Ecological Research Series
Chemistry of  Organomeircurials
In Aquatic Systems

I
55
                                                    UJ
                                                    CD
                                    National Environmental Research Center
                                    Office of Research and Development
                                    U.S. Environmental Protection Agency
                                    Corvallis, Oregon 97330

-------
            RESEARCH REPORTING SERIES
Research reports of the  Office  of  Research  and
Monitoring,  Environmental Protection Agency, have
been grouped into five series.  These  five  broad
categories  were established to facilitate further
development  and  application   of   environmental
technology.   Elimination  of traditional grouping
was  consciously  planned  to  foster   technology
transfer   and  a  maximum  interface  in  related
fields.  The five series are:

   1.  Environmental Health Effects Research
   2.  Environmental Protection Technology
   3.  Ecological Research
   4.  Environmental Monitoring
   5.  Soci©economic Environmental Studies

This report has been assigned  to  the  ECOLOGICAL
RESEARCH  series.   This series describes research
on the effects of pollution on humans,  plant  and
animal   species,  and  materials.   Problems  are
assessed   for   their   long-   and    short-term
influences.    Investigations  include  formation,
transport, and pathway studies  to  determine  the
fate  of  pollutants and their effects.  This work
provides the technical basis for setting standards
to  minimize   undesirable   changes   in   living
organisms   in   the   aquatic,   terrestrial  and
atmospheric environments.

-------
                                                    EPA-660/3-73-012
                                                    September 1973
          CHEMISTRY  OF ORGANOMERCURIALS  IN AQUATIC SYSTEMS
                                  by

                          George L. Baughman
                            John A. Gordon
                             N. Lee Wolfe
                            Richard G. Zepp
             Southeast  Environmental Research Laboratory
          Natipnal  Environmental Research Center-Corvallis
                      1   Athens, Georgia 30601
                          Project 310301  QQG
                        Program Element 1BA023
              NATIONAL ENVIRONMENTAL  RESEARCH CENTER
                 OFFICE OF RESEARCH AND  DEVELOPMENT
               U.  S.  ENVIRONMENTAL PROTECTION AGENCY
                      CORVALLIS, OREGON  97330
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, B.C. 20402- Price $1.30

-------
                             ABSTRACT
Kinetics in water of some chemical and photochemical reactions
postulated as key transformations in the environmental mercury cycle
were investigated.  Decomposition of dimethyImercury (DMM) and
diphenylmercury  (DPM) by acids and mercuric salts was shown to be
pH dependent and too slow to be significant under most environmental
conditions.  Degradation of organomercuric salts by acid is even
slower.  Theoretical evidence indicates that loss of elemental
mercury or DMM at the air-water interface can be important in turbu-
lent systems.

DimethyImercury, methylmercuric chloride, methylmercuric hydroxide,
and methylmercuric  ion were not decomposed by sunlight, but phenyl-
mercury and  sulfur-bonded methylmercuric species were readily
decomposed  to inorganic mercury.  Detailed equilibrium calculations
indicate  that the sulfur-bonded methylmercuric  species are the
predominant  species in natural waters.  Quantum yields for these
reactions  are presented  along with  a technique  for  calculating
sunlight  photolysis rates  from  laboratory  data.

The report  also  includes  a review of the  chemical  literature  con-
cerning the  kinetics of  chemical  and photochemical  decomposition of
organomercurials.
                                 ii

-------
                             CONTENTS







                                                              Page





ABSTRACT .	„   ii




LIST OF FIGURES	    v




LIST OF TABLES	  vii




ACKNOWLEDGMENTS   .............  	  .   ix




Sections




I         CONCLUSIONS	    1




II        RECOMMENDATIONS  .	    3




III       INTRODUCTION 	    5




IV        BACKGROUND	    8




          EQUILIBRIA OF MERCURY REACTIONS IN WATER .....    8




          CHEMICAL DEGRADATION OF ORGANOMERCURIALS .....   13




          PHOTODECOMPOSITION OF ORGANOMERCURIALS ......   16




V         MATERIALS AND METHODS	   18




          MATERIALS	   18




          ANALYTICAL PROCEDURES  	   18




          APPARATUS	   19




          ACIDOLYSIS OF DIMETHYLMERCURY, METHOD  1  .....   21




          ACIDOLYSIS OF DIMETHYLMERCURY, METHOD  2  	   21




          MERCURIC SALT CLEAVAGE OF DIMETHYLMERCURY  ....   21




          ACIDOLYSIS OF DIPHENYLMERCURY  	   22





          MERCURIC SALT CLEAVAGE OF DIPHENYLMERCURY  ....   22
                                 ill

-------
                                                             Page

         PHOTOCHEMICAL PROCEDURES	22

         PHOTOLYSIS OF METHYLMERCURIC SULFIDE ION	23

         SUNLIGHT PHOTOLYSIS OF METHYLMERCURY COMPOUNDS  .  .   23

         ACETONE-SENSITIZED PHOTODECOMPOSITION OF
         PHENYLMERCURIC ION AND PHENYLMERCURIC HYDROXIDE .  .   23

         QUENCHING STUDIES OF PHENYLMERCURY COMPOUNDS  ...   23

         SUNLIGHT PHOTOLYSIS OF PHENYLMERCURY COMPOUNDS  . .   24

VI       RESULTS AND DISCUSSION	25

         CALCULATED EQUILIBRIUM  CONCENTRATIONS OF METHYL-
         MERCURIC  SPECIES  IN AQUATIC SYSTEMS  	   25

         KINETICS  OF  ORGANOMERCURY REACTIONS  IN WATER   ...   28

         Theoretical  Considerations   	  28

         Acidolysis  of Dimethyl- and  Diphenylmercury ....   32

         Desymmetrization  of  Dimethyl- and Diphenylmercury .   41

         DimethyImercury Stability  to  Oxygen and Base   ...  49

         Evaporative  Loss	49

         KINETICS  OF  ORGANOMERCURY  PHOTODECOMPOSITION   ...  49

         Photodecomposition of DimethyImercury, Methyl-
         mercuric  Ion,  MethyImercurie  Hydroxide, and
         MethyImercurie Halides   	   53

         Photoreactivity of Sulfur-bonded  MethyImercury
         Complexes	56

         Photocleavage  of  Phenylmercury  Compounds   	  60

VII      REFERENCES	72

VIII     PUBLICATIONS   ........ 	   78

IX       APPENDICES  .  . .	79
                                 IV

-------
                              FIGURES


No.                                                           Page

 1   Model of the environmental mercury cycle  .......    6

 2   pH dependence of the hydrolysis of mercuric ion ....   10

 3   pH dependence of the hydrolysis of methylmercuric
     ion	   11

 4   pH dependence of the hydrolysis of phenylmercuric
     ion	   12

 5   Enclosed bomb for kinetic studies	   20

 6   Relative concentrations of CHsHgS" in systems
     containing multiple chemical species  .........   26

 7   Relative concentrations of (CH3Hg)2S in systems
     containing multiple chemical species  	   27

 8   Relative concentrations of CH3HgCl and CH3HgSR
     in systems containing multiple chemical species,
     excluding reduced sulfur  ...............   29

 9   Relative concentrations of methylmercury complexes
     in systems containing multiple chemical species,
     excluding reduced sulfur species and organic thiols .   •   30

10   Half-life of A vs. concentration of B for second-order
     reactions when TjB"l > > > [A"]	 . .   .   33

11   Acidolysis of dimethyImercury 	 .....   35

12   Acidolysis of diphenyImercury 	 ......   42

13   Desymmetrization of dimethyImercury by mercuric
     perchlorate .... 	 .............   47

14   Estimated solar irradiance at solar zenith
     angle = 0°	   50

15   UV absorption spectra of dimethyImercury, methyl-
     mercuric ion, and methylmercuric hydroxide in
     water	   54

-------
No.                                                           Page

16   UV absorption spectra of methyImercurie sulfide
     ion and bis(methylmercuric) sulfide ..........   57

17   UV absorption spectra of methylmercury-thiol
     complexes .	   58

18   Effect of quencher, 2,4-hexadien-l-ol, upon
     quantum yields for photocleavage of methylmercury-
     thiol complexes	  .   62

19   UV absorption spectra of phenylmercury compounds  ...   64

20   Quenching of diphenylmercury photolysis by
     cis-l,3-pentadiene in acetonitrile  	 .....   69
                                VI

-------
                              TABLES
No.

 1        KINETIC DATA FOR ORGANOMERCURY LIGAND EXCHANGE
          REACTIONS  .....................  14

 2        DIMETHYLMERCURY ACIDOLYSIS RATE DATA MEASURED
          BY GLC	  36

 3        DIMETHYLMERCURY ACIDOLYSIS RATE DATA DETERMINED
          CONDUCTOMETRICALLY	  37

 4        DIMETHYLMERCURY ACIDOLYSIS IN THE PRESENCE OF
          NUCLEOPHILIC SPECIES   ...............  39

 5        DIPHENYLMERCURY ACIDOLYSIS RATE DATA   .......  40

 6        INCREASING EASE OF ACID CLEAVAGE OF ARYL AND
          ALKYL GROUPS IN ORGANOMERCURIALS  .........  43

 7        CALCULATED EQUILIBRIUM CONSTANTS FOR THE
          REACTION OF DIMETHYLMERCURY WITH MERCURIC
          HALIDES ......................  43

 8        pH DEPENDENCE FOR DESYMMETRIZATION OF DIMETHYL-
          MERCURY AT 27°	  45

 9        pH DEPENDENCE FOR DESYMMETRIZATION OF DIPHENYL-
          MERCURY BY MERCURIC  PERCHLORATE AT 27°   ......  46

 10        EXPERIMENTAL PHOTOLYSIS RATES FOR FOUR  METHYL-
          MERCURY COMPOUNDS IN SUNLIGHT ...........  55

 11        QUANTUM YIELDS FOR PHOTODEGRADATION OF  SULFUR-
          BONDED METHYLMERCURY COMPLEXES AT 313 nm IN
          DISTILLED  WATER ..................  59

 12        EFFECT OF  QUENCHERS  UPON PHOTODECOMPOSITION  OF
          SULFUR-BONDED METHYLMERCURY COMPLEXES IN WATER   .  .  61

 13        CALCULATED SUNLIGHT  PHOTOLYSIS RATES FOR SULFUR-
          BONDED METHYLMERCURY COMPLEXES AT 25° IN WATER   .  .  63

 14        QUANTUM YIELDS FOR DIRECT PHOTOLYSIS  (313  nm) OF
          PHENYLMERCURY COMPOUNDS  ........  	  .  67
                                 VII

-------
'No.

15        QUANTUM YIELDS FOR PHOTOSENSITIZATION OP
          PHENYLMERCURIALS  	 ........  68

16        COMPARISON OF CALCULATED AND EMPIRICAL RATES
          FOR PHOTODEGRADATION OF PHENYLMERCURY COMPOUNDS
          IN SUNLIGHT	 .  71
                                viii

-------
                          ACKNOWLEDGMENTS
We express our appreciation for the support and assistance of
Dr. Walter M. Sanders III, Chief, National Pollutants Fate Research
Program, Southeast Environmental Research Laboratory, Environmental
Protection Agency, Athens, Georgia.

We also wish to thank Mrs. Helein Bennett of NASA's J. F. Kennedy
Space Center for her assistance in X-ray diffraction analysis, and
the staff of the National Water Contaminants Characterization Research
Program, Southeast Environmental Research Laboratory, for their help
and the use of their instrumentation.

-------
                             SECTION I

                            CONCLUSIONS
1.  The composition of dissolved methylmercuric species in aquatic
systems is dependent on the type and concentration of complexing
agents present and the pH.  An analytical expression is derived that
gives the relative percent of the methylmercuric species as a function
of concentration and pH in the presence of eight complexing agents.

2.  At pH's and concentrations expected in the environment, methyl-
mercuric ion is not the predominant methylmercuric species.  Because
natural systems contain much organic and inorganic sulfur, methyl-
mercury will exist predominantly as the sulfide or thiol complex.  In
the absence of sulfide or thiols, the hydroxide or chloride complex
would predominate.

3.  Reaction of hydrogen ion with dimethyImercury and diphenylmercury
gives methane and benzene, respectively, and the corresponding organo-
mercuric salt.  The reaction is first-order with respect to both
hydrogen ion and organomercury concentrations.  The second-order rate
constant extrapolated to 25° for dimethyImercury is 7.33 X 10~5 -L/mole
sec and 9.67 X 10"  -L/mole sec for diphenylmercury.  With the known
rate constants and kinetic expression, the acidolysis half-lives for
these compounds can be calculated at various pH's.

4.  DimethyImercury and diphenylmercury react with mercuric salts in
aqueous solution to give methylmercuric and phenylmercuric salts,
respectively.  The reaction rate shows a strong pH dependence, increas-
ing dramatically as the pH is decreased.  At alkaline pH's and mercuric
species concentrations common to the aquatic environment, this would
not be a significant degradative pathway.  At acidic pH's the observed
rate constant is higher and the reaction may be significant.  The half-
lives can be calculated from mercuric salt concentrations and pH data.

5.  Formation of dimethyImercury by symmetrization reactions of methyl-
mercuric ion or methylmercuric hydroxide occurs at a rate too slow to
be significant under environmental conditions.

6.  DimethyImercury in aqueous solution does not react with dissolved
oxygen, hydroxide, sulfide, iodide, or albumin at a rate fast enough
to be a significant pathway for degradation.

7.  Evaporative loss of dimethyImercury and elemental mercury from
aqueous solution may be significant for turbulent systems.  Half-lives
for evaporative loss can be estimated from literature data and

-------
calculated transfer coefficients.  Based on these calculations
elemental mercury is lost from solution about twice as fast as
dimethyImercury.

8.  The low sunlight absorption rate constants for dimethyImercury,
methylmercuric ion, and methylmercuric hydroxide preclude photodecom-
position as a significant pathway for degradation.  Neither photo-
sensitization nor singlet oxygen effects their decomposition.

9.  DiphenyImercury absorbs sunlight in aqueous solution and photolyzes
with a disappearance quantum yield of 0.27.  The photolysis results in
carbon-mercury bond cleavage to give elemental mercury and phenyl
radicals.  The minimum photochemical half-life determined in sunlight
experiments is 8.5 hours.

10.  Dissolved phenylmercuric salts undergo photochemical decomposi-
tion with pH independent disappearance quantum yields of 0.24.  The
photochemical reaction results in carbon-mercury bond cleavage and
formation of mercurous salts and phenyl free radicals.  The experi-
mentally determined minimum half-lives (~ 17 hours) show that photo-
decomposition of these compounds may be environmentally significant
under certain conditions.

11.  Methylmercuric thiol and methylmercuric sulfide ion complexes
undergo photodecomposition in sunlight.  The methylmercuric thiol
complexes have quantum yields from 0.12 to 0.16 with minimum photo-
chemical half-lives of 46 to 120 hours.  Methylmercuric sulfide ion
has a quantum yield of 0.65 and a minimum half-life of 0.43 hours.
The major products are methane and inorganic mercury species.
Although oxygen lowers the quantum yields, photochemical reaction
may still be a significant degradative pathway.

-------
                            SECTION II

                          RECOMMENDATIONS
1.  Studies are needed to elucidate the rates and mechanisms of the
release of dimethylmercury and elemental mercury from sediments to
overlying water.  Because these materials are non-ionic and have a
significant vapor pressure, they may play an important role in
mercury cycling through loss at the air-water interface.

2.  Calculations presented here indicate that elemental mercury and
dimethylmercury are likely to volatilize from the aquatic environment
Because little is known about the behavior of gaseous mercury species
in the atmosphere, research should be undertaken to determine the
fate of these materials.

3.  The chemical and physical behavior of mercury in the environment
can be determined by species with which the mercury complexes.
Characterization of these complexes is essential for evaluation of
the chemical, physical and biological processes in the mercury cycle.
Therefore, water from sediments and from the water column should be
analyzed to identify the mercury species present.

4.  Redox properties of natural aquatic systems should be investi-
gated with particular emphasis on the rates and mechanisms of
oxidation-reduction reactions.  An understanding of these properties
is required before general statements can be made about the effect
of redox potentials on organomercuric or other pollutants.

5.  Ligand exchange rates should be examined in detail for complexes
of mercury and other heavy metals.  Our literature survey indicates
that some exchange reactions with ligands common to natural waters
are fast enough to approach equilibrium under environmental condi-
tions.  This study should include humic acids as well as proteins
and other biological ligands.

6.  Concentrations of reduced sulfur species and sulfhydryl-containing
organics should be measured in natural waters.  These measurements
should include samples of interstitial water from sediments as well
as samples from the water column.  Sensitive techniques for analysis
of these substances should be developed because very low concentra-
tions can have important effects upon the complexation of organo-
mercuric and other metallic species.

7.  Light absorption and light scattering characteristics of
natural waters should be determined for the wavelength region 290-700

-------
ran.  In conjunction with solar radiation intensities, these data can
be used to calculate the penetration of sunlight into natural waters.

8 .  A general mathematical model is needed for photochemical
processes in natural waters. Such a model should be capable of
predicting integrated photolysis rates in natural systems based on
quantum yields, ultraviolet absorption spectra, solar radiation
intensities, and turbulence levels.

9.  Applicability of current information on sensitization and quench-
ing of photoreactions is limited by lack of knowledge concerning the
photochemical properties of natural waters.  Research is needed to
define the nature of the sensitization process, quantum yields, and
conditions under which sensitization occurs.  These studies should
also attempt to identify chemical and physical properties of natural
waters that indicate the presence of potential sensitizers or
quenchers.

