon Order* of 7i Orp;anlcs
51».
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
Name
Dichloronl tromethane
4-Methy J -2-pentanone
Dimethyld! sul fide
(2,3-Th!abutane)
NI troflchlorome thane
Tol uene
1,1, 2-Tr Ichlo roe thane
2, 6 rD I methyl -4-heptanone
TetraChloroethene
Chlorodi bromome thane
Chlorobenzene
Dichloroiodome thane
Ethy Ibenzene
Bromotr i chloroethene
TrI bromomethane
Molecular
We ! gh t
129
100
9i*
163
92
132
142
164
206
112
210
106
208
250
Number
of
Carbons
1
6
2
1
7
2
9
2
1
6
1
8
2
1
Roil In*
Point(C)
107
119
117
111
111
111*
168
121
119
132
132
136
150
Ions Used For
Mass
Chromatograms
m/e
83,30
43,100
94,79
30,117
91
97,83
57,85
166
129
112
83,175
91,106
129,208
173
Relative
Retention
Time +
1.40
1.41
1.43
1.48
1.49
1.50
1.52:}:
1.54
1.64
1.86
1.90
1.93
2.10
2.31
Source of
Standards§
**
c
e
e
c
c
b
b
d
c
**
c
**
c
(Bromoform)
68. Azulene
128
10
128
-------
Elution Ordpr* of 72
Name Molecular Number Boiling Ions Used For Relative
Weight of Point(C) Mass Retention
Carbons Chromatograms Time +
m/e
69.
70.
71.
72.
Chlorotol uene, an Isomer of 126
1,3-Dichlorobenzene 146
(m-D ichlorobenzene)
1, l»-Dichlorobenzene 1^6
(p-D Ichlorobenzene)
1,2-DIchlorobenzene 1U6
7 159-162 91,126
6 172 H»6
6 17k 146
6 176 1^6
2.59
3.12
3.22
3.56
Source of
Standards §
c
c
c
c
(o-DIchlorobenzene)
* 10 Ft. Chromosorb 101 Column
+ Retention time of trIchloromethane Is approximately t minutes.
Time zero equals the beginning of data acquisition; 'elatlve retention time
of chloroform = 1.00.
^ 10 Ft. Tenax GC Column.
§ Company: a. Air Products and Chemicals Inc.
b. Aldrlch Chemical Co.
c. Chemical Services Inc.
d. Columbia Organic Chemicals Co.
e. Eastman Organic Chemicals Co.
f. Fischer Scientific Co.
g. HSHHA S. C. Co.
h. Burdick and Jackson Inc.
I. Matheson Gas Products Co.
j. PCR Inc.
** No Source Found
-------
ORGANIC COMPOUNDS COMMON TO ALL 5 CITIES
ACETALDEHYDE
BROMODICHLOROMETHANE*
2-BUTANONE
CHLORODIBROMOMETHANE*
CHLOROMETHANE*
DICHLOROMETHANE*
ETHANOL
3-METHYLBUTANAL
3-METHYL-2-BUTANONE
2-METHYLPROPANAL
PROPANAL
2-PROPANONE (ACETONE)
TRICHLOROMETHANE*
•HALOGENATED
-------
SUMMARY OF 72 VOLATILE ORGANIC COMPOUNDS
RANGE AVERAGE MEDIAN
BOILING POINT (C) -32 to 176 78 77
MOLECULAR WEIGHT 32 to 250 101 92
CARBON NUMBER 1 to 10 3 2
-------
QUANTITATIVE ANALYSIS OF SELECTED ORGANICS
FROM FIVE-CITY SURVEY
CONCENTRATION UG/L) (PPB)
CINCINNATI MIAMI OTTUMWA PHILADELPHIA SEATTLE
BENZENE
BROMODICHLOROMETHANE
CHLOROBENZENE
CHLORODIBROMOMETHANE
1,1-DICHLOROETHENE
(VINYLIDENE CHLORIDE)
CIS 1,2-DICHLOROETHENE
TRANS 1,2-DICHLOROETHENE
NITROTRICHLOROMETHANE
(CHLOROPICRIN)
TETRACHLOROETHENE
(TETRACHLOROETHYLENE)
TOLUENE
TRICHLOROETHENE
(TRICHLOROETHYLENE)
TRICHLOROMETHANE
0.3
15
0.1
3
+
<0.1
•
3
0.3
<0.1
0.1
38
<0.1 <0.1
73 +
1 •
32 +
0.1
14
1 •
0.4 *
<0.1 0.2
* •
0.3 <0.1
301 1
0.2
20
<0.1
5
<0.,
0.1
•
2
0.4
0.7
0.5
65
.
