USDA
 EPA
   United States
   Department of
   Agriculture
Northeast Watershed
Center
University Park PA 16802
   United States
   Environmental Protection
   Agency
Office of Environmental
Processes and Effects Research
Washington DC 20460
EPA-600/7-84-032
March 1984
            Research and Development
H
III
 - en
   Atmosphere and
   Temperature Within a
   Reclaimed Coal-
   Stripmine and a
   Numerical Simulation of
   Acid Mine Drainage from
   Stripmined Lands

   Interagency
—Energy/Environment
   R&D Program
   Report

-------
 ATMOSPHERE AND TEMPERATURE WITHIN A  RECLAIMED
  COAL-STRIPMINE AND A NUMERICAL SIMULATION  OF
    ACID MINE DRAINAGE FROM STRIPMINED LANDS
                      by
D. B. Jaynes, A. S. Rogowski, and H.  B.  Pionke
      U.S. Department of Agriculture, ARS
      Northeast Watershed Research  Center
      University Park, Pennsylvania 16802
                EPA-IAG-D5-763
                Project Officer

                Clinton W. Hall
    Office of Energy, Minerals and  Industry
           Washington, D. C.  20250
      Office of Research and Development
     U.S. Environmental Protection Agency
           Washington, D. C.  20250
                                U S. Environmental Protection Agency
                                Region 5, library CPU 2J)
                                77 West Jackson Boulevard, 12th Ftoqr
                                Chicago, IL  60604-3590

-------
                                                                     ii
                               DISCLAIMER






     This report has been reviewed by the Office of Energy, Minesoils and




Industry, 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.
                                   ii

-------
                                 FOREWORD






     The Federal Water Pollution Control Act Amendments of 1972, in part,




stress the control of nonpoint source pollution.   Sections 102 (C-l),  208




(b-2,F) and 304(e) authorize basin scale development of water quality




control plans and provide for area-wide waste treatment management.  The




act and the amendments include, when warranted, waters from agriculturally




and silviculturally related nonpoint sources, and requires the issuance of




guidelines for both identifying and evaluating the nature and extent of




nonpoint source pollutants and the methods to control these sources.




Research program at the Northeast Watershed Research Center contributes to




the aforementioned goals.  The major objectives of the Center are to:






     • Study the major hydrologic and water-quality associated




       problems of the Northeastern U.S. and




     • Initial emphasis is on land uses in the Northeast which




       most severely impact surface and subsurface hydrology




       and water quality.






     Within the context of the Center's objectives, stripmining for coal




ranks as a major and hydrologically severe land use.  In addition, often




the site is reclaimed and the conditions of the mining permit are met,




stripmined areas revert legally from point to nonpoint sources.  As a




result, the hydrologic, physical, and chemical behavior of the  reclaimed
                                    iii

-------
land needs to be understood directly and in terms of control practices

before the goals of Sections 102, 208 and 304 can be fully met.
                         Signed:
                         Harry B. Pionke
                         Director
                         Northeast Watershed
                           Research Center
                                   iv

-------
                            ABSTRACT







     Oxygen, 0_, carbon dioxide, CO 2 and temperature were measured




with depth along a transect of an acid, reclaimed, coal stripmine




over a two year period.  Spoil-atmosphere 0_ concentrations decrease




with depth but approach zero only in a small portion of the transect.




Most of the mine remains well oxygenated (0_ > 10% by volume) down




to 12-meters depth.  C0? concentrations ranged from near atmospheric




levels to greater than 15%.  At some locations, especially within 2




meters of the surface, variations in 0» and C0_ are correlated with




changes in the spoil temperature.  Spoil temperatures in layers




below 3 meters remain in a range conducive to iron-oxidizing




bacterial activity year around.  Flux ratios of C0_ and Q~ and the




source/sink rates of the two gases indicate that carbonate neutral-




ization of the acid produced by pyrite oxidation is the dominant




source of CO™.




     In a second phase of the study, a numerical model describing




the production and removal of acid and acid by-products from




reclaimed coal-stripmines is presented.  Both direct oxygen and




bacterially catalyzed pyrite oxidation is considered.  The pyrite




oxidation rate is assumed to be controlled by first-order, solid-




liquid kinetics and simple diffusion of oxidant into reactive,




coarse, stone fragments.  Oxygen supply into the reclaimed profile




is considered to be governed by one-dimensional, gas diffusion.
                                   v

-------
Activity of the iron-oxidizing bacteria is controlled by the ferrous-

ferric ratio, from which they obtain their energy, and from a combination

of potentially inhibiting factors—the oxygen and hydrogen concentrations

and the spoil temperature.  Under all conditions modeled, excess acid, in

the form of free hydrogen, must be removed if bacteria are to play an

important role in accelerating pyrite oxidation.  Leaching of the spoil

by normal precipitation is insufficient in removing the excess hydrogen.

Neutralization and buffering by the host rock is required for prolonged

bacterial activity—the degree of hydrogen removal determining the

maximum sustainable activity.  At optimum conditions, bacteria can greatly

increase pyrite oxidation.

     As an offshoot of the diffusion portion of the study a comparison of

Pick's Law for the diffusion of gas, i, in a vapor to the equations based

on the kinetic theory of  gases was made.  The comparison shows that only

for certain special conditions is the Fickian diffusion coefficient, D^  ,
                                                                      Fi
independent of the mole fraction of i and the diffusional fluxes of the

other gases.  These conditions include the diffusion of a trace concen-

tration of a gas through  a gas mixture of any composition and equi-molar

counter-current diffusion in a binary gas mixture, or in a  ternary

mixture where the third gas is stagnant.  In an 0  , CO., N« atmosphere,

variations in the diffusion coefficient of at least 10%  from the tracer

values are possible with  variations in the mole fraction or flux ratio.

The temperature and pressure dependence of the  diffusion coefficient must

also be recognized.
                                     VI

-------
                             CONCLUSIONS








      Oxygen, carbon dioxide and temperature measurements within a




 reclaimed coal-stripmine indicate that at this site, the spoil




 atmosphere is seldom depleted of oxygen  despite the continuing




 oxidation of pyrite within the spoil."  Increases in carbon dioxide




 concentrations were correlated with decreases in oxygen concentra-




 tions at most depths of each site.  Oxygen and carbon dioxide con-




 centrations-were significantly correlated with spoil temperature




 at  several locations but in general the correlation was weak.




 Average spoil temperatures to at least 60 cm deep are, in general,




 higher than temperatures found in a natural soil.  Temperatures in




 the spoil below 300 cm remain within the tolerance range of iron-




 oxidizing bacteria year round.



     We have presented a comprehensive model describing the in situ




 oxidation of pyrite and the removal of the oxidation products from




 reclaimed coal—stripmines-  The  rate of pyrite  oxidation is assumed




 to be controlled by both the reaction  kinetics  and  the diffusion




 rate of the products and reactants  to  the reaction  site, where both




 oxygen and ferric iron may serve as the oxidant.  Ferric iron' con-




 centrations are controlled by iron  complexation and precipitation




reactions which are assumed to be rapid.  The only  important source




of ferric iron is oxidation of ferrous iron  by  iron-oxidating
                                 vii

-------
bacteria.  This model is unique in that the "activity" of the bacteria,




which determines the rate of ferric iron production, is calculated from



the energy available from the energy substrate and the deviation of the



bacterial environment from ideal growing conditions.  Pyrite oxidation



and bacterial "activity" are linked through their modification of a



shared environment.



     Simulations, using the model, indicate that the lack of oxygen is




the primary rate controlling factor in both pyrite oxidation and atuo-



trophic bacterial "activity."  In zones where oxygen is not limiting,



autotrophic activity can greatly increase the rate of pyrite oxidation.



Whether or not bacteria are important in these zones, depends on the



solution pH.  The oxidation rate of pyrite will be affected by



bacterially produced ferric iron only if the solution pH can be



maintained in the range between reduced bacterial efficiency (pH > 2.0)




and reduced ferric iron solubility (pH < 3.0).  The interaction between



hydrogen, produced by pyrite oxidation, and the rock matrix appears to



be crucial in establishing the pH of the spoil solution.



     Extensive testing of the model was not possible at this time,



although its behavior seemed consistent with our current knowledge



of acid  formation in spoils.  Further refinement of the model is



necessary.  The model is very sensitive to changes in the gas dif-




fusion properties of the spoil.  Although this property has been


                                                    2
measured for spoils in the laboratory (Colvin, 1977) , field-based



measurements, capable of incorporating the great heterogeneity of



reclaimed spoil, would be desirable.  The values of E'. , x_, x TT
                                                     min   T   pH


and x_  , although based on experimental data, represent only best





estimates.  Laboratory experiments designed specifically to





                                viii

-------
determine these values and the form of their mathematical relationship




would help establish the validity of the autotrophic "activity" model




and the magnitude of the model parameters.




     In view of the extreme sensitivity of the autotrophic "activity" to




solution pH, the interaction of rock matrices with acid needs further




study.  These interactions are complex and varied mainly due to the wide




range of minerals and reactions possible and the slow reaction kinetics




associated with most interactions.  It may be possible to develop a




simple analytical technique, similar to that used to measure the acid




potential of spoil (Caruccio, 1968), that combines all of the




neutralization reactions into a simple empirical representation.




Carbonate neutralization was considered separately in this model because




of the great potential these materials represent for acid consumption.




Further research on the reaction kinetics of carbonates, incorporating




the effects and the development of iron coatings and diffusion control




of reactants needs to be pursued.
                                 IX

-------

-------
                                CONTENTS

                                                                    Page

Foreword ............................     iii
Abstract . . . .........................       v
Conclusions ...........................     vii
Figures .............................     xii
Tables .............................     xvi
Notation for Section 2 .....................   xviii
Notation for Section 3 .....................     xix
Notation for Section 4 .....................      xx
Notation for Appendix A .....................     xxv
Notation for Appendix B .....................   xxvii

     1.  Introduction ......................       1
              References ....................       4

     2.  Spoil Atmosphere and Temperature in a Reclaimed
         Coal Stripmine .....................       6
              Introduction ...................       6
              Materials and Methods ...............       8
              Results ......................      16
              Discussion ....................      31
              References ....................      36

     3.  Flux and Production Rates of Oxygen and Carbon
         Dioxide in a Reclaimed Coal-Stripmine .........      39
              Introduction ...................      39
              Theory ......................      42
              Materials and Methods ...............      44
              Results and Discussion ..............      48
              Conclusions ....................      55
              References ....................      58

     4.  A Numerical Model of Acid Drainage from Reclaimed
         Coal-Stripmines ....................      60
              Introduction ...................      60
              Description of Model ...............      62
                   Basic Reactions ...............      62
                   Basic Equations ...............      63
                   Oxygen Diffusion ...............      69
                   Chemical Species ...............       83
                        Ferric Precipitation  ..........       84
                        Ferric Complexes  ............       85

-------
                                                                    Page

                   Removal of Reaction Products	 .      86
                        Acid Neutralization. . .  .	      86
                        Leaching	      90
                   Solution Method 	      93
              Results and Discussion 	      97
                   Role of Iron Oxidizing Bacteria	      97
                   Model Sensitivity and Versatility 	     112
                        Effective Diffusion Coefficient	     112
                        Pyrite Distribution	     119
                   Inhibition of Bacteria	     123
                   Carbon Dioxide Generation 	     125
              References	     129
Appendices
     A.  Applicability of Pick's Law to Gas Diffusion	     135
     B.  Parameter Values Used in the Pyrite Oxidation
         Model .	     160
     C.  Computer Listing and Example Output of Acid Mine
         Drainage Model	     173
                                   XI

