&EPA
           United States
           Environmental Protection
           Agency
           Environmental Research
           Laboratory
           Duluth MN 55804
EPA-600 3 8(
           Research and Development
Direct Photolysis of
Hexacyanoferrate
Complexes

Proposed
Applications to the
Aquatic Environment

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency,  have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination  of traditional grouping was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.   Environmental Health Effects Research
      2.   Environmental Protection Technology
      3.   Ecological Research
      4.   Environmental Monitoring
      5.   Socioeconomic Environmental Studies
      6,   Scientific and Technical  Assessment Reports (STAR)
      7.   Interagency  Energy-Environment Research and Development
      8.   "Special" Reports
      9.   Miscellaneous Reports

This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on  the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed  for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine 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.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia  22161.

-------
                                             EPA-600/3-80-003
                                             January 1980
DIRECT PHOTOLYSIS OF HEXACYANOFERRATE COMPLEXES
Proposed Applications to the Aquatic Environment
                       by

  Steven J. Broderius and Lloyd L. Smith, Jr.
Department of Entomology, Fisheries, and Wildlife
             University of Minnesota
            St. Paul, Minnesota 55108
               Grant No. R805291
                Project Officer

               John E. Poldoski
       Environmental Research Laboratory
            Duluth, Minnesota 55804
       ENVIRONMENTAL RESEARCH LABORATORY
      OFFICE OF RESEARCH AND DEVELOPMENT
     U.S. ENVIRONMENTAL PROTECTION AGENCY
           DULUTH, MINNESOTA 55804

-------
                                  DISCLAIMER

     This report has been reviewed by the Environmental Research Laboratory-
Duluth, U.S. Environmental Protection Agency, and approved  for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or  recommendation
for use.
                                       11

-------
                                   FOREWORD

     This work represents an effort  to  characterize  the  effect  of  sunlight  and
various other environmental parameters  on  the  decomposition  of  iron-cyanide
complexes to give hydrogen cyanide.  Past  work by  the  authors indicate  that
hydrogen cyanide can be highly  toxic to aquatic  organisms.   Therefore,  this
work should provide additional  insight  for making  hazard assessments  of
iron-cyanide discharges into the  environment.
                                       J. David Yount
                                       Acting Director
                                       Environmental  Research  Laboratory-Duluth
                                      111

-------
                                   ABSTRACT

     The theory and computations described by Zepp and  Cline  (1977) were
experimentally tested in predicting  the direct  photolysis  rates  of dilute
hexacyanoferrate (II) and (ill) solutions in the aquatic environment.
Essential information for these calculations includes the  quantum yield  for
the photoreaction, molar extinction  coefficients of  the complex  ions  for
wavelengths > 295 nm, solar irradiance data used to  calculate specific
sunlight absorption rates, and the assumption that the  photolysis reaction
obeys a first-order kinetic rate expression.  Direct photolysis  rates of the
irreversible photochemical reactions are calculated  as  a function of  the time
of year, latitude, time of day, meteorological  conditions,  and depth  in
natural water bodies.  Light of wavelengths < 480 nm is active in the
photolysis reactions, and pH, temperature, and  concentration  all affect  the
reaction to varying degrees.  Assuming first-order kinetics,  in  which the rate
constant was approximately concentration independent within the  range of
25-100 ug/1 total cyanide, the minimum quantum  yields of HCN  formation were
0.14 and 0.0023 for the iron (II) and (III) complexes,  respectively.  These
values correspond to minimum, nearsurface, midday half-lives  at  midsummer of
about 18 and 64 min at St. Paul, Minn.  The photolysis  rate at various  fixed
depths in a natural water column, when compared with that  at  the surface,
decreases exponentially with depth.  It is suggested that  the photolysis
reactions are enhanced by suspended material in turbid  waters because of the
forward scattering of light when compared with  that  theoretically calculated
from beam attenuation coefficients.  Hexacyanoferrate (II)  and (III)  solutions
of equal initial total cyanide concentration respond photochemically  quite
differently from one another in solutions prepared with deionized water,  but
respond in a similar manner for solutions prepared with natural  waters.   The
potentially rapid photodecomposition of iron-cyanides with formation  of  HCN
suggests that this phenomenon may be of toxicological importance under  certain
environmental conditions.

     This report was submitted in fulfillment of Grant  No.  R805291 by the
Department of Entomology, Fisheries, and Wildlife, University of Minnesota,
under the sponsorship of the U.S. Environmental Protection Agency.  This
report covers a period from July, 1977 to August, 1978, and work was  completed
as of August, 1978.

-------
                                   CONTENTS

Foreword	ill
Abstract	   iv
Figures	   vi
Tables	viii
Acknowledgments	   ix

     1.  Conclusions	    1
     2.  Recommendations 	    3
     3.  Introduction	    4
     4.  Materials and Methods 	    8
              Analytical methods  	    8
              Test procedures and apparatus	    8
              Computational approach	   10
     5.  Results and Discussion	   12
              Volatilization of HCN	   12
              Free cyanide and HCN determinations	   14
              Kinetics  of photodecomposition 	   14
                   Molar extinction coefficients 	   14
                   Specific sunlight  absorption rate  ..'........   16
                   Properties of  photochemical reaction   	   16
                   Quantum yield  	   23
                   Time of year and latitude	   23
                   Diurnal change  	  ......   27
                   Attenuation by natural waters 	   27
                   Sky  conditions and photolysis rate  .........   39
                   Photolysis in  natural waters  ....... 	   39

References .............  	   45
Appendix

     A.  Beam Attenuation Coefficients	   48

-------
                                   FIGURES

Number                                                                    Page

   1  Electronic absorption spectra of hexacyanoferrate  (II)  and  (III)
        solutions	17

   2  Specific sunlight absorption rates of hexacyanoferrate  (II)  and
        (III) complexes as a function of wavelength at midday and  mid-
        summer, latitude 40° N	19

   3  Relative photolysis rate constants normalized to the values
        determined for solutions with an initial  total cyanide concen-
        tration of 25 ug/1 CN and as a function of the average iron-
        cyanide concentration during the rate determination period  ...   22

   4  Midday half-lives for 100 Mg/1 CN hexacyanoferrate  (II)  solutions
        at near surface depths for different times of the year under
        full sunlight conditions at St. Paul, Minn	24

   5  Midday half-lives for 100 ng/1 CN hexacyanoferrate  (III)  solutions
        at near surface depths for different times of the year under
        full sunlight conditions at St. Paul, Minn	25

   6  Midday half-lives for direct photolysis of  pure water hexacyano-
        ferrate (ll) and (III) solutions (near surface) as a  function of
        the time of year for several northern latitudes.  Values are
        relative to July 1 rate at 45° N latitude	26

   7  Diurnal variation of direct photolysis rates of pure water hexa-
        cyanoferrate (II) solutions (near surface) relative to
        photolysis rates at midday on July 1 at latitude  45°  N,
        longitude 93.2° W	28

   8  Diurnal variation of direct photolysis rates of pure water hexa-
        cyanoferrate (III) solutions (near surface) relative  to
        photolysis rates at midday on July 1 at latitude  45°  N,
        longitude 93.2° W	29
      Time of day dependence of direct photolysis rates of  pure water
        hexacyanoferrate (II) solutions (near surface) relative to
        photolysis rates for midday at St. Paul, Minn, on October 20,
        1977.  Theoretical relationship indicated by  smooth line  ....   30
                                      VI

-------
Number                                                                    Page

  10  Time of day dependence of direct  photolysis  rates  of  pure water
        hexacyanoferrate  (III) solutions  (near  surface)  relative  to
        photolysis rates  for midday at  St.  Paul, Minn,  on October 21,
        1977.  Theoretical  relationship indicated  by  smooth line  ....    31

  11  Attenuation coefficients relative to  deionized  water  for natural
        water samples  collected  in north-central United  States  	    32

  12  Influence of turbidity on  the experimentally determined  to
        theoretically  calculated  relative photolysis  rate  of hexa-
        cyanoferrate (II) and  (III) solutions	    35

  13  Calculated depth-dependence of  the direct photolysis  of  hexa-
        cyanoferrate (II) at midday and midsummer  for latitude 40° N
        when using beam attenuation coefficients and  assuming  complete
        mixing of the  water column	    38

  14  Relationship between  photolysis rate  and  solar  radiation,  both
        normalized to  that  predicted  for a  clear day	    40
                                      VII

-------
                                    TABLES

Number                                                                    page

   1  Rate of HCN loss from sodium cyanide .solutions prepared with
        deionized water and in open cylindrical jars under  laboratory
        conditions	    13

   2  The HCN concentration estimated by the indirect colorimetric
        method and that determined by the direct vapor phase equili-
        bration procedure for hexacyanoferrate (ll) and  (ill) solutions
        exposed to sunlight and initially containing 100  ug/1 CN  .  .  .  .    15

   3  Molar extinction coefficients of hexacyanoferrate  (II) and  (III)
        solutions	    18

   4  The linear relationship between exposure period (X) near midday
        for different meteorological conditions and calculated log
        iron-cyanide concentration (Y), for solutions with  initial
        complex concentrations up to 200 pg/1 as total cyanide, as
        indicated by the regression correlation coefficient  	    20

   5  Experimentally determined and theoretically computed  relative
        photolysis rates for hexacyanoferrate (II) and (ill) solutions
        as normalized to deionized water controls and affected by
        bentonite and wind-blown silt.  Computed rates are based on
        beam attenuation coefficient measurements  	    34

   6  Physical properties of various natural waters and  the theoretical
        and determined rate of decrease in the depth-dependent direct
        photolysis rate of hexacyanoferrate (II) and (III)  solutions
        prepared with deionized water and exposed to natural light at
        specific fixed depths at latitude 45° N, longitude 93.2° W . .  .    37

   7  Determined photolysis rates for hexacyanoferrate (II) solutions
        prepared with different waters relative to the rates for
        similarly exposed solutions prepared with deionized water  ...    41

   8  Determined photolysis rates for hexacyanoferrate (ill) solutions
        prepared with different waters relative to the rates for
        similarly exposed solutions prepared with deionized water  ...    43

   9  Ratio of determined photolysis rates for hexacyanoferrate (III)
        to (ll) solutions of equal initial total cyanide  concentration
        prepared with different water types and exposed  to  the same
        natural light conditions 	    44

                                     viii

-------
                               ACKNOWLEDGMENTS

     The authors wish to express  our  appreciation  to Drs.  Richard  Zepp,  Donald
Bahnick, and William Swenson  for  reviewing  this  manuscript and  to  both Dr.
Zepp and David Cline of the Southeast Environmental  Research Laboratory,  U.S.
Environmental Protection Agency,  Athens,  Georgia for technical  assistance and
use of their computer program for  calculating  the  theoretical photodecompo-
sition of iron-cyanide complexes.  We also  wish  to thank Mr. Walter Koenst  for
his assistance in preparing figures  for  the final  report.
                                       IX

-------
                                  SECTION 1

                                 CONCLUSIONS

1.  The evaporative  loss  of  HCN from aqueous  solutions  is  directly  related  to
    the initial concentration  and test  temperature.   This  loss  from natural
    waters  is  relatively  slow  when compared with the potential  photolysis
    rate of hexacyanoferrate (II)  and (ill) complexes for  near  surface
    conditions.

2.  The colorimetric  method  for the determination of free  cyanide gave  the
    highest estimates  of  HCN in photodecomposed  iron—cyanide  solutions.
    These values were  essentially constant  at about  1.18 and  1.56 times  the
    HCN determined by  the volatilization method,  regardless of  the  extent  of
    photodecomposition,  for  the hexacyanoferrate (II) and  (III)  solutions,
    respectively.

3.  Only light of wavelengths  less than about 420 and 480  nm  are active  in
    the photodecomposition of  hexacyanoferrate (ll)  and (ill) complexes,
    respectively.  The maximum absorption of  solar radiation  by these
    complexes  occurs  at about  330 nm for hexacyanoferrate  (II)  solutions and
    330 and 420 nm for hexacyanoferrate (ill) solutions.

4.  The test pH, temperature,  and concentration  all  have a varying  affect  on
    the photolysis reaction  of both iron-cyanide complexes.

5.  The essentially  irreversible photolysis reaction of hexacyanoferrate (II)
    and (ill)  complexes  can  be approximately  described  by  first-order
    kinetics for concentrations up to 100 pg/1 as total cyanide,  with minimum
    quantum yields of  HCN formation determined to be 0.14  and 0.0023,
    respectively.

6.  The minimum, near  surface,  midday direct  photolysis half-lives  for hexa-
    cyanoferrate (II)  solutions containing  100 Mg/1  CN  ranged from  about 50
    min in  late fall  to a minimum of about 18 min in midsummer  at St. Paul,
    Minn.  In  comparable  hexacyanoferrate (ill)  solutions  the midday
    half-lives ranged  from about 160 min in late fall to a minimum  of about
    64 min in midsummer.

7.  Midday haIf-lives  for the  direct photolysis  of hexacyanoferrate (II) and
    (III) complexes near  the surface of an aqueous solution and  the amplitude
    of their time of year variation should  increase  with increasing northern
    latitude.

8.  The intensity of  incident  collimated sunlight that  penetrates natural
    waters is attenuated  through absorption and  scattering, thus  diminishing
    the photodecomposition rate of iron-cyanides  by  reducing  the  amount of
    light available.                  1

-------
 9.  The logarithm of photolysis rate at a  specific  depth  in  a  water  column,
     as compared to the rate at the surface, was observed  to  linearly decrease
     with depth in natural waters.

