Halocarbon Production from Oxidative
Biocides in Estuarine Waters
Maryland Univ.
College Park
Prepared for

Environmental Research Lab.
Gulf Breeze, FL
Feb 81

-------
                                     P8*l-t57977
                                     EPA 600/3-81-010
                                     February 1981
HALOCARBOK PRODUCTION FROM OXIDATIVE
    BIOCIDES IN ESTUARINE WATERS
                 by
           George R. Helz
            Rong Yew Hsu
            Richard Sugam.
       University of Maryland
    College Park, Maryland 20742
         Grant R803839-01/02
           Project Officer

          William P. Davis
     Bears Bluff Field Station
Wadmalav  Island, South Carolina 29487
  ENVIRONMENTAL RESEARCH LABORATORY
 OFFICE OF RESEARCH AND DEVELOPMENT
U.S.. ENVIRONMENTAL PROTECTION AGENCY
        GULF BREEZE, FLORIDA
                32561
       •rmwrt* IT

-------
                              TECHNICAL REPORT DATA
                        /Please read Instructions an the rri-enc be/orr
1  REPORT NO
  EPA-600/3-81-010
                                                   3. RECIPIENT'S ACCESSION NO.
                               1579?7
4. TITLE ANO SUBTITLE

  Halocarbon  Production from Oxidative Biocide?
  in Estuarine  Waters
                        5. REPORT DATE

                         FEBRUARY  1981 _ISS:JING_SAT£;
                        6. PERFORMING ORGAN NATION CODE
7. AUTHORISI

  G.R. Helz,  R.Y.  Hsu and R. Sugam
                                                   8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME ANO AOORESS
                                                   10. PROGRAM ELEMENT NO.
  University  of Maryland
  College  Park, Maryland
20742
                        11. CONTRACT/GRANT NO.
                                                    R803839-01/02
 12. SPONSORING AGENCY NAME ANO AOORESS
  Environmental Research Laboratory
  Office of  Research and Development
  U.S. Environmental Protection Agency
  Gulf Breeze,  Florida, 32561
                        13. TYPE OF REPORT ANO PERIOD COVERED
                         Final 8/1/75-7/31/77
                        14. SPONSORING AGENCY COOE
                          EPA/600/04
 IS. SUPPLEMENTARY NOTES
| ,6. ABSTRACT
     The  formation of halo-organic  compounds by chlorination of estuarine
  waters  has  been investigated  under both laboratory and field conditions.
  Haloforms are readily generated  in the laboratory with chlorine doses of
  1 to  10  mg/1, the range employed  by many coastal power plants.   At sali-
  nities  above 1 g/kg, Br is the principal halogen in the haloform products
  On a  molar  basis, more than 4% of the chlorine was converted to haloforms
  in some  tests.  Ozone in the  laboratory also generated haloforms in
  estuarine water; the yields were  similar to those obtained from chlorine.
  However, only traces of haloforms were found in a power plant field site,
  where apparently haloform-bypassing reactions consume free chlorine much
  faster  than in the laboratory.   Identification of these reactions is un-
  certain, but they may involve formation of stable halogenated macromole-
  cules.
     A  large  sewage treatment plant served as a volatile halocarbon source
  to study the fate of these compounds.  The major loss mechanism appears
  to be volatilization to the atmosphere.  Rates for this process are esti-
  mated.   However, there appears tcr be some loss under  winter i&£e cover,
  perhaps  because of chemical or biological degradation.
 17.
                            KEY WORDS ANO DOCUMENT ANALYSIS
                DESCRIPTORS
                                        b.IDENTIFIERS/OPEN £NO£D TERMS  C. COSATI [• Icld/Cfoup
Halogens
Halogen organic compounds
Halogenation
Chlroination
Estuaries
Ozone
Volitization
                                        Haloform Products
                                        Sewage Treatment Plant
                                        Chemical Biological Degra-
                                         dation
                                     06 /A
                                     07/C
 is. DISTRIBUTION
                                        19. SECURITY CLASS i
                                    21. NO. OF PAGES

                                       34
  Release  Unlimited
                                        20 SECURITY CLASS/ 7.
                                                               J2. PRICE

-------
                                 DISCLAIMER

     This report has been reviewed and approved for publication by the Gulf
Breeze Environmental Research L^'ucratoi /, U.S. Environmental Protection
Agency.  Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trades names or commercial products constitute endorsement or
recommendation for use.
                                     11

-------
                               FOREWORD
     The protection of our estuarine and coastal areas from damage caused
by toxic organic pollutants requires that regulations restricting the
introduction of these compounds into the environment be formulated on a
sound scientific basis.  Accurate information describing dose-response
relationships for organisms and ecosystems under varying conditions is
required.  The Environmental Research Laboratory, Rulr Breeze, contributes
to this information through research programs aimed at determining:


     "the effects of toxic organic pollutants on individual
      species and communities of organisms;


     "the effects of toxic organics on ecosystem processes
      and components;


     "the significance of chemical carcinogens in the estuarine
      and marine environments.
     The production and fate of halo-organic compounds that are by-products
of .disinfection/biofouling control processes are investigated in the ra-
search project described in this report.
                                     r      (I
                                    Henry F.^Enos
                                    Director
                                    Environmental Research Laboratory
                                    Gulf Breeze, Florida

-------
                                  ABSTRACT

     The formation of halo-organic compounds by chlorination of estuarine
waters has been investigated under both laboratory and field conditions.
Haloforms are readily generated in the laboratory with chlorine doses of 1 to
10 mg/1, the range employed by many coastal power plants.  At salinities
above 1 g/kg (i.e., one part per thousand), Sr is the principal halogen in
the haloform products.  On a molar basis, more than 4% of the chlorine was
converted to haloforms in some tests.  In the laboratory, ozone also generated
haloforms in estuarine water; the yields were similar to those obtained from
chlorine.  However, only traces of haloforms were found at a power plant field
site, where halofonn-bypassing reactions apparently consume free chlorine much
fatter than in the laboratory.  Identification of these reactions is uncertain,
but they may involve formation of stable halogenated macromolecules.

     A large sewage treatment plant served as a volatile halocarbon source to
study the fate of these compounds.  The major loss mechanism appears to be
volatilization to the atmosphere.  Rates for this process are estimated.  How-
ever, there appears to be.some loss under winter ice cover, perhaps because
of chemical or biological degradation.

     This report was submitted in fulfillment of U.S.. Environmental Protection
Agency Grant No. R803839-01/02 to the University of Maryland.  It covers the
period August 1, 1975 to July 31, 1977; the work was completed December 14,
1977.
                                     iv

-------
                                  CONTENTS
Foreword	   ill
Abstract	    iv
Figures	    vi
Tables	   vii
Acknowledgments	viii
     1.  Introduction .....  	    1
     2.  Conclusions  	    2
     3.  Laboratory Experiments  on Volatile Halocarbon
         Production	    3
            Analytical Method for Volatile Halocarbons  	    3
            Laboratc—y Chlorination Experiments 	    5
            Comparison of Clilorination and Ozonation	   11
     A.  Field Studies	   14
            Chalk Point Power Plant 	   14
            Back River Wastewater Treatment Plant	   21
     5.  Other Activities 	   28
            Workshop	   28
            Publications  	   29
References	   30
Appendix	   34

-------
                                   FIGURES

Number                                                                 Pa8e

   1     Map showing sample locations 	  5

   2     Bromine as a percentage of total halogen in haloform
         products	1

   3     Total haloform yield as a percentage of chlorine dose  ....  8

   4     Haloform yield as a percentage of  chlorine dose vs.  total
         chlorine consumed in 10 minutes as a percentage of
         chlorine dose	  9

   5     Calculated time for 99% consumption of Tree chlorine
         by reaction 1	.10

   6     Decay of chlorine produced oxidants in Patuxent River
         water	15

   7"    ' Effect of temperature on decay in  estuarine water chlori-
         nated in the laboratory	17

   8     Effect of removing organic matter  on decay .......... 19

   9     Back River estuary, Md., showing locations of sampling
         points	21

  10     Halocarbon patterns in Back River	24
                                    vi

-------
                                 TABLES
                                                                  ?age

       Concentration (yg/1) of 100 ug/1 halofora solutions
       one day after addition of adsorbing agents .........  4
2      Halo forms produced in selected estuarine and  coastal
       water samples  .......................   6

3      Yields of bromoform in estuarine water treated  with
       chlorine (as OC1~) and ozone ................  12

4      Exanples of possible oxldant consumption reactions  in
       estuarine water  ...............  .  ......  18

5      Macromolecular organic matter  .... ...........  20

6      Back River halocarbons ...................  22

7      Major industrially produced halocarbons ranked  by release
       rate ............................  25

8      Loss rates for volatile halocarbons  ............  26
                                  vii

-------
                               ACKNOWLEDGMENTS

     We are grateful to a number of our colleagues in the Department of
Chemistry at the University of Maryland, including David Anderson, Michael
Failey, Kenneth Ferri and Franz Kasler.  We owe a special debt to Ronald
Block for his assistance and inspiring enthusiasm.  Potomac Electric Power
Co. personnel were extremely helpful.   Finally, Dr. R. A. Saunders, of the
Naval Research Laboratory, graciously  lent time on his GC-MS instrument and
shared his expertise.
                                   viii