-------
                        LIST OF FIGURES
Figure                                                        PaSe

   1     Location of monitoring sites 1-6 and the depth of
         the gas sample-chambers (crossbars) at each site
         in relation to the strip-mine stratigraphy ..... .    11

   2     Schematic of gas sample-chambers design ......  .  12

   3     Schematic of CL measuring-chamber used in the field    14

   4     Variation of temperature { C) over time for six
         depths at site 2.   Depths shown are (A) 30cm,
         60 cm, (o) 170 cm,  (o)  320 cm,  (A) 625 cm and  (•)
         945 cm .  ......................    20

   5     Variation of temperature ( C) and the mole fractions
         of 0_ and C02 over  time for the composite record at
         site  1.   The temperature plot is shown as a line
         connecting the individual data points for clarity .    22

   6     Variation of temperature ( C) and the mole fractions
         of Q£ and C02 over  time for the composite records  at
         site 2.  The temperature plot is shown as a line
         connecting the individual data points for clarity .    23

   7     Variation of temperature ( C) and the mole fractions
         of ©2 and C02 over  time for the composite record at
         site 3.  The temperature plot is shown as a line
         connecting the individual data points for clarity .    24

   8     Variation of temperature ( C) and the mole fractions
         of 02 and C02 over  time for the composite record at
         site 4.  The temperature plot is shown as a line
         connecting the individual data points for clarity  .    25

   9     Variation of temperature ( C) and the mole fractions
         of 02 and C02 over  time for the composite record at
         site 5.  The temperature plot is shown as a line
         connecting the individual data points for clarity  .    26

  10     Variation of temperature ( C) and the mole fractions
         of 02 and C02 over  time for the composite record at
         site 6.  The temperature plot is shown as a line
         connecting the individual data points for clarity  .    27
                                    XII

-------
List of Figures (Continued)
Figure                                                        Page

  11     Schematic of spoil profile showing the division of
         the profile into layers as determined by the depths
         of the gas concentration measurements, Y   and Y _ .
         Fluxes, Nn  and N . ,  are calculated for 2        2
                  °2      C°2
         the top of each layer and assumed to' be 0.0 at the
         profile bottom.  Source terms,  Q.  and Q . , are
         calculated for each layer. .  .  .  2 •  .  •  • 2  •  • •  •    ^'

  12     Cross section of a coarse fragment containing
         pyrite ........................    64
  13     Plots of the inhibition factors (solid lines) .
         a)  XT- versus termperature (°C) ; dashed line is
         normalized data from Malouf and Prater, 1961;
         (o) normalized data from Silverman and Lundgren,
         1959; (A)normalized data from Belly and Brock,
         1974.  b) Xpg versus pg; (») normalized data from
         Silverman and Lundgre, 1959; (A) normalized data
         from Schnaitman, et al . , 1969.   c) XQ  versus
         oxygen mole fraction ..... .  ...  2 .......    '°

  14     Neutralization curves as calculated from equation
         in text.  Gg equals 1.0 in both plots.  G^ equals
         2.5 in plot 1 ( •) and 2.8 in plot 2 (• ) ......    91

  15     Example leaching curve shows the normalized con-
         centration versus water— filled pore volumes.
         Profile was 10 meters deep, divided into 20 layers
         with a water-filled pore volume of 0.13 and an
         infiltration rate of 50 cm/year ...........    94

  16     Fraction of pyrite oxidized within entire profile
         versus time for Run 1 and Run 2 ...........    98

  17     The rate of leaching out of the profile of all  iron
         species versus time for Run 1 and Run 2 .......
                                xiii

-------
List of Figures (Continued)


Figure                                                        Page
  18     Oxidation rate of pyrite for each layer after 5
         years of  oxidation.   Fine cross-hatching is total
         oxidation rate for Run 1.  Open cross-hatching is
       .  rate of oxidation by direct osygen reaction with
         pyrite in Run 2.   While cross-hatching on black
         is oxidation rate due to ferric iron reaction
         with pyrite  in Run 2.   Below 1 meter, ferric
         oxidation is negligible.  Below 4 meters, oxida-
         tion rate of Run  1 and Run 2 are identical .....    102

  19     The ratio of ferric  iron to ferrous iron in the
         water leaving the profile for Run 1 and Run 2.. . .    103

  20     Fraction  of  pyrite oxidized within entire
         profile versus time  for Runs 2,  3 and 4 ......    105

  21     The rate  of  leaching out of the profile of all iron
         species versus time  for Runs 2,  3 and 4 ......    107

  22     The rate  of  leaching out of the profile of the
         potential and active acid for Runs 3  and 4 .....    110

  23     Fraction  of  pyrite oxidized within entire profile
         versus time  for Run  3  and Run 5  ..........    HI

  24     Fraction  of  pyrite oxidized within entire profile
         versus time  for Runs 3,  6 and 7  ..........
 25     -The rate of leaching out of  the profile of all
        iron species for Runs 3, 6 and 7 ..........     116

 26     The mole fraction of oxygen within the. profile
        versus depth (m) after 5 years for Runs 1 and 6  .  .     117

 27     The mole fraction of oxygen within the profile
        versus depth (m) after 5 years for Runs 3 and 7  .  .     118

 28     The distribution of pyrite (wt./wt.) with depth  (m)
        for a) Run 3; b) Run 8; c) Run 9 ..........     120
                               xiv

-------
List of Figures (Continued)


Figure                                                     Page
  29     Fraction of pyrite oxidized within entire profile
         versus  time for Runs 3,  8 and 9	121

  30     The mole fraction of oxygen within the- profile
         versus  depth (m)  after 10 years for Runs 8 and 9.  122

  31     The rate of leaching out of the profile of all
         iron species for  Run 7 (no inhibition)  and Run
         10  (complete inhibition  after 2 years)	124

  Al     Values  of the reduced Fickian diffusion coeffici-
         ent,  Dp ,  in a binary gas mixture plotted against
         the gas^ mole fraction of component i.   Curve
         labels  are the values of the flux ratio, r.., for
         which the curve was calculated	  143

  A2     Values  of the Fickian diffusion coefficient,  D ,
         for 0?  in an O^-CO^-U- atmosphere (in m /sec) .
         The N_  mole fraction is  held constant at 0.790,
         while the 02 mole fraction,  plotted on  the
         abscissa,  is varied between 0.0- to 0.21.  C07
         comprises the balance of atmosphere.  Values
         are for 20°C and  101 KPa with the N,,  flux being
         set to  0.0.  Curve labels are the values of the
         flux ratio r _ ,0,,.   Point represents value cal-
         culated in    2    text	149

  A3     Values  for the Fickian diffusion coefficient, D_,
         for C02 in an 02~C02-N2  atmosphere (in  m^/sec).
         The N2  mole fraction is  held constant at 0.790,
         while the 02 mole fraction,  plotted on  the
         abscissa,  is varied between 0.0 and 0.21.   C0~
         comprises the remainder  of the atmosphere.
         Values  are for 20°C and  101  KPa with  the N2
         flux  being set for 0.0.   Curve labels are the
         values  of the flux ratio r^- ,  0-•  Point repre-
         sents value calculated in   2 text	150
                                xv

-------
                         LIST OF TABLES

Table                                                         Page

  1      Mean and range of selected properties of the spoil
         and drainage waters  from the reclaimed stripmine
         used in this study	      9
         Average,  maximum and minimum values for 0~ and C0_

         concentrations and temperature.for each depth of
         sites 1-6	     17

         Correlation,  r,  and regression coefficients,  a and
         b,  between the mole fractions of CO-  and decreased
         02  (0.210-YQ  ) for each depth at sites 1-6	     29

         Correlation coefficients,  r,  between  the mole frac-
         tion  of CO- and temperature  between the mole
         fraction  of C02 and temperature for each depth at
         sites 1-6.  ,,,,.,,,.,,,,,,,	     30

         Pyrite and  sulfate concentrations  expressed as
         percent by  weight  sulfur for  ten representative
         spoil layers  from  research site (Rogowski,  1977).
         Also,  the C0_ equivalent of  the  carbonate content
         required  to neutralize  the acid  produced from
         columns 2 and 3  and the measured carbonate content
         expressed as  C0_ (columns  4 and  5).   The last column
         gives the ratio  CO- measured/C02 required  in  percent.   41

         Means and coefficients  of  variation,  C.V.,  for 0~
         and C0£ fluxes at  the surface of the  six sites  and
         for all sites combined.  The  ratio  of  the  fluxes,
         r  =  N _  /Nn  , was  calculated from  the  mean values. .   49
         "     OUn  U_
         Mean  values and  coefficients  of  variation,  C.V., of
         the source  terms for D£  and C02  and for the ratio
         rq =  QQQ  at  the individual depths  at  each  site.
         The correlation  between  Q  n   and Q  ,  r,  is also
         listed	.2.  .  . U.2	   52

         Mean  values and  coefficients  of  variation,  C.V., for
         the production rate of 0»  and CO- and  for  their
         ratio, r_ = QCQ  /QQ .  Rates  are for the entire spoil

         profile at  each  site averaged  over  time	    56
                                    xvi

-------
List of Tables (Continued)


Table                                                         Page

  9      Coefficients used to calculate typical values of
         t  and t, . . .	    68
          c      d

 10      Measured IL, values and calculated X_, Xn , X „ and
                   M                        T   0_   pH

         E .   values from phase II of experiment run by
          mm         4
         Bailey (1968)*	    81

 11      Measured and calculated coefficients for the
         bacterial "activity" model from phase I of experi-
         ment run by Bailey (1968)^	    82

 12      Iron complexes and log of the equilibrium constants
         for  their formation and for sulfuric acid dissoci-
         ation and amorphous ferric hydroxide precipitation.    87

 13      Ratio of the flux of CO  to 0^ for Run 12 (hetero-
         trophs active in layer I only;, Run 13 (hetero-
         trophs active in layer 20 only)  and Run 14 (acid
         neutralization by carbonates in layer 20) 	   128
                                       *
 Al      Coefficients for calculating D..  for gas pairs of
         interest in research (after Marrero and Mason,
         1972).  Coefficients were fit to  data over the
         temperature range'shown.  The uncertainty in the
         calculated values is for the lower end of the
         temperature range listed	   154

 Bl      K  values for Fe   and 09 systems	   164
          5                      ^
                                   xvii

-------
                    NOTATION FOR PART II





a,b            empirical constants,



a              activity of H-CO, moles

 H2C03

pCO-           partial pressure at CO-,  atmosphere (101 kPa)



r              correlation coefficient,  -



T              temperature, K



T_n            temperature at 30 cm  depth,  K



X              mean value, (variable)



Yn?            mole fraction of oxygen,  -



Y -            mole fraction of carbon dioxide,  -
 co2
                                xviii

-------
                    NOTATION FOR PART  III
C.V.           coefficient of  variation,  -