10.  The photolysis reaction in turbid waters  is enhanced  by  suspended
     material due to forward scattering of  light when  compared  with  that
     theoretically calculated from beam attenuation  coefficients.

11.  The relatively small concentration of  HCN resulting from photolysis  of
     hexacyanoferrate (II) is of minimal toxicological  importance  below depths
     of about 50 to 100 cm in most well-mixed natural waters  likely  to receive
     this complex as a pollutant.

12.  The photodecomposition of hexacyanoferrate  (II) and (III)  solutions  is a
     direct function of natural light intensity, under  varying  meteorological
     conditions.

13.  The hexacyanoferrate (ll) and (ill) complexes photochemically decompose
     quite differently from one another in  deionized water solutions  but
     produce similar amounts of HCN in solutions prepared  with  natural waters.
     Because the photolysis reaction of the iron (II)  complex was  observed  to
     be unaffected by water type, effluents containing  these  iron-cyanides,
     when discharged into receiving waters  similar to  those tested,  are
     expected to respond, with regards to HCN production,  like  that  of the
     hexacyanoferrate (II) complex.

14.  The maximum amount of total cyanide that  can be photochemically released
     as HCN from dilute hexacyanoferrate (ll) and (III) solutions  prepared
     with deionized water was determined to be about 85 and 49%, respectively.
     This indicates that for every mole of  iron  (II) and (III)  complex, each
     containing 6 moles CN, only 5 and 3 moles of CN,  respectively,  can be
     released from the complex anions to form  free cyanide through a
     photolysis reaction in solutions prepared with  deionized water.

15.  This study indicates that sunlight photodecomposition of iron-cyanides
     may provide an environmentally important  pathway  under certain  conditions
     for their conversion into toxic HCN.

-------
                                  SECTION  2

                              RECOMMENDATIONS

1.  Studies are needed  to determine  the penetration  of  sunlight  into  natural
    waters as a function of  light absorption and  scattering  characteristics
    for wavelengths greater  than  295 nm.  These measurements could  then be
    used in mathematical models for  predicting photolysis  rates  as  a  function
    of depth.

2.  Data collected during this  study indicate  that hexacyanoferrate complexes
    may undergo photodecomposition  in  natural waters which can be of  ecolog-
    ical importance under certain circumstances.   Since little  is known about
    the behavior of these complexes  in natural aquatic  environments,  research
    should be initiated to determine the  fate  of  these  compounds and  HCN in
    different receiving waters.   This  work  could  also be used to test the
    proposed photolysis model  in  actual representative  polluted  waters.

3.  A greater understanding  of  oxidation-reduction reactions involving hexa-
    cyanoferrate complexes,  as  related to redox properties of natural waters,
    is needed.

4.  The toxicity of hexacyanoferrate (ll) and  (ill)  complexes to aquatic
    organisms has been  demonstrated  to be essentially due to the presence of
    HCN as derived from the  photodecompositon  reaction.  Therefore, a
    suitable analytical method  for  the determination of HCN in the  ug/1 range
    is needed in order  to determine  the actual adverse  effects  of cyanide on
    aquatic organisms in waters receiving iron-cyanide  wastes.

5.  Nearly all of the lethal and  sublethal  effects  of HCN on aquatic
    organisms have been derived from continuous and  constant toxicant
    exposure tests.  However,  it  is  reasonable  to expect HCN concentrations
    in most natural waters to  fluctuate because of  intermittent  waste
    discharge, and due  to the  loss  of  cyanide  and limited formation of HCN
    from the photodecomposition of  iron-cyanides  at  night.  Therefore,
    additional information on  the toxicity  of  fluctuating HCN concentrations
    is needed in order  to more  realistically  establish  the adverse  effects of
    cyanide on aquatic  populations  in natural  waters.

-------
                                   SECTION 3

                                  INTRODUCTION

      Appreciable amounts of ferro- and ferricyanides  (i.e., hexacyanoferrate
 (II) Fe(CN)£~,  and (ill) Fe(CN)jT) may occur in effluents  from
 color film photographic processes, electrostatic conversion washes,  the manu-
 facture of iron and steel, the cracking of oil, and from plants which
 manufacture these iron salts.  It has also been established that  the toxicity
 to many aquatic organisms of aqueous solutions of various  simple  cyanides  and
 metallocyanide  complexes is determined virtually alone by  the concentration of
 undissociated molecular HCN and not by the concentration of the cyanide ion
 (CN~) or of most metallocyanide anions (Doudoroff, 1976; Broderius et al.,
 1977).  Hydrocyanic acid (HCN) is formed in hexacyanoferrate (II) and (III)
 aqueous solutions mainly through photodecomposition of the iron-cyanide
 anions, which are otherwise highly stable and relatively nontoxic, and through
 hydrolytic reaction with water of the cyanide ions so liberated.  Fish kills
 in a New York stream (Burdick and Lipschuetz, 1950) and a Japanese river
 (Kobayashi and  Mori, 1973) were associated with the discharge of  iron-cyanides
 in industrial effluents at concentrations less than generally accepted as
 nonlethal.  The rapid development of a toxic situation was demonstrated to
;result from photodecomposition of the iron-cyanides by bright sunlight and
 release of free cyanide (i.e., HCN + CN~).  Similar findings were reported
 by Myers (1950) for seepage from a ferromanganese blast furnace and  for
 chemicals from  photographic processing by West (1970) and Terhaar et al.
 (1972).  The influence of sunlight and pH on degradation of iron-cyanides  in a
 polluted natural water was reported by Kongiel-Chablo (1966).

      The photochemical behavior of potassium ferro- and ferricyanide in
 aqueous solutions and the kinetics of their decomposition and reverse reaction
 has been investigated extensively for various irradiation wavelengths in  the
 absorption spectral region.  Solutions of both complexes when kept in the  dark
 or in diffused  light are stable but when exposed to light of certain
 wavelengths, decomposition occurs with the formation of aquopentacyanoferrates
 and reduction of ferricyanide to ferrocyanide.  According to Ohno and
 Tsuchihashi (1965) and Ohno (1967), irradiation of aqueous hexacyanoferrate
 (II) solutions  with light of 366 nm causes a ligand substitution  reaction
 yielding aquopentacyanoferrate (II) (Fe(CN)5*H20^~) and cyanide ion.
 Light of 253.7  nm resulted in a photo-oxidation reaction producing
 hexacyanoferrate (III) and reducing species.  Upon continued irradiation  of
 ferrocyanide solutions with wavelengths longer than 300 nm, Fe^+ and HCN
 are also formed and the pH slowly increases during irradiation, as a result of
 the hydrolysis  of the cyanide ions produced, and after some time  tends to
 reach a constant value (Asperger, 1952).  When the light is removed, the
 reaction is apparently reversed under certain circumstances and the  original
 pH restored if  the irradiation has not been too prolonged.  The reverse
 reaction apparently depends on the iron-cyanide concentration and is caused by
                                       4

-------
a  slower  secondary reaction between the aquo salt formed and the parent
compound  (MacDiarmid and Hall, 1953).  Asperger ^t jQ. (1960)  stated that the
reaction  is  apparently reversible only when no appreciable decomposition of
the aquopentacyanoferrate (II) ion occurred.  Under prolonged  irradiation on
aerated and  relatively concentrated solutions, Fe(OH)3 (in alkaline
solutions) and  Prussion blue (in acid solutions) are formed (Balzani and
Carassiti, 1970).

     A mechanism for the photodecomposition of potassium ferrocyanides can be
represented  according to Asperger (1952) and Mitra _et al. (1963) by the
primary reaction

                            Fe(CN)g~ + Fe(CN)|~ + CN~

                             CN- + H20 5 HCN + OH~

                                 + H20 t Fe(CN)5*
                               " + 2 H20 t Fe(CN)5'(H20)3~ + HCN + OH~     (1)

Asperger  (1952) and  Balzani  and Carassiti (1970)  have stated that in the
second phase  the  thermally and photochemically unstable FetClOg*(H20)3~
is decomposed by  prolonged exposure to light with a  progressive
photosubstitution reaction with increase in pH and release of Fe2+ plus
CN~ ions.  In the presence of oxygen Fe(OH)-j can  be  formed.

     There are a  number  of publications on the photochemistry of aqueous
ferricyanide  solutions,  but  Balzani and Carassiti (1970)  stated that the
complicated photochemical  behavior of this complex is not yet completely
understood.  Moggi £t  a^.  (1966)  showed that the  photochemical behavior of
Fe(CN)^  was  qualitatively the same regardless of the wavelength of
irradiation (254, 313, or  405 run light).  Spectral changes suggested that
Fe(CN)5*(H20)2~ was  formed as the main product with  the hydrolysis
step probably proceeding according to

                  2Fe(CN)|~  + 2H20 t 2Fe(CN)5'(H20)2~ + 2CN~              (2)

In the dark the complex  slowly underwent an oxidation-reduction reaction which
was accelerated by light and the presence of CN~  according to

         2Fe(CN)5*(H20)2-  +  CN~ + 20H~ t 2Fe(CN)5'(H20)3- + CNO~ + H20    (3)


with the overall  hydrolysis  reaction

            2Fe(CN)|~  + H20  + 20H~ t 2Fe(CN)5'(H20)3~ + CN~ + CNO~        (4)


Upon irradiation  by  light  the rates  of the above  reactions (equations 2 and  3)
increase, Fe2+ ions  are  formed,  and the pH of  the solution initially
increases.

-------
     Balzani and Carassiti  (1970) stated  that  in  addition  to a
photosubstitution reaction, a direct photoreduction  of  FeCCN)^"  to
Fe(CN)5*(H20)^~" could occur.  The oxidation-reduction reaction  of hexacyano-
ferrate (III) ions was demonstrated by Adamson  (1952) to be  rapid whenever  the
process involves a simple electron transfer and to be slow and  of complex
mechanism if such a step cannot occur.  The major product  of the rapid  reaction
was suggested to be hexacyanoferrate (II)  ion  (i.e., Fe(CN)&~ +  e~ ->• Fe(CN)6~
0.36 volt) rather than an aquocyanide and  no intermediate  ions with  a different
number of coordinate cyanide groups is involved.  Since the  nature and  yields
of the products for photochemical reactions involving iron-cyanides  are not
completely defined, the overall chemical  changes  and the reaction mechanisms
are not well established.

     The first step in direct photolysis  is the absorption of a  light quantum
(photon) resulting in an electronically excited state of the molecule.   There-
fore, only radiant energy which is absorbed by  a  molecule  can be effective  in
producing photochemical changes.  Each light quantum absorbed activates one
molecule so that the efficiency of the photochemical process can be
represented by the quantum  yield.  This yield  is  defined by  the  ratio of the
number of molecules undergoing chemical reaction  divided by  the  number  of
photons absorbed by the reactants.  In most cases this  is  experimentally
measured by the rate of the chemical reaction  divided by the number  of  quanta
absorbed per second (Balzani and Carassiti, 1970).

     According to a review  by Balzani and  Carassiti  (1970),  the  reported
quantum yields for the photodecomposition  of hexacyanoferrate (II) and  (III)
complexes, as obtained by various authors  using different  experimental
conditions and techniques,  are generally  in disagreement.   Upon  irradiation of
aqueous hexacyanoferrate (II) solutions with wavelengths of  about 313 or 365
nm, the ligand substitution quantum yield  for aquopentacyanoferrate  (ll)
formation was reported to be 0.1 using pH  measurements  in  neutral unbuffered
solutions (Carassiti and Balzani, 1960) and 0.44  for a  0.5 M solution at pH
9.9 (Emschwiller and Legros, 1954).  A value of 0.89 was determined  from the
amount of Fe(CN)5*(H20)   formed at pH 4.0 (Ohno  and Tsuchihashi,
1965).  The quantum yield was reported by  Balzani and Carassiti  (1970)  to be
almost independent of Fe(CN)g~ concentration,  temperature, light
intensity, and pH in the range 7-10.  However, Ohno  (1967) stated that  the
quantum yield decreased with an increase  in the hexacyanoferrate (II)
concentration.

     The quantum yields for the formation  of Fe(CN)5*(H20)2~ and Fe(CN)^~
from irradiation of hexacyanoferrate (ill) solutions with  light  of 365  nm were
reported by Balzani and Carassiti (1970)  to be 0.009 and 0.014 at pH 4, and
0.0065 and approximately 0.018 at pH 10,  respectively.  However, these  results
should be taken with reservation since questionable  procedures were  used in
their determination (Balzani and Carassiti, 1970).

     The lack of correlation between photodecomposition tests conducted with
natural and artificial light can be due to the difference  in the ultraviolet
spectral energy distribution and intensity of  sunlight, and  the  various
artificial light sources.   Because solar  radiation at the  earth's surface has
negligible intensity at wavelengths less  than about 295 nm (Bener, 1972), the

-------
chemical of interest must have  appreciable  absorptivity at  greater  wavelengths
if significant photoreaction  is  to  occur  in sunlight.   Most past  studies  of
the photodecomposition  of hexacyanoferrate  (II)  and  (ill)  complexes were  not
prompted by environmental considerations, and  thus much of  the  information
derived from them cannot be extrapolated  to natural  water  environments.   The
primary purpose of this study was  to  determine the photochemical  response of
iron-cyanides under natural light  conditions  in natural bodies  of water,  both
near the surface and as a function  of depth.   Such a study  allowed  us  to
evaluate the relative importance of these compounds  as  sources  of toxic
hydrocyanic acid (HCN)  in natural waters.   The approach used for  predicting
the photochemical decomposition by  solar  radiation of  such  compounds in
aquatic environments was based  on principles  and equations  described by Zepp
and Cline (1977).