-------
                                  SECTION 1

                                 INTRODUCTION

     Chlorine is injected directly  into marine waters by large coastal power*.
plants that use chlorine to prevent biofouling of their heat exchangers.
Chlorinated effluents are also  injected into marine waters by wastewater
treatment plants and certain  industrial enterprises.  Since the discoveries
of haloforms in chlorinated drinking waters by Rook (1976) and a number of
halo-organics in chlorinated  sewage treatment plant effluents by Jolley (1974) .
there has been concern about  whether such compounds adversely affect aquatic
environments because large volumes  of chlorinated discharges are entering
coastal environments.  For example, in the northern Chesapeake Bay, 2Z of the
non-saline throughput is now  treated wastewater effluent (Brush, 1972).
Based on past rates of growth,  this fraction will reach 30Z or more in another
century.  Power plants currently circulate a volume greater than 10Z of the
non-saline throughput; if past  technology and trends continue, a volume
approaching this throughput will be circulated by the end of this century
(Bongers, 1973).  It is likely  that similar statistics could be cited for
many other major estuaries.   Thus, any persistent, bioactive compound in-
jected into the aquatic environment from these sources may be capable, in
principle, of producing massive alterations in our coastal waters.

     When this project began, no previous work had been done on the formation
of halo-organics through chlorination in marine waters.  Yet there was sub-
stantial evidence that the behavior of chlorine in marine waters differs
from- its behavior in fresh waters because of the involvement of Br~ (which
is ubiquitous in seasalt) in  the halogen chemistry.  Similarly, some
differences were known ';o exist between the organic matter in marine and
fresh water systems.  Thus, research on halo-organic formation in marine
waters seemed warranted.  A distinctive feature of this project is the in-
clusion of a major field study  to explore the production and fate of halo-
carbons in the natural environment.

-------
                                  SECTION 2

                                 CONCLUSIONS

1.   Trihalonethanes (CUClj, CHBrCl2r CHBr2Cl, and CHBr3) were generated at^
the 10 to 1000 pM (~3 to 300 ppb) level when fluvial arvl estuarinc water
samples were chlorinated under laboratory conditions.

2.   When salinity exceeded 5 g/kg (%o) bromoform was virtually the only
trihalomethane product.

3.   No volatile halocarbons other than trihalotnethanes in estuarine water
of 14 g/kg salinity were found at the pM level by headspace analysis.

4.   When both ozone and chlorine were rpplied in estuarine water of 14 g/kg
salinity to produce after 10 minutes a residual of 22 pequiv/liter (0.78 ppm
as chlorine), they produced identical trihalomethene yields.

5.   Field studies at an estuarine power plant that employs continuous
chlorination revealed only traces (-Ipg/liter) of volatile halocarbon pro-
ducts.

6.   No "free chlorine" (as measured amperometrically at pH 7) was observed
beyond the condensers at this plant, and the negligible yields of trihalo-
methanes are tentatively ascribed to the extremely rapid consumption of
free chlorine in the condensers.  More work should be done to test this
hypothesis.

7.   It was possible to trace the halocarbon plume from a wastewater treat-
ment plant more than 5 miles seaward under winter ice, but only a fraction
of this distance in spring when the ice was gone.

-------
                                   SECTION  3

                      LABORATORY  EXPERIMENTS ON VOLATILE
                             HALOCARBON PRODUCTION

     Haloforms,  the  most  abundant  known halo—organic products  from chlori-
nation of  freshwaters,  were a major concern in this project.   In this section,
we describe  our  analytical method  and provide some basic  experimental data on
the nature of  the  halofonns produced by chlorinating estuarine water!?.

ANALYTICAL METHOD  FOR VOLATILE HALOCARBONS

     In  the  field, 40.0 ml of sample was placed in a Hypovial  (nominal size
50 ml; actual  volume when sealed cs described, 60.19 + 0.72 ml; Pierce
Chemical Co.).   If oxidant was present, it was quenched with a drop of
concentrated Na2S202 solution.  Then, a Teflo..-to-rubber  laminated septum was
placed over  the  mouth and sealed in place wit'i a crimped  aluminum ring.  The
sample was stored  on ice  and normally analyzed within 48  hours.  In labora-
tory chlorination  studies, the samples were first treated with the desired
amount of  NaOCl  and  then  sealed  as above.  For analysis,  the vials were
equilibrated at  60°C in a shaking  water bath; 1 ml of the vapor was extracted
through  the  septum cap  with a gas-tight hypodermic syringe, and injected into
a gas chromatograph  equipped with  a Hall Electrolytic Conductivity Detector
(Tracor) .  The following  conditions were normally used: carrier gas, 30 ml
He min~l;  column packing, 60-80  mesh Tenax GC (Applied Science Labs, Inc.);
temperature  in injection  port, 200°, in column, 120°-200° programmed at 8°
min~^, in  transfer line,  250°, in  detector furnance, 850°; detector reaction
gas, 50 ml H2  min~l; detector electrolyte, 50Z n-propanol-water.  Standards
were prepared  in distilled water using aliquots of a primary standard that
contained  1  gram each of  the compounds of  interest, then  brought to 100 ml
with methanol.   Coefficients of  variation, based on repeated analysis of
standards  in the 10^ to 10^ nM range, were about 42.  In  early stages of the
work, some identifications veve  checked by mass spectrpmetry, but the high
selectivity  of  the method for volatile chloro- and bromocarbons minimizes
the danger of  nisidentifying gas chromatograph peaks.

     When  saline samples  are analyzed, using a standard curve prepared from
solutions  of the analyte  in distilled water, the apparent concentration C^,
obtained from  the  standard curve must be corrected for salting cut in the~
sample.  If  CT and CA are the true and apparent original  concentrations of a
compound in  ~tne  tes~t~solution, Vy  and V^ are the volumes  of vapor and liquid,
Hf is the  Henry's  Law constant "expresses' as a concentration ratio, ^ is the
"salinity,  and £  is the  Setchenow coefficient of the compound in water, then
the relationship between  the true  and apparent concentration is:

-------
          c,
'T
         -as  , —
         e   + V.
                           L
                    1 +
                         VL
                                    x C
                                                                        (1)
                                       'A
For the haloforms, Hep is 0.45, 0.25,' 0.17 and 0.02 for CHCls, CHBrCl2,
CHBr2Cl and CHBr3, respectively.  A reasonable numerical value for £ is
0.008 when salinity is expressed in g/kg.  Using these figures in
Equation (1), we found that CT is less than C^ in 35 g/kg seawater by
about 25% for CHC13 and by aBoiit 30% for CHBr3-  The correction factor
diminishes with salinity, and is within the range of analytical uncertainty
for salinities below about 5 g/kg.

     Our headspace method provides a measure of the amount of analyte that
is free and unassociated with molecules other than water.  If part of the
analyte is adsorbed or bound in some non-volatile form, that fraction will
go effectively undetected.  Table 1 presents some data on potential adsorb-
ing agents.  These were added at concentrations higher than would be

          TABLE 1.  CONCENTRATION (yg/1) OF 100 ug/1 HALOFORM SOLUTIONS
          	ONE DAY AFTER ADDITION OF ADSORBING AGENTS	


          Haloform                     Adsoroing Agent
                        Huoic Acid (a)      Fe(OH)3 (b)      Clay (c)

          CHC13           98                  101             99

          CHBrCl2         99                   93             96

          CHBr2Cl        101                   93             95

          CHBr3          100                  , 89             92

          luAldrich Chem. Co. lot no. 121137; 10 mg/1.
          b.  Precipitated from FeCl3 and aged 1 hr before exposure to
              halofonrs; 10 mg/1.
          c.  Wyoming Na-Montmorillonite, Clay Minerals Repository,
              University of Missouri; 20 ug/1.

encountered typically in coastal waters.  Inasmuch ~s the measurement un-
certainty is about 4%, the data for humic acid do not show significant
absorption.  In the presence of Fe(OH>3 and montmorillonite, there is
significant absorption for the heavier haloforms.  However, in most
estuarine and marine waters where the concentrations of these adsorbing
agents would be many times less, the small effects seen in Table 1 would

-------
become negligible.  Thus, we conclude that volatile halocarbons are likely
to be mainly in a free state in natural waters; therefore, concentrations
obtained by a headspace procedure, when corrected by Equation 1, approximate
total concentrations.