                                                               2

D              Fickian diffusion coefficients for component,  cm /sec

  i



DZ             weighing factor,  cm


                                                 -2     -1
N.             flux rate of component  i,  moles-cm -sec



P              total pressure,  kPa



Q.             source term for component  i, moles-cm   -sec



r              flux ratio, r  = N n  /N  ,  -
 n                          n     oU-   u_


r              source ratio, r  * Q  n  /Qn ,  -
 q                            q   CU2  U2



R              gas  constant, kPa-cm  -mole  -k



t              time



T              temperature, K



X              mean



Y.             mole fraction of component i



Z              depth, cm



T              tortuosity, -


                                               3   3
$              air-filled porosity of  spoil,  cm /cm
                                 xix

-------
                   NOTATION FOR PART IV


A              fragment surface area per volume spoil water,  cm

b              stoichiometric ratio between pyrite and oxidant
               consumption, -

C ff           normalized concentration of solute, in effluent, -

C.             normalized concentration of -solute in layer i, -

C              normalized concentration of solute in influent, -
 in

C              oxident concentration at the fragment surface,
 ox            moles /cm-*

d.             distance between layers, cm

D              effective diffusion coefficient for coupled oxidant-
 c          '   product counter diffusion, cm^/sec

Dn             effective diffusion coefficient for 0- in an Q~,  N~
  2            and C0» atmosphere, cm^/sec

D              diffusion coefficient at 0~ water, moles-cm -sec

*Vi  rn  ^n     binary diffusion coefficients for the gas pairs 0»
 °2'  2'  2'    and C00, 0~ and N0  and C00 and N0, respectively
   _             o,z/      /        z      /
   D _         cm^/sec
E              electromotive force,  volt

E              energy per unit time  available  to  bacterial
               population,  W

E              energy per unit time  available  to  bacterial popu
               lation for growth,  W

E              energy per unit time  available  to  bacterial popu
               lation for maintenance,  W

E        ,      energy per mole of  electrons  oxidized available
               to bacterial population,  measured  as a driving
               force, volts
                                xx

-------
Notation for Part Iv  (Continued)

 '             minimum energy per mole of electrons oxidized re-
 min           quired to sustain bacterial population, measured
               as a driving force, volts

F              the Faraday, C

  3+
Fe             total iron concentration (all species), moles/L

FFR            fraction of coarse fragments in spoil, g/g

FPY            fraction of pyrite in coarse fragments, g/g

G ,G           empirical constants, -

I              ionic strength,  moles/kg

K              first-order surface reaction rate constant for pyrite
               oxidation per unit surface area of pyrite, cm/sec

K              equilibrium constant for chemical reactions,  -

K_             first-order reaction rate coefficient for bacterial
               oxidation of ferrous iron,  sec

K              first-order surface reactioiurate constant for
               carbonate dissolution, L-cm  -sec"

K              solubility constant for amorphous iron hydroxide, -
 sp
IC.,K2          reaction coefficients_for chemical oxidation  of
               ferrous iron, mole -L  -sec"1 and sec~l,
               respectively

H              one-half thickness of fragment,  cm

MW             molecular weight of pyrite, g/mole

Nn , N _ ,Nn    gas flux of oxygen, carbon dioxide and nitrogen,
  2    22   respectively, moles-cm~2-sec~l

P              total pressure,  kPa

Q              oxygen source term, due to iron-oxidizing bacteria,
               moles-cm~3_sec-l
                                xxi

-------
 Notation for Part IV (Continued)
 Q              rate of  carbonate consumption per surface area of
                carbonate,  moles-cm~^-sec~^

 Q              oxygen source term due to chemical oxidation of
                ferrous  iron, moles-cm~3-sec~^-

 Q              oxygen source term due to heterotrophic respira-
                tion,  moles-cm~3-sec~l

 Q              base rate of  oxygen consumption from heterotrophic
  H KB                   .       _      	o    _1
                respiration,  moles-cm -'-sec -1-

 Q              total  oxygen  source term,  moles-cm  -sec

 0              oxygen source term due to direct pyrite oxidation
  *^            by  oxygen,  moles-cm~3_sec~l

'r ,r2          ratios of the fluxes of CO ,  0_ and N2  to 02,
                respectively, -

 R              gas constant, cm   -KPa-mole  -K

 R              rate at  which bacterial population oxidizes  iron,
                moles  of electrons/sec

 R              radius of carbonate particles,  cm
  D

 t              time,  sec

 t_             time required for complete pyrite oxidation  if
                chemical reactions are the rate controlling
                process, sec

 t              total  time  required for complete pyrite oxidation
                if  diffusion  processes are the  rate limiting
                steps, sec

 T              temperature,  K

 x.              general  inhibition factor,  -

 x ,x TT,X_       inhibition  factors dependent  on temperature, pH,
     ^    2      and oxygen  concentration,  respectively,  -

 x              fraction of pyrite remaining  in a fragment, —
                                 xxii

-------
Notation for Part IV  (Continued)
x              fraction of carbonate particle remaining,  -
 U


Y              mole fraction of oxygen in spoil atmosphere,  -



z              d ep th,  cm



z.           .  charge  of the 9th chemical species,  -



a              surface area of pyrite per unit volume fragment,  cm




T..,T2,Y_       activity coefficients for  H ,  Fe   and Fe   ,

               respectively



8              effective thickness  of pyrite  within which pyrite

               is  oxidized



A£",            thickness of water film covering fragments, cm



AG             energy  released in iron oxidation,  J



AG             change  in standard free energy, J
  K.


AH             actual  increase of H  in solution, moles



AH             H   produced through  chemical reactions, moles



II              product operator



p              bulk density of spoil,  g/cm
 B


p_             molar density of carbonate in  carbonate spheres,

               moles/cm^
p             molar  density  of  oxygen  in  the vapour phase,

  2           moles/cm
P             molar  density  of  pyrite  in  fragments, moles/cm



I.             summation operator



T             tortuosity of  gas diffusion path,  -
                                xxiii

-------
Notation for Part IV (Continued)
<|>              air-filled porosity, cm  /cm
 A.

                                        3    3
              water-filled porosity cm /cm
 w


[]             concentration operator,  moles/L
                                 xxiv

-------

-------
                                                                XXV
                   NOTATION FOR APPENDIX A
a,b,c          semi-empirical constants



C.             concentration of gas component k, kg m



v.,v.          average velocity of gas i and j, respectively,  m sec



D              Fickian diffusion coefficient for component i,

  i            m^ sec~l


                                              2    -1
D..            binary diffusion coefficient, m  sec



D. .            temperature dependent, pressure independent binary

  ^            diffusion coefficient, kPa m^
                                                  2    -1
D              coefficient of thermal diffusion, m  sec



F.,F.          external force field acting on gas i and j,  N/Kg



m.,m.          mass of molecules of gas i and j, respectively,  Kg



M              Avagadrops number,  mole -1
 o


n.,n.          number density of gas i and j, respectively,  m



nm             = n.-hn.
 T                i  j

                                           -2    -1
N.             molar flux of gas i, moles m   sec



P              total pressure,  KPa


                                       _2    —]_
q.             mass flux of gas i, kg m   sec



r              direction vector, m



R              universal gas constant,  m  KPa mole   K



r..            flux ratio,  - N.N.
 Ji                           Ji


t              time, sec



T              absolute temperature,  K



Y.             mole fraction of component i
                                 xxv

-------
z              distance or depth, m

a              combined air-filled porosity,  tortuosity term

II              product operator

p.»P.»PT       density of gas i, j and the combined gas density
    ^          respectively,  Kg m~3                            '

£              summation operator
                                 xxvi

-------
                    NOTATION FOR APPENDIX B





b              shorthand notation, see text


                                      3

C              concentration, moles/cm

                                              3

C.,C-          integration constants,  moles/cm



C              concentration at the leached rim - oxidation  zone

               interface, moles/cm^



C, C1,  C-      shorthand notation, see text



C.             same as C_ ,  moles/cm
 jL                      &

                                                              2
D              diffusion coefficient in the spoil fragment,  cm /sec
 c


FFR            fraction of spoil mass  composed of coarse  fragments, -



FPY            fraction of fragment mass that  is pyrite,  -



K              first order reaction coefficient for  pyrite oxidation

 s             per  unit area, cm/sec



£              one-half fragment thickness,  cm



£'             thickness at  leached rim,  cm


                                                              -3    -1
R              reaction rate per fragment volume,  dx, moles-cm -sec



R              total reactionrate per unit surface  area  of  fragment,

               moles-cm  -sec



X              thickness,  cm



a              surface area  of pyrite  per unit volumn of  fragment, cm

                                          3

p              bulk density  of spoil,  g/cm
 B


p              molar density of pyrite in fragment,  moles/cm
 py
                                               3  3
4>A             air-filled porosity in  spoil, cm /cm


                                                 3  3
              water-filled  porosity in spoil,  cm /cm
 w


T              tortuosity of diffusion path is spoil, -
                                xxvii

-------

-------
                                SECTION 1






                              INTRODUCTION






     The United States has the most extensive deposits of coal in




the world.  The rate at which we mine this coal must accelerate in




the next century if we are to satisfy our increasing energy appetite




and decrease our reliance on foreign sources of energy.  Much of




this coal will be mined by surface operations, such as stripmining.




Stripmining will disturb millions of acres of land; land that must be




restored if it is to become productive again.  Unfortunately,




reclaimed stripmines in the humid eastern United States have a




tremendous potential for both acute short-term and significant long-




term pollution of surface streams and lakes and subsurface aquifers




(Collier et al., 1970).  Water leaving many reclaimed stripmines




contains acid, iron, sulfates and other constituents at levels far




exceeding their concentration in natural waters.  Agnew and Corbett




(1973) found a large increase in the acidity of water draining from




stripmined lands, especially during "flush outs" (large stream flows)




caused by periods of heavy rain.  Striffler (1973) found similar




results for eastern Kentucky stripmine waters and concluded that




primary watersheds, those directly draining stripmined areas, have a




serious problem.  Caruccio (1973), among others, has measured acid




water draining from stripmine sites to have pH's as low as 2.8.  At




such low pH's, Warner (1973)  found a severe decrease in stream




invertebrates and algae populations and Kinney (1964) documented the




adverse effects on fish and other vertebrates.  Even though acid

-------
pollution of streams caused a decrease in the stream biota, it does not




directly cause any visual degradation.  Iron pollution, on the other




hand, can be readily observed when it precipitates in surface waters to




color them from red to yellow-orange.  These stairiings, commonly




referred to as "yellow-boy," can greatly impact the scenic value of




streams in the Appalachian coal region.




     Fortunately, the amounts of pollutants leaving a mined area can be




considerably reduced if proper reclamation techniques are used (Grimm




and Hill, 1974).  However, we can intelligently develop these proper




reclamation techniques or best management practices only if we under-




stand the basic physical and chemical processes that lead to acid




drainage.  Researchers have been intensively investigating acid drainage




problems for more than 50 years.  A synopsis of their finding can be




found in many articles including review papers by The Ohio State




University Research Foundation (1971) and Rogowski et al.  (1977).  A




crucial area of the acid drainage problem in reclaimed stripmines that




has not received sufficient attention is the oxygen status of the mine




environment (Ohio State University Research Foundation, 1971).  The




first portion of this research was to investigate the oxygen status of




a  reclaimed stripmine.  The results  from this part of the  study are




presented in Part II.  Part II expands upon the oxygen concentration




findings and presents calculations of the oxygen and carbon dioxide




gas  fluxes and  the  related  rates of  consumption and production of these




gases within  the reclaimed  mine spoil.