-------
                                   SECTION  4

                            MATERIALS AND  METHODS

ANALYTICAL METHODS

     Aliquots of hexacyanoferrate  (II) and (III) solutions  were  tested
periodically for free  cyanide  (i.e., HCN + CN~)  by  the  pyridine-pyrazolone
colorimetric method  (APHA, 1971).  Molecular HCN concentrations  were
calculated from free cyanide and pH measurements, and dissociation  constants
derived from pKgcN = 3.658 + 1662/T where  T is  temperature  in  degrees
Kelvin (Broderius and  Smith, 1979).  This  procedure worked  satisfactorily for
both the deionized and natural water solutions  when appropriate  corrections
were made.  A method similar to that proposed by Broderius  and Smith  (1977)
for the determination  of ^S and by Broderius (1973) for  studies with
metallocyanide complexes was incorporated  for the direct  determination  of
molecular HCN in the pg/1 range.   This procedure utilizes a glass bead
concentration column coated with 0.1 N NaOH for  collecting  displaced HCN,
which is determined  colorimetrically.  The separation of  HCN from a sample  by
means of displacement  with N2 bubbled through the solution  does  not upset
the chemical equilibrium involving the various  cyanide  forms.

     Turbidity was measured on well-shaken samples with a Hach model 2100A
turbidimeter* and was  reported in  Nephelometric  Turbidity Units  (NTU).   Total
and filterable residue, pH, and dissolved  oxygen were also  determined
according to standard  procedures (APHA, 1971).

     Light was measured with a LI-COR model LI-185 quantum  meter**, a LI-192S
underwater quantum sensor with a spectral  response based  on photon  absorption
between 400 and 700 nm, and a mv recorder.  For  experiments relating different
meteorological conditions with photolysis  rate,  these measurements  were
integrated with exposure period to determine change in  energy  associated with
radiation incident to  test solutions.

     Molar extinction  coefficients were determined with a Beckman DB-GT
spectrophotometer from electronic  absorption spectra of the hexacyanoferrate
(II) and (III) complex anions  in deionized water.  The  light from the
spectrophotometer was  insufficient to cause any  measurable  photochemical
reaction.
TEST PROCEDURES AND APPARATUS

     All chemicals were of reagent-grade quality and were  used without
additional purification.  Iron-cyanide solutions were usually prepared by
* Hach Chemical Company, Ames, Iowa.
**Lambda Instruments Corporation, Lincoln, Nebraska.
                                      8

-------
weighing out potassium  salts  immediately  before  use  and  dissolving  in
deionized water saturated with  dissolved  oxygen.   Appropriate  further  dilution
under laboratory  light  gave  the desired  concentrations.   Iron-cyanide
solutions with concentrations of 25  to 200  Mg/1  as total CN  were  generally
used.  The pH in  solutions prepared  with  potassium iron-cyanide and deionized
water was maintained with phosphate  or borate  buffer concentrations of about
10~^ M.  Nearly all  test  solutions prepared with deionized water  were
buffered at pH 7.8.

     In photochemical  experiments the reaction cells through which  the light
travels to reach  the dissolved  reactant  must be  transparent  to the  exciting
radiation.  This  was accomplished by conducting  the exposure experiments  in
open vessels or in  tightly sealed tubes  which  were irradiated  by  sunlight  on
the roof of our laboratory in St. Paul,  Minn,  (latitude  45°  N, longitude  93.2°
W).  The test solutions were  exposed to  sunlight in 25 x 150 mm Teflon lined
screw-cap Pyrex test tubes (Corning  No.  9826)  unless stated  otherwise. A
series of tubes for  each  concentration was  filled to the brim  and submerged in
40 mm of deionized  water  in  a water  bath thermostatically controlled,  usually
at 20 _+ 0.1°C.  The  tubes were  positioned horizontally above a black
background in a shallow tray  to minimize reflected light, or for  field
experiments at various  fixed  depths  in a  natural water column.  It  was
determined that the  photolysis  rates for solutions prepared  with  deionized
water and exposed in test  tubes were not  significantly different  (P =  0.05)
from those determined  for 8  1 of test solution in an open 2-gal Pyrex  acid
battery jar (21 cm  diameter  x 25 cm  high, Corning 6942)  blackened on the
inside and tilted directly towards the  sun's rays.  Therefore, change  in
photolysis rates  due to internal reflection of sunlight, and  slight  absorption
of active light wavelengths  in  the Pyrex test tubes submerged  in  water
appeared to be insignificant  factors. Losses of HCN to the  atmosphere from
open vessels were also minimal  during the relatively short exposure periods.
The concentration of free  cyanide determined in an open vessel was  the same
for solutions which  were  stirred occasionally or undisturbed.

     The shadowing  effect  on the test solutions  by the walls of the test  jars
during low sun angles  prevented using them in an upright position  for  the
determination of  actual photolysis rates.  However, the jars can  be used  for
comparative type  experiments.  To determine the effect of turbidity and
materials dissolved  in natural  waters on the photodecomposition reaction,
hexacyanoferrate  (II)  and  (ill) solutions in test jars were  exposed to natural
light filtered by about 40 mm of various solutions placed in open Pyrex
filters (27 cm square  x 5  cm high) placed directly over the  jars.  The
photolysis rates  calculated  for these iron-cyanide solutions were divided by
the rates determined for  comparable  solutions subjected for  the same exposure
period but in jars  under  light  that  had  been filtered through 40  mm of
deionized water.   In this  way relative photolysis rates were calculated.

     During the photolysis experiments,  samples from test jars were frequently
taken or test  tubes  from a certain series were removed periodically and
analyzed spectrophotometrically for  free cyanide.  Control solutions kept in
the dark showed no measurable formation  of free cyanide during the exposure
period of similar solutions.  Based  on the fraction of hexacyanoferrate (II)
and  (III) that disappeared,  as  measured  by the formation of  free cyanide or

-------
HCN, photolysis rate  constants and half-lives  were  calculated assuming first-
order kinetics.
COMPUTATIONAL APPROACH

     The rate of HCN loss  from  open acid  battery  jars  is  proportional to the
remaining hydrogen cyanide concentration  at  any time.   According to Palaty and
Horokova-Jakubu (1959),  this  loss  can  be  expressed  by  a first-order rate
equation.  Therefore, the  half-life, or time required  to  reduce the original
concentration of hydrogen  cyanide  to one-half,  is given by the equation
                              0.693     _   _   0.693  t                       ...
                         \ = ~~k~ °r \ =  ln(C0/C)                       C5)


where CQ is the initial  and C the  final HCN  concentration after time t.  If
In CQ/C calculated from  the experimental  data is  plotted  as a function of
HCN removal time, equation (5)  is  an expression of  a straight line passing
through the zero coordinates.   The slope  of  the line equals the rate constant
(k=ln(C0/C)/t) of HCN removal.  The value of k is a function of the factors
affecting HCN loss and consequently it can be used  to  characterize and compare
these effects.

     Quantitative calculations  of  direct  sunlight photolysis rates and half-
lives were made from quantum yields and electronic  absorption spectra of
hexacyanoferrate (II) and  (III) aqueous solutions assuming first-order
kinetics.  Equations defining the  intensity  and path lengths of direct and sky
radiation in air and water as described by Zepp and Cline (1977) were
employed.  With this approach,  photolysis rates under  average intensities
derived from full sunlight and  the whole  sky on cloudless days were calculated
as a function of latitude  and longitude,  time of  year,  time of day, depth in a
water body, and other influencing  parameters. The  input  data required to
calculate the theoretical  photolysis rates are

     (1) molar extinction  coefficients of the hexacyanoferrate (II) and (III)
         complex anions  at wavelengths >  297.5 nm;

     (2) the attenuation coefficient or absorbance  per cm for the reaction
         medium (a refractive index of 1.34  is assumed);

     (3) the quantum yield for  the direct photolysis reaction;

     (4) the angular height of  the sun as determined from the solar declin-
         ation, solar right ascension, and sidereal time  for the date of
         interest as obtained from the American Ephemeris and Nautical
         Almanac;

     (5) the latitude and  longitude; and

     (6) the average atmospheric ozone layer thickness that pertains to the
         season and location of interest  (a  value of 0.30 cm was assumed -
         London, 1963).

                                       10

-------
     The average  sunlight  photolysis  rate at  a certain wavelength (x) is
proportional to the  quantum  yield  (;\)  for the reaction of the weakly
absorbing system.  The  kinetic  expression for this formula as discussed by
Zepp and Cline (1977) is
where [P] is the hexacyanoferrate  (II)  or  (ill)  concentration in moles/liter
and ka,  is the specific  sunlight absorption rate as a function of wave-
length.  The ka, term  is equal  to  the wavelength specific average rate of
light absorption per unit volume (la>)  times the molar extinction coeffi-
cient of the iron-cyanide (e^)  divided  by  the attenuation coefficient of the
water body ("A) times  a  constant (i,  6.02  x 10  ) which converts the
intensity expressed  in photons  cm  s    into units compatible with [P]
(Zepp and Cline, 1977).   Assuming  that  the quantum yield for the photolysis
reaction in solution  is  not  wavelength  dependent then the rate expression is
where kfl equals  the  sum of the ka»  values integrated over all wavelengths
of sunlight absorbed  by the reactants.  This expression conforms to a first-
order rate equation  in which the photolysis rate constant (<{>ka) is expressed
in units of reciprocal time.  The value of ka changes during the day because
the surface spectral  flux distribution of solar radiation is a function of the
solar zenith angle.   For natural waters ka is also a function of competitive
light absorption  by  the water itself along with other absorbing materials and
light scattering.  Because in most natural waters attenuation due to light
scattering is  less important than that due to absorption, scattering was
initially ignored in  the following computations.  Surface reflection and
scattering from water bodies was also ignored because it is small (about 5 to
6 percent) for most  solar zenith angles (Hutchinson, 1957).

     For a first-order reaction the half-life (tu), or time required for one-
half of a compound to react, is independent of the initial concentration and
can be expressed  by

                          ti  = —T~,	 or t.  =
where t  is  the  time  of exposure, and PQ is the initial iron-cyanide concen-
tration  expressed  as HCN with that after exposure (P) equal to PO minus the
determined  HCN  concentration.  The half-lives represent the period of time
required  to decompose the iron-cyanides with release of half the HCN that can
be potentially  formed from the specific complexes.  The photolysis rate
constant  (ka)  can be calculated from the expression ln(Po/P)/t.

     A computer program written in Fortran IV, that uses the above mentioned
theory and  equations to compute direct photolysis rates, is available on
request  from the Environmental Research Laboratory-Athens, Georgia, U.S.
Environmental Protection Agency (Zepp and Cline,  1977).

                                       11

-------
                                   SECTION  5

                            RESULTS AND  DISCUSSION

VOLATILIZATION OF HCN

     The volatilization of HCN  from 8  1  sodium cyanide solutions prepared with
deionized water and placed in 2 gal acid battery  jars  was determined in the
laboratory as a function  of temperature  and  concentration.  This rather slow
loss of cyanide, determined from 6 hr  exposure periods,  is apparently not
independent of the  initial concentration as  indicated  by the variation in the
first-order rate constants in Table 1.   This  correlation between rate constant
for HCN loss and initial  concentration implies that  the  evolution of HCN from
an open aqueous solution  does not  follow the  first-order kinetics "as antici-
pated.  However, from  the limited  number of  tests conducted, it is concluded
that the loss is approximately  first-order,  especially when one considers the
possible analytical error introduced  into  the determinations for solutions
with the lowest initial cyanide concentrations.

     At a given initial cyanide concentration a decrease in test temperature
resulted in higher  residual cyanide concentrations.  This is indicated by a
reduction with decreasing temperature  in the  rate constant for HCN loss to the
atmosphere from the surface of  comparable  solutions.   Over the temperature
range 10 to 25° C,  this decreased  rate is  approximately  a linear function of
temperature such that  for a 10° C  decrease the rate  constant is reduced by
about 55%.  Experiments conducted  with solutions  placed  in jars on the roof of
our laboratory demonstrated that the  rate  of  HCN  loss  was about twice that
determined in the laboratory.   This increase  is most  likely a result of
increased loss due  to  wind action, since it was demonstrated that HCN was not
subjected to photolysis by natural light.  Also,  solutions placed on the roof
and loosely covered with  a sheet of glass  showed  approximately the same rate
of HCN loss when compared with  similar solutions  in  the  laboratory.

     Since the loss of HCN from a  solution into the  atmosphere is through the
phase borderline, the  rate of cyanide  decrease is a  function of the surface
area to solution volume ratio.  For our  containers with  perpendicular walls
the HCN removal rate is inversely  proportional to the  solution depth.  Thus
the time required for  the removal  of  HCN to  1/2 its  original concentration is
directly proportional  to  the solution depth,  since the same type of container
and solution volume was used in all HCN  rate  loss experiments, temperature and
initial concentration  are the only variables  for  which a comparison of HCN
removal can be made.