     In concluding  this section,  it should be noted that Kopfler et al.  (1976)
and Nicholson et al.  (1977) have  provided evidence for the existence of  a
long-lived halo-organic intermediate in chlorinated natural waters that
decays to release haloforms upon  heating.  Thus, in the case of decay of such
an intermediate, the  quantity of  haloforms detected by our technique may be
greater than the quantity in the  sample at the time it was collected.

LABORATORY CHLORINAT10N EXPERIMENTS

Sampling Sites

     To study halocarbon production by chlorination of coastal waters, a
group of seven  samples was collected from coastal, estuarine and fluvial
sites in Maryland  (Figure 1).  The sites were selected to avoid local sources
of either haloforms or chlorine.  The samples were filtered with 0.45 urn
                                                          DELAWARE
                                                             BAY   (:
                Figure 1.   Map showing  sample  locations.  See Appendix
                for latitude and longitude.

-------
Hillipore filters, and then stored in a cold room  (4°C) prior  to use  in exper-
iments.  Cblorination in the laboratory was conducted with calibrated NaOCl
stock solutions, rather than with Cl2 gas; under the conditions  of the tests,
Cl2 hydrolyzes almost instantly to form OC1~, so our results should be the
same as would be obtained with Cl2-  Total nitrogen data for these samples
were obtained by a UV oxidation-Cd reduction method (Sugam, 1977) . Total
organic carbon determinations were done by a commercial laboratory (Martel
Laboratories, Inc.), using an Oceanography International organic carbon
analyzer.

Results

     Table 2 presents the results of a series of chlorination  experiments in
which samples selected to represent a full range of water conditions, from
fresh water to ocean water, were treated with two  concentrations of NaOCl in

     TABLE 2.  HALOFORMS PRODUCED IN SELECTED ESTUARINE AND COASTAL" WATER
	SAMPLES (1J~9 moles/l)a	

                  	        Sample Number	
Salinity (g/kg)
PH
Org-C (UM)
Tot-N (uM)b
0
7
350
185
.00
.21


1.06
7.94
270
68
3
7
270
58
.68
.94


7
7
280
34
.90
.83


10
7
270
31
.27
.82


30.35
7.36
240
13
31.54
8.10
510
NAC
     NaOCl dose 14 pM (=1 mg/l):d

CHC13            20      15     ND       3      ND     NDC     NA
CHBrCl2          ND      17      7       1      ND     ND      NA
CHBr2Cl          ND      15     28      13       7     ND      NA
CHBr3            ND      22    183     211     122     54      NA
Yield (%)&        0.43    1.48   4.67    4.88    2.76   1.15
Cl Demand (%)*   18      22     74      78      77     55      NA
     NaOCl dose 140 pM (=10 mg/1) :
CHC13
CHBrCl2
CHBr2Cl
CHBr3
Yield (%)
763
287
87
ND
2.43
41
163
564
753
3.25
10
11
126
970
2.39
4
3
42
1160
2.59
8
1
27
900
2.00
6
ND
9
245
0.56
ND
ND
16
880
1.92
                                               .    .         .  ,
    sample.
b,  Tot-N - N0~ + N0~ 4- NH- + Org-N.
c.  NA = not analyzed; ND - not detected.
d.  1 yM NaOCl has the same oxidizing capacity as 1 yM Cl?.
e.  Yield = moles of halogen in haloforms as percent of moles of hypochlorite
    added.  Note that carbon is in excess in these tests.
f.  Cl Demand = percentage of initial dose detectable after 10 min by ampero-
    metric titration at pH 4.0 with excess KI added.
                                     ..6..

-------
the laboratory.  These levels  span  the range of  doses typically  applied by
power plants  and treatment plants.  Haloforms were produced in each case;  no
volatile halocarbons other than halofonns were produced at-the nM level in
any of the  experiments.  In the fresh water sample (No. 1), CHC13 was the
dominant haloform, as has been found in most drinking waters.  Conversely,
CHBr3 overwhelmingly dominated in the ocean samples.  There was  a systematic
shift, as the salinity increased  (i.e., left to  right in Table 1), in the
proportion  of bromine present  in  the haloform products.  Figure  2 shows that
bromine dominated  the products at salinities above about 5 g/kg.
        100-

        80
      IO
     X
     geo

     ^40
     i_
     m
     55 20

          0
V
                         o   14  pM   DOSE
                         •  140 pM  DOSE
            0
             IO     15    20    25
           SALINITY   (g/kg)
30  35
                Figure 2.  Bromine as a percentage  of total halogen in
                haloform products.

     The total haloform yield, expressed as a percentage of the  chlorine
dose (i.e., the amount added initially), varied considerably (Figure 3), but
generally seemed to be higher at low-to-intermediate salinities.  Absolute
haloform yields were roughly proportional to chlorine dose, so that percentage
yields were of similar magnitude for both dosages tested.  Although on a
molar basis, organic carbon graatly exceeded chlorine dosage in both sets of
tests, its availability appeared nevertheless to exert some minor influence
on yield.  (Samples 6 and 7 are very similar, except for organic carbon con-
tent; sample 7, with higher organic carbon, yielded much more haloform).

-------
                                       o   14  jjM   DOSE
                                       o   140  JJM  DOSE
                          10     15    20     25
                          SALINITY  (g/kg)
                                     30
              Figure 3.  Total halofonn yield as a percentage  of chlorine
                        dose; chlorine added as NaOCl.

     It  is particularly interesting to note (Figure 4) the correlation
between  the percentage of the chlorine dose that disappeared within 10
minutes  and the yield of halofonr.s produced.  This suggests that organic
matter plays a major role in chlorine consumption, even though  only a small
fraction of the consumed chlorine is recovered in the form of volatile
halocarbons.

Discussion

     When chlorine is added to seawater, bromide which is present at a
concentration of about 840 yM is rapidly oxidized to HOBr (e.g. Duursma
and Parsi, 1976; Eppley et al., 1976):
    HOC1 + Br
HOBr + Cl
                                                                   (2)

-------
            Q
            _J
            LJ
              ro
            X

            o  I
                  0     20     40     60    80     100

                          CHLORINE  DEMAND
                Figure 4.  Haloform yield as a percentage of chlorine dose
                          vs. total chlorine consumed  in 10 min, as a
                          percentage of chlorine dose; ilose was 14 yM.

Using the ionization constant data of Sugara and Helz (1976) and the rate  con-
stant data of Farkas et al. (1949), wi  calculated the time needed for HOC1  to
be consumed by reaction 2  (see Figure 5).  It was assumed in these calcula-
tions that the chlorine dose was small  compared to the  available Br~ so chlo-
rine consumption could indeed reach 99% completion.  For typical doses, this
assumption must break down at low salinities.  Also at  pH of 9 and higher,
equilibrium between OC1~ and OBr~ is reached before 99Z of the hypochlorite
has been consumed.   Nevertheless, the figure clearly shows that under near-
neutral conditions  at salinities abave  about 5 g/kg, consumption of HOC1  by
•Br~ will occur within less than l.min;  in many cases, the conversion time
will actually be less than 10 sec.  Thus, the dominance of Br in haloform
products formed in  waters that contain  more than 5 g/kg seasalt is readily
explained.  As pointed out in a later section, these results imply that the
organic substitution reactions which give rise r.o haloforms are slow compared
to reaction 2.

     The fact that  haloforros are the only major volatile halocarbon product
of the direct chlorination of marine waters (as with drinking waters) implies
that a unique mechanism exists for their synthesis.  Recently, Rook (1977)
suggested that halogen attack begins at the 2nd carbon  position in a 1,3-

-------
               1000
               Figure 5.
     1C        20       30
   SALINITY  (g/kg)
Calculated time for 99%  consumption of free
chlorine by reaction 2 (see text).
dihydroxybenzene (resorcinol)-type unit in fulvic acid,  thus leading to ring
cleavage and release of  a haloform molecule.  In freshwaters, 60-80Z of the
dissolved and particulate organic carbon typically consists of humic sub-
stances, including fulvic acid  (Reuter and Perdue, 1977).   Although a number
of workers have pointed  out differences between organic  matter in freshwater
and seawater (Kalle, 1966; Khaylov, 1968; Sieburth and Jensen, 1968; Stuenner
and Harvey, 1974;  Kerr and Quinn, 1975), sufficient similarity apparently
exists,  nevertheless,  to justify common use of the term, fulvic acid, to
describe marine dissolved-organic matter.  Our results indicate that, fluvial
and marine organic matter share the ability to produce haloforms.
                                     10

-------
     It is clear from ttese and other results (Rook, 197A) that the organic
nwtter that supplies carbon for the halofonns reacts preferentially with Br,
rather than Cl.  For example, sample 3 which contained 86 yM Br~ (estimated
from salinity) could have consumed a maximum of only about 60% of the 140 uM
chlorine dose via reaction 2.  Yet the excess chlorine is scarcely represented
in the haloform products (Table 2).