10.  A mathematical model should be developed describing the aquatic
mercury cycle.  The model should have sufficient detail to permit
determination of the relative importance of the various proposed
transport processes.

-------
                            SECTION III

                           INTRODUCTION
Although the occupational hazards of mercury have been known for
many years, its environmental impact only recently became apparent.
The recent interest in the environmental implications of mercury
stems primarily from two incidents in widely separated countries.
First, a massive case of poisoning in Japan—the Minamata Bay
incident—was attributed to industrial discharge of organomercurials.
Secondly, declining bird populations in Sweden was blamed on the use
of phenyl and methylmercurial pesticides as seed dressings.   Pio-
neering work, particularly in Sweden, has since shown that mercury
accumulates in fish to concentrations much higher than that in
surrounding waters and that it exists in fish predominantly in a
methylated form.3  The widespread occurrence of methylmercury
compounds remained a mystery until it was shown by Jernelov that
inorganic forms could be biologically methylated in aquatic systems.4
No direct proof exists, however, that the mercury in fish is
methylated prior to uptake.3  Several extensive reviews of the
overall mercury problem are available and should be consulted for
more detail.3' 5

Tentative descriptive models similar to Figure 1 were offered in
explanation of the transformations and cycling of mercury in
aquatic systems.4  The phenylmercurials were included primarily
because of their widespread release into waterways as a result of
their use as fungicides.

While prior research has clearly demonstrated the importance of
organomercurials, few data are available concerning the mechanisms
or rates of their reactions.  And although some work has been done
on the non-aqueous chemistry of organomercurials, very few studies
were carried out in water or, in the case of photochemical reactions,
in light of wavelengths characteristic of solar radiation.  The  lack
of relevant information may partially explain the strong reliance
of environmental chemists on thermodynamics to predict the aquatic
chemistry of mercury and mercury compounds.  The environmental
literature, as a result, contains many erroneous conclusions about
the forms and reactions of organomercurials as pollutants.

The present report includes an extensive background section with a
discussion of pertinent equilibria and a review of the chemical  and
photochemical degradation reactions of mercury and mercury compounds.
Because the chemical and photochemical reactions occur by different
mechanisms and require different experimental techniques, they are
discussed separately.

-------
      (C6H5)2Hg
                                          CH3Hg+
       C6H5Hg+
           Hg1
Hg
(CH3)2Hg
CH3OCH2CH2Hg+
     Figure 1.  Model  of the environmental mercury cycle.

-------
A comprehensive kinetic study is presented of some chemical and
photochemical reactions expected to result in degradation and cycling
of organomercurials in the aquatic environment.  Study was limited to
the simple phenyl and methyl derivatives because they are known to
enter aquatic systems and because higher homologs are less stable.
The following reactions were studied.
                                  Inorganic mercury species

               (C6H5)2Hg   """^   C6H5Hg+  	1

The rate equations are the result of laboratory studies and should
predict the rates of the respective reactions in water.  They will
not predict the net rate of the reaction in natural systems unless
the reaction in question is the rate determining step.

-------
                            SECTION IV

                            BACKGROUND
Chemists have long been intrigued by reactions of mercury compounds,
and a vast body of literature on this subject has accumulated during
the past century.  Since a review of that work is x^ell beyond the
scope of this report, the interested reader is referred to several
excellent books that have recently been published.   Also, annual
reviews of mercury chemistry have been compiled since 1968 by
Professor Dietmar Seyferth and his coworkers,7"9  Bass and Makarova
have recently reviewed the photochemistry of organomercurials.1

Since most past studies of mercury chemistry were not prompted by
environmental considerations, relatively few studies were carried
out under conditions that could be extrapolated to the environment.
A review of the environmentally pertinent literature is presented
with particular emphasis on publications related to kinetics of
organomercury degradation.

EQUILIBRIA OF MERCURY REACTIONS IN WATER

Early studies of aqueous solutions of mercuric,1  alkylmercuric,12
and phenylmercuric13 compounds indicated that many of these com-
pounds react with water to form acidic solutions.  Subsequent
investigations 4~18 showed that reactions 1-5 account for the acid
formation.  The symbol X represents any electron-withdrawing ligand
that forms an ionic bond with mercury, and R stands for an organic
group such as methyl or phenyl.  Mercury compounds with ionic Hg-X
bonds are often designated as mercury "complexes" or "salts" in the
literature.

          HgX2  i±  Hg2 + + 2X~                                   (1)
               + H20  ^  HgOH+ + if                              (2)


                + H20  z±  HgfOH)2 + if                           (3)


          RHgX  ^  RHg+ H- X~                                     (4)


          RHg+ + H20  z±  RHgOH + H*                              (5)

-------
Although spectroscopic studies have  shown  that mercuric1"'  and  organo
mercuric30 ions, like the hydrogen ion, exist as hydrated  forms  in
water, these species will be represented for convenience as Hg2+,
RHg+, and tf" , respectively.

Mercuric31 and organomercuric14""18 perchlorates, nitrates, and
sulfates are very ionic and are  thus completely dissociated in
aqueous solution (eq 1 and 4).   Other,  less ionic mercury  compounds
only partially dissociate at higher  concentrations,  the degree of
dissociation being concentration dependent.23  Thermodynamic
studies14""18'    have shown that  the  dissociation tendencies of
mercuric  (HgX2) and organomercuric  (RHgX)  compounds  depend upon  the
nature of X" as follows:

     F~ > OCOCHg- > HP042~ ~ Cl" > Br~ > Nil, > OH~ > SR" > S3~
Thiol  and sulfide  compounds  are particularly  stable,  i.e.,  they have
very low tendencies  to dissociate.

Mercuric14 and organomercuric15""18  ions  react rapidly with  water
(hydrolyze)  to form  corresponding hydroxides  (eqs  2,  3,  and 5).
These  reactions are  pH-dependent and will not occur to a significant
extent in acidic water  (pH < 3-4).  However, within pH ranges usually
found  in natural waters  (pH  5-9), the  ions are  almost completely
hydrolyzed  (Figures  2-4).  Experiments discussed later in this
report indicate that these hydrolysis  equilibria have important
effects upon the rates of organomercury  reactions  in  water.

Natural waters contain a variety of chemical  species  that can under-
go "ligand exchange  reactions" with mercuric  or organomercuric
complexes.   Ligand exchange  reactions  involve reaction of mercuric
or organomercuric  complexes  with some  chemical  species,  Y,  to form
a new  complex  (eqs 6-8).

          HgX2 -f Y T±  HgXY  + X                                  (6)


          HgXY + Y *±  HgY2  + X                                  (7)


          RHgX + Y ?±  RHgY  + X                                  (8)

Ligand exchange reactions have been studied by  a variety of techniques.
Early  attempts to  measure the rate  of  reaction  of  mercuric  salts
with human serum mercaptalbumin, a  protein that contains sulfhydryl
groups, were unsuccessful because of the rapid  rate of the  reaction.23
Interest in  the function of  sulfhydryl groups in enzyme  catalysis34
prompted a number  of kinetic studies of  the reactions of protein-SH
groups with  organomercuric salts.   Pioneering studies by Boyer35

                                 9

-------
o

o


o
 0
 20
 40
 60
 80
100
                                                         Hg(OH)2
                                                                                              d
                                                                                              o
                                                                                             •H

                                                                                              O
                                                                                             •H
                                                                                              !-l
                                                                                              3
                                                                                              O
                                                                                         o

                                                                                         CO

                                                                                         CO
                                                                                             M-l
                                                                                              O
                                                                                              O
                                                                                              13
                                                                                              CU
                                                                                             T3

                                                                                              0)
                                                                                              &
                                                                                              CU
                                                                                             T)
                                                                                             CN4
                                                                                         CD
                                                                                         S-i
                                                                                         Ml
                                                                                         •r-l

-------
  100
   80
    60
O
Q£
UJ

Q_
    40
    20
     0
CH3Hg'
                                   CH3HgOH
       01234567

                             PH
                                  10
 Figure 3.   pH dependence of the hydrolysis of methylmercuric  ion.
                            11

-------
       100
        80
     31
     O
     en
     O
     O
     CfL
         60
         40
         20
            0
C6H5Hg+
                              8
Figure  4.  pH dependence of the hydrolysis of phenylmercuric ion.
                              12

-------
showed that the reaction of £-chloromercuribenzoate (PMB) , a substi-
tuted phenylmercury compound, with sulfhydryl groups could be used
to distinguish kinetically different types of -SH groups in proteins.
Subsequent studies by other groups have established that second-
order rate constants for reactions of PMB with protein sulfhydryl
groups range from about 102 -t/mole sec ("masked" -SH groups)25" 8 to
105 -I/mole sec ("free" -SH groups)38 at temperatures in the 20-30° C
range.  Rapid exchange also occurs with complexed divalent inorganic
mercury.  For example, the sterically hindered reaction of the mercury
dimer of serum albumin (ASHgSA) shown below proceeds with a rate
constant of 5 X 103 I/mole sec at 24°C.29

          ASHgSA + Hg(HPQi)2  -  2ASHg(HP04)                     (9)

Presumably, unhindered thiol-mercury complexes would exchange much
more rapidly.

Recent work has provided kinetic data on the very rapid exchange
reactions involving non-sterically hindered methylmercury complexes.
Using the temperature-jump method, Eigen, Geier, and Kruse have
determined forward and reverse rate constants for reaction 10.30

                       kf
          CH3HgX + OH"  5±  CHgHgOH + JT                         (10)
                       kr

Simpson31 determined rate constants for ligand exchange using the
nuclear magnetic resonance technique of Gutowsky and Holm.32  Rate
constants for the exchange reactions frequently were found to be
close to diffusion-controlled, with the slowest rate constants  about
10 -t/mole sec (Table 1).  These kinetic data indicate that achieve-
ment of ligand exchange equilibrium in systems such as natural
waters that contain multiple chemical species and mercury compounds
is quite rapid even at the very low concentrations observed in  the
aquatic environment.

Relative equilibrium concentrations of reduced and oxidized inorganic
mercury have been calculated by Hem for a model system containing
chloride and sulfate.3   Hem used these calculations to predict  the
most stable forms of inorganic mercury in the aquatic environment.
No calculations of relative equilibrium concentrations of organo-
mercury complexes under environmental conditions have been reported
in the literature.  Calculations that we have carried out for
methylmercury complexes are discussed later in this report.

CHEMICAL DEGRADATION OF ORGANOMERCURIALS

Kinetic studies of chemical degradation of organomercurials
generally have not been carried out in water.  Nonetheless, a few
relevant studies have been gleaned from the literature,

                                13

-------
Table 1.  KINETIC DATA FOR ORGANOEERCURY LIGAND  EXCHANGE REACTIONS
                               kf
                   CHgHgX + Y~  ^  CHgHgY + X~
X
cr
Br~
I"
SOT
SOS2~
CN~
CN~
CN~
or
Y
OH"
or
OH"
OH"
OH~
OH~
cr
S032'
S2Cb2~
Log kf
8.18
8.08
7.61
8.70
6.70
4.2
1.1
2.8
3.5
Log k,.
4.04
5.34
6.84
5.30
5.40
8.9
9.9
8.8
6.7
Reference
30
30
30
30
30
31
31
31
31
                                14

-------
Acidolysis reactions of non-ionic organomercury compounds  (eq 11) and
organomercuric salts (eq 12) have been studied by several  investiga-
tors.

          R2Hg + HX  -  RHgX + RH                               (11)
          RHgX + HX  ->  HgX2 + RH                               (12)

Zimmer and Makower reported that HBr reacted with diphenylmercury
(R = CsHg, eq 11) in 5% water-methanol twice as fast as did HC1.34
Under these conditions, sulfuric, perchloric, nitric, acetic, and
trichloroacetic acids reportedly did not react.  In aqueous tetra-
hydrofuran and dioxane solvents, second-order kinetics were obeyed
when HC1 was the acid.35  Reutov and coworkers reported that
acidolysis of dibenzylmercury (R = C6H5CH2, eq 11) was first-order
in dibenzylmercury and first-order overall,36 but Jensen  and Rickborn
have strongly criticized Reutov1s experimental technique  and inter-
pretation. 7  Kinetic data for dimethylmercury acidolysis  (R =  CHg,
eq 11) in water are not available in the literature.  Dessy and
coworkers reported that the acidolysis of dimethylmercury by HC1 in
dimethylsulfoxide-dioxane  (10:1) was too slow to measure  at 40°C.38
This limited and sometimes confusing information emphasized the need
for kinetic studies of acidolysis in water by dissociated  acids.

Cleavage  of organomercuric salts in water by acids has been inves-
tigated.  Kreevoy reported the second-order rate constant  for
acidolysis of methylmercuric iodide  (R = CHg , X = I),, eq  12, by
1 M H2S04. 9  Extrapolation of these data from higher temperatures
to 25°C gave the very low rate constant, 3 X 10~9 -t/mole  sec.   Under
Kreevoy's conditions the calculated half-life was 3 X 103  days  at
25°C.  Acidolysis of phenylmercuric chloride  (R = C6Hg , X = Cl),
eq 12, in water containing 10% ethyl alcohol was investigated by
Brown.40  Extrapolating Brown's results to 25° gave a second-order
rate constant of 1.7 X 10~~6 -t/mole sec.  Comparison of Kreevoy1 s
and Brown's data indicates that methylmercuric salts undergo
acidolysis at much slower rates than phenylmercuric salts.

Another possible route for chemical degradation of organomercuric
salts is  the so-called "demercuration reaction" shown below.

               H20
          RHgX  ->  ROH + Hg° + X~                               (13)

Ouellette41 has shown that reaction 13 is kinetically first-order
and that  the reaction rate is very rapid for branched alkylmercuric
salts, such as Jt-butylmercuric and cyclohexylmercuric halides.
However,  demercuration of methylmercuric salts is very  slow.  Extra-
polation  of Ouellette's data for CH3HgX to 25° gives the  first-order
                                  15

-------
rate constant, 8 X 10"13 sec"1 or a half-life of about 3 X 104 years.
No data on demercuration of phenylmercuric salts are available from
the literature.

PHOTODECOMPOSITION OF ORGANOMERCURIALS

Most of the abundant literature on the photochemistry of organo-
mercury compounds is concerned with product studies.    Upon
irradiation, the carbon-mercury bond of organomercurials is cleaved
to give alkyl or aryl free radicals and inorganic mercury (eqs 14-17)

               hv
          R2Hg  -  R- + -HgR                                   (14)
           'HgR  -  Hg° + R-                                     (15)

               hv
          RHgX  ->  R* + -HgX                                    (16)
          2-HgX  -  Hg2X2                                       (17)

That alkyl and phenyl free radicals are formed from photolysis  of
organomercuric salts has been demonstrated by electron spin resonance
studies of Janzen and Blackburn.42

Quantitative mechanistic studies of organomercury photoreactions are
difficult to find in the literature.  However, the photolysis of
dimethyImercury has been intensely studied.  In the vapor phase,
dimethylmercury (R = CH3, eqs 14 and 15) photodecomposed at 30°
with a quantum yield of unity to give methyl radicals and elemental
mercury.43  In the liquid phase, a significant fraction of the  methyl
radicals underwent cage recombination to form ethane, and the
remainder abstracted hydrogen atoms to form methane.    Although
Fagerstrom and Jernelbv have suggested that photodegradation of
dimethylmercury is important in the environment,46 the insignificant
absorption of (CH3 )2Hg at wavelengths > 280 rim   indicates that its
sunlight degradation must be very slow.  Experiments of Strausz,
Do Minh, and Font have shown that photodecomposition of a dialkyl-
mercury compound is extremely slow when wavelengths available from
solar radiation., i.e., Pyrex-f iltered light  (> 290 nm) , are
employed.

Although no quantum yield data were found  for the photodecomposition
of phenyImercury compounds or alkylmercuric salts, a few publica-
tions on the sunlight stability of these compounds are available.
Takehara and coworkers irradiated several  phenylmercuric and
alkylmercuric compounds with several light sources.47  Irradiation

                                 16

-------
of several phenylmercuric salts by an intense 2537A low-pressure
mercury lamp or sunlight resulted in extensive photodecomposition.
Photoreaction was more rapid when the compounds were dissolved in
water than when they were irradiated as pure solids or in dust
formulations.  Sunlight irradiation of aqueous phenylmercuric
acetate caused 25% decomposition in 10 hours.  Major products from
photolysis of phenylmercuric acetate and phenylmercuric chloride in
water were reported to be Hg20 and Hg2Cl2, respectively.  Shiina
and coworkers reported results that seemingly conflicted with
Takehara's findings.  These workers reported that irradiation of
phenylmercuric acetate, phenylmercuric chloride, methylmercuric
iodide, and several other phenylmercuric salts resulted in little
decomposition after light exposure "equivalent to 7 summer days" of
sunlight.48  The apparent conflict in the two publications was
probably due to differences in analytical procedures.  Takehara
analyzed directly for the residual organomercury compounds, whereas
Shiina measured the "residual pesticide effect" of the photolyzed
organomercurials upon a fungus culture.  Comparison of the two
publications suggests that the photoproducts also act as effective
fungus growth inhibitors.

Taken together, the published studies indicate that sunlight photo-
decomposition of organomercury compounds may provide an important
pathway for conversion of such compounds to inorganic mercury.  The
lack of quantitative data on the rates of these photoreactions
prompted the studies included in this report.
                                17

-------
                             SECTION V

                       MATERIALS AND METHODS
MATERIALS

Reagent grade dimethylmercury and methylmercuric salts were purchased
from various commercial sources.  Dimethylmercury was purified by
distillation:  bp 94-95°  (760 mm).  Diphenylmercury and phenylmercuric
salts were obtained from the Perrine Primate Laboratory, U. S.
Environmental Protection Agency, Perrine, Florida.  Diphenylmercury
was chromatographed on Woelm basic alumina (activity I), and
recrystallized from 95% ethyl alcohol.  Phenylmercuric acetate was
recrystallized from 570 acetic acid in water and phenylmercuric
nitrate was recrystallized from 95% ethyl alcohol.  Reagent grade
mercuric acetate was purified by recrystallization from glacial
acetic acid.