4
•
3
•
•
•
•
•
•
•
21
(CHLOROFORM)
+ . DETECTED BUT NOT QUANTIFIED
• - NOT DETECTED BY GC/MS
-------
RESIDUAL CHLORINE AND pH MEASUREMENTS
OF SPIKED SEATTLE WATER
AMOUNT OF
CHLORINE
INITIALLY
ADDED
mg/l
0
1.12
1.12
REACTION
TIME
0 MIN
15 MIN
4 DAY
REACTION
TEMPERATURE
C
25
25
25
pH
6.6
5.0
5.5
TOTAL
CHLORINE
RESIDUAL
mg/l
0
1.08
1.02
-------
FIGURES:
1. 500-ml VOA (first purge at 95C) of (A) Miami finished water and
(B) blank water. C through G are mass chromatograms of (A):
in (C) m/e=29, 43, and 68; CD) m/e=47; (E) m/e=61 and 62;
(F) m/e=77 and 91; (G) m/e=79 and 81.
2. Percentage Occurrence of 72 Organics in Finished Water from
Seattle (S), Ottumwa (0), Philadelphia (P)» Cincinnati (C),
and Miami 01) .
3. Reconstructed GC/MS Chromatograms of (A) Seattle Finished Water
Spiked with Six Organics and Ferric Chloride (1 mg/1) and (B)
with Chlorine (1.12 mg/1) added to (A).
4. Mass Spectrum (143-140) of Dichloronitromethane, which was formed
by the Reaction of Nitromethane (50 yg/1) and Chlorine (1.12 mg/1)
in Seattle Finished Water.
5. Mass Spectrum (148-145) of Trichloronitromethane, which was
Formed by the Reaction of Nitromethane (50 yg/1) and Chlorine
(1.12 mg/1) in Seattle Finished Water.
£. Mass Spectnun (223-216) of an isoraer of ChlflrOxyUne, which was
formed by the Reaction of m-Xylene (50 ug/1) and Chlorine
(1.12 mg/1) in Seattle Finished Water.
7. Mass Chromatograms Derived by Extracted Ion Current Searches
-------
1 — 1 — I — 1 — I — 1 — 1 — T-
A
n — i —
A
T — i — i — i — i — i — i — i — i — i — r i
(G) MC FOR BROMOMETHANES
t\
*
(f) MC FOR AROMATICS AND
1. BROMOMETHANES
\ AA
^^V
(E) MC FOR CHLOROETHANES AND
CHLOROETHENES
(D) MC FOR CHLOROMETHANES
(C) MC FOR ALCOHOLS, ALDEHYDES,
KETONES AND NITRILES
TiTRGC OF BLANK
(A) RGC OF MIAMI FINISHED WATER
_y\ —————=====.
400 450 500
T 1 1 1 1 1 I I
250 300 350
SPECTRUM NUMBER
500-ml VOA (FIRST PURGE AT 95 °C) OF (A) MIAMI FINISHED
WATER, (B) BLANK WATER, AND MASS CHROMATOGRAMS
C THRU G OF (A) FOR (C) m/e 29, 43, AND 68, (D) m/e 47,
(E) m/e 61 AND 62, (F) m/e 77 AND 91, AND (G) m/e 79
AND 81.
-------
100
90,
w—
U 80
X
u
< ™
U, 70
z
1— 60
Z
UJ
to
ill
Ui
OC 50
0.
»- 40
O
n
•••
O 30
H-
z
UJ
-------
c
D
20 40 60 80
SPECTRUM NUMBER
100 120 140 160 180 200 220 240
Fig-3
RECONSTRUCTED GC/MS CHROMATOGRAMS OF (A) SEATTLE
FINISHED WATER SPIKED WITH SIX ORGANICS AND FERRIC
CHLORIDE (1 mg/l), AND (B) WITH CHLORINE (1.12 mg/l)
ADDED TO (A).