     One likely reason why chlorine in haloforms is disproportionately low
is that reactions involving ammonia, organic amines and amino acids rapidly
tie up chlorine in chloramine compounds:

     HOC1 + k-NH2 -»• R-NHC1 + H.O

     HOC1 + R-NHC1 -* R-NC12 + H^O

These reactions are extremely fast (Morris, 1967).  The products are oxidants,
but probably cannot react vith dissolved organic matter to produce haloforms.
Analogous bromamines also can form in marine waters, but, at least in the
case of the inorganic bromamines, their formation is reversible (Johnson and
Overby, 1971).  Thus, chlorine that becomes nitrogen-bound may be permanently
removed from the haloform generation process, whereas nitrogen-bound bromine
may continue to be available.

COMPARISON OF CHLORINATION AND CZONATION OF ESTUARINE WATER

     A few exploratory experiments were conducted in cooperation with Dr.
Ronald Block, of the Chesapeake Biological Laboratory, to compare the halo-
form-generating performance of chlorine and ozone.  Those experiments employed
his bioassay system and natural estuarine water.

Methods and Results

     Data on the water chemistry are given in Table 3.  Chlorinated, ozonated,
and untreated control solutions were examined in parallel experiments run
simultaneously to insure that prior to treatment the water quality was the
same.  For the chlorine experiments, Ca(OCl)2 stock solution (10.6 equiv/1)
was metered at 0.5 ml/min into a 3.8 1/min stream of estuarine water that
flowed continuously through an 80-1 glass tank; thus, the dosage was 139
Hequiv/1, or about 5 ppm Cl.  Turnover time of water in this tank was roughly
20 min.  The system was allowed to run for several hours until a steady-state
level of oxidant in the tank was observed.  Then, a sample was collected in
a glass bottle and sealed so that no air-space remained in the bottle.

     A second sample was obtained at the same time, but dechlorinated with
^28703 before sealing.  Next, a dozen white perch (Morone americana) were
introduced into th« tank; after 30 min, two additional water samples (one of
which was quenched with Na2S203) were taken and sealed.  In the ozone experi-
ments, the above procedure was conducted in a similar fashion, except that
the flowing water was treated with the oxygen-ozone gas effluent from a
Welsbach T-816 ozonator; dose was adjusted to give approximately the same
                                     11

-------
 oxidant level as in the chlorine testing tank.   Turnover  time of water  in  the
 ozone tank was 8 rain.   With the controls,  all procedures  were followed  (i.e.
 taking of a quenched and an unquenched sample and testing both in the pres-
 ence and absence of fish).   The samples were analyzed for volatile halo-
 organics 18-24 hours after  collection  by transferring them to  septum vials
 and then following the method described previously.

 Results

      As shown in Table 3, bromoform concentrations in the neighborhood  of
 20 ug/1 were found in  all the samples  that had  been  Created with hypo-
 chlorite or oznne.  Furthermore,  bromoform was  not detectable in the con-
 trols.   This indicates that bromoform  is produced by the  oxidants and is
 not introduced by sample handling or contamination.

      TABLE 3.   YIELDS  OF BROMOFORM IN  ESTUARINE WATER* TREATED WITH
 	CHLORINE (AS PCI") AND  OZONE.	

                                     Fish Absent          Fish Present

 Hypochlorite (u-equiv/1)               23.8                   21.2
   Bromoform Yield (ug/1)
     Na2S203 Quench                     24.2                   20.1
     No Quench                          25.8                   19.8
 Ozone (u-equiv/1)                      23.2                   21.8
   Bromoform Yield (ug/1)
     Na2S?03 Quench                     19.0                   17.0
     No  Quench                          25.6                   19.0
 Controls
    Na2S203 Quench                      <1                      <1
    No Quench                           <1                      <1

 *Cheraical  characteristics of the water are: salinity «* 14 g/kg,  pH = 8.0,
 T  = 15°C,  NH3  < 0.1 mg/1, initial dissolved oxygen * 10 mg/1  (dissolved
 oxygen  in  ozonated tank = 17.7 mg/1).   All data are  based on  a single
 analysis for each parameter in each sample.

     In both the hypochlorite  and the  ozone tests, introduction  of fish
lowered the steady-scate oxidant  concentration and the amount of bromoform
observed.  In the  unquenched samples,  the yield  of bromoform  is  essentially
the same for both  hypochlorite and ozone.   Furthermore, in the hypochlorite
experiments, quenching with Na2S203 had little effect upon bromoform yield.

 Discussion

      The virtually identical haloform  yields  in the  quenched  and unquenched
 chlorine samples indicate that the substitution reactions must be fast
 compared to turnover time of water in  the  experimental tank (i.e.,  20
 minutes, or roughly 1000 seconds)  because this is the average time between
 addition of chlorine and addition of Na2S203  to the  quenched  sample.  On
 the other  hand, the substitution reactions  must  be slow compared  to reaction

                                     12

-------
"(I)  because bromoform is the only significant product.  Based on Figure 5,
 reaction (1) would reach completion in roughly  20  seconds at  the salinity
 and  pH of these experiments.  Thus, these observations bracket the order of
 magnitude of the time needed to complete the halogen substitution.


      It Is important to distinguish between the substitution reactions dis-
 cussed above and the haloforro production reactions.  Laboratory studies of
 halofonn production from chlorinated solutions of  fulvic or huraic acids,
 believed to be the sources of carbon in the haloforms, suggest that this
 production requires hours to reach completion (Stevens et al., 1976;  Rook,
 1976 and 1977).  Further, Kopfler et al.  (1976) and Nicholson et al.  (1977)
 have provided evidence for the existence of a long-lived halo-organic inter-
 mediate that decays to release haloforms upon heating.  Thus, the substitu-
 tion of halogen atoms into organic structures in water, and the release
 of haloform molecules from those organic structures, appear to be separate
 processes with quite different rates.  The release process apparently does
 not require the presence of an oxidant residual, implying that in seawater,
 it would be impossible to limit haloform production by instituting dechlori-
 nation practice ut power plants unless the entire chlorination-dechlorination
 cycle can be eccoraplished in a very short time span.

      The excellent agreement in bromoform yield from ozone and chlorine is
 somewhat surprising.  In the case of the unquenched samples, the yield of
 bromoform is the same within experimental error.  For the quenched samples,
 slightly less bromoform was obtained from ozone than from chlorine, probably
 reflecting simply the shorter turnover time of water in the ozone-treated
 tank.  It would seem from these data that replacement of chlorine by ozone
 as an antifoulant in marine power plants offers little advantage for mini-
 mizing volatile halo-organics, unless it can be shown that ozone is basically
 more effective than chlorine and can be used at lower doses.

      The chief purpose of placing fish in the test tanks after the first set
 of samples had been collected was to determine if reaction of the oxidants
 with excretory products, body, or tissues produced a large increase
, in the haloform yield or in the Cl/Br ratio in the products.  A large en-
 hancement in yield could make haloforms potential toxic factors in bioassay
 experiments intended to assess the acute toxicity of chlorine and ozone to
 fish.  However, the data indicate small decreases in bromoform content in
 the tanks when fish are present.  A small decline in the residual oxidant is
 also observed.  One possible explanation is that fish produce compounds
 which react rapidly with oxidative biocides, but which generate no volatile
 halo-organic compounds.  Ammonia, urea, an! other nitrogenous products are
 likely candidates.  Probably they compete with the naturally occurring
 bromoform precursors for the available halogen and thus lower bromoform
 yield.  An alternate hypothesis for the decrease in bromoform may be uptake
 of this compound by the fish, either through absorption into body slimes or
 transport into the body via the oouth or gills.
                                      13

-------
                                SECTION 4

                              FIELD STUDIES

     This section extends the previous work to the field and includes some
evidence regarding the fate of volatile halocarbons in the natural
environment.

CHALK POINT POWER PLANT

Site Description

     Field work was done at a coal-fired power plant located on the Patuxent
River near Benedict, Maryland.  Although the upper part of the Patuxent
drainage basin contains numerous residential suburbs and has a population
density of approximately 500 persons/tan^ (data from Maryland Environmental
Service, 1974), the area within 30 km of the plant, both upstream and
downstream, is dominantly rural with a population density closer to.50
persons/km^.  The power plant is the only major industrial installation in
the lower part of the basin.

     The salinity in the vicinity of the plant ranges with season from
virtually zero to about 15 g/kg (Cory and Nauman, 1971).  Ammonia is
typically less than 0.1 mg N/l (Sugam, 1977).

     The power plant contains two 355 megawatt generating units that each
circulate 16,000 I/sec (560 cfs)  of estuarine water through twin single-pass
condensers.  This water,  which is continuously chlorinated, is discharged
into a 2.6-km-long cooling canal.   At the head of the canal, the condenser
effluent is sometimes diluted with unchlorinated estuarine water by as much
as a factor of two to meet regulations limiting thermal loading.