Water that was distilled, passed through ion exchange columns, and
redistilled was used in all experiments except those carried out
with natural waters.  The natural water samples were obtained from
a pond near Athens, Georgia, and a western North Carolina stream
located at the Coweeta Hydrologic Laboratory.

Benzene was acid-washed1,  dried, and distilled.  Other reagent grade
solvents were used as received.  Cis-l,3-pentadiene was distilled
(bp 42-43°, 760 mm), and stored under nitrogen at -20°C.  Reagent
grade thioglycolic acid and 2-mercaptoethanol were used as
received and stored at -20°C.  Cysteine hydrochloride and Na2S-9H20
were used without purification.  Reagent grade mercuric oxide, acids,
and bases were used as received.

ANALYTICAL PROCEDURES

Acidolysis of dimethylmercury was followed by two analytical
techniques:  (1) gas liquid chromatography (glc) of dimethylmercury
on a Porapak QS column using cetrahydrofuran as an internal
standard, and (2) a conductometric technique, described in the
literature.38  Aqueous solutions of methylmercuric and phenylmercuric
salts were analyzed by the dithizone method of Gran.49  Phenylmercuric
salts were also analyzed by ultraviolet spectroscopy. °  Diphenyl-
mercury acidolysis in water was followed by ultraviolet spectroscopy,
as described by Kaufman and Corwin.51  Kinetics of the mercuric-salt
cleavages of aqueous dimethylmercury and diphenylmercury were
determined by measuring the disappearance of mercuric salt by
flameless atomic absorption  (AA) spectroscopy.52

                                18

-------
Actual yields for sensitized photolysis of diphenylmercury in benzene
were determined by glc on a 37» OV-1 column.  Yields for photodecompo-
sition of methylmercury-thiol complexes were measured by nuclear
magnetic resonance (nmr) spectroscopy and gas liquid chromatography.
Mercury yields in the precipitates resulting from photolysis of
phenylmercuric acetate and methylmercury-thiol complexes were
measured by digesting the precipitates with aqua regia (3:1 HN03-HC1)
and determining the mercury by flameless atomic absorption spectros-
copy.  YieldSi of benzene from the photolysis of phenylmercury
compounds were measured by glc on a Porapak PS column.  Product
yields in the chemical actinometer were determined by gas liquid
chromatography as described elsewhere. 4

APPARATUS

Kinetic studies of methylmercury and phenylmercury compounds were
conducted in a thermostated oil bath that regulated temperature
within + 0.05°C.  One of the major problems in working with DMM was
its volatility (vapor pressure 50 mm Hg at 20.5°C),  a which made it
difficult to accurately weigh and transfer small quantities.  In
addition, DMM dissolved slowly (several hours) in water.  Special
bombs were constructed for the studies of dimethylmercury chemical
reactions and photolysis (Figure 5).  The closed bombs were almost
completely filled with solutions to minimize loss of gaseous dimethyl-
mercury.  When the stopcock of the bomb was opened, aliquots of the
reaction solution were removed by inserting a syringe needle through
the septum.  The aliquots were analyzed by glc as described above.
Control experiments showed that dimethylmercury did not volatilize
from the bomb at the elevated temperatures used in the kinetic runs.
Dimethylmercury acidolysis was also followed with procedures and
apparatus described by Davis and McDonald.5  ' c  Solutions for the
conductance studies were prepared and added to the cell under a
nitrogen atmosphere in a glove box.

Quantum yield studies were carried out on a photochemical apparatus
consisting of a rotating turntable assembly contained in a water bath.
A Hanovia 450-W medium pressure mercury lamp, positioned in the
center of the turntable, was employed as the light source.  The
apparatus has been described in detail elsewhere.    Samples for
photochemical studies were degassed by several freeze-thaw cycles
under vacuum.

Glc analyses were performed on a Tracer MT-220 gas chromatograph,
equipped with flame detectors and a Ni-63 electron capture detector.
Glc peaks were integrated by a Varian Model 477 Digital Integrator.
Mass spectra were obtained on a Hitachi-Perkin Elmer RMU-7 Mass
Spectrometer.  Nmr spectra of the methylmercury-thiol complexes were
measured on a Varian HA-100 NMR Spectrometer.  Flameless AA analyses
were carried out using a Perkin-Elmer 403 Atomic Absorption

                                 19

-------
                 SEPTUM
                 TEFLON STOPCOCK
                  D!METHYLMERCURY
              /  SOLUTION
             V
                 GLASS TUBE
Figure 5. Enclosed bomb for kinetic studies,
                20

-------
Spectrophotometer.   Ultraviolet spectra were measured by a Perkin-
Elmer 352 Ultraviolet Spectrophotometer and uv analyses were
carried out on Bausch and Lomb Spectronic 505 and Beckman DU Spec-
trophotometers.  X-ray diffraction studies were carried out at J. F.
Kennedy Space Center, Florida.

ACIDOLYSIS OF DIMETHYLMERCURY, METHOD 1

A weighed portion of dimethylmercury was added to a volumetric
flask containing 0.0040 M tetrahydrofuran (the glc internal standard)
in water and a magnetic stirring bar.  The resulting mixture was
stirred for 12 hours to dissolve the dimethylmercury, which was
found to have a solubility limit of 0.02 M in water at room tempera-
ture.  An aliquot of standardized acid solution was added to the
aqueous dimethylmercury solution, and the resulting solution was
transferred by a syringe to the kinetic bomb (Figure 5), which was
then totally immersed in the thermostated oil bath.  Periodically,
the bomb was removed from the bath and cooled by immersing in water
at room temperature.  After a 1.0-microliter aliquot was removed
from the bomb and analyzed by glc, the bomb was returned to the
bath.  Second-order rate constants for acidolysis were calculated
                                                  t-i "7
by computer using a least-square analysis of data.

ACIDOLYSIS OF DIMETHYLMERCURY, METHOD 2

Dimethylmercury was added to water in a volumetric flask containing
a magnetic stirring bar.  The flask was placed inside the glove box
under nitrogen and the dimethylmercury was dissolved by stirring for
12 hours.  An aliquot of standardized acid solution was added to the
dimethylmercury solution under nitrogen in the glove box, and the
resulting solution was transferred to the conductance cell.  The cell
was immersed in the thermostated oil bath, and the conductance of
the solution was recorded at appropriate time intervals.  Reactions
were carried out with a 100-fold excess of dimethylmercury, and
pseudo-first-order kinetics were observed.  Pseudo-first-order rate
constants were calculated by computer employing a least-squares  fit
of the data.57

MERCURIC SALT CLEAVAGE OF DIMETHYLMERCURY

Aqueous solutions of mercuric perchlorate (10™6 to 10~9 M) were
prepared by the addition of perchloric acid to mercuric acetate
solutions.  The pH was measured with a pH meter and the initial
mercuric salt concentration was measured by flameless AA spectro-
photometry.  The mercuric perchlorate solution was then transferred
to a reaction bomb and the bomb was immersed in the oil bath
thermostated at 27.0°.  An aliquot of aqueous dimethylmercury
solution was added and, at appropriate time intervals, aliquots  of
the reaction solution were removed and the mercuric salt concentration


                                 21

-------
was measured by flameless AA spectrophotometry.  Control experiments
established that the mercuric salt concentration was not decreased by
adsorption on the glass walls of the reaction vessel during the time
periods of the kinetic runs.  Second-order rate constants were calcu-
lated as described above.

ACIDOLYSIS OF DIPHENYLMERCURY

Aqueous solutions of diphenylmercury (~ 1CT6 M) were prepared by
adding weighed amounts of diphenylmercufy to water and stirring for
several days.  Exact concentrations were determined by uv analysis.
An aliquot of standardized perchloric acid was added and the rate of
disappearance of diphenylmercury was determined by uv spectroscopy.

MERCURIC SALT CLEAVAGE OF DIPHENYLMERCURY

The procedure for the kinetic runs was essentially the same as
described above for dimethylmercury with the exception that the
mercuric perchlorate and diphenylmercury concentrations were lower
(10~9 to 10~10 M).  Attempts to follow the reaction by ultraviolet
spectroscopy with 1CT6 M reactant concentrations were unsuccessful,
because complete cleavage occurred within a matter of seconds in
neutral and acidic media.

PHOTOCHEMICAL PROCEDURES

Solutions of organomercury compounds were irradiated by broad-band
(> 290 nm) and monochromatic (313 nm) light from the mercury lamp.
Light from the mercury lamp was filtered through a Pyrex sleeve for
the broad-band studies and through a Pyrex sleeve and 1.0 cm of A
solution of 0.001 M potassium chromate in 2% aqueous potassium
carbonate to isolate the 313 nm line.  Procedures for preparing
and degassing sample and actinometer tubes have been previously
described. 4

A valerophenone actinometer58 was used for the studies with 313 nm
light and a benzophenone-cis-1,3-pentadiene actinometer59 was used
for the broad-band irradiations.  After irradiating the organo-
mercurial and actinometer solutions in parallel in the quantum
yield apparatus, quantum yields were calculated by comparing actual
yields of photoreaction in the solutions.54

Sensitized photolyses were carried out with sufficient concentrations
of sensitizers to absorb > 99% of the light.  Singlet oxygen was
generated by methylene blue sensitization.60  Quenching studies
were carried out under conditions where no light was absorbed by
the quenchers.  Disappearance quantum yields for phenylmercurials
and sulfur-bonded methylmercury complexes were independent of the
extent of reaction to at least 35% of completion.  All quantum
yields were determined at 25° C.

                                22

-------
PHOTOLYSIS OF- METHYLMERCURIC SULFIDE ION

Aqueous solutions of methyImercurie sulfide ion (0.010 M) were pre-
pared by reacting methylmercuric hydroxide with a two-fold molar
excess of sodium sulfide.  Photolysis by > 290 nm light caused
formation of gas and a black precipitate.  The gas was characterized
by glc and the precipitate was identified by X-ray diffraction.

Yields of the mercuric sulfide precipitate were determined as
follows:  (1) The photolyzed solutions were diluted 1:100.  Mercuric
sulfide is soluble in concentrated sodium sulfide and therefore did
not completely precipitate prior to dilution.61  The diluted solutions
were allowed to stand in the dark for 12 hours to ensure complete
precipitation.  (2) The supernatant liquid was partially decanted
and the remaining mixture was centrifuged.  (3) After centrifugation,
the precipitate was washed with water, recentrifuged, and dissolved
in aqua regia.  (4) Mercuric ion concentration was determined as
usual by atomic absorption spectrophotometry.

SUNLIGHT PHOTOLYSIS OF METHYLMERCURY COMPOUNDS

Aqueous solutions  (1.00 X 10~4 M) of methylmercuric hydroxide,
methylmercuric chloride, methylmercuric bromide, and methylmercuric
iodide were degassed as usual in 10.0 + 0.1 mm quartz tubes.  The
tubes were sealed under vacuum and irradiated by September sunlight
between 9 AM and 3 PM on the roof of the Southeast Environmental
Research Laboratory (SERL). Athens, Georgia.  Methylmercuric iodide
was irradiated for 3.7 hours and the other methylmercury compounds
were irradiated for 17.1 hours.

ACETONE-SENSITIZED PHOTODECOMPOSITION OF PHENYLMERCURIC ION AND
PHENYLMERCURIC HYDROXIDE

Degassed aqueous solutions of phenyimercuric perchlorate  (1,00 X 10"3 M)
and acetone (0.60 M) adjusted to pH 2.3 and pH 10.2 were irradiated
by Pyrex-filtered light.  Benzene was determined by glc and precipi-
tates were analyzed for mercury content by flameless AA. spectropho-
tometry.  Nearly quantitative yields of inorganic mercury precipitated
under basic conditions  (0.95 mole Hg per mole C6H5HgOH decomposed)
and lower yields (0.55 mole Hg per mole CgHgHg"1" decomposed) precipi-
tated under acidic conditions.  Analysis of the supernatant of the
acidic solution showed that it contained an additional 0.4 mole
Hg per mole C6HgHg+ decomposed, presumably in the form of mercurous
and/or mercuric ions.

QUENCHING STUDIES OF PHENYLMERCURY COMPOUNDS

Solutions containing the phenylmercury compound and various con-
centrations of quencher (0-4 M) were prepared, degassed  (unless
                                23

-------
molecular oxygen was quencher), and irradiated in parallel at 313 run
on the photochemical apparatus.

SUNLIGHT PHOTOLYSIS OF PHENYLMERCURY COMPOUNDS

Air-saturated solutions of diphenylmercury (1.0 X 10~6 M) and
phenylmercuric salts (4.0 X 10~5 M) in water were sealed in 13.0
+0.1 mm quartz tubes.   The tubes were irradiated by sunlight in an
exposed area on the roof of the Southeast Environmental Research
Laboratory, Athens, Georgia.  In a preliminary experiment, tubes
were irradiated for 18 days during July (average temperature, 30°C
during daylight hours).  In a second experiment, tubes were irra-
diated for two days in August (average temperature 31°C).  During
this period the weather was mostly fair, and the tubes received
20.0 hours of sunlight.  Unphotolyzed controls showed no decrease
in phenylmercurial concentration during the irradiation periods.
Based on the fraction of phenylmercurial that disappeared, photolysis
rate constants and half-lives were calculated assuming first-order
kinetics.62' 63
                                24

-------
                            SECTION VI

                      RESULTS AND DISCUSSION
CALCULATED EQUILIBRIUM CONCENTRATIONS OF METHYLMERCURIC SPECIES IN
AQUATIC SYSTEMS

Although the literature is devoid of quantitative information about
formation of organomercury complexes in the environment, previous
studies have provided abundant information concerning equilibrium
constants for methylmercury complexes in water,15' 18  We have devised
a technique   utilizing these equilibrium constants to calculate
relative concentrations of methylmercury complexes in aqueous systems
containing chemical species found in natural waters, i.e., hydroxide,
chloride, hydrogen sulfide and its dissociated forms, thiols  (RSH),
amines  (RNH2), phenols (humic acid), ammonia, and orthophosphate.64
Bicarbonate and organic carboxylic acids were not considered because
they form very weak complexes with the methylmercury moiety.  To
render  these calculations relevant to the aquatic environment, we
used recently measured concentrations of chemical species in Lake
Erie65 where high concentrations of mercury have been found in fish.6

The Project Hypo study of the Lake Erie central basin hypolimnion
showed  that high concentrations of hydrogen sulfide are present in
the water column near the bottom sediments.67  Formation of locally
high concentrations of hydrogen sulfide is a phenomenon observed in
many natural waters.68  Complexation of the methylmercury group by
reduced sulfur species (H2S, SH"~ , and S2~) results in the formation
of two  species, methylmercuric sulfide ion (CH3HgS~) and bis(methyl-
mercury) sulfide ((CH3Hg)2S),18  The two species are in equilibrium
as shown in equation 18.

          ZCHgHgS" + H+  5±  (CH3Hg)2S + SH~                    (18)

Our calculations indicate that these two species should account for
very high fractions (> 95%) of the methylmercury complexes in
natural waters that contain reduced sulfur species (Figures 6 and 7).
Relative concentrations of CILjHgS" and (CH3Hg)2S depend upon  the pH
and total concentration of reduced sulfur species.  At high concen-
trations of reduced sulfur species (10~3 to 10~4 M), CH3HgS~  accounts
for virtually all of the complexed methylmercury in the pH 5-9 range
(Figure 6).  As the reduced sulfur concentration drops,  (CHgHg)2S
becomes the predominant methylmercury complex in acidic waters
(Figure 7) and CHgHgS" remains the predominant complex in basic
waters  (Figure 6).
                                 25

-------
en
3:
 ff\
a:
CTl
    100
    90   -
    80   -
    70
    60
    50
    40   -
    30   -
    20   -
     10   -
              [CH3H9JIOI
                      [HPO
              [ci -
              [NH/] + [NH,]
              [RNH,+] +  [RNH;]
              [RSH]  + [RS-]
              [ArOH] + [ArO'
              [H2S]  + [SH-]
                                                                    10
   Figure 6.   Relative concentrations  of CHgHgS"  in systems
                containing multiple chemical  species.
                                  26

-------
     100
      70
      60
      50
      40
       30
       20
       10
[HjFO,-] * [HP04
[ci-
[NH,+] * [NH,]
[RNH,+]
[RSH]  + [RS~]
[ArOH] + [ArO-
[HjS]  * [SH~]
                         [S
Figure 7,   Relative concentrations of  (CH3Hg)2S in  systems
             containing multiple  chemical  species.
                                27

-------
In solutions containing the chemical species at concentrations
indicated in Figure 6, but no reduced sulfur, nearly all the methyl-
mercury would be complexed by organic thiols, e.g., sulfhydryl-
containing proteins, in the pH 5-9 range (Figure 8).  Thus, when
either reduced sulfur species or organic thiols are present in
natural waters, methylmercury is quantitatively complexed by
sulfur-bonded ligands.

Exclusion of reduced sulfur species and organic thiols from the
system, permits complexation by more weakly binding chemical species.
The plot of calculations shown in Figure 9 indicates that in the
absence of sulfides and thiols, methylmercuric hydroxide is the
major complex in basic waters  (pH 7-10) and methylmercuric chloride
predominates in acidic waters.