     The total loss of cyanide  from natural  waters occurs not only by vola-
tilization into the atmosphere  but also  through chemical reactions and
biological oxidation to ammonia and C02«  The above  calculated rates
represent only minimal losses since the  oxidative loss is negligible in
                                      12

-------
    TABLE 1.  RATE OF  HCN  LOSS  FROM SODIUM CYANIDE  SOLUTIONS  PREPARED WITH
  DEIONIZED WATER AND  IN OPEN CYLINDRICAL  JARS  UNDER LABORATORY  CONDITIONS*

Initial free
cyanide cone.
(ug/1 CN)

25
50
100
200

25
50
100
200

10
Rate Constant
0.00624
0.0103
0.0136
0.0149
Half Life,
111.0
67.3
50.8
46.5
Temperature
15
, kChour'1)
0.0139
0.0117
0.0135
0.0173
tu (hour)
49.7
59.4
51.5
40.1
(°c)
20

0.0194
0.0225
0.0215
0.0254

35.8
30.8
32.2
27.3

25

0.0257
0.0263
0.0299
0.0315

27.0
26.3
23.2
22.0
*Test solution pH was  about  7.9.
                                      13

-------
deionized water solutions as well as  the  loss  through  agitation  since  the
solutions were unstirred.

     Lure and Panova (1964) determined  that  for  river  water  test  solutions
initially containing 10 to 25 mg/1 CN and exposed  in open  vessels,  the
concentration of cyanide during standing  declines  after  7  days  to about  10%  of
the original with the complete disappearance of  cyanide  occurring in 10-12
days.  These authors concluded that in  the natural water they  investigated,
the loss of cyanide by volatilization is  the most  significant  of  the means of
HCN decline.

     The rate of HCN loss has been demonstrated  to depend  on temperature but
Palaty and Horokova-Jakubu (1959) observed that  the intensity  of  agitation and
ratio of the solution volume to surface area are also  very important factors.
In fact, they determined the removal  rate in vigorously  air-agitated solutions
to be one order of magnitude greater  (12  to  14 times)  than the  nonagitated
solutions.  However, even though the  decomposition of  simple cyanides
dissolved in natural and deionized water  occurs  differently, it  is  likely that
the rapid loss of HCN under the most  favorable circumstances is  still  consid-
erably slower than the formation of HCN from photolysis  of hexacyanoferrate
(II) and (III) complexes for midday and near surface conditions.


FREE CYANIDE AND HCN DETERMINATIONS

     It has been assumed by previous  workers (Burdick  and  Lipschuetz,  1950)
that the pyridine-pyrazolone method for the measurement  of free  cyanide  does
not upset the chemical equilibria by  liberating  cyanide  from the  hexacyano-
ferrate complex ions or intermediate  photolysis  forms.   Comparison  of  the
direct vapor phase method and the indirect colorimetric  methods  for the
determination of HCN (Broderius, 1973) demonstrated that the indirect  pyridine-
pyrazolone method gives a relatively  high estimate of  free cyanide  in
iron-cyanide solutions.  This may be  due  to  some cyanide which  is liberated
from the hexacyanoferrate complex ions  or intermediate photolysis forms  by the
pyridine-pyrazolone method.  In five  tests with  each complex,  however,  the
ratios between HCN concentrations determined by  the direct vapor  phase
equilibration procedure and those estimated  by the indirect  colorimetric method
were relatively constant, regardless  of the  degree of  decomposition, at  about
84.6% and 64.2% for the hexacyanoferrate  (II) and  (ill)  solutions,  respectively
(Table 2).  Therefore, the indirect colorimetric method  for  the  estimation of
HCN concentrations can be used when the relative (normalized)  photolysis rates
of hexacyanoferrate (II) and (ill) solutions exposed to  natural  light  are
determined.  When the proper correction is made  the results  from  the indirect
colorimetric method can also be used  to determine  the  actual photolysis  rate of
these complexes as indicated by HCN formation.


KINETICS OF PHOTODECOMPOSITION

Molar Extinction Coefficients

     The electronic absorption spectra  of freshly  prepared and  dilute  potassium
Fe(CN)g  and Fe(CN)g  aqueous solutions were measured  in the wavelength
                                      14

-------
  TABLE 2.  THE HCN CONCENTRATION ESTIMATED BY THE  INDIRECT  COLORIMETRIC
    METHOD AND THAT DETERMINED BY THE DIRECT VAPOR  PHASE EQUILIBRATION
    PROCEDURE FOR HEXACYANOFERRATE (II) AND (III) SOLUTIONS  EXPOSED  TO
              SUNLIGHT AND INITIALLY CONTAINING  100 ng/1 CN*

Test

1
2
3
4
5


1
2
3
4
5
Determined HCN concentration, yg/1

A B Percentage
Indirect Direct vapor phase B of
colorimetric equilibration initial total Ratio
method method cyanide B/A
Ferrocyanide solutions
33.9 29.9
38.8 31.4
60.4 51.2
83.1 68.9
98.2 84.8

Ferricyanide solutions
25.7 17.2
45.2 27.7
47.2 31.7
60.9 37.5
64.8 41.4

29.9 88.2
31.4 80.9
51.2 84.8
68.9 82.9
84.8 86.4
mean 84 . 6
^2.9

17.2 66.9
27.7 61.3
31.7 67.2
37.5 61.6
41.4 63.9
mean 64.2
*Test temperature was  25°  C  and pH 7.9.
                                     15

-------
region close to the  strong ultraviolet  absorption bands.   As seen in Figure 1,
these complexes are  photosensitive at wavelengths <420 nm for the ferro and
<480 nm for the ferri  ions.  Therefore,  photolysis rates  of these complexes
would be expected to reflect fluctuations  in  intensity of the shorter
wavelength component of  sunlight.  The  molar  extinction coefficients (e)
(Table 3) derived from the absorption spectra  of  freshly  prepared aqueous
solutions of potassium Fe(CN),  and Fe(CN)^   (approximately
10   M) agreed closely with  those reported by  Ibers and Davidson (1951),
Adamson (1952), and Asperger (1952).  A possible  reaction product,
Fe(CN)5*(H20)3~, has also been  reported to absorb photoactive
spectral light (Asperger, 1952).  However, this product is a transient
intermediate and since absorption spectra  of  reaction solutions  were quite
similar in shape and reductions in these spectra  were proportional  to the
degree of decomposition, it  was assumed that  the  predominant cyanide
components contributing  to the  total absorption were the  hexacyanoferrate (II)
and (III) ions.

Specific Sunlight Absorption Rate

     The rate of a photochemical reaction  in  an aqueous solution'is dependent
upon the solar spectral  irradiance at the  solution surface,  radiative transfer
from air into the solution,  and the transmission  of sunlight in  the solution.
Through absorption by components in the atmosphere,  the intensity of sunlight
is decreased such that the ultraviolet  and visible region decreases with
decreasing wavelength so that essentially  no  light is transmitted to the
earth's surface at wavelengths <295 nm  (Bener, 1972).  Molar extinction
coefficients of hexacyanoferrate (II) and  (III) ions can  be coupled with
actinic irradiance data  of Leighton (1961) and Bener (1972)  to calculate the
specific rate of sunlight absorption (kfl)  (Zepp and Cline, 1977).  The
magnitude of these rates depends upon the  degree  of spectral overlap between
the electronic absorption spectra and the  spectrum of sunlight at the earth's
surface.  The specific sunlight absorption rates  (ka*) of hexacyanoferrate
(II) and (III) complexes were computed  as  a function of wavelength  for shallow
depths and apply to midsummer and midday at latitude 40°  N.   Only wavelengths
of less than about 480 nm are photometrically  important (Figure  2).  The
maximum interaction with solar  radiation occurs at about  330 nm  for ferro-
cyanide solutions and at 330 nm and 420 nm for ferricyanide solutions.  The
ratio of integrated ka, values  (ka) for the wavelength region of 297.5 -
490 nm indicates that the sunlight absorption  rate constant  for  ferricyanide
is about 17 times larger than that for  ferrocyanide (Figure 2).

Properties of Photochemical  Reaction

     The photolysis  of hexacyanoferrate (II)  and  (III) complexes in solutions
of certain initial total cyanide concentrations follows first-order kinetics
as indicated by typical  linear  regression  correlation coefficients  in Table 4.
These values are for the relationships  between midday exposure period and
calculated log iron-cyanrde  concentration  for  solutions with initial complex
concentrations up to 200 ug/1 as total  cyanide.   It has been proposed that the
reaction is reversible in darkness (Asperger  et^ _a_l., 1960).   However, deter-
minations of the amount of released cyanide that  is  recombined or lost showed
that the reaction is irreversible in the dark  for ferricyanide solutions, but

                                      16

-------
                o
                OC
                UJ
                Q.
                UJ
                U
                CO
                a:
                o
                
-------
TABLE 3.  MOLAR EXTINCTION COEFFICIENTS OF HEXACYANOFERRATE (II) AND
                           (III) SOLUTIONS

Wavelength
(nm)
297.5
300.0
302.5
305.0
307.5
310.0
312.5
315.0
317.5
320.0
323.1
330.0
340.0
350.0
360.0
370.0
380.0
390.0
400.0
410.0
420.0
430.0
440.0
450.0
460.0
470.0
480.0
490.0
Extinction coefficient,
Fe(CN){T
483.7
428.5
387.9
363.2
346.9
335.3
330.6
329.0
328.5
328.5
326.4
309.5
252.7
174.8
105.3
55.8
25.8
11.0
5.3
2.6
1.6
1.6
1.0
1.0
0.5
0.5
0.0
0.0
liter/mole- cm
Fe(CN)^~
1589
1673
1710
1677
1575
1410
1279
1215
1193
1195
1175
938
571
334
336
434
577
770
945
1007
1053
932
597
261
80
18
4
0
                                 18

-------
                  10
                   -s
                  10'
                   -5
                 10
                   -6
                     I//
_J	I	1_
                       375  300  335  350  375  40O  425  450  475  500

                                 WAVELENGTH  (nm)
Figure 2.  Specific sunlight absorption rates of hexacyanoferrate (II)  and
           (III)  complexes as a function of wavelength  at midday and mid-
           summer,  latitude 40° N.
                                      19

-------
TABLE 4.  THE LINEAR RELATIONSHIP BETWEEN EXPOSURE PERIOD  (X) NEAR MIDDAY FOR
     DIFFERENT METEOROLOGICAL CONDITIONS AND CALCULATED LOG  IRON-CYANIDE
   CONCENTRATION (Y), FOR SOLUTIONS WITH INITIAL  COMPLEX CONCENTRATIONS  UP
  TO 200 ug/1 AS TOTAL CYANIDE, AS INDICATED BY THE  REGRESSION  CORRELATION
                                 COEFFICIENT

Total Exposure
cyanide, period,
iig/1 CN min
25 0
15
45
75
50 0
15
45
75
100 0
15
45
75
200 0
15
45
75
25 0
15
45
75
135
50 0
15
45
75
135
100 0
15
45
75
135
200 0
15
45
75
135
Determined Calculated
HCN iron-cyanide
concentration, concentration as
ug/1 CN ug/1 CN
0
4.95
13.0
15.1
0
11.2
27.6
36.2
0
36.8
69.4
82.7
0
68.4
128.8
156.3
Fe(CN)|~
1.87
5.87
7.96
10.8
0
4.46
11.5
16.2
22.3
0
8.22
21.6
31.5
43.9
0
28.9
63.7
82.0
94.9
25.00
20.05
12.0
9.9
50.0
38.8
22.4
13.8
100.0
63.2
30.6
17.3
200.0
131.6
71.2
43.7
25.00
23.13
19.13
17.04
14.2
50.00
45.54
38.5
33.8
27.7
100.00
91.78
78.4
68.5
56.1
200.00
171.1
136.3
118.0
105.1
Linear
correlation
coefficient,
r

0.962



0.999



0.992



0.990




0.972




0.981




0.986




0.897


                                      20

-------
 is  slightly reversible though incomplete for ferrocyanide solutions that have
 undergone only minimal decomposition.  After prolonged exposure the reaction
 becomes  irreversible, even for ferrocyanide solutions.

      The photolysis reactions for both iron-cyanide complex solutions
 initially containing 100 ug/1 CN were virtually unaffected by dissolved oxygen
 in  the range 2-8 mg/1.  There was a moderate effect of pH on the photolysis
 reactions with the rates increasing with a decrease of test pH over the range
 9.0 - 6.6.   The relative rates normalized to that determined at pH 6.6
 decreased to about 0.84 and 0.70 at pH 9.0 for the ferri- and ferrocyanide
 solutions,  respectively.  The decrease in relative rate with increase in pH is
 linear over the pH range tested for ferricyanide solutions but decreases in a
 linear manner only from pH 6.6 to about 8.0 and then remains fairly constant
 to  pH 9.0 for ferrocyanide solutions.

      The photochemical reaction rate for iron-cyanide solutions initially
 containing  100 pg/1 CN and at pH 7.8 was measured at 5.6, 12.1, and 23.5° C.
 The slight  negative temperature relationship between the relative photolysis
 rate  (Y)  normalized to the rate determined at 23.5° C, as a function of test
 temperature in °  C (X), can be represented by
          Log Y =  -0.104 + 0.00464 X (r = 0.962)
      and Log Y =  -0.136 + 0.00574 X (r = 0.999)
 for hexacyanoferrate (ll) and (ill) solutions,  respectively.  Balzani and
 Carassiti (1970)  stated that this observed temperature dependence may be due
 to  secondary thermal reactions which contribute to the overall quantum yield.