-------
100-
< 80-
LU
O.
LU ""
«/i
2 60-
UL
O
LU
< 40-
^
Z
u
£ 20-
•MM
0-
30
1 '
1
10 30
M/E
1
1
8
48
j
,1
II
I
3
85
87
1 1 1 1 1 1 1 1 1 I 1 1
-15
_j
*^
>—
-10 2
u_
0
LU
O
H-
z
— 5 uj
ai
o.
— 0
50 70 90 110 130 150 170 190
MASS SPECTRUM (143-140) OF DICHLORONITROMETHANE WHICH
WAS FORMED BY THE REACTION OF NITROMETHANE (50 jjg/l)
AND CHLORINE (1.12 mg/l) IN SEATTLE FINISHED WATER.
-------
100 -,
< 80-
a.
LU —
ca 60-
LL.
O
< 40-
z
LU —
U
£ 20~
0-
1
*
h
JO
I
0 30
M/E
47
III ,
ll,
1 1 1
50
,
11
. .. •ill
III!
1 ll .
17
119
121
123
1 1 1 1 1 1 1 1 1 1 1
70 90 110 130 150 170 IS
-10
^
O
u.
O
-5 g
Z
LU
U
LU
a.
-0
>0
SPECTRUM (148-145) OF TRICHLORONITROMETHANE WHICH
WAS FORMED BY THE REACTION OF NITROMETHANE (50 pg/l)
AND CHLORINE (1.12 mg/l) IN SEATTLE FINISHED WATER.
-------
105
lOO-i
< 80-
LU
OL
LU ""
V)
2 60-
Um
O
LU
2 40-
<
H-
Z
LU
u
£ 20-
Q.
«•
0-
«;i
ta
39
27
1
10
J
I
30
w •
77
63
I i
1
50
il i n,i
i *l III 1
• ^^ «^
140
125
1 l!27 ,
1 1 1 1 1 1 1
70 90 110 130
142
,1
1 1 1 1
-20
•
"15S
o
u_
o
-10 ^
o
1—
z
LU
U
-5 £
Q.
150 170 190
M/E
MASS SPECTRUM (223-216) OF AN ISOMER OF CHLOROXYLENE
WHICH WAS FORMED BY THE REACTION OF m-Xylene (50 jjg/l)
AND CHLORINE (1.12 mg/l) IN SEATTLE FINISHED WATER.
-------
RGC
I || TETRACHLOROETHENE
1 Ml
MC30
I DICHLORONITROMETHANE
NITROTR 1C HLO ROM ETHANE
(CHLOROPICRIN)
TTTr TTr »|"TT,I,,..
0160 0170 0180 0190 0200 0210
CHROMATOGRAMS DERIVED BY
EXTRACTED ION CURRENT SEARCHES
-------
APPENDIX:
Additional confirmation of organics identified in 500-ml
VGA tap water samples and blank water was carried out by direct
aqueous injection (DAI) (5yl) and head space analysis (HSA).
The procedure of Harris, I.E., Budde, W.L,, and Fichelburger,
J.W. (Anal Chem, 46_, 1912 (1974) was used for DAI. A similar
procedure was followed for HSA except that 2.5 ml of head space
air from 140-ml serum bottles (2.5 ml of head space above water)
was injected into the GC/MS with a valve controlled 5-ml gas syringe
(Precision Sampling Corporation). For additional HSA sensitivity,
3c
larger serum bottles (500 ml with 4# mm crimp-on seals) are
available from Wheaton Scientific, 1000 North 10th Street, Millville,
N.J. 08332.