One noteworthy aspect of the calculations is their prediction that
commonly occurring chemical species such as orthophosphate, ammonia,
phenolic groups in humic acid, and amino groups in protein have
relatively little impact upon complexation of methylmercuric ion
under environmental conditions.  Moreover, the calculations indicate
that methylmercury can be freed from sulfur-bonded complexes by
strong acidification.  Experiments have also shown that methylmercury
can be freed from sediments or biological samples by treating the
samples with strong acid.69

Equilibrium calculations are strictly applicable only to closed
systems, and equilibrium is closely approximated in natural waters
only if the rate of approach to equilibrium is more rapid than the
rate of change of environmental conditions.70  Kinetic studies of
ligand exchange reactions discussed previously have shown that
achievement of equilibrium is very rapid for such reactions.  More-
over, the Lake Erie Time Study of Kramer and coworkers indicated
that the rate of change in concentration of the chemical species in
Lake Erie considered in the above calculations is relatively slow
even during periods of high biological activity.66  Thus, near
approach to the calculated equilibrium concentrations of methyl-
mercury complexes is probable in local regions of lakes and slower
moving rivers and streams.

Although the literature does not contain a great deal of information
concerning equilibrium constants for phenylmercury complex formation,16' l7
sufficient data are available to indicate that relative stabilities
of phenylmercury complexes parallel those of methylmercury complexes.

KINETICS OF ORGANOMERCURY REACTIONS IN WATER

Theoretical Considerations

The reaction mechanism, the rate expression, and the rate constant
are required to evaluate the significance of a reaction in the
                                28

-------
en
    80
    60
UJ
o
Q_
               [CH3HgCI]
                [CH3Hg]T
                2
[CH3HgSR]
 [CH3Hg]TOT
                                      SPECIES
                     com. M
                      10
                        -5
[CH3Hg]TOT              
-------
 o
 I—
 Gn

IT,
0_j
i__
^
UJ
C£
UJ
20
 0
    0
        io2
    [CH3HgCI]
     [CH3Hg]
                    TOT
III!
                                      10Z
[CH3HgOH]
 [CH3Hg]T01
                                          SPECIES
                                           1 TOT
                            [CH3HgJ
                            [H2P04~J  +  [HP04=]
                            [cr]
                            [NH4+] +  [NH3J
                            [RNH3+] + [RNH2J
                            [RSH]  +  [RS-]
                            [ArOH] +  [ArCT]
                            [H2S]  +  [SH~]
                                                   CONC, M
                                                      [S=]
                              I    |Xj    i
                     6
                    pH
                                          8
                  12
                        10'6
                        io-4
                        ID'5
                        1Q-6
                         0
                        io-5
                         0
   Figure 9.  Relative concentrations of methylmercury complexes
             in systems containing multiple chemical species,
             excluding reduced  sulfur species and organic  thiols.
                               30

-------
degradation or transformation of a pollutant.  Also determination of
rate constants at different temperatures will allow calculation of
activation parameters, which are indicative of the reaction pathway,
and will allow extrapolation of the rate constant to other tempera-
tures.  It is desirable to carry out these studies in water when
possible because of pronounced solvent effects on rates and products.

Since rate constants are sometimes difficult to relate to the life-
                                                                    rv -i
time of a reactant, the half-life (t^) expression is often employed.
The half-life is defined as the time required for the concentration
of a reactant to be reduced to one-half its initial concentration.

The tj, expression is especially convenient for first -order71
reactions (eq 19) .  The rate of disappearance of reactant A is given
by differential equation 20.

             ki
          A  -  B                                              (19)


                                                               (20)
             dt

The tj, expression derived from the integrated form of equation 20 is
given in equation 21.

               0.693
The tj, is independent of the concentration of reactant A and
dependent only on the magnitude of the rate constant  (k^ ) .

The ti, expression for second-order71 reactions requires a statement
of reactant concentrations.  The disappearance of reactant A  (eq 22),
given by equation 23, includes the concentrations of both reactants.
          A + B  -  C                                           (22)


               1= k2[A][Bj                                     (23)
             dt

The derivation of the half -life expression by integration of equation
23 gives an equation too complex to be of practical value.  However,
in the special case of equal initial concentrations of reactants,
the tj, expression may be expressed as
                                31

-------
          tj. = ~ —                                            (24)
where [A] = [B]

Another special case results when the effective concentration of one
reactant does not change with time (pseudo-first-order reaction) .
This occurs when one reactant is present in large excess or its
effective concentration does not change with time because of buffering.
Half-life of reactant A is dependent only on the concentration of  the
reactant in excess [B] and the magnitude of the rate constant (k2)
(eq 25}.
                                                                (25)
when  [B~l > > > [A]

A  graphical representation of this relationship is given in Figure  10
in which tx is plotted as a function of concentration for  specific
rate  constants.

The application of kinetics in terms of the reactivity or  transforma-
tion  of a pollutant requires an understanding of the reaction  process
and reaction conditions.  This is particularly true in extrapolation
of laboratory to the aquatic environment.

Acidolysis of Dimethyl- and Diphenylmercury

Cleavage of the carbon-mercury bond by protic acid in dialkyl- or
diarylmercury compounds is referred to as acidolysis  (eq 26) ,6a' b
This  reaction has been proposed by several investigators4'  Ba' 72  to
be a  pathway for the chemical transformation of dimethylmercury to
methylmercury derivatives (Figure 1) in the aquatic environment.

                             H20
                   + H1" + X~  -  CH3Hg+ + X~ + CHi              (26)
          X = Cl  , Br~ , I", ClQj", N03~

However, as discussed  earlier  (Section IV),  the  rate  constant  for
acidolysis of dialkyl- or  diarylmercury  compounds  in  water  has  not
been determined.  Even with compounds for which  the kinetics have
been determined in organic solvents,  the reaction  mechanisms are
not clearly defined.

Preliminary work was done  to evaluate a  glc  method for  following
the acidolysis kinetics.   Although the organomercuric salts were

                                 32

-------
    41.2
    4.12
    0.41
   0.041
   0.0041
        10
         -5
                                     1000
                                     100
                                     10
                                                                 13
                                                                 O
                                                                 .c
                                     1.0
                                     0.1
                                                             0.01
KT6              10'7

    [B], moles/1
10
                                                            -8
Figure  10.  Half-life of A vs. concentration  of B for  second-order
             reactions when "Us"] > > > [A],
                                  33

-------
found to decompose during chromatography,73 DMM was found by combined
GC-MS to be stable to glc conditions employed.  The kinetics of
acidolysis were therefore determined by following the disappearance
of DMM by this technique.  The methyl- and phenylmercuric salts were
shown to be stable under the acidolysis reaction conditions, as
reported earlier.4 > 41

A plot of the second-order rate expression71  (Figure 11) indicated
that the dimethyImercury acidolysis reactions obeyed second-order
kinetics through 50-757o of the reaction.  The variation in slope
with temperature illustrates the temperature dependence of the
reaction.

The kinetic data for HC1, HBr, and HC104 in Table 2 demonstrates
that the second-order rate constants are independent of the nature
of the acid.  In water at the low reactant concentrations indicated,
strong mineral acids are completely dissociated and no anion depen-
dence is observed.  For HC1 the extrapolated rate constant at 25°C
is 7.33 X 1CT5 -I/mole sec.

Verification of the kinetics was obtained by employing a 20-fold
excess of HC1 compared to DMM and determining the pseudo-first-order
rate constant at 40° .  The second-order rate constant was obtained
by dividing the pseudo-first-order rate constant by the HC1 concen-
tration.  The constant so obtained, 4.99 + 0.13 X 10~4 -I/mole sec,
agrees with the extrapolated value of 5.08 X  10~4 t/mole. sec obtained
under second-order conditions.

The glc method of determining the reaction velocity was verified by
a conductance method55 '  ° employing a large molar excess of DMM  compared
to HC1.  The reaction was followed through 25% to 5070 completion by
monitoring the decrease of conductance with time.  This was possible
because of the high specific conductance of IT compared to DMM or
MMX.74

The conductometrically determined rate constants for HC1, HC104, and
HN03 are given in Table 3.  The calculated second-order rate constants
agree with the values reported in Table 2  (glc method) for HC1 and
HClQj .

The above results definitively show the reaction to be first-order
with respect to both DMM and acid  (an electrophilic substitution
reaction  (Sg 2)).  Based on stereo-chemical studies, these reaction?
have previously been thought to proceed by a  four-centered or
similar type mechanism (Sg i), as shown in equation 27.
                            R
                     R---Hg
          R-Hg-R
             •&'
           HC1
  R     R
•  I   +  I                        (27)
  H     Hg

        Cl

 34

-------
03
     48
     40
     32
     24
     16
          o
                                                          o
                40      80      120      160     200     240     280     320
                               TIME, minutes

              Figure 11.   Acidolysis of dimethylmercury.
                                  35

-------
                 Table  2.   DIMETHYLMERCURY ACIDOLYSIS RATE DATA MEASURED BY GL(f
Acid
HC1
HC1
HC1
HC1
HC1
HC1

HC1
HC1
HBr
HBr
HC104
HC104
Cone. Acid
M
8.27 X 1CT3
8.27 X 10~3
8.27 X 1(T3
8.27 X 10~3



1.0 X 10"1
1.9 X 10"1
8.24 X 10~3
8.24 X 10~3
8.20 X 10"3
8.20 x 10~3
Cone. DMM
M
3.46 X 10~3
5.43 x 10~3
4.90 X 10~3
4.60 X 10~3



6.16 X 10~3
5.28 X 10" 3
3.47 X 10"3
6.82 x 10~3
4.79 x 10~3
6.08 X 10"3
Temp.
°C
65
65
85
85
40°
c
25

40
40
85
85
65
85
k -I/mole sec
(7.19 + 0.32) X 10"3
(7.43 + 0.19) X 10~3b
(4.89 ± 0,07) X lO"2
b
(4.87 ± 0.39) X 10~2
5.08 X 10~4
7.33 X 10~5
d
5.12 X 10~4
4.87 X 10~4
-------
                 Table 3.  DIMETHYLMERCURY ACIDOLYSIS RATE DATA DETERMINED CONDUCTOMETRICALLY21
Acid
HG1
HC1
HClQt
HClCx
HNOg
HNOg
Cone. Acid
M X 10*
7.13
7.13
7.20
7.20
6.90
6.90
Cone. DMM
M X 102
1.16
1.07
0.501
1.01
1.15
0.52
Temp.
°C
65
85
65
85
65
85
Pseudo-First Order
Rate Constant
k sec"1 X 104
0.792
4.12 -f 0.0005
0.756 + 0.01
1.94 + 0.01
0.937 + 0.48
2.70 + 0.01
Second-order
Rate Constant^
k I/mole sec x 102
0.70
4.0
0.78
4.1
0.81
5.4
U)
        aSee Section V.
         Obtained from pseudo-first-order rate constants.

-------
In the case of protic acid cleavage of DMM in water, where the strong
mineral acids are completely dissociated, the reaction probably takes
place by a pure Sg 2 mechanism (eq 28).
          R-Hg-R       R-Hg         R       R
            +           ;        -   I    +   I                    (28)
            H+          H+          H       Hg+

Because a myriad of chemical species exist in the aquatic environment,
several reactants that might be expected to affect the mechanism were
investigated to determine their effect on the rate constant  (Table 4).
None of the salts tested appreciably affected the rate constant.
The small increase noted in some cases may be attributed to  a salt
effect.  Iodide ion, a better nucleophile, complicated the reaction
because CH3HgI precipitated.  However, based on the first 20% of the
reaction, a rate constant of 7.6 X 10~2 I/mole sec at 85° indicated
no apparent rate alteration.

In the presence of S=, the reaction could not be followed because
NasS addition resulted in interfering glc peaks.  However, glc
analysis -iid reveal that DMM was stable to Na2S (1CT2 M) at  85° for
24 hours in water.

Since proteins are common to the aquatic environment, an experiment
was designed to check for any effect they might have on acidolysis.
Cysteine, a thiol-eontaining amino acid was chosen as a representa-
tive compound.  The rate constant for the acidolysis of DMM  by  cysteine
hydrochloride is given in Table 4.  The kinetics were complicated by
the inability to accurately determine the actual H1" concentration,
The acid concentration was obtained by measuring the pH at 65°
(pH = 2.32).  The rate constants determined were about 50% lower
than those obtained with HC1; the possibility of a large rate
alteration by proteins was therefore ruled out.

The rate constant for DPM acidolysis by HC104 was determined in both
aqueous ethanol and in water (Table 5).  Aqueous ethanol was used
because the solubility of DPM in water  (10~6 M) was too low  for
convenient determination of the rate constant under second-order
conditions,  Product studies indicated that the reaction proceeded
according to equation 29.

                                  H20
                 sHe + if + C104~  ->  C6H5Hg+ + C104~ + C6He    (29)
The reaction obeyed second-order kinetics through 50%  to  75%
completion in both 30% and 40% ethanol-water  solutions.
                               38

-------
         Table 4.  DIMETHYLMERCURY ACIDOLYSIS  IN  THE PRESENCE  OF  NUCLEOPHILIC SPECIES3
Acid
HC1

HC1

Cysteine-HCl1'

Cysteine-HCl1'

HC1

HC1
Cone. Acid
M
8,27 X 1(T3
8 .27 X 1CT3
7.11 X 1CT3
7 .,84 X 10~3
8,27 X 10~3
1.0 X 1CT2
Cone „ DMM
M
3.40 X 10"3
5.72 X 10"3
6.36 X 10~3
5.60 X 10~3
3,50 X 10"3
5.05 X 10~3
Temp.
°C
85
85
65
85
85
85
k -t/mole sec
(5.13 + 0.05) X 10"2
(4.90 + 0.07) X 10~2
(4.54 + 0.14) X 10*3
(3.48 + 0.08) X 10"2
(7.61 + 0.02) X 10~2

Nucleophile
1.03 X 10~2 M NaCl
__
__
1.02 X 10"2 M Nal
Na2S
aGas-liquid ehromatography technique  (see  Section V).
*pH 2.23.

-------
                               Table 5.  DIPHENYLMERCURY ACIDOLYSIS RATE DATA
Temp.
°C
70
70
70
70
70
50
25
% Ethanol
in Water
40
40
30
30
0
0

Cone. DPM
M
8.80 X 10"5
8.80 X 10~5
8.80 X 10~5
8.80 X 10"5
1.02 X 10~5
1.02 X 10~5

Cone. HC104
M
2.40 x 10~4
2.40 X 10"4
2.40 X 10"*
2.40 X 10~4
1.37 X 1CT3
1.37 X ID"3

k sec"1 X 104




(7.80 + 0.3)
(1.53 + 0.02)

k -I/mole sec x 101
1.88 + 0.07
1.69 ± 0.07
1.44 + 0.04
1.31 + 0.05
5.69 + 0.19
1.12 + 0.02
9.67 X 10"3a
AH*
kcal/mole




18.1

AS*
eu




-7.0

-p-
o
       'Extrapolated value.

-------
When the pseudo-first-order rate constant was determined in water
using a
kinetics
  i       	__  _„_.	—
 large  excess of HC104,  the  reaction  obeyed  first-order
;  through one half-life  (see Figure 12). x
The rate constant increased as the ethanol concentration was reduced
from 30% to 0%.  Although the rate increase was small for such a
large solvent change, qualitatively it agrees with that expected
from the increase in acidity with decreased ethanol concentration. 5
The small decrease in rate constant in going from 40% to 30% ethanol
was probably due to solvation effects that could not be evaluated
without further study.

Extrapolation of the rate data obtained at higher temperatures to 25°
gives a constant of 9.67 X 10~3 -I/mole sec for water.

The acidolysis half-lives for DMM and DPM can be calculated at 25°
from these data (Tables 2 and 5).  Assuming that the acid concentra-
tion remains constant because of buffering, pseudo-first-order
conditions would prevail (see eq 25 where [B] = [H4"^)'  At pH's
commonly found in the environment, DMM would have a relatively long
half-life.  For example, at pH 5, t^ would be 33 years at 25°.  For
DPM under the same reaction conditions, it would be 0.25 years.
Thus, acidolysis may be important for DPM under certain conditions,
but for DMM it would be important only at low pH's.

The literature contains limited quantitative data on the relative
rates of acidolysis for dialkyl- and diaryImercurials.  However, a
qualitative order of reactivity is given by Kharasch and Grafflin76
(Table 6).  They did not measure actual rates, but rather determined
relative rates using unsymmetrical organomercurials in competition
studies.  They also demonstrated that the relative order of reactivity
did not change for a variety of acids and solvent systems.  Our data
for DMM and DPM are in agreement with this order.

Since the phenyl group is the most reactive moiety listed in Table 6
and acidolysis of DPM is low under environmental conditions, other
organomercurials containing the alkyl groups listed would react even
slower.

Desymmetrization of Dimethyl- and Diphenylmercury

Carbon-mercury bond cleavage of dialkyl- or diarylmercury compounds
by mercuric salts is termed desymmetrization a'   (eq 30).  The
equilibrium for this reaction generally lies far to the right as
shown in Table 7.

                       desymmetrization
          RgHg + HgX3          ^         2RHgX                  (30)
                        symmetrization

                                 41

-------
0.5
0.4
0.3
0.2
0.1
                       4         6
                      TIME, sec x K
10
        Figure 12.  Acidolysis  of  diphenylmercury,
                            42

-------
    Table 6.  INCREASING EASE OF ACID CLEAVAGE OF ARYL
                                                  ry £•
              AND ALKYL GROUPS IN ORGANOMERCURIALS
 1.   Methyl

 2.   Ethyl

 3.   T|-Butyl

 4.   Tj-Propyl
5.   Benzyl

6,   Cyclohexyl

7.   Phenyl
Table 7.  CALCULATED EQUILIBRIUM CONSTANTS FOR THE REACTION
          OF DIMETHYLMERCURY WITH MERCURIC HALIDES
                    (EQ 30, R =
 Salt

 HgCl2

 HgBr2

 HgI2
         K
    3.5 X 10

    2,5 X 109

    3.4 X 105
            ,11
                            43

-------
Kinetic data for the desymmetrizations of DMM and DPM by Hg(C104)2
are summarized in Tables 8 and 9, respectively.  Rate constants were
obtained by use of the integrated second-order rate expression,71
which was obeyed through two half-lives as shown in Figure 13.