-------
     Another productive avenue for developing best reclamation techniques




is to develop and use mathematical models that describe the acid drainage




process.  A model, based on basic chemical and physical processes, can




serve as a useful tool for identifying and gaining insight into the




important, rate controlling processes of acid drainage.  A model that




accurately predicts the behavior of actual stripmines can be used to




develop alternative reclamation techniques where the consequences of each




modification can be tested rapidly and inexpensively.  Part IV presents




the development of a general model for acid formation and leaching from




reclaimed coal stripmines.




     Included in the appendices is a review of the process of gas diffu-




sion in the vapor phase and a listing of the magnitude and derivation of




some of the parameters used in the model for acid drainage.  Also listed




in the appendices is a listing of the FORTRAN program of the model and




example output for the program.

-------

-------
                              REFERENCES




1.  Agnew, A. F. and D. M. Corbett.  Hydrology of a Watershed Containing




    Flood-Control Reservoirs and Coal Surface-Mining Activity,




    Southwestern Indiana.  In:  Ecology and Reclamation of Devastated




    Land, R. J. Hutnik and G. Davis, eds.  Gordon and Breach, New York,




    1973.  pp. 159-174.




2.  Caruccio, F. T.  Characterization of Strip-Mine Drainage by Pyrite




    Grain Size and Chemical Quality of Existing Groundwater.  In:




    Ecology and Reclamation of Devastated Land, R. J. Hutnik and G.




    Davis, eds.  Gordon and Breach, New York, 1973.  pp. 193-226.




3.  Collier, W. R., R. J. Pickering, and J. J. Musser.  Influences of




    Strip Mining on the Hydrologic Environment of Parts of Beaver Creek




    Basin, Kentucky, 1955-56.  U.S. Geological Survey Professional




    Paper 427-C, 1970.




4.  Grimm, E. C. and R. D. Hill.  Environmental Protection in Surface




    Mining of Coal.  EPA-670/2-74/093, Environmental Protection




    Technology Series, U.S. Environmental Protection Agency,




    Cincinnati, Ohio,  1974.  pp. 1-277.




5.  Kinney, E. C.  Extent of Acid Mine Pollution in the United States




    Affecting Fish and Wildlife.  U.S. Department of the Interior,




    Bureau of Sport Fisheries and Wildlife Circular 191, 1964.




6.  Ohio  State University Research Foundation.  Acid Mine Drainage




    Formation and Abatement.  Water Pollution Control Research Series




    Program 14010  FPR, U.S. Environmental Protection Agency,




    Washington, D.C.

-------
7.   Rogowski,  A.  S. ,  H.  B.  Pionke, and J. G. Broyan.  Modeling the Impact




    of Strip Mining and Reclamation Processes on Quality and Quantity of




    Water in Mined Areas:  A Review.  J. Environ. Qual., 6:237-244, 1977.




8.   Striffler, W. D.   Surface Mining Disturbance and Water Quality in




    Eastern Kentucky.  In:   Ecology and Reclamation of Devastated Land,




    R. J. Hutnik and G.  Davis, eds.  Gordon and Breach, New York, 1973.




    pp. 175-192.




9.   Warner, R. W.  Acid Coal Mine Drainage Effects on Aquatic Life.  In:




    Ecology and Reclamation of Devastated Land, R. J. Hutnik and G.




    Davis, eds.  Gordon and Breach, New York, 1973.  pp. 227-238.

-------
                            SECTION 2
             SPOIL ATMOSPHERE AND TEMPERATURE  IN A
                   RECLAIMED COAL STRIPMINE
                          Introduction


     Within the past two decades, the problems caused by acid-

drainage from reclaimed coal stripmines in the eastern United States

have been the focus of numerous studies.  These studies have shown

the primary source of acid to be the oxidation of  the iron sulfide

mineral, pyrite.  Two mechanisms for pyrite oxidation have been

demonstrated (Ohio State University Research Foundation, 1971) .  The

first involves oxygen reacting directly with pyrite and water and

is summarized in Eq . [1] .
               + 3.5 02 + H20 + Fe   + 2S0" + 2H              [1]


     The second mechanism, shown in Eq. [2], uses ferric iron  as the

electron acceptor.

          14Fe3+ + FeS2 + 8H2
-------
 1968).  Thus Eq.  [2] is considered  to  be  bacterially  catalyzed

 within acid stripmine spoils because the  ferric  iron  is  produced  by

 Eq.  [3]:
                             bacteria
      14Fe   + 14H  + 3.5 02 	>  14Fe   + 7H20              [3]


 Summing Eqs. [2] and [3] gives Eq.  [1], thus regardless  of mechanism,

 the result of pyrite oxidation is the  same; two  moles of H  are

 produced for every three and one-half  moles of oxygen consumed..

 Moreover, significant oxidation of  pyrite cannot take place far from

 the presence of gaseous 02 because  of  the slow diffusion rates of Q~

 in water (Ohio State University Research Foundation, 1971).  Thus,

 the presence of oxygen in the spoil atmosphere is critical for in

 situ oxidation of pyrite.

     The temperature of the spoil environment is also critical in

 determining the oxidation rate of pyrite.  The rate of direct pyrite

 oxidation by 0» and thus 0? consumption increases with temperature

 (Clark,  1965).   Autotrophic bacteria are also sensitive  to temperature

 showing  maximum activity at temperatures near 30 C and almost no

 activity for temperatures  below 4 C and above 55 C (Malouf and

 Prater,  1961;  Ehlrich'and  Fox," 1967; Cathles and Ap'ps, 1975).

     Although the mechanics of pyrite oxidation  and leaching have

been studied for years,  almost no information is available on the

 environmental status of  reclaimed stripmines (Ohio State University

Research Foundation,  1971).   As part of a comprehensive  investigation

of the chemical and physical processes of a reclaimed stripmine, we

measured temperatures and  spoil atmosphere compositions  over an ex-

 tended period of time and  for the entire depth of the mine site.

-------
                         Materials and Methods






     Oxygen, carbon dioxide and temperature profiles were measured




within a reclaimed bituminous coal stripmine.  The sparsely grassed




mine-site, located in Clearfield County, west-central Pennsylvania,




has been described in detail elsewhere (Pedersen et al., 1978;




Pedersen et al., 1980; Pionke and Rogowski, 1980).  The mine was




developed in Pennsylvanian-age deposits.  Two coal seams were mined




at the site, the middle Kittanning (C) coal seam and the lower




Kittanning  (B) seam that underlies the C.  Noncalcareous, thin bedded




sandstones and shales overlay the middle Kittanning seam while




predominantly silt stones and shales overlay the lower seam.  High




pyrite contents are usually associated with the shale layers  (Rogowski,




1977; Lovell et al., 1978).  Typical values of some of the important




properties of the mine spoil and drainage waters are shown in Table 1.




All the parameters vary greatly, but the water draining the site has




been consistently low in pH and high in acidity indicating that pyrite




oxidation and leaching are taking place.




     In the first 10 months of the study, the spoil atmosphere composi-




tion was monitored at four locations along a transect of the mine




(sites 1-4, Fig. 1).  Site 1 was located in an area between two strip-




mines composed of 440 cm of spoil overlying unmined, coal-bearing




strata.  Site 2 was placed in 1250 cm of spoil from Kittanning B-seam




and sites 3 and 4 were placed in 700 and 1060 cm of spoil from the




Kittanning C-seam.  In general, the shale and pyrite content of the




spoil decrease from sites 1-4, while the vegetation cover increases




from virtually 0.0 to 3.2 metric tons/ha (Pionke and Rogowski, 1980).

-------
Table 1.   Mean and range for selected properties of the spoil and
          drainage waters from the reclaimed stripmine used in
          this study'.'

Spoil property x
% pyritic S 0.18
(st/wt)
bulk density 1,560
(kg/m3)
% coarse fragments 78
(wt/wt)
% carbon 4.4
(wt/wt)
pH 4.9
(1:1 spoil-water
extract)
acidity 16.4
(meq/100 g)
Drain-water property x
PH 2.8
acidity 15 . 0
(meq/L)
range
0.02
640-1,790
54-93
0.4-22.8
3.7-6.0
4.0-24.0
range
2.3-3.7
9.0-19.0

 'Data from Pedersen,  et al., 1978; Pedersen, et al., 1980;
 Rogowski, 1977;  Rogowski, et al., 1982; and Rogowski, A. S.,
 personal communication.

-------
     At each site,  a 15 cm (6 in)  diameter bore-hole was drilled with

a standard well-drilling rig down to or slightly deeper than the

deepest extent of the stripmining process; a depth of between 870 and

1250 cm.  Five or six gas-sampling chambers were placed at selected

depths between 150 cm of the surface and the bottom of each bore-hole

(Fig. 1).  After location of the chambers, the bore-holes were

backfilled a few centimeters at a time; each layer being packed with

a metal rod.  Clean, coarse, silica sand was placed around each chamber

with a 10-cm thick bentonite berm placed above the sand.  The remaining

space between samplers was filled with spoil material that had been

screened to remove stone fragments.  Two additional holes were drilled

at sites 1-4 with a Giddings soil auger .  The holes were 5 cm in

diameter, 30 and 50 cm deep.  Each hole was fitted with a gas sampler

and backfilled as described above.

     After the first 10 months of monitoring, two additional sites

were added.  Site 5 was located close to the buried highwall of

the Kittanning B-seam.  Site 6 was placed in an adjacent area

overlying the unmined Kittanning B seam where the top 30 cm were

composed of disturbed spoil material.

     Design of the gas samplers is shown in Figure 2.  The samplers

consisted of an 8.25-cm chamber of either 2.5- or 5-cm  (1- or 2-in)

diameter pvc pipe.  A series of 0.48-cm  (3/15-in) holes were drilled
 The mention of  trade names does not constitute an endoresement of
 the product by  the U.S. Department of Agriculture over other
 products not mentioned.
                                  10

-------
   340
                                 EZ2 Unmlned-Dliturbed
                                 CD B-Stripped
                                    C-Strlpped, B-0«ep Mined
           Reclaimed  Spoil\
   300
Figure 1.   Location of monitoring sites 1-6  and the depth  of the gas sample-chambers
            (crossbars) at each site in relation to the stripmine stratigraphy.

-------
                                  Nylon Tubing


                                Wire Lead
                                  to Meter
                \	1.1	r'
— Plastic Fastener

   Rubber Stopper


 —5.1cm OD PVC
      (2 in)
                          /Rubber-backed Tape


                            -PVC Coupling



                            -Rubber Stopper

                           -0.48cm (3/16") Holes

                           —Thermocouple
                           -Nylon Screen

                          —PVC End Cap
Figure 2.  Schematic of gas sample-chambers  design.
                           12

-------
in each sample-chamber and were screened with a medium-mesh, nylon




screen.  The chambers were coupled to lengths of pvc pipe so that




they could be placed at the desired depth.  The sample-chambers were




connected to the surface by a 0.64-cm (0.25-in) diameter nylon tube.




At the surface, this tube was capped with a rubber septum except




when gas samples were withdrawn.  Each sample-chamber was also




equipped with a thermocouple so spoil temperatures at each site could




be measured.  The thermocouple consisted of a 0.082-cm (0.032-in)




diameter, copper/constantin, low-temperature element (Omega Engin-




eering, Type T) silver soldered to 24-gauge, copper/constantin,




unshielded wire (Omega Engineering, Type T Poly Rip Control).