Figures A and B illustrate the results of DAI and HSA of
Miami water as compared with the 60 organics identified by 500-ml
VGA (see 500-ml VGA chromatograms in attached paper by Kopfler
et. al.) The quantitative results in the attached paper by
Coleman e£ al. (Table 8) illustrate the comparative sensitivity
of the three methods. For example, <2-£s-l,2-dichloroethene
(14 yg/1), trichloromethane (301 yg/1), and bromodichloromethane
(73 yg/1) were identified in DAI samples of Miami tap water
above levels present in blank water (Figure A). In addition to
these three compounds, 1,1-dichloroethene (0.1 yg/1) and
trichloroethene (0.3 yg/1) were also identified in HSA samples
-------
2.
above blank water levels (Figure B). Therefore, for the compounds
identified in Miami tap water, we realized a sensitivity of
14 yg/1 using DAI, 0.1 yg/1 using HSA, and less than 0.1 yg/1
using 500-ml VOA for some selected organics.
The overall sensitivity for each method is dependent upon
the physical properties (water solubility, vapor pressure, etc)
of each organic compound. We observed no effect in the 140-ml
VOA quantitation of 50 yg/1 of trichloromethane solutions that
varied in Na£S04 salt concentration between 0 and 500 yMHOS/cm
of conductivity.
The occurrence of organics in tap water above blank water
levels from the five surveyed cities using DAI, HSA and 500-ml
VOA methods is given in the attached table.
-------
TABLE
List Of Compounds Detected By 500-ml VOA
Compound Name
1.
2.
3.
4.
5.
6.
7-
8.
9.
10.
11.
12.
13.
14.
15-
16.
17.
18.
19.
20.
21.
Acetaldehyde (Ethanal)
Acetic acid, methyl ester
(Methyl acetate)
Azulene
Benzene
Bromochlorome thane
Bromodichlorome thane
Bromoethyne
(Bromoacetylene)
Bromomethane
Bromotrichloroethene
1-Butanol
2-Butanone
2-Butenal
Carbon disulfide
Chlorobenzene
Chlorodibromome thane
Chloroethane
Chloroethene
(Vinyl chloride)
Chloroethyne
(Chloroacetylene)
Chloromethane
2-Chloropropane
Chlorotoluene, An
City*
M S 0 P C
S
s
M PC
M
MfS 0 PtCt
M
M C
M C
C
M S 0 P C
M
M PC
M PC
MtS 0 P C
M S PC
]V[t p
M PC
MfS 0 P C
C
M
Confirmatory Mol
Techniques§ Wt .
abed
abed
c d
abed
abed
abed
c d
c d
d
abed
abed
c d
abed
abed
abed
abed
abed
c d
abed
abed
a c d
44
74
128
78
128
162
104
94
208
74
72
70
76
112
206
64
62
60
50
78
126
Isomer of
-------
Compound Name City* Confirmatory Mol.
Techniques § Wt.
22. Cyanogen chloride . MOPC abed 6l
23. Dibromomethane M abed 172
24. 1,2-Dichlorobenzene M PC abed 146
(a-Dichlorobenzene)
25. 1,3-Dichlorobenzene M PC abed 146
(m-Dichlorobenzene)
26. 1,4-Dichlorobenzene M PC abed 146
(p-Dichlorobenzene)
27. 1,1-Dichloroethane M abed 98
28. 1,2-Dichloroethane M PC abed 98
29. 1,1-Dichloroethene Mf PC abed 96
(Vinylidene chloride)
30. cis 1,2-Dichloroethene Mt PC abed 96
31. trans 1,2-Dichloroethene Mf abed 96
32. Dichloroiodomethane M P C c d 210
33. Dichloroethyne M P C c d 94
(Dichloroacetylene)
o
34. Dichlorofltfromethane M C abed 102
35. Dichloromethane MSOPC abed 84
(Methylene chloride)
36. Dichloronitromethane C c d 129
37. Diethyl ether M PC abed 7^
(Ethyl ether)
38. Dimethyl ether SPC abed 46
(Methyl ether)
39. Dimethyldisulfide 0 abed 94
(2,3-Thiabutane)
40. 2,6-Dimethyl-4-heptanone P c d 142
41. Ethanol MSOPC abed 46
42. Ethylbenzene M PC abed 106
-------
Compound Name City* Confirmatory Mol.
Techniques § wt.
