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

-------