Soldered connections were protected by ceramic insulators.  Each




thermocouple was calibrated before installation.  Temperature was




measured directly using a Leeds and Northrop detector (Model 934




Numatron Digital Thermocouple Recorder,  Type T, battery operated).




An ice-water bath was used as a reference temperature when measure-




ments were made.




     Oxygen, carbon  dioxide and temperature were measured at each




sampler periodically (approximately every month). -Initially, oxygen




and carbon dioxide concentrations were measured in the field




independently of each other.  A gas sample was obtained by connect-




ing the gas chamber access tube to a rubber bulb equipped with one-




way valves.  The bulb was used to pump air out of the gas samplers.




For oxygen measurements,  gas was pumped into a measuring chamber




(Fig. 3) equipped with a  dissolved oxygen probe (Lazar Research Labs,
                                 13

-------
f
5
<
c
a
Temperature
Probe -j
Rubber | V
Stopper^ II
1
j
3
/
i
^



1

! ' '/
l i \L
i i y -

/ ' '
J
n '
1 \:
                              To
                               Millivolt Meter
                                       ,Dissolved-Oxygen  Probe
                                            SAMPLE CHAMBER
                       I  I
7.5cm 00  PVC
     (3*)
                                                           0.6cm Plastic
                                                               Tubing
0


-
>
o


— Holes
	 GAS TRAP
- nil

                                                             25 ml Vial
                                                    Tubing
                     From Pump
Figure  3.   Schematic of C>2 measuring-chamber used  in the field.
                                  14

-------
 Inc.,  Model  00-166 Dissolved  Oxygen Probe).   Sufficient  gas was




 pumped from  the gas  sampler to  ensure a representative sample in the




 measuring chamber as indicated  by steady 0_  readings.  The required




 gas volume was approximately  1.6  liters,  which  corresponded to




 removing gas from a  9.4-liter spoil volume  (1.5 g/cm  bulk density,




 70% saturation) or a 26-cm diameter sphere centered at each probe.




 Oxygen levels were read directly  from a millivoltmeter (Orion




 Research, Model 407A), which  had  been calibrated for 0_  concentrations




 in air (21% 0_) and  bottled N_  gas  (0% 0-).   C02 concentrations  were




 measured by pumping  a gas sample  directly into  a Lab-Line  CO-  Analyzer




 (Lab-Line, Inc.) which gave the CO 2 levels directly in percent.




     For the past twelve months of  the study, CO- and 0,., concentra-




 tions  were measured  in the laboratory.  Field measurements  were  dis-




 continued because the technique did not allow for measurements in the




 winter.  For laboratory measurements,  gas samples were collected in




 air tight bottles (150 ml volume), returned to the lab and measured




 with a gas chromatograph (Varian  Aerograph,  Model 1820) as  described




 by Bollag and Barabasz (1979).  The laboratory  and field techniques




 were correlated with each other and  found to be in good agreement




 for both 0-  and C0~ measurements.




     Statistical analysis was  performed using standard procedures.




 Duncan's new multiple-range test, adjusted for  unequal sample size,




was used to  compare  the means  of  0~  and C0_  for each layer at each




 site (Steel  and Torrie, 1960,  p.  114).  Homogeneity of regression




 equations was tested as described by  Steel and Torrie (1960, p. 319).
                                 15

-------
                             Results




     Average, maximum and minimum values for O-* CO 2 and temperature




for each depth at the six sites are shown in Table 2.  The average




mole fraction of 09 ranged from 0.010 to 0.207, while the range in




individual values was between atmospheric levels (0.210) to below




the detection limit (< 0.001).  Considerable variation in measured




values was found for most layers.  Several layers near the surface




had variations in 0- levels of less than 0.02 but most layers showed




variations of greater than 0.05 mole fraction.  In general, the




average values decreased with depth at any one site, as did the values




measured at any one time.  At sites 1, 2 and 3, the 0_ concentrations




showed a tendency to decrease with depth until a minimum was reached




and then start to increase again.  This pattern was consistent over




time.   Site 1 showed the extreme decrease of 02 with depth, where




average 07 values dropped to 0.01 within 410 cm of the surface.  The




low 0_ levels in the deeper layers of sites 1 and 6 were distinctly




different from the other values as indicated by Duncan's new




multiple-range test.  Other than at sites 1 and 6, however, the oxygen




levels were consistently high indicating that most of the reclaimed




mine site was well oxygenated.




     Oxygen and carbon dioxide concentrations in the surface layers




were in general not very different from atmospheric levels.  One




reason for this may have been the contamination of gas samples by air




from the soil surface.  The gas samplers used in this study were




primarily designed for obtaining samples at deeper depths.  The large
                                  16

-------
Table 2.  Average, maximum and minimum values  for  0_  and  CO,
          depth at sites 1-6.                       2        2
                                                                    concentrations and temperature for each
Site 1
Site 2
Site 3
Depth
cm
30
60
105
260
410
565
30
60
170
320
625
945
1250
30
60
150
305
455
610
810

X


.205
.206
.147
.022
.010
.016
.190
.194
.178
.171
.143
.152
.156
.207
.203
.191
.194
.183
.155
.188




at
a
g
j
j
j
abed
abc
cd
de
g
fg
efg
a
a
abc
abc
bed
efg
abed
°2
max


.210
.210
.210
,.082
.066
.115
.210
.207
.210
.209
.187
.199
.204
.210
.210
.207
.209
.205
. .194
.208
co2
min
ole fracl
.170
.190
.090
n.d.
n.d.
n.d.
.143
.154
.129
.126
.103
.090
.121
.185
.195
.169
.149
.140
.116
.153
X


.0045
.0046
.040
.132
.144
.147
.011
.010
.017
.023
.048
.040
.039
.0022
.046
.013
.012
.013
.030
.012



a
a
ef
i
ij
j
ab
ab
be
cd
fg
ef
ef
a
a
abc
ab
abc
de
ab
max

-
.025
.020
.104
.168
.187
.187
.034
.033
.045
.052
.082
.082
.060
,009
.014
.024
.027
.033
.055
.033
min


n.d.T
n.d.
.002
.062
.104
.088
n.d.
.001
.001
.002
.027
.013
.006
n.d.
n.d.
.004
.002
.002
.011
.004
Temperature
X


14.2
16.0
13.9
12.3
12.2
10.8
14.5
16.6
12.6
11.6
11.8
12.2
12.3
13.9
13.8
12.6
12.2
11.4
11.3
11.5
max
o..
	 C —
25.2
24.0
24.5
17.5
14.8
13.4
23.7
24.5
20.4
16.1
13.6
12.9
13.3
25.2
23.5
21.2
18.7
15.3
13.2
12.4
min


-2.6C
1.28
-2.5
3.0
5.3
7.6
-2.3
-1.8
1.5
6.6
10.4
11.2
11.2
-1.4
-1.0
0.8
4.4
7.0
9.5
9.9

-------
Sire 4





Site 5





Site 6

30
60
185
335
490
790
30
145
325
630
935
1240
150
305
.201 a
.179 cd
.194 abc
.199 ab
.192 abc
.189 abed
.197 abc
.193 abc
.186 abed
.186 abed
.179 cd
.170 df
.114 h
.046 i
.210
.207
.204
.209
.206
.208
.210
.210
.204
.208
.207
.190
.156
.141
.160
.097
.177
.181
.175
.141
.169
.152
.131
.155
.134
.139
.040
n.d.
.0032
.012
.010
.008
.015
.018
.005
.011
.014
.015
.022
.032
.056
.098
a
ab
ab
ab
abc
be
ab
ab
abc
abc
bed
de
8
h
.018
.024
.022
.019
.028
.050
.019
.033
.036
.034
.044
.048
.103
.136
n.d.
.002
.002
n.d.
.002
.001
n.d.
.001
.004
.002
.004
.020
.015
.059
14.0
14.2
11.9
12.6
12.5
12.2
10.0
10.2
10.8
11.5
11.6
11.8
9.9
11.1
23.4
23.4
20.6
18.6
16.4
13.7
25.0
19,7
15.4
13.0
12.6
12.4
19.7
14.7
-0.9
-0.1
3.1
6.1
8.3
10.4
-3.0
1.2
6.2
10.1
10.5
10.6
0.0
6.1

•I-
 Numbers followed by the same letter in a column are not significantly different at p = 0.05.




 Below detected limits (<  .001).




§Data during the winter is missing.

-------
volumes of air required to obtain a sample may have permitted short-




circuiting of surface air through large pores to the sample chambers




located near the surface.




     Table 2 also shows the mean and extreme values of CO- for each




site and layer.  Average CO- mole fractions ranged from 0.009 to




0.147.  Individual measurements ranged from less than 0.001 to 0.187.




High CO- concentrations (< 0.10) were observed only at sites 1 and 6.




The high average values of CO- at those two sites and for the 625 cm




layer of site 2 were .tested to be distinctly different than the other




values.  Site 4 consistently had the lowest CO- concentrations with




no average value over 0.02 mole fraction for any layer.  The surface




layers at every site normally showed the lowest CO- concentrations




with the concentrations increasing with depth.  As in the case of




0-, sites 2 and 3 showed a tendency to reach maximum CO,, concentra-




tions at an intermediate depth and then start to decrease with depth.




     Mean and extreme values for temperature are also shown in




Table 2.  Mean values for sites 5 and 6 are considerably lower than




sites 1-4 because temperatures were measured over a period including




two winters and only one summer, while the data for sites 1-4 span




a period including two summers and winters.  As expected, spoil




temperatures showed an annual cyclic fluctuation, with the amplitude




of the fluctuation decreasing and the phase lagging as depth increased.




This behavior is shown in Figure 4, where the temperature is plotted




for five depths at site 2 for a portion of the data record.




Temperature showed little variation at depths below 500 cm.
                                 19

-------
                        30
                      O
                        20
Ni
O
                      W
                      H
                      «<: 10
                      cs
                      W
                      (X.
                      S
                      W  0
                      H
                        -10
                             n   i   i    i   i    i   i    i   i    i    I   i    i   i    i   i    i   i    i    i   r
                              i   i    i   i    i   i    i   i    i   i    i   i    i   i    i   i   i    i   i    i   i
                            J   A  S   O  N   D
                                   1980
J   FMAMJ   JA   BOND
                                                                   1981
J   F   M   A
                                              -1982-
                                                               DATE, MO8.
            Figure  4.   Variation of temperature  (  C)  over time for six  depths at site 2.  Depths shown are

                       30 cm, (D) 60 cm,  (•) 170 cm,  (o)  320 cm, (A) 625  cm,  and (•) 945 cm.

-------
     Although the mean values shown in Table 2 give a good indication




of the spoil atmosphere characteristics, the oxygen and carbon




dioxide levels also showed distinct seasonal tendencies.  These




tendencies are best shown in Figures 5-10 where 0« and C0_ mole




fractions and temperature are plotted against time for each layer at




the six sites.  Data spanning three years are superimposed on a




single Julian year.  In Figures 5-10 temperature is shown as a con-




tinuous line connecting the individual data points for clarity.