Results

     A large amount of data was collected on the inorganic aqueous  chemistry
at this site (mostly under other sponsorship).  Data are reported elsewhere
(Sugam, 1977) and only will be summarized here.  As illustrated by  the lower
two curves in Figure 6, more than 90% of the oxidant dose disappears by
the time the cooling water emerges from the plant.  Comparable initial
decay has been observed by others CSnoeyink and Markus, 1974; Eppley, 1976;
Bongers et al., 1977), but rarely results in loss of as much of the initial
dose a* we observed at this site.  We will refer to this phase of the decay
process as the fast phase.  \t the pover plant, it occurs mostly under non-
photic conditions inside conduits and condenser plumbing.  We could not
study this phase in detail because of inadequate access to the water inside
the plant.


                                     14

-------
              10
           <
           Q
             O.i
                                                               1.0
                                                                  CM

                                                                 G
                                                              0.1
o>
E
                                                              0.01
                 0     30     60    90    120    150
                          TIME    (min.)
                Figure 6.  Decay of chlorine-produced oxidants  in Patuxent
                          estuarine water.  Curve A - intake water chlori-
                          nated with NaOCl in the laboratory at room
                          temperature.  Curve B - samples collected from
                          discharge canal at night.  Curve C - samples col-
                          lected from discharge canal during the day.  Analy-
                          ses by amperometric titration at pH  4.0 with excess
                          KI.

     When samples from the intake  canal were chlorinated in the laboratory
with doses comparable to those used in the plant (upper curve,  Figure  6),
the amount of chlorine lost  during the fast phase was much less than observed
at the power plant canal.   Samples extracted from the plumbing  upstream of
the condensers inside the  plant  resembled those chlorinated in  the labora-
tory; that is, they did  not  yet  display the large loss of chlorine subse-
quently observed in the  canal.
                                    15

-------
     The fast phase is followed by a slow decay phase.  In the discharge
canal, this always appeared to be first order  in  total oxidant  concentration
with tk - 30-100 min.  In figure 6, the dayime decay rate appear:, to be
slightly faster than the nighttime decay rate, suggesting that photochemical
processes might play a minor role during this phase.  However, in other runs,
the night decay rates were faster so the general-  importance of photochemical
processes is questionable.

    .No free chlorine (determined by amperoroetric procedures) was ever
detected in the canal.  On the other hand, in laboratory-dosed or precon-
denser samples, for which the fast dec^y of total oxidant was much less than
in the field, free chlorine persisted for substantial periods of time.

     Probably because of the rapid loss of free chlorine, only traces of
haloforms (~1 pg/1) were found in discharge canal wccers thnt were collected
in gas-tight septum bottles and analyzed 24 to 48 hours later.  Yet,  10-100
yg CEEr-j/1 was readily generated in intake water by chlorinating it tilth the
same dose used by the plant and then processing it as the discharge canal
samples were processed.   Precondenser samples also yielded haloforms.  For
example, one sample extracted from the conduit between the chlorine injection
point and the condenser inlet was found to contain 40 yg CHBT3/1 £.nd
traces of CHB^Cl when analyzed the next day.  However, its total oxidant
residual after 1 hr was 11 yN, much higher than that observed in the canal
after a comparable period.

Discussion

     The finding of only traces of haloforms in the discharge canal at Chalk
Point was a very surprising result in view of the -ease rith which haloforms
could be generated under laboratory conditions.  Thus, some explanation is
required.

     One suggestion might be that haloforms actually uid form, but were lost
(possibly by volatilization) before the water reached the sampling point.
In this report, some data are presented on the rates and mechanisms of loss
of volatile halocarbons from estuarine waters.  These studies suggest no
mechanism that could remove halocarbons sc rapidly as tc result in their
absence from the canal only minutes downstream of the plant.

     Instead, the negligible haloform yields are probably a consequence of
the remarkably rapid oxidant loss in the condensers.  This then raises the
question, what controls the oxidant loss?  Since  the fast phase of decay
consumed more oxidant witnin the. Chalk Point plant than in experiments that
matched the plant's oxidant dose, one might suspect initially that some of
the oxidant consumption under the field conditions was caused by reaction
with the condensers and plumbing, or with accumulated slimes within them.
However, a thorough review of the replacement history of both the condensers
and the sacrificial anodes, along with estimates  from metallurgical data of
the likely corrosion rate of Cu-Ni condensers, suggests that metallic
corrosion probably consumes at most only a small percentage of the chlorine
used annually.  It is also possible to dismiss reduction by slimes or by-
products of slime organisms as a major factor with the following argument:

                                      16

-------
Because this plant chlorinates continuously  5 months a year, biosynthetic
production of a hypothetical reducing  substance per unit volume of water
would have to equal at least the rate  of  chlorine consumption per unit
volume.  This exceeds 1 mg C12/1 during the  ~5 min period that the water is
Inside the plant.  If reductant production did not match oxidant consumption,
the store of the reducing substance would be depleted rapidly.  Biosynthesis
of enough material from each liter of  water  to accomplish this much reduction
is inconceivable in the short time available.

     It is more reasonable to argue that  the greater fast oxidant consumption
observed in the power plant can be explained by thermal and hydrodynamic
effects in the water itself.  The rate of chlorine decay in estuarine  water
is strongly temperature-dependent, as  seen in Figure 7.  Thus, the elevated
temperatures in the condensers should  enhance decay processes.  Further,  it
              0.01
                          30     60     90     120
                              TIME  (minj
               Figure 7.  Effect of temperature  on decay in estuarine water
                          chlorinated in the laboratory.


will be shown that colloidal organic matter undergoes substitution and
deamination reactions with chlorine.  It is quite likely that these reactions
are partly diffusion-controlled and thus subject to acceleration by the
intense agitation occurring inside the condensers.
                                     17

-------
     The search for reducing agents to consume the oxidant probably should
be directed toward the water itself.  For most well-oxygenated estuarine
waters, it is possible to reject common inorganic reducing agents (e.g.
Fe2+, Mn2+, S2~, N02~) as major chlorine consumers based  upon one or both of
the following grounds (Sugam, 1977): either the concentration of the agent
in its reduced form is small compared to typical chlorine doses, or
available kinetic data suggest that the redox reaction is too slow to explain
the large chlorine decay that occurs in the fast phase.   The  second argument
is always subject to the objection that there may be unrecognized catalytic
processes occurring under field conditions.  Nevertheless, we have come to
believe that most chlorine consumption in estuarine waters (especially con-
sumption during the fast phase) must involve reactions mainly of the types
listed in Table 4; this list is intended only to provide  examples of reaction
types, not a definitive list of possible reactions.

     TABLE 4:  EXAMPLES OK POSSIBLE OXIDANT CONSUMPTION REACTIONS IN
     	ESTUARIES. (X - Cl or Br)	

     Carbon Reactions (Substitution, Oxidation)

                                R_

                  3H  +  nHOX  -»•
2.  R-C=0 + HOX -»• R - C=0 + HX


      H                H
3.  R-i-COOH + HOX -»• R-C=0 + NH_ + CO- +



Nitrogen Oxidation;

4.  2NH, + 3 HOX •*• N, + 3 HX + 3H20


Self-Decomposition;

5.  HOX -»• HX + %0,
     6.   nHOX -•• (n-1)  HX + HXO
                              n
                                              HX
                                      18

-------
Reactions with Organic Carbon

     That organic carbon plays a major role  in the fast decay processes seems
Inescapable.   As  seen in Figure 8,  the rate  of chlorine decay in an estuarine
water sample  can  be significantly slowed  if  the sample is first subjected
to ultrafiltration (Amicon UM 2) before chlorination.   Analysis shows that
the material  removed by this treatment is mostly organic.   The rate is more
dramatically  reduced if the sample  is  first  passed through an XAD column to
remove organic molecules.  Eppley et al.  (1976) reported that most of the"'
fast decay in Southern California coastal water could  be eliminated if the
water was treated by a UV oxidation procedure to remove organic carbon
prior to chlorination.  Crane (personal communication, EPA Bears Bluff
Field Station) has found an excellent  correlation between water color (pre-
sumably a measure of dissolved humic compounds) and chlorine  demand in a
South Carolina estuary.
                                   ULTRAF1LTERED
                          30    60    90     120
                             TIME  (min.)
               Figure 8.   Effect of removing organic matter on decay.
                          Ultrafiltration employed Amicon UM-2 membrane
                          filters.
                                    19

-------
     However! it is inconsistent to argue on the one hand,  that oxidant is
consumed mostly by reaction with organic carbon while also  arguing that
negligible haloform production occurred because of the rapid disappearance
of oxidant.  One way to resolve this conflict is to hypothesize that the
combination of elevated temperature and intense agitation in the condensers
increases the availability of reactive sites on the entrained organic floes
and particles.  Thus, instead of getting multiple halogen substitution at
relatively few carbon atoms, a necessary condition for haloform production,
one gets single halogen substitution at many different carbon atoms.  Accord-
ing to this hypothesis, the resulting halogenated macromolecules are re-
latively stable and fail to split-off low molecular weight  fragments which
would be detectable by methods based on gas chromatography.