      Varying the  initial concentration of iron-cyanide salt affected the
 photodecomposition rate constant.  The dependence of this constant on concen-
 tration  demonstrates, according to Asperger (1952), that the decomposition is
 due not  only to the light energy, but also to collision of molecules.
 Therefore,  photolysis of the hexacyanoferrate (II) and (ill) complexes is not
 truly first-order.  For practical purposes,  however,  we can assume that the
 reactions  follow  first-order kinetics for complex concentrations that are
 likely to be found in natural waters.  The relative photolysis rate constants
 normalized  to those determined for solutions at 20° C with an initial total
 cyanide  concentration of 25 Mg/1 CN and as a function of the average iron-
 cyanide  concentration during the rate determination period are represented in
 Figure 3.   The pH of all  solutions was about 7.8.  The rate constants of the
 decomposition reaction decrease with increasing initial cyanide complex
 concentration up  to some  apparent limiting concentration.  The relative
 photolysis  rate constant  normalized to that  determined for 25 pg/1 total
 cyanide  (Y)  was a function of average hexacyanoferrate (II)  concentration (X)
 up to 1000  Ug/1 CN and can be represented by the empirical expression,  Y =
 2.099 x~0-222" (r = -0.954).   For hexacyanoferrate (ill)  solutions the
 above relationship can be represented by Y = 1.390 x~°-1538 (r = -0.904).
 Photolysis  rates  were determined for comparable hexacyanoferrate (II)  and
 (ill) solutions prepared  with deionized water and exposed to the same light
 conditions.   The  ratio of these  rates for ferro- to ferricyanide solutions
 over  the  initial  total cyanide concentration range of 25  - 200 yg/1 CN
averaged  2.05 _+ 0.32 for  32 such comparisons.   However,  the rate of change in
 photolysis  rate constants (Figure 3) is such that at  initial total cyanide
 concentrations  of approximately  2.0 mg/1  CN  and greater the decomposition rate

                                       21

-------
       o
       o
       UJ
       I
      CO
      CO
     1.0

     0.9

     0.8

     O.7

     0.6

     0.5

     O.4

     0.3

     0.2

     O.I

      0
                                            Y-2.099 X
                                               r-0.954
Fe(CN)*"
                                                    -0.2220
                    100
                          200
                                  300
                                         400
                                                500
                                                       600
                                                              700
                                                                     800
                                                                            9OO
Q,

UJ

I
_l
UJ
           1.0

           O.9

           O.S

           O.7

           O.6

           0.5

           0.4

           0.3

           O.2

           O.I

            0
                                     Y-1.390  X
                                        r« 0.904
                                             -0.1538
                                    Fe(CN)
                                         3-
                    IOO    200    300    400    500    600     700     800

                    AVERAGE  IRON-CYANIDE  CONCENTRATION   (fJG/L CN)
                                                                     9OO
Figure 3.   Relative  photolysis  rate constants normalized to the values
            determined  for solutions with an  initial total cyanide concentration
            of 25 ng/1  CN and as  a function of the average iron-cyanide
            concentration during  the rate determination  period.
                                         22

-------
for ferricyanide solutions  is  faster  than  for  comparable  ferrocyanide
solutions.

     The maximum amount  of  total  cyanide that  could  be  photochemically
released as HCN from prolonged exposure of dilute  hexacyanoferrate  (II) and
(III) solutions was determined to be  about 85% and 49%, respectively.  This
indicates that for every mole  of  iron (II) and (ill)  complex,  each  containing
6 moles CN, only 5 and 3 moles of CN, respectively,  can be  released as free
cyanide from the complex anions  through a  photolysis  reaction.   These results
are not consistent with  the reaction  pathways  for  the photodecomposition  of
hexacyanoferrate complexes  as  proposed in  the  introduction  (equations 1 and
4).  This supports the contention of  Balzani and Carassiti  (1970)  that the
overall chemical changes and the  reaction  mechanisms  for  the  photolysis of
hexacyanoferrate complexes  is  not well defined.

Quantum Yield

     Minimum direct photolysis rates  of hexacyanoferrate  (II)  and  (ill)
complex solutions prepared  with  deionized  water were  determined  empirically  at
near-surface depths and  midday for different  times of the year under full
sunlight at St. Paul, Minn. The  experimental  midday  half-lives  for
hexacyanoferrate (II) solutions  initially  containing 100  yg/1  CN at pH 7.8 and
20° C ranged from about  50  min in late fall  to a minimum  of about  18 min  in
midsummer (Figure 4).  For  comparable hexacyanoferrate (III)  solutions the
midday half-lives ranged from  about 160 min  in late  fall  to a  minimum of  about
64 min in midsummer (Figure 5).

     The quantum yields  (4>) for  the photodecomposition of the iron-cyanide
complexes were estimated by a  visual  best  fit  analysis  of observed date-
dependent midday half-lives, determined  from the  change in  HCN concentration
(eq. 8, p. 11), to the theoretical curves  calculated for  certain $  values and
specific sunlight absorption rates (kfl)  of the reactants  as determined from
the Zepp and Cline computer program.   From our investigation  it  was experi-
mentally determined that the iron-cyanide  disappearance quantum  yield as
indicated by HCN formation  for the hexacyanoferrate  (II)  and  (III)  complexes
are approximately 0.14 and  0.0023, respectively  (Figures  4  and 5).   This
calculation assumes that the quantum  yields  are wavelength  independent in the
region of sunlight absorption.   Our experimentally determined quantum yields
are in reasonable agreement with  those reported  in the review by Balzani  and
Carassiti (1970) (see p. 6).

Time of Year and Latitude

     The intensity and spectral  distribution of  sunlight  on a horizontal
surface generally decreases with  decreasing  angular  height  of the  sun.
Therefore, intensity decreases from midday to^sunset, from  summer  to winter,
and from the tropics to  higher latitudes.   Midday  half-lives  for direct
photolysis of hexacyanoferrate (ll) and  (III)  complexes near  the surface  of  an
aqueous solution and as  a  function of time of  the  year and  latitude are  shown
in Figure 6.  The results  were computed  as relative  values  with  the half-life
of each complex on July  1  at latitude 45°  N  assigned a value  of  unity.  For
both complexes the photolysis  half-lives at  the midlatitudes  are predicted  to

                                       23

-------
        CO
        UJ
        UJ
        u,
        _l
        I
        u.
        _J
        o
        Q
               JFMAMJJASONDJ
                               TIME   OF   YEAR
                                                                -40
                                                                 30
                                                                -20
                                                                - 10
Figure 4.  Midday half-lives for 100 yg/1 CN hexacyanoferrate (II)  solutions

           at near surface depths for different  times  of  the  year under  full

           sunlight conditions at St. Paul,  Minn.
                                      24

-------
         tu
         =}
         UJ
         <
         X
         o
         Q
              zoo
               180
              I6O -
              I4O-
                                                               200
              IOO •
                JFMAMJJASONDJ
                               TIME  OF   YEAR
Figure 5.  Midday half-lives for 100 ug/1 CN hexacyanoferrate  (ill) solutions
           at near surface depths for different  times of  the year under full
           sunlight conditions at St. Paul, Minn.
                                      25

-------
             o
             Q
            UJ
            <
            X
            UJ
            CE
                                                             -0.8
                   JFMAMJ   JA
                                                  0   N
                                 TIME  OF   YEAR
Figure 6.  Midday half-lives for direct photolysis of pure water hexacyano-
           ferrate (II) and (III) solutions (near surface) as a function of
           the time of year for several northern latitudes.  Values are
           relative to July 1 rate at 45° N latitude.

                                      26

-------
be minimal  during the summer and maximum during the winter months.  Both the
half-lives  and  the amplitude of the time of year variation increase with
increasing  northern latitude.  In the tropical zone photolysis rates should be
relatively  constant throughout the year.  Variations in rates during the
summer  are  also expected to be minimal with less than a 1.3-fold increase from
the equator to  latitude 60° N.  The relative half-lives for both iron-cyanide
complexes are quite similar at all latitudes and times of the year.  During
the ice-free months,  the variation in relative midday half-lives is expected
to be less  than a value of 3 on sunny days and for latitudes to 60° N.

Diurnal Change

     The direct photolysis rate of iron-cyanide complexes changes diurnally as
the intensity of sunlight increases and then decreases throughout the day,
with maximum rates occurring at midday.  This variation in computed rates for
shallow depths  and relative to photolysis rates at midday on July 1 at
latitude 45° N, longitude 93.2° W is illustrated for the first of various
months  in Figures 7 and 8.  Experimentally determined photolysis rates
relative to midday values and as determined at different times of the day for
pure water  iron-cyanide solutions were close to the theoretically computed
values as indicated in Figures 9 and 10.

Attenuation by  Natural Waters

     Light  falling directly on the surface of a water body is both reflected
at an angle equal to  the angle of incidence and penetrates with a change in
direction due to refraction.  The fraction of direct sunlight and sky
radiation that  is reflected is small when compared with that which penetrates
(Wetzel, 1975).   The  reflected light is essentially the same in spectral
composition as  incident light but the spectrum of light penetrating the
surface of  the  water  is altered.  The intensity of incident collimated sun-
light that  penetrates natural waters, when compared with that of pure water,
is attenuated through absorption and scattering.  The reduction in trans-
mission of  light and  shift in absorption selectivity depends on the
wavelengths of  the incident light and the depth to which the light has
penetrated.  In inland surface waters absorption is due mainly to dissolved
metallic io<.s and natural organics.  Attenuation coefficients (e^) were
measured for different water bodies with a Beckman DB-GT spectrophotometer  and
are presented in Appendix A.  In all cases the attentuation of light was wave-
length dependent and  varied considerably from one water body to another
(Figure 11).  This is especially true for the ultraviolet region where
attenuation of  light  intensity increases with decreasing wavelength.

     Both suspended and dissolved materials within a water body induce
variations  in the depth of light penetration.  Increasing turbidity decreases
transparency and shifts maximum wavelength transmission to longer wavelengths.
Substances  in a solution that absorb or scatter light should diminish the
photodecomposition rate of iron-cyanides by reducing the amount of light
available.  This effect was demonstrated by using turbid suspensions prepared
with deionized  water  and bentonite colloidal clay or wind-blown silt from
melted snow.  The experiments were conducted with an apparatus consisting of a
Pyrex glass filter containing 40 mm of the suspension and horizontally

                                       27

-------
            I.Or
                         8      10      12      2      4

                            TIME  OF  DAY  (C.S.T.)
Figure 7.  Diurnal variation of direct photolysis rates of pure water hexa-
           cyanoferrate (II) solutions (near surface) relative to photolysis
           rates at midday on July 1 at latitude 45° N, longitude 93.2° W.
                                     28

-------
                          8      10      12      2      4


                             TIME   OF  DAY  (C.S.T.)
Figure 8.  Diurnal variation  of  direct  photolysis  rates  of  pure water  hexa-
           cyanoferrate  (III)  solutions (near  surface) relative to  photolysis
           rates at midday  on July  1  at latitude 45°  N,  longitude 93.2° W.
                                      29

-------
            LU
            Sc
            K
            >
            5
            111
            en
                  I.Or
                 0.9-
                 0.8
                 0.7 •
                 0.6
                 0.5
                0.4
                O.2 -
                               8      10      12      2


                                 TIME  OF  DAY  (C.S.T.)
Figure 9.  Time  of day-dependence of direct  photolysis rates of pure water
           hexacyanoferrate (II) solutions  (near surface) relative to
           photolysis rates for midday at St.  Paul, Minn, on October 20,  1977,
           Theoretical relationship indicated  by smooth line.
                                       30

-------
                i.o
               0.9
               0.8
               0.7
               0.6
           Ul   0.5
           UJ
           tr
               0.4
               0.3
               0.2H
               O.I
                          FC (CN)
                                     10
                                            12
                                TIME  OF  DAY  (C.S.T.)
Figure  10.   Time of day dependence of direct photolysis rates  of pure water
             hexacyanoferrate  (III)  solutions (near surface)  relative to
             photolysis rates  for midday at St. Paul, Minn, on  October 21, 1977,
             Theoretical relationship indicated by smooth line.
                                       31

-------
CO
NJ
                      E
                      u
                      o
                      UJ
                      o
                      u

                           .eoor
                           .900
                           .400-
                           .300
                           .200
                           .100
                                      MISSISSIPPI
                                 V  \  RIVER
                                  EL£PHA
                                     CAK
                                      LAKE PHALEN

                                      SQUARE LAKE
                                 300     320     340
                                                       360     360     400     420    440
                                                                                             460
                                                                                                     480
                                                           WAVELENGTH (nm)
             Figure  11.  Attenuation  coefficients  relative  to  deipnized  water  for  natural  water samples
                          collected in north-central United  States.

-------
positioned  over  an open cylindrical jar which contained the iron-cyanide
solutions at  pH  7.8 and 20°  C.   In this manner,  the suspensions were not in
direct  contact with the iron-cyanide solutions and all of the light reaching
an iron-cyanide  solution had to first pass through a filter.  During the short
exposure periods a negligible amount of suspended material settled out on the
bottom  of the  filters.