Much of the discontinuity in the temperature plot is due to the




superimposition of the three years of data.  Figures 5-10 illus-




trates that, at least for some depths at each site, there is a strong




tendency for the C0» levels to increase during the summer months




(150-273 days), while remaining low during the remainder of the year.




The increase in C0~ is accompanied by a decrease in 0~ concentrations.




This behavior is especially well illustrated at the 105 cm depth of




site 1 (Fig. 5) and all layers of site 2 (Fig. 6).  Near the surface,




the increase in CO- and decreases in 0- coincide with the maximum




in the spoil temperatures.  However, at the lower depths these ex-




tremes in CO- and 0? still occur during the summer while the tempera-




ture either experiences a maximum later in the year (320 cm, site 2,




Fig. 5) or exhibits almost no variation over time (625 cm and deeper,




site 2, Fig. 5).




     A relationship between 02 and C0» concentrations and between 0-




and temperature and C0_ and temperature is suggested by Figures
                                  21

-------
   O
   <
   
-------
Z
o
o:
u.

LU
_J
o
     d -
     d -
             J	1	1	1	1	1     '	I      i     I     I
                                               CM

              . . . a   '   "• "   «
             4D CM
                            .8)   « . B
                . . .1
                                 n a
            320 CM
         "    f
         ft«   I t t
             i • ' • i	r
       0.    tfl.   122.
             n	nr
                  JL!	L.
                                         a«   I
                                                                 go
                                               CM
                                                                       - d
                            12PO CM
                                                         0° Og.
                                           	1	1	1	

                                           61.    122.   172.
                                                                         -   O
Ul
cc
o
H-

(T
UJ
CL
                                                        - d
                                                                            UJ
                                                                            I-
                                                                30?.
                  30?.


                  JULIAN
                                         DATE
   Figure  6.
Variation of temperature and the  mole fractions of  CL

(X) and  CO^ (0) over  time for the composite record  at

site 2.   The temperature plot is  shown as a solid line

connecting the individual points  for clarity.
                                    23

-------
o


-------
        
-------
     d -
     d .
 »-   d -
o
                 I	I
     d -
                                       £30 CM
                                                               . d
                                                               . d
                                                        >  o
                                      12-tfl CM
                                               000
                                                                d  —
                                                               - d
- d




 d

 i
                                                                   <
                                                                   cr
                                                                   Q.
                                                                   s
                                                                   LLJ
                                                               - d
         I	jj	«_,	,	,	,	«_|	,	,	,	,	1	

       0.   (51.  122.  155.  2-M.  3DS.   166.  6\.   122.  153.   2-M.  3D?.  364. '


                             JULIAN  DATE




Figure  9.   Variation  of  temperature and the mole fractions of  £)„


            (X) and  CO, (0) over  time for the composite record  at


            site  5.  The  temperature plot is shown as a solid  line


            connecting the individual points for  clarity.
                             26

-------
cr
LU-

LL)
_j
o
              a -
                      IPO CM
                     3DP CM
                                                      - =5
                                                       a
                                                       Ml
                                                       •CM
                                                            O
                                                            LU
                                                            tr
                                                            Z)
a:
LL)
a.
S
UJ
I-
                           JUL1AN  DATE
Figure 10.  Variation of temperature  and  the  mole  fractions of 0_

            (X) and C0_  (0) over  time for the composite record at

            site 6.  The temperature  plot is  shown as a solid line

            connecting the individual points  for clarity.
                              27

-------
5-10.  To test the relationship between 02 and C02, regression  and



correlation coefficients were calculated for the relation:



                    YCQ  = a + b  (0.210 - YQ )                   [4]




where Ypr.  and Y_  are the mole fraction of C09 and 0_, respectively
       C02      U2                            2.      /


and a and b are regression coefficients.  Table 3 shows the values



for a, b and r, the correlation coefficient, for 0- and C02 at  the



different layers of each site.  Correlations between Y__  and the



term 0.210 - Yn  range between -0.29 and 0.97, although the smallest



absolute value for r, significant at the 5% level, is -0.51.  Values



for b range from 0.19 to 0.88 and for a. from -0.014 to 0.074.



Although the degree of correlation is high in most cases, the frac-



tion of the variation in Yrn  that can be attributed to variation in


                     2      2
Yn  (represented by r ,  Steele and Torrie, 1960, p. 187) ranges from



0.26 to 0.94 when the r  value is significant at the 5% level.



     To test the relationship between temperature and Yn  and Y _ ,

                                                       °2        2

correlation coefficients were calculated for temperature, T,and Y_ ,



T and Y n ,  and also between temperature at the 30 cm depth for



sites 1-5 and Yn  and Y^,.. .   These r values are listed in Table 4.

               °2      C°2

Correlations between Y-   and T are in general rather poor with only

                       2

5 values having an absolute value greater than 0.75 (excluding site



6).   Correlations between Y__  and T are even poorer with only 6



layers having an absolute correlation greater than 0.75 (excluding



site 6).   Correlations between T and Yrn  and Yn  are much stronger
                                      co2      o2


for  most depths when the temperature at the 30 cm depth of each site,



T-,njis used.  Thirteen layers have an absolute correlation greater
                                 28

-------
    Table 3.  Correlation, r, and fegression coefficients,  a and b,  between the mole fractions
              of C0_ and decreased 0  (.210-Y  )  for  each depth at sites l-6f.
K)


Site 1





Site 2






Site 3






Depth
(cm)
30
60
105
260
410
565
30
60
170
320
625
945
1250
30
60
150
305
455
610
810
a

.0029
.0016
.0030
.050
.074
.028
.0018
.0016
.0015
.0004
.0192
.0141
.0075
.0013
.0016
.0033
.0050
.0025
.0098
.0026
b

.38
.88
.59
.43
.35
.62
.48
.58
.50
.60
.43
.46
.59
.34
.51
.51
.44
.39
.32
.42
r

.51* Site 4
.68**
.87**
.57**
.29
.59**
.10** Site 5
.97**
.91**
.96**
.77**
.81**
.93**
.74** Site 6
.72**
.85**
.81**
.77**
.80**
.84**
Depth
(cm)
30
60
185
335
490
790
30
145
325
630
935
1240

150
305





a

.0009
.0061
.0021
.0016
.0026
.0057
.0014
.0013
.0052
.0029
.0081
.014

-.0145
.028





b

.27
.19
.55
.61
.70
.58
.31
.56
.39
.52
.47
.46

.74
.43





r

.76**
.69**
.87**
.91**
.93**
.82**
.72**
.98**
.89**
.95**
.78**
.83**

.90**
.89**





Significant at the 5% level.
Significant at the 1% level.
     - a + b (.210-Y- ).
     t.

-------
Table 4.  Correlation coefficients, r, between the mole fraction of 0,
          and temperature and between the mole fraction of CO- and
          temperature for each depth at sites 1-6.


Site 1





Site 2






Site 3






Site 4





Site 5





Site 6

Depth
(cm)
30
60
105
260
410
565
30
60
170
320
625
995
1250
30
60
150
305
455
610
810
30
60
185
335
490
790
30
145
325
630
935
1240
150
305
02-T

.40
-.27
-.92**
.11
.13
.08
-.72**
-.53*
-.79**
-.54**
.34
.28
-.30
.40
-.31
-.70**
-.49*
-.56**
-.13
.16
.50*
.36
.56**
-.09
.11
-.26
-.80**
-.80**
-.30
.77**
.55
.13
-.83**
-.82**
C02-T

.05
.56**
.84**
-.06
.08
.03
.74**
.57*
.85**
.60**
.12
-.06
.39
.01
.38
.74**
.80**
.74**
.19
-.19
-.30
.03
-.58**
-.02
-.21
.44*
.88**
.84**
.53
-.73**
-.62*
-.41
.72**
.80**
VT30

—
-.28
-.91**
.34
.02
.34
—
-.53*
-.84**
-.91**
-.90**
-.80**
-.93**
—
-.32
-.66**
-.42
-.77**
-.78**
-.51
—
.34
.58**
.25
.58**
-.13
—
-.86**
-.78**
-.87**
-.78**
-.82**


•W

—
.53*
.84**
-.16
.10
-.13
—
.55*
.87**
.92**
.70**
.68**
.88**
—
.40
.68**
.70**
.56**
.65**
.56*
—
.06
-.50*
-.29
-.58**
.20
—
.88**
.82**
.91**
:57
.92**



*Significant at the 5% level
**Significant at the 1% level
^Correlation coefficients between oxygen and tenperature; carbon di-
oxide and temperature; oxygen and the temperature at the 30 cm depth,
T  •  and carbon dioxide and T,_, respectively.

                                 30

-------
than 0.75 between Y   and T,n, while eight layers have absolute

values greater than 0.75 for Y _  and T,n.  The number of signifi-
                              OU ~      jU
cant correlations at the 5% level increases from 12 to 17 for T__

and Yn  and from 15 to 19 for T,n and Yr_ .  For layers where the
     U-                        JU        9
correlation is significant at the 5% level, the correlations are

positive between Y _  and temperature and negative between Y   and
                  C°2                                       °2
temperature.  Two layers at site 4 exhibit the reverse behavior but

the correlations for these two layers are fairly weak (1r|  < 0.6)'.
                            Discussion


     Oxygen concentrations remained fairly high for sites 2-5 during

the entire period of this study.   This is a fairly surprising result

considering the abundance of pyrite within the spoil at this site

and is contrary to earlier calculations and observations of in situ

pyrite oxidation in reclaimed mines and refuse dumps where only
                                                                  2
limited penetration of 02 was calculated or measured (Colvin, 1977 ;

Good,  1970 ).   The increase in 0- concentrations with depth after

reaching a minimum value at several sites indicates that gaseous ex-

change ibetween the spoil and atmosphere may also be accompanied by

exchange between the spoil and a  deeper layer, perhaps the water

table  or more  likely the zone of  deep-mined, B coal (Fig. 1).

Carbon dioxide showed the same behavior with depth.
 Colvin,  S.  L.   1977.   Oxygen diffusion in strip-mine soils.  Un-
 published M.S.  Thesis, Iowa State Univ., Ames, Iowa.
3
 Good,  D.  M.   1970.   The relation of refuse pile hydrology to acid
 production.   Unpublished M.S. Thesis, The Ohio State Univ., The Ohio
 State  Univ.,  Columbus, Ohio.
                                 31

-------
     Only at sites 1 and 6 did Y   approach 0.0.  At both of these sites,



trace levels of methane were detected during spring measurements when



Y   was below detection levels indicating that anaerobic conditions



existed at least during part of the year.  Low CL levels at site 6 can



be attributed in part to slower diffusion rates than at the other sites



since this site was located in undisturbed shale-overburden.  Low CL



concentrations at site 1 must be due to higher consumption rates than



at sites 2-5 with perhaps impeded diffusion at the lowest depth where



again undisturbed material was encountered.  The low 0  concentrations



at sites 1 and 6 were accompanied by very large increases in CO .  We



have not found any other reports of such high CO. values in soil or



spoil materials.  At all sites, the decrease in 0_ is strongly



correlated to the increase in CO .  This behavior suggests that



respiration of plant roots or soil organisms is responsible for the



changes in 0  and CO .  The high carbon content of these spoils



(Table 1) may support considerable activity by heterotrophic bacteria.