     Support for this hypothesis was obtained by isolating  (by ultrafiltra-
tion) the macromolecular organic matter from samples collected within a few
minutes of each other in the intake and the discharge canals at Chalk Point.
The macromolecular matter was then analyzed, the halogens being determined
by neutron activation analysis.  The results (Table 5) indicate that both
samples are very similar except for the bromine content in  the sample from
the discharge canal.  We infer that this bromine is bound to the macro-
molecules as a result of chlorination.  The data also show  significant
chlorine in both these samples, but we believe this is simply residual salt,
rather than carbon-bound chlorine because it does not increase from the
intake to the discharge canal.

                  TABLE 5.  MACROMOLECULAR ORGANIC MATTER

Item
C %
H %
•K %
Cl %
Br %
I %
0 + S (difference) %
Ash %
C/N (molar)
Intake
Canal
39.72
5.80
5.69
0.81
0.04
0.0030
47.94
100.00
6.40
8.14
Discharge
Canal
39.31
6.18
4.77
0.85
1.60
0.0026
47.29
100.00
6.42
9.61
              Amount Recovered
               (mg 1-^
1.31
1.29
     Reactions with organic carbon, although undoubtedly important, can be
over-emphasised.  Other reactions listed in Table 4 probably are also im-
portant, especially in waters that contain very little organic carbon, but
since these reactions are not directly relevant to the subject of this re-
port, they will not be discussed here.
                                     20

-------
BACK RIVER WASTEWATER TREATMENT PLANT
     When first encountered, the very low halofonn yields at Chalk Point
caused concern that perhaps haloforms were being produced, but were lost
rapidly or masked by other components in the water.  To evaluate this possi-
bility, we decided to study a halocarbon source for which we would not have
to depend upon halocarbon synthesis from chlorine.  We chose a large sewage
treatment plant located on the Back River estuary, east of Baltimore,
Maryland, where we established that a suite of volatile halocarbons was pre-
sent in the effluent even before chlorination.  Thus,'the vagaries of the
halocarbon production process were eliminated from the experimental design.

     Figure 9 is a map of Back River, a shallow, 12-km-long tributary estuary
to the Chesapeake Bay with a salinity range of about 0.4 g/kg.  Its mean
depth is about 1 m and it is well mixed vertically.  Near its upper end,
Back River receives 1.5 to 1.9 x 108 1 of wastewater/day from Baltimore's main
sewage treatment plant; the waste discharges often exceed freshwater flow
from the watershed by a factor of 2 (Helz et al., 1975).  The plant provides
1002 secondary treatment (mostly by the trickling filter process) to domestic
and commercial wastes.  The effluent is chlorinated before discharge.
                BACK   RiVER
                 Figure 9
Back River estuary,  Maryland,  showing  locations
of sampling points.   See Appendix A for latitudes
and longitudes.

         21

-------
     The first series of samples  from Back River  (Numbers  8  to 12)  were
collected in early  February, 1977, after the northern Chesapeake Bay had
been covered with ice for more  than one month.  The only uncovered  area was
a 0.2 km diameter patch of water  immediately above the underwater diffusers
at  the discharge point in mid-river.  The second  set of samples (numbers
13  to 23) was collected in early  May, 1977, six weeks after  the spring thaw.

Results

     The vo]atile halocarbons foand in Back River (Table 6)  are very dif-
ferent from those produced by chlorinating estuarine waters  (Table  2).  The
Back River assemblage of compounds must be derived from the  numerous
commercial and light-industrial firms which the treatment  plant serves.

     Although the same assemblage of compounds was found in  February and
May, the amounts of each compound were much lower in ^he latter month.  This
difference may be due to a slower rate of halocarbon disposal  into  the
.sevage system in May or to an increased halocarbon volatilization rate during
the trickling filter process.   The first seems less likely because  we know
of  no reason why the disposal rate should change  drastically with season.
The large differences between the February and May in-plant  samples (8 and
14) are reflected by all the samples from the receiving waters; this rules
out the possibility that the February in-plant sample was  spurious  because
a halocarbon slug was passing by  chance through the plant  when the  sample
was taken.
     	TABLE 6. HALOCARBONS (uH) IN BACK RIVER	

     Sample*    Salinity    CH2C12    CHC13    CCl^   CC12=CHC1  CC12=CC12

     February 1977  (Ice cover);

     8           -           775       413     186       230         369
9 0.57 512
10 0.69 304
11 1.42 152
12 2.92 ND
13 3.16 ND
May 1977 (Open water);
14b - 47
15 0.34 ND
16 0.36 24
17 0.35 9
18 0.34 ND
330 97
238 60
207 57
18 ND
ND ND

106 14
3 ND
71 3
8 2
1 ND
19-23 No volatile halocarbons detected seaward of
113
88
50
2
ND

106
4
75
14
2
station 18
356
337
114
5
ND

50
ND
35
4
ND
.

a. Sample numbers refer to
Figure 10.
b. Samples 8 and 14 were taken in the treatment
final chlorination.


plant just


before

                                     22

-------
     In a seaward direction,  the  concentration of volatile halocarbons
decreases systematically.  The dominant process  causing this decrease is
simply tidal mixing of the effluent with saline  Chesapeake  Bay water.  How-
ever, if dilution were the only process (i.e. , if halocarbons  mixed  conser-
vatively),  then  a plot of halocarbon  concentration vs. specific  conductance
(measured at 25°C) would be linear  (Boyle et al.,  1974; Helz et  al., 1975).
However, such plots were non-linear in winter  (Figure 10).  Since Back River
was almost  entirely covered with  ice, the non-conservative behavior  cannot
be ascribed to volatilization.  When May samples were collected, there was
not sufficient contrast between the specific conductance of the  effluent and
of the water off the mouth of Back River to allow these data to  be treated
as in Figure 10.  However, it is  probably significant that  the seaward de-
cline in halocarbon concentrations was much more abrupt in  May.  Values near
or below the detection limit wera encountered  within 2 km of the outfall in
May.  In contrast, in the February  series, values at station 12  (about 4 km
from the outfall) were only 20-50% of the effluent values.  More rapid loss
in the later samples may be due either to volatilization in the  absence of
ice cover or possibly to more active biodegradation.

Discussion

     Table  7 lists the major  industrially produced, volatile halocarbons
ranked according to rate of release to the environment.  Release rates
were determined  by Nelson and Van Duuren (1975)  by estimating for the
United States the fraction of production plus  imports used dispersively,
plus the fraction of production lost during manufacture.  Of the 11 compounds
listed, three  (CC12F2, CC^F  and  CH2Br-CH2Br)  would not be  expected in waste-
waters because they are released  directly to the atmosphere.  The two fluoro-
carbons are used mainly as refrigerants and aerosol propellants, and CH2Br-
CH2Br is used mainly as a gasoline additive.   A  fourth compound, CH2=CHC1
(vinyl chloride), is probably too volatile to  be retained in wastewater
during treatment, even if it  were somehow present initially.  Of the remaining
seven compounds  in Table 7, five  were found at the nM level in Back River.
It is not known whether the other two, CH2C1-CH2C1 and CH3-CC13, are absent
because they are not discharged into  the sewage  system in significant
quantities, or because they are degraded or adsorbed from the wastewater
before leaving the plant.  Octanol-water partition coefficients, estimated
by the method of Leo (1975),  have been included  in Table 7.  These co-
efficients  are considered a good  measure of the  lipophilic  character of a
compound and have been shown  to correlate with the tendency of a compound
to be bioactive  and to bioaccumulate  (Neely et al., 1974).  However, neither
of the two  missing compounds  in Table 7 is particularly distinctive with
respect to  its partition coefficient or to its Volatility.  Interestingly,
no clear relationship exists between the partition coefficient and the de-
gree of curvature in the concentration vs. salinity curves in Figure 10.
CH2C12 (with the greatest curvature) and CCl2=CCl2 (with the least)  have
essentially the same partition coefficient.
                                      23

-------
                       CH,CIa
    600

    400

    200

     0
V
             CHCU
    200
    100'
i
                       CCL
    200
             CCLjCHCl
    300

    200

    100
      0123456
       SPECIFIC  CONDUCTANCE (mmho)
Figure 10.  Halocarbon patterns in Back River.   Vertical
            scales are all In nanomolar units  for the
            indicated halocarbon.