     Relative  photolysis rates  were defined as those normalized to rates
determined  for iron-cyanide  solutions for which  the overlying Pyrex filters
contained 40 mm  of deionized water.  The ratios  of the experimentally
determined  and theoretically computed relative photolysis rates of hexacyano-
ferrate (II) and (ill)  solutions as affected by  suspensions of benttmite clay
and wind-blown silt are presented in Table 5.  The relationships between the
ratios  expressed as logarithms  and turbidity were determined, as represented
in Figure 12,  to be linear at least to 40 NTU.  These lines can be charac-
terized for the  bentonite solutions by a slope function (g) of 0.0618 and
0.0502  for  the hexacyanoferrate (II) and (ill) solutions, respectively.  The
slope for the  relationship determined when light had to pass through wind-
blown silt  solutions before  reaching hexacyanoferrate (ll) solutions is
0.0456.

     The total attenuation of radiance passing through a medium is due to
absorption  and to a redirection or scattering of some of the beams radiance.
Determination  of the forward scattering function requires artificial light
which yields a collimated beam  (Tyler and Preisendorfer, 1962).  Therefore,
this parameter could not be  measured from our data.  However, from Figure 12
it is apparent that the photolysis reaction in iron-cyanide solutions is
enhanced by suspended materials when compared with that theoretically
calculated  from  beam attenuation coefficients.  In fact, one might expect up
to a threefold to fourfold increase in photolysis rates when compared with
those calculated for iron-cyanide solutions in natural waters with a turbidity
of about 10 NTU.  The photolysis reaction for ferrocyanide solutions is
affected to a greater degree by the presence of  bentonite than for
ferricyanide solutions.  The presence of wind—blown silt had less of an
enhancement effect  than that of bentonite on the photodecomposition of
hexacyanoferrate (II) solutions (Figure 12).  This may be due to a reduction
in the  forward scattering of light by silt solutions when compared with those
prepared with bentonite.

     The decrease in photolysis rate with increasing depth is dependent upon
the relative magnitude  of the attenuation coefficients of the water body, the
molar extinction coefficients of the iron-cyanide complex, and the intensity
of sunlight, all as a function  of spectral wavelength.  In general, the
shorter wavelengths are more readily removed by  the surface layer of most
natural waters.   Thus,  the photochemically active light which decomposes iron-
cyanides is expected to penetrate only a short distance in natural waters.

     Field  experiments  on the photolysis of iron-cyanide solutions in natural
waters were conducted by suspending test tubes containing cyanide solutions
prepared with pH 7.8 phosphate  buffered deionized water for a specific time
period at known  depths  in a  water column.   Using beam attenuation coefficients
and assuming photolysis at a specific depth,  the depth-dependence of the

                                       33

-------
TABLE 5. . EXPERIMENTALLY DETERMINED AND THEORETICALLY COMPUTED  RELATIVE
   PHOTOLYSIS RATES FOR HEXACYANOFERRATE (II) AND  (ill) SOLUTIONS AS
 NORMALIZED TO DEIONIZED WATER CONTROLS AND AFFECTED BY BENTONITE AND
    WIND-BLOWN SILT.  COMPUTED RATES ARE BASED ON  BEAM ATTENUATION
                       COEFFICIENT MEASUREMENTS

Test
solution and
concentrat ion ,
(mg/1)

Bentonite
Bentonite
Bentonite
Bentonite
Silt
Silt
Silt

Bentonite
Bentonite
Bentonite
Bentonite
Bentonite
Bentonite
Bentonite

75
250
600
750
—
-
-

75
100
200
250
400
500
600
Relative photolysis rate
normalized to controls
Turbidity
NTU

6.7
24.0
57.0
68.0
23.0
43.0
63.0

5.6
9.9
18.5
23.0
36.4
46.0
54.0
A
Determined
FERROCYANIDE
0.916
0.632
0.326
0.260
0.403
0.186
0.0824
FERRICYANIDE
0.958
0.960
0.805
0.872
0.528
0.603
0.421
B
Computed

0.365
0.0155
0.000147
0.0000264
0.0431
0.00197
0.000107

0.471
0.240
0.0906
0.0405
0.00936
0.00244
0.00111
Ratio of
A to B

2.51
40.8
2218
9848
9.35
94.4
770.1

2.03
4.00
8.88
21.5
56.4
247.1
379.3
                                  34

-------
        «5
        QC
        UJ
        UJ
        QC
        O
        UJ
        ID
        O

        <
        O
        O
        UJ
        a:
        UJ
        h-
        UJ
        O
                                             = 0.0456
                                                        4-
                                               •   Fe(CN)K  BENTONITE
                                                   Fe(CN)g" SILT


                                                   Fe(CN)g  BENTONITE
                       10      20      30      40      50


                                 TURBIDITY   (NTU)
60
       70
Figure 12.   Influence of  turbidity on the experimentally determined  to

             theoretically calculated relative  photolysis rate of hexacyano-
             ferrate (II)  and  (III) solutions.
                                        35

-------
 direct photolysis of hexacyanoferrate (II) and (ill) solutions relative  to
 near-surface rates were computed with the Zepp and Cline (1977) computer
 program for latitude 45° N, longitude 93.2° W.  The photolysis rates
 normalized to that at the surface decreased exponentially with depth.  The
 linear regression lines defining these relationships were calculated by
 forcing the line through the log relative rate of 1.0 at depth 0 cm.  The
 theoretical and determined rates of decrease in the depth-dependent photolysis
 of hexacyanoferrate (II) and (ill)  solutions are presented in Table 6.  The
 regression lines apply to relative  photolysis rates as low as one-hundredth
 of that determined at the surface.   The photolysis rate for hexacyanoferrate
 (II)  solutions  decreases more rapidly with increasing depth than the rate for
 hexacyanoferrate (III) solutions.  The rates differ because the predominant
 wavelengths affecting photolysis of the iron (II) complex are shorter and thus
 penetrate  less  into natural waters  than those activating the iron (III)
 complex.

      The ratio  of the theoretical to determined rate of decrease in photolysis
 rates  varied from near 1.0 for Square Lake and bog waters of low turbidity to
 3.52  for the turbid Minnesota River.  The ratios  are apparently a function of
 turbidity  or suspended solids,-since for Square Lake and bog water the
 observed and theoretical rates of decrease were almost identical.  The
 theoretical computations assume that the natural  materials  in a water body act
 only  as photochemically inert sun screens.  Reactions by "sensitizers" and the
 bouncing of light or light scattering in turbid solutions which may increase
 photolysis  over  that predicted is ignored.  However,  it was observed that in
 the more turbid  waters the photolysis reaction was relatively enhanced with
 depth  because of light scattering and thus the rate of decrease in photolysis
 rate with depth  is  less  than theoretically predicted  by the Zepp and Cline
 model.  These results confirm the relationship between turbidity and
 photolysis  rate  as  presented in Figure 12.

     By using the attenuation coefficients- in Appendix A and assuming complete
 mixing of the water  column of interest,  the depth-dependence of the average
 direct photolysis rates  for hexacyanoferrate (ll)  solutions at midday and
 midsummer relative  to near-surface  rates  were calculated  for latitude 40° N by
 the computer program of  Zepp and Cline (1977).  These relationships are shown
 in Figure 13.  From  the  results  in  Table 6 for  photolysis  rates  at  a point for
 various depths in a  water  column, it is  proposed  that  the theoretical
 relationships actually overestimate the  rate at which  the  relative  photolysis
 rate decreases when  mixing is  assumed.   For waters like the Minnesota River,
 the rate of decrease  may  in fact be overestimated  by  a  factor  of  about 3.5
 (Table 6).   For  the  natural  waters  tested  the average  observed decrease is
about one-half as rapid  as  that  theoretically predicted in  Figure 13.   If this
 factor is applied to  the  computed depth-dependence curves where  mixing of the
water column was  assumed  (Figure 13),  it  is  proposed  that below a depth of
about 50 -  100 cm photolysis  of  hexacyanoferrate  (II)  is  insignificant from a
 toxicological standpoint  in  all  tested natural waters.  A relationship for the
hexacyanoferrate  (ill) complex  was  not  included since  the concentration of HCN
 formed by photolysis  of  this  complex in  solutions  prepared  with  natural  waters
was approximately the  same  as  that  produced  in  comparable hexacyanoferrate
 (ll) solutions.
                                      36

-------
TABLE  6.  PHYSICAL  PROPERTIES OF  VARIOUS NATURAL WATERS AND  THE THEORETICAL AND  DETERMINED RATE OF
     DECREASE IN THE DEPTH-DEPENDENT DIRECT PHOTOLYSIS  RATE  OF HEXACYANOFERRATE  (II) AND (ill)
       SOLUTIONS PREPARED  WITH DEIONIZED  WATER  AND EXPOSED TO NATURAL LIGHT AT SPECIFIC  FIXED
                                DEPTHS AT LATITUDE 45° N,  LONGITUDE 93.2° W
Secchi
disk
Water measurement,
body cm

Square Lake
Lake Phalen 202
Lake Como 79
Elephant ISO
Lake
Mississippi -
River
Minnesota 45
River
Bog water

Square Lake
Lake Phalen 202
Lake Como 79
Elephant 180
Lake
Mississippi
River
Minnesota 45
River
Bog water
Residue, mg/1 Linear regression slope values*
Turbidity,
NTU

0.43
2.7
5.4
2.9

3.6

14.0

0.9

0.43
2.7
5.4
2.9

3.6

14.0

0.9
Total Total non A
filterable filterable Total theoretical
FERROCYANIDE
-0.0118
223 3.0 226 -0.0364
190.5 11.0 201.5 -0.0884
74 <1 74 -0.0886

-0.130

495 24.5 519.5 -0.206

-0.290
FERRIC* ANIDE
-0,00516
223 "3.0 226 -0.0201
190.5 11.0 201.5 -0.0590
74 <1 74 -0.0338

-0.0939

495 24.5 519.5 -0.121

-0.112
B
determined

-0.0107
-0.0190
-0.0392
-0.0453

-0.0850

-0.0586

-0.260

-0.00482
-0.00956
-0.0314
-0.0226

-0.0461

-0.0444

-0.120

R2

0.984
0.990
0.998
0.927

0.947

0.939

0.978

0.956
0.925
0.998
0.972

0.976

0.996

0.982
Ratio of
A to B

1.10
1.92
2.26
1.96

1.53

3.52

1.12

1.07
2.10
1.88
1.50

2.04

2.73

0.93
  *Linear regression analysis for  log relative photolysis rates (Y)  normalized to those  determined for  comparable solutions at the surface
   and depth (X) in era.  Regression lines are forced through the log relative photolysis rate of 1.0 at depth of 0 cm.  Log Y = gX with g =
    I (Steel and Torrie, p. 179,  1960).

-------
                                                     A - PURE WATER
                                                     B - SQUARE LAKE
                                                     C - LAKE PHALEN
                                                     0 - LAKE COMO
                                                     E-ELEPHANT  LAKE
                                                     F - MISSISSIPPI RIVER
                                                     G - MINNESOTA RIVER
                                                     H - BOG  WATER
                                       DEPTH  (CM)
Figure 13.   Calculated  depth-dependence of the  direct photolysis  of hexa-
             cyanoferrate (II) at midday and midsummer for latitude 40° N
             when using  beam attenuation coefficients and assuming complete
             mixing of the water column.
                                         38

-------
     In making the depth-dependent  computations,  it  was  assumed  that  the  iron-
cyanide is isotropically distributed  and  all  that  is in  a  water  layer  is
exposed to the same amount  of  light during  a  given time  period.   This
assumption has been demonstrated  according  to Zepp and Cline  (1977)  to  be
valid at depths  in which a  small  fraction of  the  incident  light  is  absorbed.
However, if at a given depth almost all of  the light is  absorbed in the upper
part of the water column, the  assumptions are valid  only if mixing  is  more
rapid than entry of iron-cyanide  into or  loss from the upper  layer  of  the
water body.  If  entry  to the upper  layer  is more  rapid than mixing,  then  the
concentration of iron-cyanide  and thus HCN  will be higher  than predicted  near
the surface.  If a situation exists in which  the  iron-cyanide is initially
uniformly distributed  in the water  column but mixing is  incomplete,  then  it
would be expected that the  photolysis rate  would  be  slower with  increasing
depth than is indicated in  Figure 13. The  above  calculations apply to near
midday and summer situations.   As the sun moves  lower in the  sky as a  function
of the time of day or  latitude,  the underwater path  length of direct  sunlight
is longer and depth dependence increases.

Sky Conditions and Photolysis  Rate

     The daytime radiation  received at any point  on the  earth's  surface
consists of direct solar radiation  or sunlight, and  indirect  solar  radiation
or scattered light of  the sky.  The spectral  distribution of  irradiance
reaching the earth's surface depends  on  the sun's  altitude and the
meteorological conditions of the intervening  atmosphere.  The relative
spectral composition of the energy  distribution of combining  solar  and sky
radiation between 315  and 800  nm is essentially constant during  the day
(Robinson, 1966).  However, a  reduction  in  atmospheric clarity contributes  to
a decrease in intensity of  light falling  on the surface  of a  water  body.
Measurements of  the relative spectral distribution of the radiation which
penetrates a cloud or  an overcast indicate that it will  be essentially the
same as that which enters (Hull,  1954, and Leighton, 1961).   Therefore, the
relative energy/wavelength  curve of sun  and sky radiation together  is  nearly
the same on a clear and overcast day. This phenomenon was tested by
determining the  photolysis  rate as  a  function of  solar radiation, both
normalized to that predicted by interpolation from results for clear days.
This relationship is depicted  in Figure  14.  Since the  experimentally
determined observations are close to  the  45°  line in Figure  14,  it  can be
concluded that the photodecomposition of  hexacyanoferrate (ll) and  (ill)
solutions is a direct  function of natural light intensity for various
meteorological conditions.