     Another possible source of C0_ is neutralization of acid by



carbonate minerals.  The acid produced by pyrite oxidation will



react with the gangue material (Rogowski et al., 1971).  If the



gangue contains carbonates or other salts of weak acids, they will



react to neutralize the acid.  For carbonate reactions in acid



stripmines (pH < 4), carbonic acid will be the dominant species



present since the first pK  for H CO  is 6.4 (Garrels and Christ,
                          3.      Z  j


1965, p. 76).  The  reaction for any carbonate species is:
                                   32

-------
                 M - C03  + 2H+ •+ M2+ + H2C°3   .                 [5]
                        s                    aq

Carbonic acid will in turn be in equilibrium with gaseous" C0_, the

relation being governed by Henry's Law (Garrels and Christ, 1965,

P, 76).

                            "                                    C6]
                  •*"*•')  O                *»  '

where:

            pCO- is the partial pressure of C0_  (atmospheres)



            a      is the activity of carbonic acid in solution,

                 where by convention, all C0_    is treated as
                                             aq
                 H2CO_   (see Kern, 1960 for further discussion
                      aq
                 on this relationship).

     Lovell, et al. (1978) have observed almost  complete carbonate

neutralization of acid in stripmine spoils and data presented by

Rogowski, et al. (1982) indicate that some carbonate neutralization

is taking place at this site.  Where carbonates  are present in

stripmine spoils, appreciable CO,, can be generated over much of the

active lifetime of a reclaimed stripmine.

     The correlation between 0. and T and C0_ and T is fairly weak.

This is similar to the result found by DeJong (1981) between C02 and

T in native and cultivated soils.  The stronger  correlation between

the temperature at 30 cm and the gas concentrations in some of the

deeper layers, particularly for sites 2, 3 and 5, indicates that

processes taking place near the surface are the  dominant processes
                                  33

-------
controlling the 02 and CO. concentrations in these deeper layers.


We should expect this to be true only if gas diffusion through the


spoil was the dominant process of gas movement and if consumption


and production processes for 0^ and CO- in the deeper layers were


considerably less than in the surface layers.


     The mean values for temperature at the 30 and 60 cm depth for


sites 1-4 were 14.2 and 15.2 C, respectively.  These values are


about five degrees higher than those measured by Carter and Ciolkosz


(1980) for a natural soil at approximately the same elevation and


location as the research site.  They found average temperatures of


about 9.2°C at 25 cm and 9.3°C at 50 cm.  Part of this difference is


due to a bias in the data used to calculate the means, since more


measurements were taken during the summer months.  However, even the


time-weighted average for mean temperature, 10.4 C at  30 cm and


11.0°C at 60 cm, is 1.2 to 1.7°C higher than the natural soil.  These


higher temperatures are probably due to the differences in vegetative


cover, allowing greater solar heating of the surface spoil materials


(Deely and Borden, 1978).  The natural soil was located under a


closed forest canopy while the mine spoil had no tree cover and only


a sparse covering of grasses and herbaceous plants (Table 1).  The


spoil showed a greater temperature variation in the top 30 cm layer


than the natural soil studies by Carter and Ciolkosz (1980), with


the temperature dropping below 4 C in the top 300 cm for a portion of

                               o
the year.  Temperatures below 4 C should inhibit autotrophic bacteria


activity if the autotrophs are present in these layers (Ehlrich and
                                 34

-------
Fox, 1967).   The deeper soil stayed within the temperature tolerance




range of iron-oxidizing bacteria, though apparently 10-20 C below the




temperature of maximum activity (Table 2, Silverman and Lundegren,




1959).
                                   35

-------
                              REFERENCES




1.  Bollag, J. M. and W. Barabasz.  Effects of Heavy Metals on the




    Denitrification Process in Soil.  J. Environ. Qual., 8:196-201,




    1979.




2.  Carter, B. J. and E. J. Ciolkosz.  Soil Temperature Regimes of




    the Central Appalachians.  Soil Sci. Soc. Am. J., 44:1052-1058,




    1980.




3.  Cathles, L. M. and J. A. Apps.  A Model of the Dump Leaching




    Process that Incorporates Oxygen Balance, Heat Balance and Air




    Convection.  Metall. Trans., 63:617-624, 1975.




4.  Clark, C. S.  The Oxidation of Coal Mine Pyrite.  Ph.D. Thesis,




    John Hopkins University, University Microfilms International,




    Ann Arbor, Michigan, 1965.




5.  Deeley, D. J. and F. Y. Borden.  High Surface Temperatures on




    Strip-Mine Spoils.  In:  Ecology and Reclamation of Devastated




    Land, R. J. Hutnik and G. Davis, eds.  Gordon and Breach, New




    York, 1973.  pp. 69-79.




6.  DeJong, E.  Soil Aeration as Affected by Slope Position and




    Vegetation Cover.  Soil Sci., 131:34-43, 1981.




7.  Dugan, P. R. and C. I. Randies.  The Microbial Flora of Acid Mine




    Water and its Relationship to Formation and  Removal of Acid.




    Project Completion Report A-002-2410, to Office of Water Resour.




    Rec., U.S. Department of the Interior, 1968.




8.  Ehlrich, H. L. and S. I. Fox.  Environmental Effects on Bacterial




    Copper Extraction from Lower Grade Copper Sulfide Ores.  Biotech.




    Bioeng., 9:471-485, 1967.
                                  36

-------
 9.  Garrells, R. M. and C. L. Christ.  Solutions, Minerals and




     Equilibria.  Freeman, Cooper and Company, San Francisco,




     California, 1965.




10.  Kern, D. B.  The Hydration of Carbon Dioxide.  J. Chem. Ed., 37:




     14-23, 1960.




11.  Lovell, H. L., R. R. Parizek, D. Forsberg, M. Martin, D. Richardson,




     and J. Thompson.  Environmental Control Survey of Selected




     Pennsylvania Strip Mining Sites.  Final Report to Agronne National




     Laboratory, Contract No. 31-109-38-3497, 1978.




12.  Malouf, E. E. and J. D. Prater.  Role of Bacteria in  the Alteration




     of Sulfide Minerals.  J. Metals., 1961:353-356, 1961.




13.  Ohio State University Research Foundation.  Acid Mine Drainage




     Formation and Abatement.  Water Pollution Control Research




     Series Program 14010 FPR, U.S. Environmental Protection Agency,




     Washington, D.C., 1971.




15.  Pedersen, T. A., A.  S. Rogowski, and R. Pennock, Jr.  Comparison of




     Morphological and Chemical Characteristics of Some  Soils and




     Minesoils.  Reclam.  Rev., 1:143-156, 1978.




16.  Pedersen, T, A., A.  S. Rogowski, and R. Pennock, Jr.  Physical




     Characteristics  of  Some Minesoils.  Soil  Sci. Soc.  Am.  J.,




     44:321-328, 1980.




17.  Pionke,  H.  B. and A.  S.  Rogowski.   Implications  for Water  Quality




     on Reclaimed Lands.   In:  Economics, Ethics  and  Ecology:   Roots




     of Productive  Conservation,  W.  E.  Jeske,  ed.  Soil  Conservation




     Society of  America,  Ankeny,  Iowa,  1981.   pp.  426-440.
                                   37

-------
18.  Rogowski, A. S.  Acid Generation within a Spoil Profile:  Pre-




     liminary Experimental Results.  In:  Seventh Symposium on Coal




     Mine Drainage Research, NCA/BCR Coal Conference and Expo IV,




     Oct. 18-20, Louisville, Kentucky, 1977.  pp. 25-40.




19.  Rogowski, A. S., H. B. Pionke, and J. G. Broyan.  Modeling the




     Impact of Strip Mining and Reclamation Processes on Quality and




     Quantity of Water in Mined Areas:  A Review.  J. Environ. Qual.,




     6:237-249, 1972.




20.  Rogowski, A. S., H. B. Pionke, and B. E. Weinrich.  Some Physical




     and Chemical Aspects of Reclamation.  Preprint, paper presented




     at the North Atlantic Region ASAE 1982 Annual Meeting, 1982.




21.  Singer, P. C. and W. Stumm.  Acidic Mine Drainage:  The Rate




     Determining Step.  Science, 167:1121-1123.




22.  Steele, R. G. D. and J. H. Torrie.  Principles  and Procedures of




     Statistics.  McGraw-Hill Book Company, Inc., New York, 1960.
                                   38

-------

-------
                            SECTION 3
               FLUX AND PRODUCTION RATES OF OXYGEN
                AND CARBON DIOXIDE IN A RECLAIMED
                         COAL-STRIPMINE

                          Introduction
     In Part II we presented data on the oxygen and carbon dioxide

status within a reclaimed coal-stripmine.  An unexpected finding was

the prevalance of C0_ within the spoil profile.  Carbon dioxide

concentrations ranging from 0.02 to 0.16 mole fraction were found in

the stripmine atmosphere.  The higher values were restricted to only

one portion of the study area, but greatly'exceed the highest values

reported for natural soils (DeJong, 1981; Russell and Appleyard,

1915).  We proposed two possible sources for the observed CO^:  i)

heterotrophic respiration by soil bacteria within the spoil bank

and ii) carbonate neutralization within the mine of the acid produced

by pyrite oxidation with subsequent release of CCL gas.  If either

of these mechanisms accounts for the observed C02> the implications

for acid generation and leaching within reclaimed stripmine is

significant.  In the first case, uptake of 0- by micro-organisms

would help reduce pyrite oxidation by reducing the 0^ concentration

in the spoil atmosphere since the oxidation rate of pyrite is

directly proportional to the gaseous 0_ concentration (Clark, 1965;

Ohio State University Research Foundation, 1971).  On the other hand,'

if the CO, is from carbonate neutralization of acid, the effect would

be to reduce the acidity leaving the stripmine, while the production
                                  39

-------
rate of pyrite oxidation products remains unchanged.  Both mechanisms




would serve to reduce the amount of acid leaving a stripmine.




However, in the first case,  the reduction is from a decrease in the




production rate of acid, while in the second case, the reduction is




from an in situ neutralization of the acid produced.  The objective




of this study was to see if  either i or ii could be identified as




the dominant process at this reclaimed mine.




     Neither mechanisms can  be ruled out at this site.  Heterotrophic




respiration in the surface layers must take place at this site since




a plant cover is well established.  Also, heterotrophic activity




appears to be occurring in the deeper depths as indicated by the




formation of methane during  the wet, presumably most poorly aerated




times of the year (Part II).  Carbonate neutralization of acid is




also a possibility at this site, despite being located in an area




low in carbonate-containing  overburden.  Water quality data presented




by Rogowski, et al. (1982) indicates that some carbonate neutraliza-




tion is taking place at this site.