In the kinetic studies, Hg(C104)2 formed by the reaction of Hg(QAc)s
with perchloric acid gave results identical to those obtained with
Hg(C104)2 formed by reacting HgO with HC104 .  The perchlorate was
chosen because it is completely dissociated in aqueous solution,
eliminating interference by any associated salt, which simplifies
the reaction kinetics  (see Section IV).

The desymmetrization rate constants for DMM and DPM show a strong
pH dependence varying  six orders of magnitude over the pH range of
1 to 9.  The reaction velocity for DPM at pH 2 was so fast (near
diffusion controlled)  that the rate constant could only be estimated
by single-point determinations.  The rate constant obtained  (2 x 108 -L/
mole sec)therefore represents a lower limit.  A calculation of
acidolysis half-lives  at the pH's employed indicates that acidolysis
was not a competing reaction under the conditions of the desymmetri-
zation reactions.

The strong pH dependency of the desymmetrization reaction may be due
to the dependence of the concentration of the various mercuric
species on pH.  The hydrolysis of mercuric ion  (eqs 31 and 32)
involves three different mercuric species  (Hg+  , HgOH1", and Hg(OH)2)
whose relative concentrations are a function of hydronium ion con-
centration.

          Hg++ + H20  «±  Hg+OH + H1"                             (31)


          Hg+OH + H20  
-------
            Table 8.  pH DEPENDENCE FOR DESYMMETRIZATION OF DIMETHYLMERCURY AT 27°'
PHb
1.2
3.1
5.8
9.0
Hg(C104)2
M
7 X 10~8
7 X 10~8
3 X 10"7
2 X 10~5
DMM
M
7 X ID'8
5 X ID"8
3 X 10"7
i x io-B
k
-t/mole sec
5 x 104
2 x 104°
1 X 103°
2 X 10"lC
aUsing atomic absorption.
 pH determined at start and end of reaction.
cCalculated by computer program.

-------
Table 9.  pH DEPENDENCE FOR DESYMMETRIZATION OF DIPHENYLMERCURY BY MERCURIC PERCHLORATE AT 27°'
PHb
2.0
4.8
5.8
6.9
8.1
9.4
Cone. Hg(C104)2
M
2.0 X 10~10
5.0 X 10~9
7.0 X 10"9
5.2 x 10'a
6.57 X 10~8
1.06 X 10"7
Cone. DPM
M
1.0 X 10~10
4.1 X lO"9
4.5 X lO'9
4.6 X 10"8
4.56 X 10"8
2.6 X 10"6
k
-t/mole sec
> 2 X 108°
3 X 10sd
3 X 1056
4 X 1046
3 X 1036
3 X 103S
aUsing atomic absorption.
bpH determined at the start and end of the reaction.
cLower limit.
 Calculated from three single point determinations.
e Calculated by computer program.

-------
     48
     40
fc,   32
     24
     16
      8
                                         _L
                10       20       30      40
                          TIME, minutes
50
60
         Figure 13.  Desymmetrization  of  dimethylmercury
                     by mercuric perchlorate.
                               47

-------
                              R
          R-Hg-R         RrHg'           R       R
            +             :              i     +  i
           Hg++            Hg++           Hg+      Hg+            (33)
          R-Hg-R         R---Hg         R       R
            +            !    !      -    I        !               (34)
          Hg-OH          ',    !           Hg  4-   Hg
         '               Hg — OH         I        |
       OH               /               0       0
                      0*                H       H

                          SE i

The high rate of electrophilic  substitution by Hg++ as compared to
Hg(OH)g represents about five orders of magnitude in increased
electrophilicity towards the carbon -mercury bond.   Although extensive
background data are not available to evaluate the desymmetrization
reaction for other mercuric species, the more electropositive the
mercury atom, the larger the rate constant.  For example, in ethanol
the order of reactivity is Hg(N03>3> Hg(OAc)2 > Hg(Br)2:i which is
also the order of decreasing ionic character of the mercury bond.

The reverse reaction, symmetrization (eq 30) has been suggested as
a possible chemical pathway for the conversion of methylmercuric ion
to DMM.45  It was proposed that the equilibrium would be shifted to
the left by removal of mercuric ion as insoluble mercuric sulfide.

A maximum value for the symmetrization rate constant can be obtained
from equilibrium data (K^ = ka es y m /ks y m ) «  Experiments showed that
the equilibrium constant (Kg q )  was > 10*; thus the rate constant for
symmetrization must be four orders of magnitude smaller than that
for desymmetrization.  Using a methylmercuric hydroxide concentration
of 1 ppb (4 X 10~9 M) and the second-order rate constants in Table 8,
calculations (eq 24) show this reaction to have a half -life of
several years.

The effect of sulfide on the symmetrization reaction was examined
employing 0.4 M sodium sulfide and 0.1 M methylmercuric hydroxide in
water at 25°.  At the end of 24 hours, the organoinercurial was 25%
reacted.  The second-order rate constant derived from these data
was ~ 10~6 -I/mole sec, too low to permit the reaction to be environ-
mentally significant.

Based on the rate constants  (Tables 8 and 9), half-lives can be
calculated, under pseudo-first-order reaction conditions, where

                                 48

-------
[Hg11] ([Hg11] = [Hg++] + [HgOtf"] + [Hg(OH)s]) is in excess.  Using
a value of 0.03 ppb for [Hg11] and k = 1 X 1(T ,  at pH 5.8 (25°), the
half-life for reaction with DMM is about 50 days.  For DPM under the
same conditions, the half-life would be four hours.

Dimethylmercury Stability to Oxygen and Base

DMM was shown to be stable to 1 M KOH.  After heating for 24 hours
at 85° , glc analysis showed that DMM had not reacted.  Likewise DMM
neither reacted with KI (five hours at 85°) nor did it react with a
saturated solution of egg albumin  (2% hours at 30°).

Secondary and tertiary dia Iky liner cury compounds are reported to
undergo slow oxidation by oxygen in organic solvents.60   Primary
dialkylmercury compounds are less  susceptible to oxidation.   DMM
reaction could not be detected by  glc analysis after standing for 20
hours at 85° in an aqueous solution saturated with oxygen.  Lack of
reaction under these stringent conditions precludes any significant
contribution to degradation in the environment.

Evaporative Loss

Evaporative loss of organomercurials from the aquatic environment is
a physical process, but because of its potential importance as a
means of transport, it is discussed here.  Many organomercurials have
a substantial vapor pressure.  Although the ionic or polar covalent
mercurial salts when dissolved in water will not be readily lost to
the atmosphere because of solvation effects, mercury and dimethyl-
mercury are non-ionic compounds and evaporative loss may be important.

An estimate of the magnitude of loss may be obtained by employing
the method and data of Tsivoglou. 8  Using the relationship between
transfer coefficients and molecular diameters (Kj/K2 = dz/di) and
van der Waals radii61 to calculate molecular diameters, the ratio
of transfer constants for DMM to oxygen is calculated to be 2.4.
Utilization of Tsivoglou's a experimental data reveals that for a
moderately turbulent river section, DMM would have an evaporative
half-life of about 12 hours.  This relationship also predicts that
dissolved elemental mercury would be lost from the river at a rate
2.3 times faster than DMM.

KINETICS OF ORGANOMERCURY PHOTODECOMPOSXTION

Since tropospheric solar radiation has negligible intensity at
wavelengths less than about 290 nm (Figure 14), 2'    systems must
have appreciable absorptivity at. wavelengths greater than 290 nm if
significant photoreaction is to occur in  sunlight.  Predictions of
sunlight photoreactivity can be derived from laboratory experiments
employing Pyrex-glass filtration of light from a mercury  lamp since
this filter transmits only wavelengths greater than 290 nm.80

                                 49

-------
E
o  3
o

o
X
         300     400     500      600      700


                       WAVELENGTH, nm
800
   Figure 14.   Estimated solar irradiance at

               solar zenith angle =  0  .
                      50

-------
Quantitative calculations of sunlight photolysis rates can be made
from specific absorption rates (ka ) and quantum yields (0) both
determined in the laboratory.81  Concentrations of pollutants such
as organomercury compounds are generally so low that absorption at
wavelengths greater than 290 nm is very weak.  For weakly absorbing
systems, rates of sunlight photodecomposition  (-d[RHgX]/dt) can be
expressed as


                    = ka0[RHgX]                                (35)
              dt

where the term [RHgX^ represents concentration of organomercury
compound.  This equation assumes that sunlight photodecompositions
follow first-order kinetics.

Specific absorption rates for the organomercurials were calculated
using solar radiation data published by Leighton62 and extinction
coefficients at wavelengths > 290 nm.  Leighton1 s data are most
appropriate for Los Angeles during August through November, the
period we used for our experiments in sunlight.  Since Athens,
Georgia, has about the same latitude and elevation as Los Angeles,
the spectral flux distribution of sunlight at Athens should closely
parallel that at Los Angeles.  The value of ka changes during the
day because the solar spectrum is a function of the solar zenith
angle, z.  For this reason, it is convenient to use an integrated
specific absorption rate constant, (ka)lntg for predictions of sun-
light photolysis rates.82
                          z = n/2
           (ka)lntB  =   j        kazdz/TT                         (36)
                       J  Z = -TT/2


In the above expression,  kaz is the absorption rate  constant at
solar zenith angle, z.  The half-life  (ti.)  for a photoreaction  can
then be expressed as
          t, .         -                                        (37)
           ^    
-------
photolysis rates of organomercurials at various depths in natural
waters by measurements of sunlight intensities.

Light absorption by an organomercury compound may be expressed by1
the equation

                  hv
           (RHgX)0  -  (RHgX)*                                   (38)

where (RHgX)0 and (RHgX)* represent the compound in its ground and
electronically excited states, respectively.  The primary quantum
yield for any photochemical process is simply the fraction  of
electronically excited molecules that undergo the process.  If the
electronically excited molecule undergoes two or more competitive
primary photoprocesses, the sum of the quantum yields for these
processes is unity, but the quantum yield for a specific process is
less than unity.  In liquid-phase photochemistry, the measured
quantum yield for photoreaction of a compound is often affected by
secondary chemical reactions.  For example, the primary quantum
yield for photocleavage of an organomercury compound (eq 39) may be
high, but the measured quantum yield may be lower because of the
secondary reaction in equation 40.

           (RHgX)*  -*  R- + -HgX                                 (39)

          R- + -HgX  -  (RHgX)0                                 (40)

Photodecomposition of weakly absorbing compounds can often  be
accelerated by addition of other compounds that absorb light more
strongly.  Such acceleration, or photosensitization, can result
from electronic energy transfer from the strong light-absorber or
sensitizer (S) to the photoreactive compound  (eq 42).84'85

               hv
           (S)* +  (RHgX)0  -   (S)° +  (RHgX)*                     (42)

Since numerous substances that absorb sunlight more  strongly  than
organomercury compounds are present  in natural waters,  one  goal  of
this study was to define those types of compounds  that  can  photo-
sensitize degradation of organomercurials.

Quantum yields for photoreactions are sometimes  lowered by  energy
transfer from the electronically excited, photoreactive molecule to
a quencher molecule  (OJ.85

           (RHgX)* +  (Q)°  .+   (RHgX)0 +  (Q)*                     (43)
                                52

-------
The effect of added quencher upon the quantum yield is defined by
the following Stern-Volmer expression,


          ^  =  1 + kq T [Q]                                  (44)
          0Q

where 00 and 0Q are quantum yields without and with added quencher,
respectively, k^ is the bimolecular rate constant for the quenching
process, and T is the lifetime of the excited, photoreactive molecule
in the absence of added quencher.  Since several excellent quenchers,
such as oxygen, are present in the aquatic environment, the effect of
known quenchers upon photodecomposition rates of organomercury com-
pounds was also examined.

Photodecomposition of Dimethylmercury, Me thyluiercurie Ion,
Methylmercuric Hydroxide, and Methylmercuric Halides

The ultraviolet absorption spectra of (CHs)2Hg, CHgHg*, and CH3HgOH
in water revealed that these species absorb virtually no light at
wavelengths greater than 290 nm  (Figure 15).  Specific absorption
rate constants for these species are extremely low, so sunlight
photolysis rates are very slow.  Spectral predictions of low sunlight
photoreactivity were corroborated by irradiating aqueous solutions of
(CHg)2Hg, CH3Hg+, and CHgHgOH with Pyrex-filtered ultraviolet light
from a mercury lamp.  Prolonged  irradiations resulted in no photo-
decomposition of these species.

Other laboratory experiments showed that the decomposition of these
methylmercury species was not photosensitized by acetone, a high
energy sensitizer (triplet energy, 80 kcal mole"1).  Moreover,
dimethyImercury was not degraded by singlet oxygen, a chemical species
important in environmental chemistry.86

Sunlight irradiation of aqueous  10~4 M solutions of methyImercurie
hydroxide, methyImercurie chloride, methylmercuric bromide, and
methylmercuric iodide resulted in rapid photodecomposition of CH3HgI,
slow photodecomposition of CH3HgBr, and negligible degradation of
CH3HgCl and CH3HgOH (Table 10).  Dark controls at the same tempera-
ture (30°) showed no change.  These and following results demonstrate
that changes in the ligand bonded to methylmercury markedly affect
photoreactivity.

Facile photodecomposition of CH3HgI could cause significant errors in
the analysis for methylmercury content of environmental samples since
several widely used procedures call for gas chromatographic analysis
of methylmercuric iodide in organic solvents.53'7 ' 87  Because methyl-
mercuric iodide is photodecomposed by fluorescent lights, precautions
should be taken to shield the CHgHgl solutions from light during the
analysis.
                                 53

-------
                            (|DH2.0)
tu
O
           CH3HgOH
           (j)H 10.0)
                                   (CH3)2Hg  (&H7.0)
              200     220     240     260
                        WAVELENGTH, nm
280     300
          Figure 15.  UV absorption spectra  of dimethyImercury,
                      methylmercuric ion,  and methylmercuric
                      hydroxide in water.
                                    54

-------
         Table  10.   EXPERIMENTAL PHOTOLYSIS  RATES  FOR FOUR
                    METHYLMERCURY COMPOUNDS  IN SUNLIGHT
Compound51
CH3HgI
CH3HgBr
CH3HgCl
CH3HgOH
l^(ka)lntx 0
hr~lf
> 230
6.2
< 1.3
< 1.3
fc%
hrsb
< 3
110
> 530
> 530
 Concentration 1,00 x 10   M in water.
'Expressed in hours of sunlight,  not actual hours.
                                55

-------
Photoreactivity of Sulfur-bonded Methylmercury Complexes

As discussed earlier, under certain environmental conditions, a
large fraction of the methylmercury species exists as sulfur-bonded
methylmercury complexes.  Earlier studies88'89 showed that mercuric
mercaptides are readily decomposed by light from a mercury lamp or
even ordinary room light.

Previously recorded ultraviolet spectra showed that some organo-
mercuric mercaptides have significantly large extinction coefficients
at wavelengths > 290 nm.25  Methylmercuric sulfide ion, bis (methyl-
mercuric) sulfide, and methylmercury-thiol complexes also absorb at
wavelengths > 290 nm (Figures 16 and 17).  Preliminary experiments
showed that Pyrex-f iltered ultraviolet light decomposed the sulfide
and thiol complexes with cleavage of the methylmercury bond.

                  hv
          CH3HgS~  ->  CHi t + HgSi                              (45)
                  hv
          CHgHgSR  -  CHtt + C2H6t

             •f Inorganic mercury precipitate ; RSSR             (46)

Products of the photolyses were methane and ethane (20:1) gases and
inorganic mercury precipitates.  The black precipitate from photolysis
of methylmercuric sulfide ion was found to be mercuric sulfide, and
the precipitate from the methylmercury-thiol photolysis was not
identified, but was shown by mass spectrometry to contain no organi-
cally bound mercury.  Yields of inorganic mercury in the precipitates
quantitatively accounted for disappearance of the methylmercury
complexes.  Under these same conditions, unphotolyzed controls
formed no inorganic mercury during the photolysis period.