43. Pluorotrichloromethane 0 P C abed 136
44. Formaldehyde, dimethyl M p abed 76
acetal (Dimethoxmethane)
45. Formic acid; methyl ester S abed 60
(Methyl formate)
46. lodomethane M abed 142
(Methyl iodide)
47. Methanol MSPC abed 32
48. 3-Methylbutanal MSOPC abed 86
(Isovaleraldehyde)
49- 3-Methylbutanoic acid, MSPC abed 83
nitrile (Isovaleronitrile)
50. 3-Methyl-2-butanone MSOPC abed 86
51. Methyl cyanide M c d 4l
(Acetonitrile)
52. 4-Methyl-2-pentanone M PC abed 100
(Methyl isobuyl ketone)
53- 2-Methylpropanal MSOPC abed 72
(Isobutyraldehyde)
54. 2-Methylpropanoic acid, MSPC abed 69
nitrile (Isobutyronitrile)
55. 2-Methyl-2-propanol MSPC abed 74
(t-Butyl alcohol)
56. 2-Methylpropenal M P c d 70
57. Nitromethane M PC abed 6l
58. Nitrotrichloromethane M PC abed 163
(Chloropicrin)
59. Pentane MSPC abed 72
60. 2-Pentanone MOPC abed 86
61. Propanal MSOPC abed 58
62. Propanoic acid, nitrile M abed 55
(Propionitrile)
-------
Compound Name
63. 2-Propanone (Acetone)
64. Propenoic acid, nitrile
(Aerylonitrlie)
65. Tetrachloroethene
(Tetrachloroethylene)
66. Tetrachloromethane
(Carbon tetrachloride)
67. Toluene
68. Tribromomethane
(Bromoform)
69. 1,1,1-Trichloroethane
70. 1,1,2-Trichloroethane
71. Trichloroethene
(Trichloroethylene)
72. Trichloromethane
(Chloroform)
City*
MfS 0 P C
M
M 0 P C
M 0 P C
Confirmatory Mol.
Techniques § Wt.
abed 58
abed 53
abed 164
abed 152
M
M
M1"
P
P
0 P
0 P
C
C
C
C
MtfS+0 P^Ct
a
a
a
a
a
a
b
b
b
b
b
b
c
c
c
c
c
c
d
d
d
d
d
d
92
250
132
132
130
118
*Cities Tested: Miami, Fla. (M)
Seattle, Wash. (S)
Ottumwa, Iowa (0)
Philadelphia, Pa. (P)
Cincinnati, Ohio (C)
tCompound also found by headgas analysis.
tCompound also found by direct aqueous injection.
§ Confirmatory techniques:
a Mass spectrum of compound matches mass spectrum of standard
compound run on same^instrument under same GC/MS conditions.
b. The relative retention time (RRT) of compound agrees with that
of a standard compound run under same GC conditions.
c. Mass spectrum agrees with spectrum found in literature
-------
d. Mass spectrum agrees with our interpretation. (In cases
where this is the only confirmatory technique, an authentic
standard was not available at the time, nor was there a
spectrum in the literature).
-------
100
20 40
SPECTRUM NUMBER
80
100
FIG./4 DIRECT AQUEOUS INJECTION RGC OF
(A) MIAMI FINISHED WATER, (B) BLANK
WATER, AND (C) A MASS CHROMATOGRAM
OF (A) WHICH SHOWS THE PRESENCE OF
CHLORINATED ORGANICS.
-------
I
20 40
SPECTRUM NUMBER
I
60
I
80
100
FIG.B. HEAD GAS ANALYSIS RGC OF (A) BLANK
WATER, (B) MIAMI FINISHED WATER, AND
(C) A MASS CHROMATOGRAM OF (B)
WHICH SHOWS THE PRESENCE OF
CHLORINATED ORGANICS.
-------
ACKNOWLEDGEMENTS:
The authors thank Mr. Donald E. Mitchell, Mr. William H.
Kaylor, and Dr. Richard K. Schoenig of this laboratory for
considerable technical assistance in sample analysis and the
assembling of data for this report. We also thank Mr. Thomas A.
Bellar of the EPA Environmental Monitoring and Support Laboratory,
Mr. Alan A. Stevens and Dr. Robert B. Dean of EPA Municipal
Environmental Research Laboratory for their technical suggestions
and moral support.
*USGPO= 1976-657-695/5413 Region 5-11
------- |