                       24

-------
  TABLE 7:  MAJOR INDUSTRIALLY PRODUCED HALOCARBONS RANKED BY RELEASE RATE
Compound
Release Rate    Boiling Point
(109 g yr-1)
                                                 log p
                                                      oct
                                                             In Back River
cci2=cci2
cn2ci-CH2ci
CHC1-CC12
CC12F2
CH2C12
CH2Br-CH2Br
CH3-CC13
CC13F
CHC13
CH2-CHC1
CCIA
267
245
195
183
167
138
129
124
76
67
48
121
83
87
-28
40
131
74
24
61
-14
77
1.27
1.48
0.96
2.08
1.26
1.76
2.50
2.52
1.96
0.34
2.96
yes
no
yes
no
yes
no
no
no
yes
no
yes
     The much less  rapid  decrease in halocarbon concentrations when Back
River was covered with  ice  suggests that volatilization  is an important loss
process.  Dilling et  al.  (1975)  and Dilling (1977)  established that a large
variety of C].-C3 halocarbons  escape from open,  stirred beakers in the labor-
atory with half lives (T) on  the order of 20 min.   Liss  and  Slater (1974)
have developed a simple model by which the rates of diffusive exchange of
gases across the air-sea  interface may be estimated.  The model, as applied
to the volatile halocarbons,  leads to the following decay equation (Mackay
and Leinonen, 1975):
                                                                        (A)
     C_ - C_ exp (-K, t/L)
Here, CQ_ is the  initial  concentration,  and Ct is the mean concentration at
tisne=t£i; L_ is  the mean depth of  the water. "Values  of  K^ are presented in
Table 8. ~The  half-life  of  ^ compound in a well  mixed  water body would be:
        0.69 L K
                -1
                                                                        C5)
                                                                      Be-
Values of T for  two arbitrary m«an depths are also given  in Table 8.
cause the validity of  the  simple Liss-Slater model may be subject to
challenge, especially  if applied to organic-rich,  contaminated water around
waste discharges,  the  half-lives in Table 8 must be regarded  as  tentative
estimates.  However, the important implication,  that relatively  shallow,
well-mixed coastal waters  will lose volatile halocarbons  to the  atmosphere
within a few days, is  probably correct.
                                       25

-------
     It Is noteworthy that the half-lives  for  all  the compounds in Table
8 are similar  in magnitude.  This implies  that although volatilization will
cause fairly rapid decline in the concentration of any one  compound, it will
cause the concentration ratio between two  compounds to change only slowly.
Thus, these ratios could be "fingerprints" for a particular discharge plume
and might be useful as plume tracers.

               TABLE 8.' LOSS RATES FOR VOLATILE HALOCARSOSS

CC1 =CC1.
2 2
CHC1=CC12
CH-C1.
2 2
CH,CC1,
3 3
CHC13
cci4
CHBrei_
CHBr2Cl
CHBr,
K. (cm hr )
10.2

11.3
13.1

11.4

11.4
10.6
8.9
6.8
4.1
T (hrs)
L = 1m
6.8

5.9
5.3

6.0

6.0
6.5
7.8
10.2
16.9

L = 5m
34

30
26

30

30
33
39
51
85

     Loss of volatile halocarbons to the atmosphere Is likely to constitute
a permanent sink from the aquatic geochemist's viewpoint.  Published
estimates for the lifetime of these compounds in the troposphere range from
a few days for the unsaturated compounds to a thousand years for CC14 (Yung
et al., 1975; Dilling et al., 1976; Galbally, 1976),  However, even in con-
taminated urban air, concentrations of these compounds are usually less than
1 x 10~9 by volume.  Since the Henry's law constants (Hilling, 1977) are
mostly less than unity, rain would contain mu"h less than 1 nM, even if
saturated.  This is below the concentrations found In this study; therefore
in contaminated coastal waters, rain will be a diltttent rather than a source.
Measurements in rainwater by Pearson and McConnell (1975) support this
conclusion.

     Whereas dilution ar.d volatilization are probably the major processes
that control concentration patterns of volatile hydrocarbons in transport
away from a discharge point, the.February Back River data show that these
compounds are consumed to some extent even when volatilization is precluded
by Ice cover.  A number of processes might contribute to this.  Hydrolysis
or similar abiotic chemical mechanisms might be involved, although limited
available information suggests that these processes will be slow.  For
example,  Zafiriou (1975) estimated that the 50% conversion of CH3C1 to
CH30H in seawater would take 2.5 yr at 20°C and 14 yr at 10°C.  Dilling et
al.  (1975) determined half lives of 9 to 18 months for 5 chlorocarbons in
oxygenated solutions.  Exposure of the solutions to sunlight increased the


                                     26

-------
"-_decay rate by a factor of 1.5 to 2.0 for olefinic compounds, but had little
  effect on saturated compounds.  The products in these reactions were not
:  determined.
,       Whether biochemical pathways offer a faster consumption mechanism for
|  volatile halocarbons is not yet clear.  A brief attempt by Jensen and
'-Rosenburg (1975) to answer this question yielded inconclusive results.

1       A potentially important removal mechanism in Back River involves trans-
j  fer into the sediments.  Although our results and those of others suggest
  that volatile halocarbons are inefficiently adsorbed by such substances as
  humic acid, ferric hydroxide, and clay (montmorillonite), there may be
'•  biologic  pathways in which the compounds become bound in fecal pellets
;  or other biogenic debris.  Furthermore, the sediments in Back River have the
  consistency of black mayonnaise and may act somewhat like a non-polar liquid,
  simply extracting the dissolved halocarbons from the overlying aqueous
|  layer.  However, two attempts to establish the presence of these compounds
!  in Back River sediments yielded negative results.  In one experiment, 5 g
j  of sediment was placed in a septum vial with 40 ml of H2O; after sealing,
 • Che vial was shaken for an hour at 60 C, and the headspace above the slurry
i  then analyzed.  In the second experiment, & Soxhlet extraction of 50 g of
  sediment was performed for 1 week with 9:1 benzene-methanol.  These experi-
  ments indicate that if volatile halocarbous are transferred into the Back
  River sediments, then they are probably degraded there, presumably by
  anaerobic biologic processes.                                                ;

-------
                                 SECTION 5

                              OTHER ACTIVITIES

WORKSHOP

     In the Spring of 1976, a workshop on the fate and effects of chlorine
in Coastal Waters was held at Solomons, Maryland, with financial support
from this grant.  The workshop was organized jointly by the principal
investigator and Dr. Ronald Block of the University of Maryland.  About 30
scientists from as far away as Great Britain participated.  Althougn there
were 15 scheduled talks in the 2\ days of the workshop, emphasis was placed
on informal discussion of problems and research needs associated with the
use of chlorine in marine waters.

     A full report on the workshop has been published (Chesapeake Science,
18, 96-160), and reprints of this report have been provided to EPA.  Among
the research needs identified in that report are a need to understand the
reaction paths and mechanisms by which chlorine disappears in marine water,
a need to account for all the products and a need to improve analytical
procedures.  Bioassay methodology was a major topic of discussion at the
Workshop and some recommendations regarding the design of bioassays have
been cade in the report.

PUBLICATIONS

     The following publications have resulted from work on this project:

Sugam,  R.,  and Helz, G.  R., 1977, Speciation of chlorine produced oxidants in
     marine waters:   theoretical aspects.  Chesapeake Science, 18:113-115.

Block,  R.,  G.  R. Helz,  and W.  P. Davis, 1977, The fate and effects of chlorine
     in coastal waters,  Chesapeake Science, 18:97-101.

Helz, G.  R.,  R. Y. Hsu,  and R. M. Block, 1977, Bromoform production by oxida-
     tive biocides in marine waters.  In: Ozone-Chlorine Dioxide Oxidation
     Products of Organic Materials, R.  G. Rice, J. A. Cotruvo and M. E.
     Browning, eds.   International,. Ozone Institute,  pp. 69-76.

Helz, G.  R.,  R. Sugam,  and R.  Y. Hsu, 1978, Chlorine degradation and volatile
     halocarbon generation in estuarine waters.  In: Water Chlorination:
     Environmental Impact and Health Effect, R. L. Jolley, D. H. Hamilton
     and,  H.  Gorchev, eds.  2, pp. 209-222.

Helz, G.  R.,  and R. Y.  Hsu, 1978, Volatile chloro- and bromocarbons in coastal
     waters,  Limnol. Oceanogr., 23:858-869.

                                     28

-------
     In addition to these publications, oral presentations have been made at
the annual meeting of the American Geophysical Union (1977), at the 2nd
Conference on Water Chlorination in Gatlinburg, Tenn. (1977) and at the
International Symposium on the Analysis of Hydrocarbons and Halogenated
Hydrocarbons in Hamilton, Ont. (1978).
                                      29

-------
                                 REFERENCES

Bongers, L.  1973.  Assessing the impact of power plants on Maryland's water-
     ways.  Record of the Maryland Power Plant Siting Act.  2:1-4.

Bongers, L. H., T. T. O'Connor, and D. T. Burton.  1977.  Bromine chloride -
     an alternative to chlorine for fouling control in condenser cooling
     systems.  EPA 500/7-77-053, U.S. Envrionmental Protection Agency,
     Cincinnati, Ohio.

Boyle, E., R. Collier, A. T. Denglar, J. M. Edmond, A. C. Ng, and R. F.
     Stallard.  1974;  On the chemical Mass-balance in estuaries.  Geochim.
     Cosmochim. Acta.  38:1719-1728.