Disappearance quantum yields for photodegradation of several
different methylmercury-thiol complexes by 313 nm light were similar
(Table 11).  Quantum yields for photodecomposition of the cysteine
and egg albumin complexes are particularly significant, since these
complexes are good models for methylmercury-thiol complexes that
form in biological systems 0155 9°  In addition to complexation by
sulfhydryl groups, the methylmercury ion was probably also complexed
by other functional groups such as amino and hydroxyl groups in the
albumin protein.90  Such complexes would not be decomposed by 313 nm
light, because methyLmercury-nitrogen and methylmercury-oxygen
complexes do not absorb light at wavelengths > 290 nm.91

Photodecomposition of the methylmercuric sulfide ion proceeded with a
high quantum yield (Table 11).  The complex was so light-sensitive
that extensive photodegradation by fluorescent lights in the
                                56

-------
100   -
 80   -
 60   -
 40    -
 20   -
            300     320     340     360     380     400     420

                          WAVELENGTH, nm
      Figure 16.  UV absorption spectra  of methylmercuric
                  sulfide ion (A)  and  bis(methylmercuric)
                  sulfide (B).
                              57

-------
30
25  -
20 -
15  -
10
 5  -
       IA
A. CH3Hg-EGG ALBUMIN
B. CH3HgSCH2COOH
C. CH3HgSCH2CH2OH
D. CH3HgSCH2CH(NH3+)COOH
                                 I
         300     320     340     360
                  WAVELENGTH, nm
                   380
400
 Figure 17.   UV absorption  spectra of methylmercury-thiol
             complexes.
                          58

-------
  Table 11.  QUANTUM YIELDS FOR PHOTODEGRADATION OF SULFUR-BONDED
             METHYLMERCURY COMPLEXES AT 313 nm IN DISTILLED WATER
Complex
CH3Hg-egg albumin

CH3HgSCH2C02H

CH3HgSCH2CH2OH

CH3HgSCH2CH(NHg+ )C02H
0.12 + 0.02

0.15 + 0,01b>

0.14 + O.of"'

0,16 + 0.02

0.65 + 0.05b
 Disappearance quantum yield in degassed solutions.
bQuantum yield shown to be concentration independent.
cQuantum yield shown to be pH independent.
                                 59

-------
laboratory occurred over a period of a few days on the desk top.  At
high concentrations (0.10 M) with excess sulfide present, the complex
was degraded by a dark reaction to form mercuric sulfide.  This
reaction was not investigated in detail, because it proceeded at a
much slower rate at concentrations < 0.01 M.  Assuming second-order
kinetics for the reaction, its rate constant was estimated to be
~ 10"6 -I/mole sec.

Photosensitization of the methylmercury-thioglycolic acid complex
(CHgHgSCHsCQsH) by acetone proceeded with about the same quantum
yield  (0.13) measured for the direct photolysis (0.15).  Presumably,
the other thiol complexes would undergo such sensitized photodecompo-
sition also, since they all have the same structure about the mercury-
sulfur chromophore.  On the other hand, humic acid, a substance likely
to be  found in natural waters, did not photosensitize the decomposi-
tion of
Quantum yields for photodecomposition of the me thy Imercury- thiol
complexes were lowered by addition of quenchers  (Table 12) .  Stern-
Volmer plots  (see eq 44) of the data were non-linear  (Figure 18),
indicating that two or more excited states are involved in the
photodecomposition of the complexes.92  Whatever the nature of the
quenching processes, concentrations of quenchers such as oxygen are
sufficiently high under certain environmental conditions to lower
quantum yields for photodecomposition of the complexes.  For example,
the concentration of oxygen in air-saturated water  (~ 3 X  10~4 M)
would lower the quantum yield for CR3RgSCRsCB.3OR from 0.14 to 0.10.

Maximum rates of sunlight photodecomposition were calculated as
described previously (Table 13) using quantum yields from  Table 11
and spectral data from Figures 16 and 17.  The rapid photolysis rate
of the sulfide complex is due to the combination of its high quantum
yield and large sunlight absorption rate constant.  Since methyl-
mercury ion is often complexed with sulfur-containing ligands in
the environment, these data suggest that sunlight photodecomposition
of sulfur-bonded me thy liner cury complexes plays an important role in
the conversion of me thy Imercury to inorganic mercury compounds.

Photocleavage of Phenylmercury Compounds

Spectroscopic studies showed that phenylmercuric hydroxide,
pheny Imercury ion, and dipheny Imercury  (Figure 19)  absorb  at wave-
lengths > 290 run.  The absorption in this spectral  region  is due to
singlet-triplet electronic transitions  that have enhanced  intensity
because of perturbation by the heavy atom, mercury.93' 94   The
singlet-triplet spectra of phenylmercuric ion and phenylmercuric
hydroxide are identical.  Thus, although changes in pH affect the
composition of phenylmercuric species in aqueous solution  (eq 47),
the rate of sunlight absorption is pH-independent.

                                60

-------
     Table  12.   EFFECT  OF  QUENCHERS UPON PHOTODECOMPOSITION OF
                SULFUR-BONDED METHYLMERCURY  COMPLEXES IN WATERa
Complex
CH3HgSCH3C02H






r^rr "LTrrQ/TT /TJ fTTJ
V_>.n.Q JTlfiO Vjilp wiJ-2 *— 'H



(TrT TT Q. O~~"

Quencher (Q)
None
2 ,4-hexadien-l-ol
2,4-hexadien-l-ol
2 ,4-hexadien-l-ol
2,4-hexadien-l-ol
sodium trans -cinnamate
oxygen
None
2,4-hexadien-l-ol
2 ,4-hexadien-l-ol
2 ,4-hexadien- l-ol
None
Oxygen
10s [Q]
M
0
86
57
38
19
20
1.4°
0
39
4.9
0.49
0
1.4C
0Hgb
0.16
0.063
0.075
0.082
0.093
0.085
0.075
0.14
0.043
0.046
0.087
0.65
0.08
 Quenching studies  carried  out at  313  nm under  conditions  in which
 no light  absorbed  by quencher.
'Quantum yield for  formation  of inorganic mercury.
'Oxygen-saturated water.
                                61

-------
                 CH3HgSCH2CHjOH
                              CH3HgSCH2COOH
                                           I
            24       6      8       10     12
                   [Qjxio2, moles/1
Figure 18.  Effect of quencher,  2,4-hexadien-l-ol,
           upon quantum yields  for photocleavage
           of methyltnercury-thiol complexes.
                       62

-------
Table 13.  CALCULATED SUNLIGHT PHOTOLYSIS RATES FOR SULFUR-BONDED
           METHYLMERCURY COMPLEXES AT 25° IN WATER
Complex
CHgHg-egg albumin
CH3HgSCH2C02H
CH3HgSCH2CH2OH
CH3 HgSCH2 CH (NHg+ ) C02H
CH3HgS-
102 (ka)lntg
hr"1
10
10
11
3.6
240
103 0(ka)lntg
hr"1
12
15
15
5.8
1600
fc%
hr
58
46
46'
120
0.43
                                63

-------

-------
                         K
                * + H20  ^   CsHgHgOH + if                       (47)
K = 6.8 X 1CT5  (Reference  16)

Irradiation of  diphenylmercury, phenylmercuric  hydroxide,  and  phenyl -
mercury ion in  degassed  solutions  caused  the  following  reactions:

                    hv
                                +  C6H5 •                         (48)
                   -  CeHg- + Hg°i                              (49)

                  hv
                   -  CsHg- + 'Hg+                              (50)
          2-Hg+  -»  Hgl+                                        (51)

                   hv
                 3E  -»  CgHg- +  -HgOH                            (52)


          2-HgOH  -  Hg2(OH)2                                   (53)


          Hg2(OH)2  ->  HgOi +Hg°l + H20                        (54)


          Cg He * H~ RH  ~~*  CgHg *T R*                              ^33ji
          ZCeHs-  -   (C6H5)2I                                   (56)

Quantitative yields of metallic mercury and phenyl  free  radicals
resulted from irradiation of diphenylmercury.  Phenylmercuric  ion
photolyzed to give phenyl radicals and mercurous  ions, and phenyl-
mercuric hydroxide yielded phenyl radicals and nearly quantitative
yields of metallic mercury and yellow mercuric oxide.  The fate of
the phenyl radicals depended upon the composition of the reaction
media.  With organic materials  (RH) present in the  aqueous media,
phenyl radicals reacted to form nearly quantitative yields of
benzene (eq 55).  In oxygen free distilled water  containing  no
additional organic substances, the most important phenyl radical
reaction was coupling to form biphenyl (eq 56).   With, oxygen present,
no biphenyl was formed, presumably because the phenyl radicals were
scavenged by the following reaction.

          CsHg- + 02  -  CGKzOz°                                (57)


                                65

-------
Previous studies by Russell and Bridger have shown that reaction 57
occurs much more rapidly than reaction 56 in cyclohexane and carbon
tetrachloride.95  Products that resulted from reactions of the phenyl-
peroxy radical were not identified because of their complexity.

Disappearance quantum yields for photodecomposition of the phenyl-
mercury compounds with 313 nm light are summarized in Table 14.
Phenylmercuric perchlorate was used in the study because it is
completely dissociated in water at the high concentrations used in
the experiments.  At pH 2.3 and pH 10.3, the phenylmercuric species
existed as phenylmercuric ion and phenylmercuric hydroxide, respec-
tively (eq 47).  However, results in Table 14 show that the quantum
yield for photodegradation of phenylmercuric salts is the same at
both pH's.  Because the broad-band (> 290 nm) quantum yield for
disappearance of phenylmercuric perchlorate (0.23 + 0.03) was about
the same as the 313-nm quantum yield, the quantum yield was assumed
to be wavelength-independent.  Benzene  (0.80 moles per mole of
phenylmercury compound decomposed) and inorganic mercury containing
precipitates were formed at both pH values in the acetone-sensitized
studies.

Photodecomposition of phenylmercurials was sensitized efficiently by
acetone (triplet energy 80 kcal/mole), but not by sensitizers with
triplet energies < 74 kcal/mole (Table 15).  The triplet-state
energy of phenylmercurials must therefore be between 74 and 80 kcal/
mole, because efficient photosensitization occurs only when the
sensitizer triplet energy is equal to or greater than that of the
photoreactive compound.96  Ultraviolet absorption spectra of the
phenylmercury compounds  (Figure 19) show that their triplet energies
are ~ 80 kcal/mole.  The similarity of quantum yields in the direct
and sensitized photolyses is not surprising since both involve
excitation of the phenylmercury compounds to their triplet states.

Photosensitized conversion of phenylmercury to inorganic mercury
compounds is unlikely to be generally important in the aquatic
environment for the following reasons:

•  Most compounds with triplet energies ^ 80 kcal mole"1 do not
absorb strongly at wavelengths > 290 nm.

a  Observed concentrations of mercury compounds in natural waters
are generally very low  (< 10~9 M).  Because much higher concentra-
tions of competing energy acceptors, such as oxygen, are usually
present in the environment, photosensitization would be negligible.

Stern-Volmer plots for the quenching of diphenylmercury photo-
decomposition were linear  (Figure 20).  In acetonitrile, the slope
of the plot, kq T, was only 0.20.  Assuming a value of 1.0 x 1010 £/
mole sec for k^ in acetonitrile,97 the indicated triplet lifetime,


                                 66

-------
         Table 14.  QUANTUM YIELDS FOR DIRECT PHOTOLYSIS
                    (313 ran) OF PHENYLMERCURY COMPOUNDS
Compound
(CsH5)2Hg
(C6H5)2Hg
(C6H5)2Hg
CsH5HgC104
CsH5HgC104
Solvent
Benzene
Acetonitrile
Water, pH 7.0
Water, pH 2.3
Water, pH 10.3
Disappearance
Quantum Yield
0.33 + 0.01
0.40 + 0.02
0.27 + 0.02a
0.25 + 0.02
0.24 ± 0.02
Extrapolated using viscosity data.
                                67

-------
                     Table 15.   QUANTUM YIELDS FOR PHOTOSENSITIZATION OF PHENYLMERCURIALS
Compound
(C6H5)2Hg
(C6H5)2Hg
(C6H5)2Hg
(C6He)2Hg
(C6H6)2Hg
CgHeHgOsCCHs
C6H5HgC104
Sensitizer
Acetone
Acetophenone
Benzophenone
Naphthalene
Anthracene
Acetone
Acetone
Triplet
Energy
(kcal mole"1 )
80
74
68
62
42
80
80
Solvent
Benzene
Benzene
Benzene
Benzene
Benzene
Water, pH 4.0
Water, pH 2.3
Disappearance
Quantum Yield
0.33
< 0.003
< 0.006
0.027
< 0.003
0.23
0.23
00

-------
2.0
1.
     0
5
                   [Pentadiene],  moles/1
Figure 20.  Quenching of diphenylmercury photolysis by
           cis-1,3-pentadiene in acetonitrile.
                       69

-------
T, for diphenylmercury is 2.0 X 1CT11 sec.  Disappearance quantum
yields in water for phenylmercury ion and phenylmercuric hydroxide
were not decreased by quencher concentrations less than 0.10 M,
indicating that T for these compounds is < 10~10 sec.  Because
concentrations of potential quenchers in natural waters are much
lower than 0.1 M, quantum yields in Table 14 should be unaffected
by quenching processes in the environment.

To determine empirically the effect of materials dissolved in natural
waters on the quantum yield for photodecomposition of a phenylmercuric
salt, air-saturated solutions of phenylmercuric acetate (1.0 X 10~3 M)
in two different natural waters and in distilled water were subjected
to equal exposures of Pyrex-filtered mercury-lamp light (> 290 run).
The phenylmercuric salt photodecomposed at the same rate in all three
solutions, in agreement with the predictions derived from the quench-
ing studies.  Dark controls showed no decomposition.

Sunlight photolysis rates calculated from laboratory data and
photolysis rates measured in sunlight are compared in Table 16.
Other experiments showed that the same salts were completely decom-
posed after: 18 days in sunlight while dark controls showed no change.
With the exception of diphenylmercury, the compounds were largely
dissociated and hydrolyzed to form 2570 phenylmercuric ion and 757o
phenylmercuric hydroxide under the experimental conditions.  Results
in Table 16 show that the calculated and empirical photolysis rates
are in reasonable agreement.
                                70

-------
              Table 16.  COMPARISON OF CALCULATED AND EMPIRICAL RATES FOR
                         PHOTODEGRADATION OF PHENYLMERCURY COMPOUNDS IN SUNLIGHT
Compound
(C6H5)2Hg
C6H5HgOCOCH3
CeHgHgNOg
CgHgHgBO,
CsHgHgGH
Calculated
108(K)lntB 0, hr-1&

3.3
3.3
3.3
3.3
Empirical
102(ka)int6 0b'c
8,1 ± 1.7
4.3 + 0.3
3.5 ± 0.2
5.0 + 0.3
4.3 + 0.3
Empirical tj,
hrs
8.5 + 1.8
16.0 + 2
20,0 + 1
14.0 + 2
16.0 + 2
aCalculated frcm solar radiation data, extinction coefficients, and quantum yields in Table 14,
 Average of three replicates irradiated by sunlight,
cCalculated assuming first-order kinetics for photodecomposition.

-------
                            SECTION VII

                            REFERENCES
 1.   T.  Takeuchi,  presented in part at the International Conference
     on  Environmental Mercury Contamination,  Ann Arbor,  Michigan,
     September 1970.

 2.   K.  Borg,  H. Wanntorp,  K. Erne, and E. Hanko, J._Appl. Ecol., J3
     (Suppl.),  171 (1966).

 3.   L.  Friberg, et al.,  Nord. Hyg. Tidskr.,  5_i> Suppl.  4 (1971).

 4.   A.  Jernelov,  "Chemical Fallout," M.  Miller and G.  Berg, Eds.,
     Charles  C. Thomas, Publisher,  Springfield, Illinois, 1969,
     Chapter  4.

 5.   (a) F. M.  D'ltri,  Technical Report No. 12 to the Michigan
     (USA) House of Representatives (1971); (b) N. Nelson, et al. ,
     Environ,  Res., 4_,  Chapter 1 (1971);  (c)  P. A. Krenkel, R. S.
     Reimers,  E. B. Shin, and W. B. Burrows,  "Mechanisms of Mercury
     Transformations in Bottom Sediments," Technical Report No. 27,
     Environmental and  Water Resources Engineering, Vanderbilt
     University, Nashville, Tennessee, 1971.

 6.   (a) F. R.  Jensen  and B. Rickborn, "Electrophilic Substitution
     of  Organomercurials," McGraw-Hill Book Co., New York, 1968;
     (b) 0. A.  Reutov  and I. P. Beletskaya, ""Reaction Mechanisms of
     Organometallic Compounds," North-Holland Publishing Co.,
     Amsterdam,  1968;  (c) L. G. Makarova and A. N. Nesmeyanov in
     "Methods of Elemento-Organic Chemistry," Vol. 4, A. N. Nesmeyanov
     and K. A.  Kocheshkov, Ed., North-Holland Publishing Co.,
     Amsterdam,  1967;  (d) L. G. Makarova in "Organometallic Reactions,"
     Vol. I,  E.  I. Becker and M. Tsutsui, Ed., Wiley-Interscience,
     New York, 1970; (e)  T. G. Brilkina and V. A. Shushunov, "Reactions
     of  Organometallic  Compounds with Oxygen and Peroxides," CRC Press,
     Cleveland,  Ohio,  1969, Chapter 2.

 7.   D.  Seyferth,  Organometal.. Chem. Rev. B,  5_, 288  (1969).

 8.   D.  Seyferth,  Organometal. Ghem. Rev. B, 6_, 239  (1970).

 9.   D.  Seyferth,  Organometal. Chem. Rev. B, _8, 425  (1971).

10.   (a) K.  C. Bass, Organometal. Chem. Rev., _!, 391  (1966);  (b) L. G.
     Makarova in "Organometallic Reactions," Vol. II, E. I. Becker
     and M.  Tsutsui, Ed., Wiley-Interscience, New York,  1971.

                                72

-------
11.   (a) H. Ley, Z. Phys. Chem., _30, 248  (1899);  (b) A. B. Garrett and
     A. E. Hinchler, J. Amer. Chem. Soc., 60, 299  (1938);  (c) J. Bjerrum,
     Dissertation, Copenhagen, 170  (1941);  (d) A. B. Garrett  and W. W.
     Howe 11, J. Amer. Chem. Soc., 6JL, 1730  (1939).