Photolysis in Natural  Waters

     The rate of photodecomposition at 15°  C  for  hexacyanoferrate (II)
solutions initially containing 100  ug/1  CN and in Pyrex  test  tube cells
exposed to natural  light was essentially  the  same when  different waters were
used in preparation of the  test solutions.   This  is  demonstrated in Table 7 by
the similarity in photolysis rates  for solutions  prepared with different
unfilterd natural waters relative to  the  rates determined for comparable
solutions prepared with deionized water.   The slight reduction (Table 7)  in
relative photolysis rate for solutions prepared with Minnesota or Mississippi

                                       39

-------
    Ill

    5
    tr
    LU



    1

    UJ
    (T
                 10   20  30   40  50   60  70   80  90   100
               PERCENTAGE   SOLAR   RADIATION
Figure 14.  Relationship between photolysis rate and solar radiation, both

          normalized to that predicted for a clear day.
                               40

-------
TABLE 7.  DETERMINED  PHOTOLYSIS  RATES  FOR HEXACYANOFERRATE
  (II) SOLUTIONS PREPARED  WITH DIFFERENT  WATERS  RELATIVE
       TO THE RATES FOR  SIMILARLY  EXPOSED SOLUTIONS
              PREPARED WITH  DEIONIZED  WATER

Type of
water
Deionized
Well
Lake Phalen
Lake Como
Elephant Lake
Mississippi River
Minnesota River
Test
PH
7.9
8.1
7.6
8.5
7.4
8.5
7.5
Relative
photolysis rate
1.000
1.044
0.988
1.096
0.991
0.819
0.960
                             41

-------
River water is probably due to  the decreased  light  penetration into the 2  cm
path length of the cells containing natural water solutions  relative to those
prepared with deionized water.  Therefore,  the  relationships presented in
previous sections, as determined for deionized  water  solutions,  are appropriate
for hexacyanoferrate (II) solutions prepared  with several  different natural
waters.

     Various natural waters and chemicals were  used to  determine empirically
the effect of materials in different waters on  the  photolysis rate of
Fe(CN)g.  These results are  summarized  in  Table 8  where the photolysis
rate at 15° C of hexacyanoferrate (III)  solutions initially  containing 100 Mg/1
CN and prepared with various  waters are  compared  to the rates determined for
similarly exposed solutions prepared with deionized water.   The Fe++ ion
from the addition of ferrous  chloride under near-zero dissolved oxygen levels
enhanced the photolysis rate, whereas Fe   ion added as ferric chloride
or that derived from the oxidation of Fe   depressed  the reaction.  The
presence of ammonia (ammonium chloride)  and nitrate (sodium  nitrate) had no
effect on the photolysis rate,  but dissolved  sulfide  (sodium sulfide)
dramatically enhanced the photodecomposition  reaction.   The  photolysis rates
for ferricyanide solutions prepared with several  different natural surface
waters and well water were about twice as great in  the  natural water as in
deionized water (Table 8).  This acceleration may be  attributed to the rapid
reduction of hexacyanoferrate (ill) to the  more photochemically active
hexacyanoferrate (II).  This  possibility is supported by the fact that the
ratios of photolysis rates at 15° C for  comparable  ferri-  and ferrocyanide
solutions initially containing  100 yg/1  CN  and  prepared with the same natural
surface water from various sources averaged 0.92  (Table 9).   If our results for
experiments conducted with natural waters are typical of other waters, then it
is believed that iron-cyanides  in effluents being discharged into various
natural waters will photochemically respond like hexacyanoferrate (II) ions,
since essentially no Fe(CN)j?~ ions will  be  present  in the  receiving
waters.

     This study has demonstrated a means by which the cyanide in relatively
nontoxic iron-cyanide complexes may, under  certain  conditions, be largely
liberated as toxic free cyanide in natural  waters.   The rate of this process
and the concentration of free cyanide produced  can  be approximately predicted
from information derived in our study and by  utilizing  the computer program of
Zepp and Cline (1977).  These calculations  can  be made  as  a  function of
longitude, latitude, time of  day and year,  and  depth  in a  natural water body.
The extent of the photolysis  reaction will  be a function of  the iron-cyanide
concentration and the degree  and duration of  illumination.   Photolysis may be
negligible in deep, turbid, or  shaded waters, and the slowly liberated
nonpersistent free cyanide may  decrease  or  escape as  rapidly as it is released.
However, in relatively clear  shallow waters,  or in  situations where effluents
are stratified in the light penetration  zone, the photodecomposition of
iron-cyanides may produce free  cyanide concentrations in the ug/1 range that
could adversely affect the distribution  and abundance of aquatic organisms
(Smith et al., 1979).
                                      42

-------
 TABLE 8.   DETERMINED PHOTOLYSIS RATES  FOR HEXACYANOFERRATE
    (III) SOLUTIONS PREPARED WITH DIFFERENT WATERS  RELATIVE
     TO THE  RATES FOR  SIMILARLY  EXPOSED  SOLUTIONS PREPARED
                    WITH DEIONIZED WATER  (DW)
Type of
water3
Deionized
DW + 0.5 mg/1 Fe+++ (DO)
DW + 0.5 mg/1 Fe++ (DO)
DW + 0.5 mg/1 Fe++ (near zero DO)
DW + 5.0 mg/1 Fe++ (DO)
DW + 0.5 mg/1 NH4+ (DO)
DW + 0.5 mg/1 N03~ (near zero DO)
DW + 0.5 mg/1 sulfide (near zero DO)
WW - after Fe removal0 (DO)
WW - after Fe removal0 (near zero DO)
WW - before Fe removal0 (DO)
Elephant Lake
Lake .Como
Minnesota River
Mississippi River
Bog
Test
PH
7.8
6.2
6.6
7.3
6.0
7.5
7.2
7.6
7.6
8.7
8.4
7.3
8.5
7.7
8.3
7.2
Relative
photolysis rate
1.000
0.790
1.486 •* 0.450b
2.995 •* 1.690b
approx. 0
1.095
0.970
2.309
1.370
1.692
, 2.228
2.299
2.157
1.988
1.758
1.528d

aDO refers to solutions  that  were aerated prior to exposure and  it is
 assumed that dissolved  oxygen concentrations were near  saturation

 Near zero DO refers  to  solutions that were stripped with  N2 prior to
 exposure and it  is assumed that dissolved oxygen concentrations were
 near zero.

"Rate decreased during  exposure period.

cRemoval of iron  from the  well water (WW) was accomplished by a
 Culligan catalytic system which converts Fe(ll)  to Fe(lII).  Iron
 concentrations were  approximately 0.3 mg/1 before and <0.05 mg/1 after
 removal.

°Low because of strong  light  absorption (Figure 11) by bog water
 relative to deionized water  solutions in test tube exposure cells.

                                  43

-------
  TABLE 9.  RATIO OF DETERMINED PHOTOLYSIS RATES FOR
   HEXACYANOFERRATE (ill) TO (II) SOLUTIONS OF EQUAL
     INITIAL TOTAL CYANIDE CONCENTRATION PREPARED
       WITH DIFFERENT WATER TYPES AND EXPOSED TO
           THE SAME NATURAL LIGHT CONDITIONS
       Type of                           Ratio  of
        water                        photolysis rates
Deionized                                 0.417

WW - after Fe removal                     0.694

WW - before Fe removal                    1.090

Lake Como                                 0.869

Elephant Lake                             0.921

Mississippi River                         0.938

Minnesota River                           0.868

Bog                                       0.987
                           44

-------
                                  REFERENCES

Adamson, W.  1952.  Electron Transfer Processes and the Oxidation-Reduction
     Reactions of Hexacyanoferrate  (ill) Ion  in Aqueous Solution.  J. Phys.
     Chem.  56: 858-862.

Alich, Sister M. A., D. T. Haworth, and Sister M.  F. Johnson.  1967.  Spectro-
     photometric Studies of Hexacyanoferrate  (III) ion and its Reaction with
     Iron  (III) in Water and Ethanol.  J.  Inorg. Nucl. Chem.  29(7):
     1637-1642.

American Public Health Association, American  Water Works Association, and
     Water Pollution Control Federation.   1971.  Standard Methods for the
     Examination of Water and Wastewater.   13th ed. Washington, D.C.  874 p.

Asperger,  S.  1952.  Kinetics of  the Decomposition of Potassium Ferrocyanide
     in Ultra-Violet Light.  Trans. Faraday Soc.   48: 617-624.

Asperger,  S., I. Murati, and D. Pavlovic.   1960.   Kinetics and Mechanism of
     the Decomposition of Complex Cyanides of Iron (II) and Molybdenum  (IV).
     J. Chem. Soc. 730-736.

Balzani, V., and V. Carassiti.  1970.  Photochemistry of Coordination
     Compounds.  Academic Press,  New York,  N.Y.  432 p.

Bener, P.  1972.  Approximate Values of Intensity  of Natural Ultraviolet
     Radiation for Different Amounts of Atmospheric Ozone.  U.S. Army Report
     DAJA  37-68-C-1017, Davos Platz, Switzerland.

Broderius, S. J.  1973.  Determination of  Molecular Hydrocyanic Acid in Water
     and Studies of the Chemistry and Toxicity  to  Fish of Metal-Cyanide
     Complexes.  Ph.D. Thesis.  Oregon State  University, Coryallis, Oregon.
     287 p.

Broderius, S. J., and L. L. Smith,  Jr.  1977. Direct Determination and
     Calculation of Aqueous Hydrogen Sulfide. Anal. Chem.  49: 424-428.

Broderius, S. J., and L. L. Smith,  Jr.  1979. Lethal and Sublethal Effects of
     Binary Mixtures of Cyanide and Hexavalent Chromium, Zinc, or Ammonia  to
     the Fathead Minnow (Pimephales promelas) and  Rainbow Trout (Salmo
     gairdneri).  J. Fish. Res. Board Can. 36: 164-172.

Broderius, S. J., L. L. Smith, Jr., and D. T. Lind.  1977.  Relative Toxicity
     of Free Cyanide and Dissolved  Sulfide Forms to the Fathead Minnow
     (Pimephales promelas).  J. Fish. Res. Board Can.  34: 2323-2332.


                                      45

-------
Burdick, G. E., and M. Lipschuetz.  1950.  Toxicity  of  Ferro- and Ferricyanide
     Solutions to Fish, and Determination of the Cause  of Mortality.  Trans.
     Am. Fish. Soc.  78: 192-202.

Carassiti, V., and V. Balzani.  1960.  Photochemistry of Complex Salts  in
     Solution:  IV.  Kinetics of the Photochemical and  Thermal Decomposition
     of Potassium Ferrocyanide.  Annali Di Chimica (Rome).   50: 782-789.

Doudoroff, P.  1976.  Toxicity to Fish of Cyanides and  Related Compounds.  A
     Review.  EPA-600/3-76-033, U.S. Environmental Protection Agency, Duluth,
     Minn.  155 p.

Emschwiller, G., and J. Legros.  1954.  The Photolysis  of Potassium Ferro-
     cyanide in Aqueous Solution.  Comptes Rendus.   239: 1491-1493.

Hull, J. N.  1954.  Spectral Distribution of Radiation  from  Sun and Sky.
     Trans. Ilium. Eng. Soc. (London).  19: 21-28.

Hutchinson, G. E.  1957.  A Treatise on Limnology Vol.  I - Geography, Physics,
     and Chemistry.  Wiley and Sons, New York, N.Y.  1015 p.

Ibers, J. A., and N. Davidson.  1951.  On the Interaction Between Hexacyanato-
     ferrate (III) Ions and (a) Hexacyanatoferrate (II) or (b) Iron (III)
     Ions.  J. A. Chem. Soc.  73: 476-478.

Kobayashi, T., and K. Mori.  1973.  Formation of Compounds Toxic to Fish from
     Ferro- and Ferricyanides by Photodecomposition  (Occurrence of Toxicity to
     Fish by Photodecomposition of Complex Cyanides).   Nihon Mizu Shori
     Gijutsu Kenkyuhai, Mizu. Shori. Gijutsu.  14(6): 575-579.

Kongiel-Chablo, I.  1966.  Behavior of Complex Cyanides in Natural Water with
     High Rate of Contamination.  Roczniki Panstwowego  Zakladu Hig.  17(1):
     95-102.

Leighton, P. A.  1961.  Photochemistry of Air Pollution.  Academic Press, New
     York, N.Y.  300 p.

London, J.  1963.  The Distribution of Total Ozone in the Northern Hemisphere.
      Beitr. Phys.*Atmos.  36: 254-263.

Lure, Y. Y., and V. A. Panova.  1964.  Behavior of Cyanide Compounds in Water
     Ponds.  Gidrokhim Materialy.  37: 133-143.

MacDiarmid, A. G., and N. F. Hall.  1953.  Illumination - pH Effects in
     Solutions of Complex Cyanides.  J. Am. Chem. Soc.  75:  5204-5207.

Mitra, R. P., D. V. S. Jain, A. K. Bannerjee, and K. V. Raghavachari.  1963.
     Role of Free Radicals in the Photo-Oxidation of Fe"1"^ £n Acidic
     Solutions of Ferrocyanide.  Nature.  200: 163-164.
                                      46

-------
Moggi, L., F. Bolletta, V. Balzani, and F. Scandola.   1966.  Photochemistry of
     Coordination Compounds-XV Cyanide Complexes.  J.  Inorg. Nuclear Chem.
     28: 2589-2597.