Brush, L. M.  1972.  Domestic and municipal waste loading to Chesapeake Bay.
     In: Annual Report, June 1, 1971 to May 31, 1972.  Chesapeake Research
     Consortium Inc., Baltimore, MD.

Cory, R. L., and J. W. Nauman.  1971.  Water quality at Patuxent River Bridge,
     MD, January 1969 through November 1969.  U.S. Geological Survey Open
     File Report.  71 pp.

Dilling, W. L.  1977.  Interphase transfer processes.  II.  Evaporation rates
     of chloro methanes, ethanes, ethylenes, propanes, and propylenes from
     dilute aqueous solutions.  Comparisons witli theoretical pr?dictions.
     Environ. Sci. Technol.  11:405-9.

Dilling, W. L., C. J. Bredeweg, and N. B. Tcfertiller.  1976.  Organic photo-
     chemistry.  Simulated atmospheric photodecomposition rates of methylene
     chloride, 1,1,1-trichloroethane, trichlorethylene, tetrachloroethylene
     and other compounds.  Environ. Sci. Technol.  10:351-6.

Dilling, W. L., N. B. Tefertiller, and G. J. Kallos.  1975.  Evaporation rates
     and reactivities of methylene chloride, chloroform, 1,1,1-trichloro-
     ethane, trichloroethylene, tetrachloroethylene and other chlorinated
     compounds in dilute aqueous solutions.  Environ. Sci. Technol.  9:833-8.

Duursma, E. K., and P. Parsi.  1976.  Persistence of total and combined chlo-
     rine in seawater.  Neth. J. Sea Res.  10:192-214.

Eppley,  R. W., E. H.  Renger, and P. M. Williams.  1976.  Chlorine reactions
     with seawater constituents and the inhibition of photosynthesis of
     natural marine phytoplankton.  Estuarine Coastal Mar. Sci.  4:147-161.

Farkas,  L., M. Lewin, and R. Bloch.  1949.  The reactions between hypochlorite
     and bromides.  J. Am. Chem. Soc.  71:1988-1991.

                                     30

-------
Galbally, I.  E.  1976.  Man-made carbon tetrachloride in the atmosphere.
     Science.  193:573-6.

Helz, G. R.,  R. J. HuggeLt, and J. M. Hill.  1975.  Behavior of Mn, Fe,.Cu,
     Zn, Cd,  and Pb discharged from a wastewater treatment plant into an
     cstuarine environment.  Water Res.  9:631-6.

Jensen, S., and R. Rosenberg.  1975.  Degradability of some chlorinated
     aliphatic hydrocarbons in seawater and sterilized water.  Water Res.
     9:659-61.

Johnson, J. D., and R. Overby.  1971.  Bromine and bromamine disinfection
     chemistry.  J. Sanit. Eng. Div. Proc. Am. Soc. Civ. Eng.  97:617-28.

Jolley, R. L.  1974.  Determination of chlorine-containing organics in
     chlorinated sewage effluents by coupled 36C1 tracer-high resolution
     chromatography.  Environ. Lett.  7:321-340.

Kalle, K.  1966.  The problem of the gelbstoff in the sea.   Oceanogr.  Mar.
     Biol. Annu. Rev.  4:91-104.

Kerr, R. A.,  and J. G. Quinn.  1975.  Chemical studies on the dissolved
     organic  matter in seawater: isolation and fraccionation.  Deep Sea
     Res.  22: 107-16.

Khaylov, K. M.  1968.  Dissolved organic macromolecules in seawater.   Geochem.
     Int.  5:497-503.

Kopfler, F. C., R. G. Melton, R. D. Lingg, and W. E. Coleman.  1976.   GC/MS
     determination of volatiles for the national organics reconnaissance
     survey (NORS) of drinking water.  In: Identification and analysis of
     organic  pollutants in water, L. H. Keith, ed.   Ann Arbor Science
     Publishers, Ann Arbor, Michigan.

Leo, A. J.  1975.  Calculation of partition coefficients useful in the evalu-
     ation of relative hazard's of various chemicals in the environment.  In:
     Structure-activity correlations in studies of toxicity and bioconcentra-
     tion with aquatic organisms, D. G. Veith, and D. E. Konasewich, eds.
     Great Lakes Research Advisory Board,  p. 151-176.

Liss, P. S.,  and P- G. Slater.  1974.  Flux of gases across the air-sea
     interface.  Nature.  42:181-4.
                                     31

-------
Mackay, D., and P. J. Leinonen.  1975.   Rate of evaporation of low-solubility
     contaminants from water bodies to  atmosphere.   Environ. Sci.  Technol.
     9:1178-1180.

Maryland Environmental Service.  1974.   The Patuxent River Basin Water Quality
     Management Plan.  Volumes 1 and 2.

Morris, J. C.  1967.  Kinetics of reactions between aqueous chlorine and
     nitrogen compounds.   In: Principles and applications of water chemistry,
     S. D. Faust, and J.  V. Hunter, eds.  Wiley and Sons, New York.   pp.  23-
     53.

Neely, W. B., D. R. Branson, and G. E.  Blau.  1974.  Partition coefficient
     to measure bioconcentration potential of organic chemicals in fish.
     Environ. Sci. Technol.  8:1113-15.

Nelson, N., and B. Van Duuren.  1975.   Final report of National Science
     Foundation workshop panel to select organic compounds hazardous to the
     environment.  Unpublished.

Nicholson, A. A., 0. Meresz, and B. Lemyk.  1977.  Determination of  free  and
     total potential haloforms in drinking water.  Anal. Chem.  49:814.

Pearson, C. R., and G. McConnell.  1975.  Chlorinated Cj and C2 hydrocarbons
     in the marine environment.  Proc.  R. Soc. London.  189:305-32.

Reuter, J. H., and E. M.  Perdue.  1977.  Importance of heavy metal-organic
     matter interactions in natural waters.  Geochim. Cosmochim. Acta.
     41:325-34.

Rook, J. J.  1974.  Formation of haloforras during chlorination of  natural
     waters.  Water Treat. Exam.  23:234-43.

Rook, J. J.  1976.  Haloforms in drinking water. J. Am. Water Works Assoc.
     68:168.

Rook, J. J.  1977.  Chlorination reactions of fulvic acids in natural waters.
     Environ. Sci. Technol.  11:478-83.

Sieburth, J. M., and A. Jensen.  1968.   Studies on  algal substances  in the
     sea.  I Gelbstoff (humic material  in terrestrial and marine waters).
     J. Exp. Mar. Biol. Ecol.  2:174-189.
                                     32

-------
Stevens, A. A.i C. S. Slocum, P. R. Seeger, and G. G. Robeck.  1976.  Chlori-
     nation of organics in drinking water.  In: Proceedings Conference on
     the environmental impact of water chlorination, R. L. Jolley, ed.  Oak
     Ridge Nat. Lab.  CONF-751096.  p. 85.

Snoeyink, V.  L., and F. I. Markus.  1974.  Chlorine residuals in treated
     effluents.  Water Sewage Works.  121:35-45.

Stuenner, D. H., and G. R. Harvey.  1974.  Huraic substances from seawater.
     Nature.  250:480-1.

Sugam, R.  1977.  Chlorine degradation in estuarine waters.  Unpublished Ph.D.
     Dissertation, University of Maryland.

Sugam, R., and G. R. Helz.  1976.  Apparent ionization constant of hypo-
     chlorous acid in seawater.  Environ. Sci. Technol.  10:384-6.

Yung, Y. L., M. B. McElroy, and S. C. Wofsy.  1975.  Atmospheric halocarbons:
     A discussion with emphasis on chloroforai.  Geophys. Res. Lett.   2:397-9.

Zafiriou, 0. C.  1975.  Reaction of methyl halides with seawater and marine
     aerosols.  J. Mar. Res.  33:75-81.
                                     33

-------
             APPENDIX




LATITUDE AND LONGITUDE OF STATIONS
Station
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
IS
19
20
21
22
23



Latitude
38°
39 o
38°
38°
38°
38°
38°
39°
39 »
39°
39 o
39°
39 o
399
39 o
39 o
39°
39°
39°
39°
39°
39 c
39 o
49'
11'
33'
25 r
20'
14'
14'
17'
17'
17*
16'
14'
15'
17'
17'
17'
17'
17'
16'
16'
15'
14'
14'
40"
40"
15"
25"
00"
50"
50"
45"
50"
25"
20"
40"
35"
45"
45"
30"
10"
00"
50"
25"
35"
45"
45"



Longitude
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
76°
42'
16'
25'
35'
20'
7'
7'
29'
28'
27'
26'
24'
23'
29'
28'
28'
27'
27'
27'
26'
26'
25'
24'
25"
45"
20"
35"
50"
55"
10"
20"
15"
55"
25"
20"
55"
20"
40"
15"
55"
40"
15"
40"
45"
50"
40"
              34

-------