12.   (a) J. L. Maynard and H. C. Howard,  J. Chem. Soc., 123,  960
     (1923); (b) I. B. Johns, W. D. Peterson, and R. M. Hixon, J. Amer.
     Chem. Soc., 34, 2218 (1930).

13.   (a) R. M. Schramm, J. Amer. Chem. Soc., _69,  1831  (1947); (b) R.
     Benesch and R. E. Benesch, J. Amer.  Chem. Soc., ^3_, 3391 (1951).

14.   S. Hietanen and L. D. Sillen, Acta Chem. Scand., 6, 747  (1952).

15.   R. B. Simpson, J. Amer. Chem. Soc.,  83, 4711  (1961).

16.   (a) T. D. Waugh, H. E. Walton, and J. A. Laswick, J. Phys. Chem.,
     59, 395 (1955); (b) S. S. Parikh and T. R. Sweet, J. Phys. Chem.,
     65_, 1909  (1961); (c) E. A. Smith, Piss. Abstr. Int. B., _32, 5075
     (1932); Dissertation, Rutgers University, 1971.

17.   L. F. Sytsma, Piss. Abstr. Int. B, 32, 6311  (1972).

18.   G. Schwarzenbach and M. Schellenberg, Helv. Chim. Acta, 48, 28
     (1965).

19.   J. D. Calvert and J. N. Pitts, "Photochemistry," John Wiley and
     Sons, Inc., New York, 1966, p. 270.

20.   (a) P. L. Goggin and L. A. Woodward, Trans. Faraday Soc., 56, 159
     (I960); (b) J. H. R. Clarke and L. A. Woodward, Trans. Faraday
     Soc., 644, 1041 (1968).

21.   F. A. Cotton and G. Wilkinson, "Advanced Inorganic Chemistry,"
     Interscience Publishers, New York, 1966, p.  616.

22.   See Appendices for Tables of equilibrium constants and additional
     discussion.

23.   H. Edelhoch, E. Katchalski, R. H. Mayberry, W. L. Hughes, Jr.,
     and J. T. Edsall, J. Amer. Chem. Soc., 75, 5058 (1953).

24.   P. D. Boyer in "The Enzymes," P. D.  Boyer, H. Lardy, and
     K. Myrbach, Ed., Academic Press, New York, 1959, Chapter 11.

25.   P. D. Boyer, J. Amer. Chem. Soc., 76, 4331 (1954).

26.   E. Chiancone, D. L. Currell, P. Vecchine, E. Antonini, and J,
     Wyman, J. Biol. Chem.. , 245, 4105 (1970).


                                 73

-------
27.  D. L. Currell and C. Toppolo, Biochim. Biophys. Acta, 263, 82
     (1972).

28.  Y. Lin, £t al.,  Biochem. Biophys. Res. Commun., 47, 1209  (1972).

29.  L. Levine and R. K. Brown, Biochim. Biophys. Acta, 25, 329 (1957).

30.  M. Eigen, G. Geier, and W. Kruse, "Essays in Coordination
     Chemistry," Experientia (Birkhauser Verlag, Basel, 1964), Suppl.
     9, p. 164.

31.  R. B. Simpson,  J. Chem. Phys., 46, 4775 (1967).

32.  H. S. Gutowsky and C. H. Holm, J. Chem. Phys., _25_, 1233  (1956).

33.  J. D. Hem, U. S. Geol. Survey Prof. Paper 713, 1970.

34.  H. Zimmer and S. Makower, Naturwissenschaften, 41, 551 (1954).

35.  F. Nerdel and S. Makower, Naturwissenschaften, 45, 491 (1958).

36.  I. P. Beletskaya, L. A. Fedorov, and 0. A. Reutov, Proc. Acad.
     Sci., USSR Chem. Sect., 163, 794  (1965).

37.  See reference 6a.

38.  R. E. Dessy, G.  F. Reynolds, and J. Kim, J. Amer. Chem.  Soc.,  81,
     2683 (1959).

39.  M. M. Kreevoy,  J. Amer. Chem. Soc., 7^, 5927  (1957).

40.  R. D. Brown, A.  S. Buchanon, and A. A. Humffray, Aust. J. Chem.,
     18, 1507  (1965).

41.  R. J. Ouellette, Ph.D. Thesis, University of California,  Berkeley,
     1962.

42.  E. G. Janzen and R. J. Blackburn, J. Amer. Chem. Soc., 91, 4481
     (1969).

43.  (a) A. Terenin,  J. Chem. Phys.,  2_, 441  (1934);  (b) J. W.  Linnett
     and H. W. Thompson, Trans. Faraday Soc., 33, 501  (1937);  (c) R.
     Corner and W. A.  Noyes, Jr., J. Amer. Chem. Soc., 71,  3390 (1949);
     (d) R. A. Holroyd and W. A. Noyes, J. Amer. Chem. Soc.,  76   1583
     (1954).                                                  ~

44.  R. E. Rebbert and P. Ausloos, J. Amer. Chem. Soc., 85, 3086  (1963)

45.  T. Fagerstrbm and A. Jernelbv, Water Research, £, 1195  (1972).
                                 74

-------
46.  0. P. Strausz, T. DoMinh, and J. Font, J. Amer. Chem.  Soc. , JIQ,
     1930  (1968).

47.  H. Takehara, M. Kotakemori, and T. Kajimura, Nippon  Nogei Kagaku
     Kaishi, _39, 448  (1965); Chem. Abstr., 6>4, 10342g  (1966).

48.  H. Shiina, R. Nishiyama, M. Ichihashi, and K. Fiyikama, Nippon
     Nogei Kagaku Kaishi, 38, 481  (1964); Chem. Abstr., _63,  10594h
     (1965).

49.  G. Gran, Svensk Paperstidn, 53, 234  (1950).

50.  A. Eldridge and T. R. Sweet, Anal. Chem., 28, 1268  (1956).

51.  F. Kaufman and A. H. Corwin, J. Amer. Chem. Soc., 77_,  6280  (1955),

52.  W. R. Hatch and W. L. Ott, Anal. Chem., 40, 2085  (1968).

53.  J. E. Longbottom, R. C. Dressman, and J. J. Lichtenberg, private
     communication.

54.  P. J. Wagner, P. A. Kelso, and R. G. Zepp, J. Amer.  Chem. Soc..,
     94, 7480  (1972).

55.  (a) W. H. Thompson _et al. , Trans. Faraday Soc., 3_2,  681  (1936);
     (b) G. Davis, Dissertation, Kansas State University,  1971;  (c)
     R. N. McDonald and G, E. Davis, J. Org. Chem.,  38.=  138 (1973).

56.  F. G. Moses, R. S. H. Liu, and B. M. Monroe, Mol. Photochem., _!,
     245  (1969).

57.  R. Livingston in "Technique of Organic Chemistry/"  Vol. VIII,
     Part  I, A. Weissberger, Ed., Interscience Publishers,  Inc., New
     York, 1961, Chapter V.

58.  P. J. Wagner and A. E. Kemppainen, J. Amer. Chem. Soc.,  94,
     7495  (1972).

59.  A. A. Lamola and G. S, Hammond, JL__Chem. Phys., 44,  2129  (1965).

60.  C. S. Foote, Accounts Chem. Res., _1, 104  (1968).

61.  R. C. Weast, Ed., "Handbook of Chemistry and Physics," 45th Ed.,
     The Chemical Rubber Co., Cleveland,  Ohio, 1964.

62.  P. A. Leighton, "Photochemistry of Air Pollution,"  Academic
     Press, New York, N. Y., 1961, pp. 26-30.

63.  E. R. Allen, R. D. McQuigg, and R. D. Cadle, Chemosphere,  I,  25
     (1972),

                                 75

-------
64.  See Appendix A of this report for more detail.

65.  J. R. Kramer, H. E. Allen, G. W. Baulne, and N. M. Burns,
     "Lake Erie Time Study," Canada Centre for Inland Waters Paper
     No. 4, Burlington, Ontario, 1972.

66.  R. L. Carr, C. E. Finsterwalder, and M. J. Schibi, Pestle.
     Monit. J., 6, 23  (1972).

67.  N. M. Burns and C. Ross, "Project Hypo," U. S. Environmental
     Protection Agency, Technical Report TS-05-71-208-24, 1972,
     Chapter 7.

68.  G. E. Hutchinson, "A Treatise on Limnology," John Wiley & Sons,
     Inc., New York, 1957, Chapter 13.

69.  L. Friberg, e_t _al., Nord. Hyg. Tidkr. , 5_2, Suppl. 4, Chapter 3,
     (1971).

70.  J. J. Morgan, "Equilibrium Concepts in Natural Water Systems,"
     Advances in Chemistry Series, No. 67, American Chemical
     Society, Washington, D. C., pp.  1-29.

71.  A. A. Frost and R. G. Pearson, "Kinetics and Mechanisms," 2nd Ed.,
     Wiley and  Sons, Inc., New York, N. Y., 1961.

72.  S. I. Shibko  and  N. Nelson, Environmental Research, 4, 23  (1971).

73.  G. L. Baughman, M. H. Carter, N. L. Wolfe, and R. G. Zepp,
     J. Chromatogr., 76_, 471  (1973).

74.  G. M. Barrow, "Physical Chemistry," 2nd Ed., McGraw-Hill,
     New York,  N.  Y.,  1966, p.  651.

75.  Roger G.  Bates, "Determination of pH," John Wiley and  Sons,  Inc.,
     New York,  N.  Y.,  1954, p.  212.

76.  M. S. Kharasch  and M. W.  Grafflin, J.  Amer. Chem. Soc.,  47,  1948
      (1925).

77.   (a) K. Hartley, H. 0. Pritchard, and  H. A.  Skinner, Trans.  Faraday
     Soc., 46,  1019  (1950);  (b) H. B. Charman,  E.  D.  Hughes,  and
     C. K. Ingold, J.  Chem.  Soc.,  2530  (1959).

78.  E. C. Tsivoglou,  "Tracer  Measurement  of Stream Reaeration,"
     USDI, Federal Water  Pollution Control Administration,  Washington,
     D. C.,  1967.

79.  M. P. Thekaekara, R.  Kruger,  and C. H.  Duncan, Applied Optics,
     8,  1713  (1969).

                                 76

-------
80.  J. G. Calvert and J. N. Pitts, Jr., "Photochemistry,"  John  Wiley
     & Sons, Inc., New York, N. Y., 1966, p.  742.

81.  Reference 62, Chapters II and III.

82.  E. R. Allen, R. D. McQuigg, and R. D.  Cadle,  Chemosphere,  I,  25
     (1972).

83.  S. Q. Duntley, J. Opt. Soc. Amer. , _53_5  214  (1963).

84.  A. A. Lamola, Photochem. Photobiol., _8,  601  (1968).

85.  N. J. Turro, "Molecular Photochemistry," W. A.  Benjamin,  Inc.,
     New York, 1967, Chapter 5.

86.  (a) I, Frankiewicz and R. S. Berry, Environ.  Sci. Technol., _6,
     365  (1972);  (b) J. N. Pitts, Jr., in "Chemical  Reactions  in
     Urban Atmospheres," L. S. Tuesday, Ed.,  American Elsevier
     Publishing Co., Inc., New York, N. Y.,  1971,  pp. 3-31.

87.  J. F. lithe, J. Solomon, and B. Grift,  J. Assqc. Offic.  Anal.
     Chem. , 5_5, 583 (1972).

88,  C. Marcker, Ann., 136, 81 (1865).

89.  R. J. Kern, J. Amer. Chem. Soc.,  75, 1865  (1953).

90.  W. L. Hughes, Ann. N. Y. Acad. Sci. , _65, 454  (1957).

91.  B. G. Gowenlock  and J. Trotman,  J. Chem.  Soc., 1454  (1955).

92.  P. J. Wagner in "Creation and Detection of  the  Excited State,"
     Vol.  1, Part A3 A. Lamola, Ed., Marcel Dekker,  New  York,  1971,
     Chapter 4.

93.  A. P. Marchetti and D. R. Kearns, J. Amer.  Chem. Soc. ,  89_,  768
     (1967).

94.  N. J. Turro, "Molecular Photochemistry/' W. A.  Benjamin,  Inc.,
     New York, 1967, pp. 48-52.

95.  G. A, Russell and R. F. Bridger,  J. Amer.  Chem_._S_qc,. ,  85_, 3765
     (1963).

96.  K. Sandros and H. L, J. Backstrom, Acta Chem. Scand. ,  16_, 958
     (1962).

97,  P. J. Wagner and D. J. Bucheck, J.___Am_er_. Chem.^ 5oc__. , £1,  5090
     (1969).

                                 77

-------
                          SECTION VIII

                          PUBLICATIONS
1.  G. L. Baughman, M. H. Carter, N. L. Wolfe, and R. G. Zepp,
    Gas-liquid Chromatography-Mass Spectrometry of Organomercury
    Compounds.  J. Chromatogr. (Amsterdam), 76, 471  (1973).

2.  N. L. Wolfe, R. G. Zepp, J. A. Gordon, and G. L. Baughman,
    Chemistry of Phenylmercury Compounds in the Aquatic Environ-
    ment.  Chemosphere (Oxford),  _1, 273 (1972).

3.  R. G. Zepp, N. L. Wolfe, and J. A. Gordon, Photodecomposition
    of Phenylmercury Compounds in Sunlight.  Chemosphere (Oxford),
    ^ (in press).

4.  N. L. Wolfe, R. G. Zepp, J. A. Gordon, and G. L. Baughman,
    Chemistry of Methylmercurials in Aqueous Solution.  Chemosphere
    (Oxford)  (in press).

5.  R. G. Zepp, N. L. Wolfe, and G. L. Baughman, Methylmercury
    Complexes in Aquatic Systems.  Environ. Lett, (in preparation).
                               78

-------
                            SECTION IX

                            APPENDICES
A.  Methylmercuric Ion Equilibrium Calculations  	    80

B.  Equilibrium Constants for Formation, Dissociation,
    and Hydrolysis of Mercury Species	    93
                                79

-------
                            APPENDIX A

            METHYLMERCURIC ION EQUILIBRIUM CALCULATIONS


The total concentration of methylmercury species present in a system
was represented as the sum of the concentrations of all complexes
incorporating the methylmercury moiety, plus the concentration of
the free ion itself.  The formation constant expressions for the
various complexes considered were solved for the concentration of
the complex.  Substitution of the concentration expression so
derived into the methylmercury concentration equation yielded an
equation for S[CHgHg] as a function of any one methylmercury complex.
Dividing both sides by the concentration of the complex and rearrang-
ing terms yields an expression for the fraction of the total methyl-
mercury that exists as a given complex.  Because the formation
constant for bis(methylmercury) sulfide contains a squared term, the
resulting expression is second-order and requires an iterative
technique for solution.

An example has been worked out in detail for the complexing agents,
OH", CI~, S=, thiol (RSH), amino nitrogen (RNH2), ammonia, humic
acids (ArOH), and orthophosphate.  The formation constants (Appendix
B) indicate that few other complexes are likely to be important in
the environment, so other complexing agents were excluded from the
sample calculation.

The total concentration of methylmercury in the system is expressed
as:

          S[CH3Hg] = [CH3HgS"l + [CH3HgOH] + 2[(CHgHg)2S]


          + [CHsHgCl] + [CH3HgNH2R+] + [CH3HgSR] + [CHaHgHPCk"]


          + [CH3HgNH3+] + [CH3HgOAr] + [CH3Hg+]                (58)

The following formation equilibrium expressions and the constants
(Appendix B) were then solved for the concentration of the respective
complexes.
          CH3Hg+
                                      '
                                80

-------
CHgHg+ + OH"  *±  CH3HgOH


   „       [CH3HgOH]       ~37
+ + CH3HgS~  s:   (CHQHg)2S


 _      [(CHsHg)sS]

   [CH3Hg+]
CH3Hg+ + Cl   ;±
           [CH3HgCl]
                    ^T =  105'
                                 >
              ?+1 [cr]



              KB
        [RNH2] [CH3Hg+]
                            8-25
CH3Eg+ + SIT  ^  CH3HgSR
      [CHgHgSR]    _    i6.is
                       _  1ni
                       '
        [CH3Hg+l
               K7
CH3Hg+ + HPCf "  ^  CH3HgHP04
              ^] [HPOg-1
                      81

-------
                       K8
                 + NH3  ^
                     [CH3HgNH3+]
                =      3     3     _ in7-60
              8 ~ [CH3Hg+l  [NH3]
          CH3Hg+ + ArO~  ;±   CH3HgQAr
                                   _   6>
                        +l  [ArCT]
Substitution of the above  in  equation 58 and division of both  sides
by [CH3HgS~l gives the  following equation:


                                      2K3
            K! [S-]    KI   [S~l     Kj.  [S=l    Kx  [S=l


            Kg [NH3 ]   Kg  [ArO" ]       i
            if  r^113!    F    fc;=l     T^  r^1^1!
            -1^1 L^1  ]    K1   |_t>  I     i\i L'S  I

Simplifying and taking the reciprocal  of both sides gives the
fraction of methylmercury  existing as  CHgHgS"".