Myers, C. S., and T-. lezzi.  1950.  A Report of Experiments on the Effects of
     Certain Toxicants on Fish Life with Especial Reference to the Effects of
     Photodecomposition.  Ind. Wastes Division, Bureau of  San. Engr., Penn.
     Dept. of Health, Harrisburg, Penn.  10 p.

Ohno, S.  1967.  The Photochemistry of Aqueous Hexacyanoferrate  (II)
     Solutions.  I.  Photo-Aquation Reaction at 3660 A.  Bull. Chem.
     Soc.  Japan.  40: 1765-1769.

Ohno, S., and G. Tsuchihashi.  1965.  The Photochemistry of Hexacyanoferrate
     (II) Ions in Aqueous Solutions.  Bull. Chem. Soc.  Japan.  38(6):
     1052-1053.

Palaty, J., and M. Horokova-Jakubu.  1959.  The Course and Rate  of Removal of
     Cyanides from Water Under Natural Conditions.  Sbornik Vscht, Prague,
     Oddil Fak. Tech. Paliv A Vody.  3(1): 83-110.

Robinson, N.  1966.  Solar Radiation.  Elsevier Publishing Company, New York,
     N.Y.  347 p.

Smith, L. L., Jr., S. J. Broderius, D. M. Oseid, G. L. Kimball,  W. M. Koenst,
     and D. T. Lind.  1979.  Acute and Chronic Toxicity of HCN to Fish and
     Invertebrates.  EPA-600/3-79-009, U.S. Environ. Prot. Agency, Duluth,
     Minn.  115 p.

Steel, R. G. D., and J. H. Torrie.  1960.  Principles  and  Procedures of
     Statistics with Special Reference to the Biological Sciences.
     McGraw-Hill Book Co., Inc.  New York, N.Y.  481 p.

Stumm, W., and J. J. Morgan.  1970.  Aquatic Chemistry. An Introduction
     Emphasizing Chemical Equilibria in Natural Waters. John Wiley and Sons,
     Inc.,  New York, N.Y.  583  p.

Terhaar, C. J., W. S. Ewell, S.  P. Dziuba, and D. W. Fassett.  1972.  Toxicity
     of Photographic Processing  Chemicals to Fish.  Photogr. Sci. Eng.  16:
     370-377.

Tyler, J. E., and R. W. Preisendorfer.   1962.  Transmission of Energy Within
     the Sea.  In:  The Sea, Vol. I, M. H. Hill, ed.   Interscience, New York,
     N.Y., pp. 397-451.

West, L. E.  1970.  Disposal of  Waste Effluents  from Motion-Picture Film
     Processing.  J. Soc. Motion Picture Tel. Eng.  79: 765-771.

Wetzel, R. G.  1975.  Limnology.  W. B.  Sanders  Co., Philadelphia, Penn.   743
     P«

Zepp, R. G., and D. M.  Cline.   1977.  Rates  of Direct  Photolysis in Aquatic
     Environment.  Environ.  Sci. Technol.   11(4): 359-366.
                                       47

-------
                                                    APPENDIX  A
                                           BEAM ATTENUATION COEFFICIENTS
oo
Water body
Turbidity (NTU)
Wave length (nm)
297.5
300.0
302.5
305.0
307.5
310.0
312.5
315
.0
317.5
320
323
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
480
.0
.1
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
Deionized
0.0028
0.0028
0.0026
0.0025
0.0024
0.0023
0.0022
0.0021
0.0020
0.0019
0.0018
0.00152
0.00122
0.00100
0.00082
0.00069
0.00056
0.00043
0.00035
0.00030
0.00026
0.00023
0.00020
0.00017
0.00016
0.00016
0.00016
Bentonite Solutions
5.6
0.118
0.114
0.113
0.110
0.108
0.107
0.105
0.103
0.102
0.101
0.099
0.096
0.093
0.090
0.088
0.085
0.082
0.079
0.076
0.072
0.070
0.067
0.065
0.061
0.060
0.057
0.056
6.7
0.122
0.121
0.119
0.116
0.114
0.112
0.110
0.109
0.108
0.107
0.106
0.102
0.098
0.096
0.095
0.092
0.089
0.086
0.082
0.080
0.076
0.073
0.070
0.068
0.066
0.064
0.061
9.9
0.221
0.216
0.213
0.208
0.203
0.201
0.199
0.195
0.192
0.189
0.187
o.iai
0.174
0.169
0.166
0.160
0.154
0.149
0.143
0.138
0.134
0.130
0.125
0.121
0.118
0.114
0.112
23.0
0.509
0.496
0.487
0.478
0.469
0.462
0.455
0.450
0.444
0.438
0.429
0.419
0.399
0.387
0.377
0.365
0.351
0.339
0.328
0.317
0.308
0.299
0.291
0.281
0.274
0.268
0.260
24.0
0.527
0.516
0.509
0.498
0.491
0.481
0.475
0.469
0.463
0.457
0.450
0.438
0.420
0.405
0.391
0.377
0.364
0.349
0.337
0.327
0.316
0.305
0.294
0.286
0.277
0.269
0.263
46.0
0.963
0.951
0.924
0.910
0.893
0.879
0.866
0.857
0.848
0.833
0.824
0.799
0.770
0.745
0.721
0.697
0.674
0.652
0.631
0.609
0.590
0.575
0.559
0.545
0.535
0.521
0.509
54.0
1.081
1.060
1.041
1.018
1.004
0.991
0.971
0.963
0.951
0.939
0.924
0.900
0.866
0.839
0.815
0.788
0.759
0.735
0.710
0.690
0.672
0.654
0.635
0.618
0.604
0.590
0.577
57.0
1.131
1.102
1.086
1.066
1.046
1.032
1.009
1.000
0.991
0.983
0.963
0.943
0.907
0.876
0.848
0.818
0.790
0.764
0.742
0.719
0.699
0.678
0.660
0.642
0.629
0.618
0.602
68.0
1.387
1.367
1.328
1.310
1.292
1.276
1.252
1.237
1.222
1.208
1.187
1.155
1.114
1.081
1.046
1.013
0.979
0.947
0.914
0.886
0.857
0.836
0.815
0.790
0.772
0.750
0.735

-------
         APPENDIX A
BEAM ATTENUATION COEFFICIENTS
Water body
Turbidity (NTU)
Wave length (nm)
297.5
300.0
302.5
305.0
307.5
310.0
312.5
315.0
317.5
320.0
323.1
330.0
340.0
350.0
360.0
370.0
380.0
390.0
400.0
410.0
420.0
430.0
440.0
450.0
460.0
470.0
480.0
Wind-blown silt
23.0
0.346
0.343
0.338
0.334
0.330
0.327
0.322
0.319
0.316
0.313
0.309
0.301
0.291
0.281
0.272
0.261
0.253
0.246
0.238
0.231
0.226
0.220
0.214
0.209
0.203
0.200
0.194
43.0
0.688
0.680
0.674
0.664
0.658
0.650
0.642
0.635
0.631
0.622
0.613
0.599
0.580
0.558
0.539
0.523
0.506
0.492
0.479
0.465
0.452
0.441
0.428
0.419
0.408
0.398
0.388
63.0
1.022
1.000
0.996
0.991
0.971
0.959
0.951
0.947
0.928
0.921
0.910
0.886
0.857
0.830
0.801
0.780
0.754
0.738
0.719
0.697
0.678
0.660
0.644
0.627
0.614
0.599
0.585
Square
Lake
0.43
0.023
0.022
0.021
0.020
0.018
0.018
0.017
0.016
0.016
0.015
0.014
0.013
0.012
0.009
0.008
0.007
0.005
0.005
0.004
0.004
0.004
0.003
0.003
0.003
0.003
0.002
0.002
Lake
Phalen
2.7
0.066
0.061
0.059
0.056
0.052
0.051
0.048
0.046
0.045
0.043
0.041
0.039
0.034
0.030
0.026
0.023
0.021
0.020
0.018
0.017
0.016
0.015
0.014
0.013
-0.013
0.013
0.013
Lake
Como
5.4
0.123
0.119
0.114
0.109
0.107
0.103
0.101
0.097
0.096
0.092
0.089
0.084
0.077
0.071
0.066
0.061
0.057
0.053
0.051
0.048
0.047
0.046
0.044
0.041
0.040
0.039
0.039
Elephant
Lake
2.9
0.171
0.164
0.156
0.149
0.143
0.137
0.131
0.126
0.121
0.117
0.111
0.102
0.087
0.074
0.064
0.056
0.047
0.041
0.036
0.032
0.028
0.025
0.022
0.019
0.018
0.016
0.015
Mississippi
River
3.6
0.206
0.197
0.190
0.182
0.176
0.170
0.166
0.159
0.154
0.149
0.143
0.131
0.115
0.101
0.088
0.078
0.070
0.063
0.057
0.052
0.048
0.045
0.041
0.039
0.036
0.035
0.032
Minnesota
River
14.0
0.283
0.275
0.268
0.260
0.251
0.243
0.237
0.230
0.224
0.218
0.210
0.198
0.178
0.161
0.147
0.134
0.125
0.114
0.108
0.101
0.096
0.091
0.086
0.083
0.081
0.077
0.075
Bog Water
0.9
0.548
0.527
0.509
0.491
0.470
0.455
0.438
0.420
0.407
0.395
0.372
0.337
0.292
0.248
0.208
0.177
0.148
0.125
0.107
0.090
0.076
0.066
0.057
0.049
0.043
0.038
0.034

-------
                                          TECHNICAL REPORT DATA
                                  (Please read Instructions on the reverse before completing)
1. REPORT NO.
  EPA-600/3-80-005
                                    2.
                                                                        3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
 Direct  Photolysis  of Hexacyanoferrate Complexes:
 Proposed Applications  to  the Aquatic  Environment
                5. REPORT DATE
                  January; 1980  issuing  date
                6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)


 Steven J. Broderius and Lloyd  L.  Smith, Jr.
                                                                        8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Department of Entomology,  Fisheries,  and  Wildlife
 University of Minnesota
 St.  Paul,  Minnesota   55108
                1O. PROGRAM ELEMENT NO.

                   1BA608
                11. CONTRACT/GRANT NO.


                   Grant No.  R805291
12. SPONSORING AGENCY NAME AND ADDRESS
 Environmental Research  Laboratory
 Office of Research  and  Development
 U.S.  Environmental  Protection Agency
 Duluth,  Minnesota   55804
                13. TYPE OF REPORT AND PERIOD COVERED
                14. SPONSORING AGENCY CODE
                       EPA/600/03
15. SUPPLEMENTARY NOTES
16. ABSTRACT
      The  theory and  computations described by Zepp and Cline (1977) were experimentally tested  in  predicting the
 direct  photolysis  rates of dilute hexacyanoferrate (II) and (III)  solutions in the aquatic environment.  Essential
 information for these  calculations includes the quantum yield for  the photoreaction, molar extinction coefficients
 of the  complex ions  for wavelengths > 295 nm, solar irradiance data used to calculate specific  sunlight absorption
 rates,  and the assumption that the photolysis reaction obeys a first-order kinetic rate expression.  Direct
 photolysis rates of  the irreversible photochemical reactions are calculated as a  function of the time of year,
 latitude, time of  day, meteorological conditions,  and depth in natural water bodies.  Light of  wavelengths < 480
 nm is active in the  photolysis reactions, and pH,  temperature, and concentration  all affect the reaction to
 varying degrees.  Assuming first-order kinetics, in which the rate constant was approximately concentration
 independent within the range of 25-100 Mg/1 total  cyanide, the minimum quantum yields of HCN formation were 0.14
 and 0.0023 for the iron (II) and (III) complexes,  respectively. These values correspond to minimum, nearsurface,
 midday  half-lives  at midsummer of about 18 and 64  min at St. Paul, Minn.  The photolysis rate at various fixed
 depths  in a natural  water column, when compared with that at the surface, decreases exponentially  with depth.  It
 is suggested that  the  photolysis reactions are enhanced by suspended material in  turbid waters  because of the
 forward scattering of  light when compared with that theoretically  calculated from beam attenuation coefficients.
 Hexacyanoferrate (II)  and (III) solutions of equal initial total cyanide concentration respond  photochemically
 quite differently from one another in solutions prepared with deionized water, but respond in a similar manner  for
 solutions prepared with natural waters.  The potentially rapid photodecomposition of iron-cyanides with formation
 of HCN  suggests that this phenomenon may be of toxicological importance under certain environmental conditions.
17.
                                      KEY WORDS AND DOCUMENT ANALYSIS
                     DESCRIPTORS
D.IDENTIFIERS/OPEN ENDED TERMS   C.  COSATI Field/Group
  Cyanides
  Photolysis
  Aquatic
  Environment
    06/F
18. DISTRIBUTION STATEMENT

 RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
   UNCLASSIFIED
21. NO. OF PAGES
      60
2O. SECURITY CLASS (This page)
   UNCLASSIFIED
                                 22. PRICE
EPA Form 2220-1 (Rev. 4-77)    PREVIOUS EDITION is OBSOLETE
                                                       50
                                                                                  a US GOVERNMENT PRINTING OFFICE: 1980 -657-146/5539

-------