          [CH3HgS~]

          2 [CH3Hg]
                                                                 (59)
       + K2[OIT]  + 2X3 [CHgHgS" 1 + ^[CF] + K^RMls] H-

             + K7[HPC|~] + KsLNHal + Kg[ArO~] -f- 1               (60)
The concentrations  of [S=],  [EHH3], [RS~], [HPCf"],  [NHg],  and [ArO"]
are difficult  to  measure and are also pH-dependent;  they must there-
fore be defined in  terms of more easily measurable variables.
Equations  67-72 redefine the variables in terms of quantities that
can be more  readily estimated.
                                  82

-------
          ES =  [H2S] +  [S~l + [SIT]                             (61)


          ERN = [RNH2]  -f  [RNH3+]                                (62)


          ERS = [RS~! + [RSH]                                  (63)


          EP04 = [H2PQi~] + [HPCl~]                             (64)


          EN = [NH3 ] +  [NHi+]                                  (65)


          EArO = [ArOH] + [ArO~]                                (66)

Substituting expressions for  acid-base  equilibria into equations
61-66 and rearranging terms results  in  the following:

                 	6.3ZS
                                                                (67)
                                                                (68)
                      SRS
                   l+K.
                                                                (70)
                      EArO
          [ArO-1 =           <-         Kh  = 109'8                 (72)
where Kn , KB , Kp , Ka ,  and Kh  are  equilibrium constants for protonation
of amino acids,  thiolate ions,  hydrogen phosphate,  ammonia, and
phenolate ions,  respectively.
                                 83

-------
The final solution is obtained by substituting  equations 67-72 into
equation 60, eliminating the [OH""] term,  and  rearranging to give


          [CH3HgS"l =   K! [S=]                                 (73)

          E[CH3Hg]    X + K,. [S=]
where X = 1 + -0-2- + 2K3 [CHsHgS~l +
                                                 1  +Kn  [If]
            1 + Ks [H" 1   1 + Kp  [tf- ]    1 + Ka  [if ]    1 + Kh


 and Ky, represents the ionization constant  of water.
Using similar procedures, the fraction  of methylmercury existing as
        S was derived  (eq 74).
                            2K
                              10
            E[CH3Hg]          [S=] ELCHaHg]                      (74)

Expressions derived for  the relative  concentrations  of other methyl'
mercury complexes are shown below.
       F3 =           = - -
            E  [CH3Hg]   X + K!  [S~l
           [CH3HgCl]
          —	-=K4  [C1-]  (F3)                             (77)
          E [CH3Hg]



           [CH3HgNH2R+] _  Kg  (ERN)  (F3)
            E  [CH3Hg]      1 +
                                                                (78)
                                 84

-------
          [CH3HgSR] _ K6  (F3) ERS

          E [CHgHg]   1 + KS [H+]


          [CHgHgOH] _ K2  (F3) K*
          E [CH3Hg]      [H+]


                    "l   K7  (F3) EP04
            E [CH3Hg]     1 +Kp


                    ]   K6  (F3) EN
           Z [CHgHg]    1 + R,


          [CHgHgOAr] = Kg (F3) ZArO

           E [CHgHg]   1 + Kh [H+]
                                                                (80)
                                                                (81)
                                                                (82)
(83)
The computer program used for  the calculations  is  shown  in  Figure  1A.
Originally developed as a Conversational Programming  System (CPS)
program, it was later: converted  to Fortran  IV for  plotting  of  the
calculated fractions.  Calculations were carried out  on  an  IBM 360/67
computer.  Calculations from the latter were plotted  on  a Tektronix
4010-1 terminal.  Typical computer plots are shown on the following
pages (Figures 2A-7A).  Concentrations of all chemical species used
in the calculations, except sulfur compounds, are  given  below.

                    = 10~10 M

          E RN = 10"6 M

          E P04 = 10"6 M
          E N = 10~5 M
          E ArO = 10~5 M
                = 10~4 M
                                85

-------
?L£3 I.'JC SEVL, GEN EHAL )
PA5r'-;?KD:
rs^zrsE
Ger;
   t.l            GET  LISTCriB, LAST, DEL T., X>J
   2.             PUT  LISTC1  ' >;
   3.             PUT  LISTC ' CH2S)+CSH-)+CS=>=', Cl >;
   3.1            PUT  LISTC ' CH2PP4-5 + CH?04=>=1., C9>;
   4.             PUT. LISTC ' CCL-)= ', C2)i
   4.1            PUT  LISTC ' CNH4O + C:JH3) = ' / C10>;
   A.r            PUT  LISTC ' CAR2H> + CA?.S-)= ', Cl 1 );
   5.             PUT  LISTC ' CR.-.'H3-O-i-CK.:]K2>= ',C3>;
   6.             PUT  LISTC ' (RSH5+CP.S- >=', O)i
   7.             PUT.  LISTC-T0TAL MEHG  C0MC=f   ^5);
   6.             PUT  LISTC  ' >;
   9.             C6=HE;
   10.             D0 H:-:K= i  T0  LAST; •»•«•
   11.             PUT  Ll STC °PH= ',LeG10C 1/C6»J
   12.             PUT  LI STC •  ' >;
   1 A.             C3=. IE- 10;
   1 5.             i;? = 6. 3*C:/(. 1E23*C6**2+. 1 El 6*C6 + 6 . 3 >J
   16.             PUT' LISTC • CS=)='^C7>;
   J7.             IF C7=0 THEN  CS=0;
   20.             Zal+10**-A.63/C6+10**5.25*C2J
   21 .             Zl=50**8- 25*C3/C 1 + 1 OT* 10*C6>'+ 1 0** T6. 12*C4/( 1 + 10**9. 52* C
);                                     6
   22.             Z2= IO**5.03*C9/C 1 + 10**6. 79*C6 )+ 1 0**7 . 6*C10/C 1 + 10**9.
6>;                                  'C
   23.             Z4=10**6.5*C11/C]+!0**9.8*C6>J
   25.     IT:      Z3=Z + ZI-*Z2+ZA + 2*10**16. 3*C8J
   30.             F!=10**2I.2*C7/CZ3+10**21.2*C7>;
   35.             C2P=F1*C5;
   36.             IF  Fl = 0 THD-J  G3 TB 32;
   37.             IF AESCCCS-CS?)/C5);
   45.             G3 TO IT;
   50.     C0MP:   F2=2*10**-4.9*CS**2/CC7*C5>J
   55.     B2:      F3= 1/CZ + Z 1 +E2J-Z4+ 1 0* *2 1 . 2*C7 + 2* 1 0**1 6. 3*C3 >;
   56.             F7=10**9. 37*F3*. IE- 1 3/C6;
   57..             F9= 10**5.03*r3*C9/< i + 10*=*=6.79^C63;
   56.             F10=10**7.6*F3*C!0/C1+10**9.42*C6)J
   59.             IF. FJ = 0 TKLN  F2=0;
   60.             F4=!0**5. 25*C£*F3J
   65.             F5=10**S. 25*C3*F3/C 1*. I SI !*C6>;
   66.             Fl 1= 1C** 6. 5*F3*CI 1/C ! + 1 0**9. S«C6 >;'
   70.             F6= 10** 16. 1 2*F3*CA/C J.+ 10»*9. 52* C6);
   71.
   75.
   SO.
   E5.
   £6.
   87.
   88.
   90.
   91.
   95.
  100.
  101.
  102.
  103.
PUT
PUT
PUT
PUT
PUT
PUT
PUT
PUT
PUT
PUT
PUT
PUT
PUT
C6»
END
G0
LI
LI
STC
STC
LISTC
LI
LI
LI
LI
LI
LI
LI
LI
LI
LI
C6*
>
TO
STC
STC
STC
STC
STC
STC
STC
STC
STC
STC
•FP-ACTI
'FRACTI
•FP-ACTI
•FRACTI
'FRACTI
' FRACTI
'FRACTI
•FRACTI
' FHACTI
•FRACTI
' SUM OF
I * } *
* < ) ;
c
u
t
rt
at
W
t.i
N
N
*}
:j
:j
r,r
3"j
2
P
•^
\j
:j
:,'
:i
*j<
EHC-
B I SM
^••T
H
M
M
M
;.;
K^
*'J
EH3
2 =
• , F
EHGS=
I )
>
j
F2)J
+='^F35;
EHG3H=',
EHG
EHG

34=
:;H3= '
EHGCL= '^
EHG
EHG
3A
R= '
HS;:R=
EKGSR= ' ,
FRACTI








aw


<:= •


F7
• ,
,F
>;
F9);
105;
F4);
,FI i >;
%
F5>;
F6>;
ir


3);


ic*+x;

Bl J


















  105.
  110.
 Lrccu;
  ::-JE  12:39:13; TIME USELi:  CPU OOiOO:C5J  TEH:-! 03:06:18; PAGE  uu-:05:36;
Figure Al.  Computer program used for  calculations in Appendix A.

                                     86

-------
(S)
 cr»
 en
IE


O


w
           i.o   -
           0.8   -
            0.6   -
           0.2   -
           0.0
                0
                                       6
8
10
12
      Figure A2.  Relative concentrations  of methylmercuric-thiol


                  complexes in systems  containing a low concentra-

                  tion of reduced sulfur species.
                                  87

-------
 CM


CT>
CNJ
 cn
1C
    W
          1.1
           0.8
0.6
           0.2
          0.0
                0
                                                   CONDITIONS;

                                                    LS^IO
                                             8
           Figure A3.  Relative concentrations of (CH3Hg)2S in

                      systems containing a  low concentration
                      of reduced sulfur species.
                                  88

-------
     CD
           1.0
           0.8
       0.6
 CD
O
W    0-4
           0.2
          0.0
             _   CONDITIONS:
                0
                                             8        10      12
        Figure A4.  Relative concentrations of CH3HgS~ in systems
                   containing a low  concentration of reduced
                   sulfur species.
                                  89

-------
          0.20
00
en
o
          0.15
     O
W  o.
          0.05
          0.00
                0
                                            CONDITIONS:

                                            LS = 10'4M
                  LRS = 10"°M
                      LRS^IO
                            4
6
8
10
12
           Figure A5.  Relative concentrations of methylmercuric-thiol

                      complexes in systems containing a high concen-

                      tration of reduced  sulfur species.
                                      90

-------
LO




IE
 en

IE

o


CM
0
       0.2
                     2
                                            6
   Figure A6.   Relative concentrations  of  (CH3Hg)2S  in  systems

               containing a high concentration  of  reduced

               sulfur species.

-------
           0.8
           0.6
    31
    O
IE
O
           0.2
          0.0
CONDITIONS:

 LS = KT4M
                                4
            10
12
        Figure A7.  Relative concentrations of QLjHgST in systems
                   containing a high  concentration of reduced
                   sulfur species.
                                  92

-------
                            APPENDIX B

 EQUILIBRIUM CONSTANTS FOR FORMATION, DISSOCIATION, AND HYDROLYSIS
                        OF MERCURY SPECIES
           Table Bl.  LOGARITHMS OF FORMATION CONSTANTS
                      FOR Hg2 + COMPLEXES IN WATER AT 25° C
Ligand Xs
-cr
-Br~
-I"
r\if~
= WIT
-NH3
-NH2R (Histidine)
-SR~ (Cysteine)
-SCN"
-CN~
[HgX2 1 N c
Lo§ [Hg2+] [ri2
13.2°
16.8°
23.8°
21.7°
17. 4d
21. 2e
41. tf
17. 46
34.7
aSign to right denotes charge of unattached ligand,
bTaken from reference 15.
0 Ionic strength = 0.5.
dIonic strength = 2.
eIonic strength = 0.15.
1 Ionic strength = 1.
6Ionic strength = 0.1.
                                93

-------
Table B2.  LOGARITHMS OF FORMATION CONSTANTS FOR
           PHENYLMERCURIC COMPLEXES  (CgHgRgX) IN
           H20, 25° C
Ligand X
-OH~
-OCOCH3"
-OCOCH3CIV
-SCeHs"
[C6H5HgX]
Lo§ [c6H5Hgn [x-i
9.89
4.82
4.51
> 16
Reference
16
16
16
17
                        94

-------
        Table B3.   LOGARITHMS OF FORMATION CONSTANTS FOR 18
                   METHYLMERCURIC COMPLEXES (CHgHgX) IN WATER
Ligand Xs
-r
-Cl"
-Br~
-I"
-OH"
-OCeHs-
-OCOCH3~
-HPO,2-
-HP032"
-S2~
-SCH2CH2OIT
-SR" (Cysteine)
-so32-
-s2o32~
-SCN"
-NH3
-NH2CH2CH2NH2
-or
rpTJ UrrVl a
L^nygllgA 1
r GH« H &"^" n r x!~ i
9.37
5.25
6.62
8.60
9.37


5.03
4.67
21.2
16.12

8.11
10.90

7.60
8.25
14.2
(9.5)
(5.45)
(6.7)
(8,7)
(9.5)
(~ 6.5)
(~ 3.6)




(15.7)


(6.1)
(8.4)


'Values in parentheses taken from reference 15  (ionic strength =
 0,5,  temp. 25°C); other values from reference  18  (ionic strength
 0.1,  temp. 20°C).
                                 95

-------
Table B4.  DEGREES OF DISSOCIATION OF 10 ORGANOMERCURY COMPOUNDS
           IN AQUEOUS SOLUTION AT 25° C
Compound
C6H5HgOCOCH3
CsHgHgSR
CH3HgCl
CHgHgBr
CHsHgl
CHaHgOCOCHg
CH3HgHP04~
CHgHgNHsR1"
CH3HgSR
CH3HgS-
Degree of
10"4 M
32
< 0.1
21
4.8
0.5
76
26
1.6
< 0.1
< 0.1
dissociati
10" 7 M
> 99
< 0.1
98
75
3.3
> 99
99
39
< 0.1
< 0.1
on (%) at
ID" 10 M
> 99
< 0.1
> 99
> 99
91
> 99
> 99
> 99
< 0.1
< 0.1
                                96

-------
   Table B5.   EQUILIBRIUM CONSTANTS FOR HYDROLYSIS OF MERCURIC,
               ALKYLMERCURIC, AND PHENYLMER'CURIC IONS AT 25° C
Reaction
Hg2+ + H2 0 «±
HgOH* + H20 «i
Hg2+ + 2H20 ;±
CHaHg* + H20 *
C2HsHg+ + H20
CsHBHg+ + H20
HgOBT*" + H*
Hg(OH)2 + H+
Hg(OH)2 + 2lf
i CHgHgOH + H*"
j± CgHgHgOH -f rf"
*± CgHgHgOH + tf"
pK (log IT1)"
3.70 + 0.07
2.60 ± 0.09
6.30 + 0.05
4.50b
4.90V
4.11b
Reference
14
14
14
18
16a
16
'Ionic strength was 0.5 for  all constants  listed.
'Standard deviations not given.
                                   97
                                                  *U.S. GOVERNMBNT PRINTING OFFICE; 1973 5
-------
  SELECTED WATER
  RESOURCES ABSTRACTS
  INPUT TRANSACTION FORM
                    1. Report No.
                                                                   J.  Accession No.
                                       iw
  4.  Title
  CHEMISTRY OF ORGANC-MERCURIALS IN AQUATIC  SYSTEMS
  7. Author(s)  Baughman, George L., Gordon, John A.,  Wolfe,  N.
              Lee,  and Zepp, Richard G.
                                        I J. Report bate
                                                                  t 5. r rformt.. f Organization
                                                                     RepoitNo.
  9.  Organization
   Southeast Environmental Research Laboratory
   National Environmental Research Center-Corvallis
   U.  S.  Environmental protection Agency
	Athens^_Geo.rg,ia J.Q6Q.1	
   12.  S; nsorm Otgae  ttion   o. S. Environmental  Protection Agency

  IS.  Supplementary Notes

        Environmental Protection Agency report number,
        EPA-660/3-73-012, September 1973.
                                        10. Project No.
                                              310301QQG
                                        11.  Contract I Grant No.
                      \lj
                                           Type I Repo  and
                                           Period Covered
                                          Final  Report
  16. Abstract
  Kinetics in water of some chemical and  photochemical reactions postulated  as key
  transformations in the environmental mercury cycle were investigated.  Decomposition
  of dimethylmercury (DMM) and diphenylmercury (DPM) by acids and mercuric salts was
  shown to be.pH dependent and too  slow to be significant under most environmental  con-
  ditions.  Degradation of organomercuric salts by acid is even slower.  Theoretical
  evidence indicates that loss of elemental mercury or DMM at the air-water  interface can
  be important in turbulent systems.
  Dimethylmercury,methylmercuric chloride,  methylmercuric hydroxide, and methylmercuric
  ion  were not decomposed by sunlight, but phenylmercury and sulfur-bonded methylmercuric
  species were readily decomposed to inorganic mercury.  Detailed equilibrium calcula-
  tions indicate that the sulfur-bonded methylmercuric species are the predominant  specie
  in natural waters.  Quantum yields for  these reactions are presented along with a
  technique for calculating sunlight photolysis rates from laboratory data.
  The  report also includes a review of  the chemical literature concerning the kinetics
  of chemical and photochemical decomposition of organomercurials.   (Baughman -
  Southeast Environmental Research  Laboratory)
  17s. Descriptors
  *Heavy metals, *Hydrolysis, *Kinetics,  Metal organic pesticides, Chemical  degradation,
  Aqueous solutions, Water chemistry,  Evaporation, Air-water interfaces
  17b. Identifiers
  *Photolysis, *0rganomercury,  Photodegradation, Complex formation, Mercury,
  Methylmercury, Phenylmercury,  Diphenylmercury, Dimethylmercury
  17c. CO WRR Field & Croup 05B
  IS.  Availability
15, Security Class,
   'Repot.)
"i. Set lityC'i *s.
   (Page)
21. No. of
   Pages

 2. Pr, «f
                                                       Send To:
                                                       WATER RESOURCES SCIENTIFIC INFORMATION CENTER
                                                       U.S. DEPARTMENT OF THE INTERIOR
                                                       WASHINGTON. O. C. 2O24O
  Ahstractor George L. Baughman
              institution  Southeast Environmental  Research Laborator '
WRSIC 102(RfcV.jUNF 1971)

-------