     Table 5 shows the pyrite and sulfate content expressed as percent




by weight sulfur for ten representative spoil layers from this




reclaimed mine (Rogowski, 1977).  The fourth column in Table 5 repre-




sents the amount of carbonate, expressed as percent by weight C00,




required to neutralize all the acid produced if all the pyrite oxi-




dizes as shown in Eq. [8] and one-half of the sulfate content is in




the form of sulfuric acid.  The fifth and sixth column show the
                                 40

-------
Table 5.  Pyrite and sulfate  concentrations  expressed  as  percent by weight sulfur for
          ten representative  spoil  layers  from research site (Rogowski,  1977).   Also,
          the CO- equivalent  of  the carbonate  content  required to neutralize the acid
          produced from columns  2 and  3  and  the measured  carbonate content.  The last
          column gives the ratio CO,,        j/COo      j   j  *n percent.
                 6                 2 measured    2  required

Layer

1
2
3
4
5
6
7
8
9
10
Pyritic S


.095
.112
.136
.088
.124
.082
.064
.060
.066
1.967
Sulfate S


.085
.079
.089
.070
.049
.059
.056
.077
.055
.738
2 required
i . |

.122
.130
.158
.108
.118
.096
.082
.094
.082
1.86
C00 ,
2 measured


.03
.01
.01
.02
.03
.03
.03
.30
.01
1.16
Potential
neutralization
%
25
8
7
19
25
31
37
319
12
62

-------
amount of CCL forming minerals present in these layers and  the per-


centage of the C02 requirement these minerals represent.  From these


data it can be seen that appreciable CO- can be generated over much


of the active lifetime of a reclaimed stripmine.





                             Theory




     One possible means of identifying whether mechanism i or ii


dominates the production of C0~ is to compare the consumption and


production rates of 02 and CO--  For aerobic hetertrophic respiration


by plant roots or microorganisms we would expect the ratio of con-


sumption of C02 to 02, r  to be near -1.0 although values between


-0.6 and -4.0 have been reported (Bridge and Rixon, 1976; Rixon and


Bridge,  1968; Bunt and Rovira,  1955).  To estimate r  for mechanism
                                                    q

ii, the stoichiometry of carbonate neutralization of acid and pyrite


oxidation must be considered.   For pyrite oxidation we know that


3.5 moles of 02 will oxidize 1.0 mole of pyrite to form 2.0 moles


of H+.


            FeS2 + 3.5 02 + H20 •+ Fe2+ + 2S02~ + 2H+            [8]



This result is true regardless  of whether 0_ reacts directly with


pyrite  or if an intermediary oxidant, such as ferric iron,  is


involved (Ohio State University Research Foundation, 1971).   The


ferrous  iron produced in Eq. [8]  may or may not oxidize and pre-


cipitate within the mine to form more acid:


            Fe2+ + 1/4 02 + 2.5 H20 -»• Fe (OH)  + 2H+            [9]
                                 42

-------
Thus, from 0.57 to 1.07 moles of H  can be produced  for  every mole

of 0_ consumed during pyrite oxidation.

     For carbonate neutralization in acid stripmines  (pH < 4),

carbonic acid will be the dominant species in solution since  the

first pKa for H.CO- is 6.4  (Garrels and Christ, 1965, p. 76).   The

equation for any carbonate species is :

                    M-C03 + 2H+ -»• M2+ + H2C03   .                 [10]
                                             aq

The carbonic acid formed in Eq. [10] will in turn be in  equilibrium

with gaseous C0», the relation being governed by Henry's Law

(Garrels and Christ, 1965, p. 76).
where a is activity, pCO. is the partial pressure of C0_  (atmospheres)

P is total pressure (kPa), and Y_n  is the mole fraction of C0_  (-) .

If we assume that the movement of water through the spoil is suffi-

ciently slow so that the water is always in equilibrium with gaseous

C0_, we can rewrite Eq. [10].

                     M-CO- + 2H+ -»• M2+ + H20 + C02               [12]
                                                  g
Thus, the neutralization of 2.0 moles of H  will produce 1.0 mole of

cor

     The areas in the mine where 0- is consumed would correspond to

the areas containing pyrite, whereas C0? production would take place

only in zones where acidic waters come in contact with carbonates.

These areas may not be the same.  Table 5 shows that the presence of
                                 43

-------
CO  producing materials does not correspond to pyrite abundance.  If


all the acid produced by Eq. [8] is neutralized by carbonates, before


it moves far in the profile, r  would equal -0.29.  If no neutraliza-


tion occurs, r  could be 0.0 and could go as high as -0.54 or higher
              q

if iron precipitates within the mine as described by Eq. [11] or if


acid formed in one area of the mine is concentrated by leaching to a


high carbonate zone where it is neutralized.  Even if this latter


case is true, the r  value for the reclaimed site on the whole will


be between 0.0 and -0.54.  This range is considerably different from


the range expected for heterotrophic respiration  (-.6 to -4.0).  The


difference in the two ranges for r  may serve as  a means of


differentiating between mechanisms i or ii, making it possible  to


identify the dominant process of C02 production.
                         Materials and Methods




     Oxygen and carbon dioxide concentrations for selected depths at


6 sites on a reclaimed coal-stripmine were presented in Part  II.  The


consumption and production rates  for 0? and C0? at  these sites  can be


calculated from the  gas concentration profiles if we assume that  the


dominant means of  gas movement through the spoil material is  by


diffusion in the vertical direction only  (Evans, 1965).  The


diffusion of gas component i  into the spoil can be  described  in one


dimension by:
                    - d/dz  (N±)  + Q±  =  d/dt  (
-------
                                             -2    -1
where N. is the flux of component i  (moles-cm  -sec   ,  positive



downwards), Q. is the source/sink (+/-) term for component  i (moles-



cnT^sec"1), $ is the air-filled fraction of the spoil  (-),  Y.  the



mole fraction (-), z is depth (cm), and t is time  (sec).  At steady



or quasi-steady state, the gas concentration per unit volume of soil



is constant or changing slowly.  The right side of Eq.  [13]  can then



be set equal to zero since <(>, p. and Y. are constant with time  and



Eq. [13] can be rearranged:



                       d/dz  (N.) = Q.                            [14]



The molar flux,  N.,  can be found from the mole fraction form of Fick's



Law '(see Appendix A) .



                       N± = - */T Dp  |^r dYi/dz                  [15]




                                                     3     —1  —1
where P is pressure (kPa), R the gas constant (kPa-cm -mole   -K ),



T is temperature (K),  T is the tortuosity (-) and D   the Fickian


                                      2
diffusion constant for component i (cm /sec, see Appendix A).   The



Fickian diffusion coefficient in an open system, D  , is adjusted

                                                  Fi

by (j>/T in Eq.  [15]  to account for the reduced air-filled cross



section and tortuous path of gas diffusion in soils (Troeh,  et  al.,



1982).  A constant tortuosity term of 5 was used for the coarse spoil



material at sites 1-5 (Cathles and Apps, 1975).   was  assumed  to be



0.12 at these  sites  based on an average bulk density of 1.57  g/cm



and a water-saturation percentage of 0.7 for the stripmine land



(Pedersen, et  al.,  1980).  Since site 6 was located in  an undisturbed



area,  values of  10 and 0.06 were used for T and , respectively.
                                  45

-------
D   was calculated by assuming that CL and CO* diffusion were  taking
  i
place in a four -component atmosphere comprised of 0-, CCL, Ar  and
,  where N   and N.   were 0.0.   Since
         N       Ar
                                            and D^    are dependent
                                                 *
on N   and Npn  (Appendix A) , an iterative scheme was used to solve
    °2      C°2
for N- ,  N-^ , D_   and D    .   Nitrogen and argon concentrations
     °2   C°2   F02      FC02

were not measured directly so their values were set by assuming

Y^T  + YA   =1 —'%,  - Y_-.  and  that the nitrogen-argon ratio was
  ?    Ar      .  U_    <-u
constant and equal in the atmospheric ratio of 0.012.  At each site,

the mine profile was divided into a number of layers corresponding

to the depths at which O^-and CO^ were measured (Fig. 11).  Flux

values for 0- and C0_ were calculated at each layer by using the

unequal-spacing,  central-difference form of the first-order space

derivation in Eq. [15] (Peaceman, 1977, p. 38).  In general:
            N.
                                                                [16]


where the subscript j refers to the depth, increasing downwards, and

DZ is the distance weighing factor;


            t)7      = 	J+l  J	_          fl71
              j+1/2   (z.-z.)(2.-z) + (z.-z.)^


At each site the flux was assumed to be 0.0 at the spoil-bedrock or

spoil-water table boundary.  Oxygen and carbon dioxide concentrations

at the spoil surface were assumed to be at atmospheric levels.  After

- V'l
Y. .
Ti '

\.i >
Y. . "
i, 3— 1
T. . T
1,3-1.
t
I
"Y< '+1 Yi •"

Ti -+1 Ti '

DZj-l/2'


                                 46

-------
layer
                            spoil surface                _  r
                                                    I _  - ~~  \J • £.J-\J .
       N      N       	:	   02,1


        °2'1   C°2'1                                Y     = 0.0003
                                                     L»Urt j J-



                          Qo2,i      Qco2,i
                                                   Y       Y
                                                   ^2,2    C02,2
                          Q02,2      QC02,2
                        water table of bedrock
                                                          YC02,4
Figure 11.  Schematic of spoil profile showing the division of the


            profile into layers as determined by the depths of the


            gas concentration measurements,  Y   and Y.   .  Fluxes
                                             0_      Cu_

            N   and N    are calculated for the top of  each layer

              2        2
            ano are assumed to be 0.0 at the profile bottom.  Source


            terms, Qn  and Q   ,  are calculated for each layer.

                    U2      C02
                                 47

-------
 the fluxes were calculated, the average Qn  and Q n  values were
                                         °2      CU2
 found for the layers between the probes (Fig. 11), by using the

 difference form of Eq. [14] :
Fluxes and source terms for CL and C0_, and the rates, r  =

Q   /Q   and r T = N   /Nn  were calculated from the data presented
 CUn  U „      N    CU.  Li-
earlier.
                     Results and Discussion


     The calculated fluxes showed considerable variation over time

at each site.   Table 6 shows the flux values, averaged over time,

for oxygen and carbon dioxide at the soil surface of the six sites

and for all sites combined.  Coefficients of variation, C.V., for

both Nn  and N_n  are also shown in Table 6.  The C.V. values range
      °2      C°2
from 37 to 390%.  Site 6 consistently showed less variation than the

other sites and was also the only site located in undisturbed bed-

rock.  This was probably because the diffusional process was more

consistent and uniform in the bedrock than in the spoil material.

This isn't surprising since we would expect greater heterogeneity of

both porosity  and pyrite content in the spoil, which would lead to

nonuniform patterns of diffusion.  In the extreme, heterogeneity

could negate our assumption of one-dimensional diffusion, although

the generally  uniform variation of CL and CO  concentrations with
                                 48

-------
Table 6.   Means and coefficients of variation,  C.V.,  for 0- and
          CO  fluxes at the surface of the six  sites  and for all
          sites combined.   The ratio of the fluxes, r  = N 0 /
          N  ,  was calculated from the mean values.          2

Site



1
2
3
4
5
6
all
v
X
moles
2
cm -sec
0.78
2.2
0.49
0.96
1.41
.45
1.2
io10
c:v.
%


160
110
140
110
130
37
130
"CO,
X
moles
2
cm -sec
-.27
-.82
-.07
-.10
-.23
-.20
-.37
xlO10
c.v.


%
210
110
270
390
200
43
170
r


™
-.35
-.37
-.14
-.10
-.16
-.46
-.31
                                 49

-------
depth (Part II) indicates that vertical flow still dominates at  this




reclaimed mine.




     No significant difference at the 10% significance level was




measured between any site for the flux of 0^ anc* COo-  For all the



                                                        —10         —2
sites the average flux of 0- at the surface was 1.2 x 10    moles-cm




-sec  ,  while for CO  the average flux was -.37 x 10    moles-cm




-sec  .   The CO^ fluxes measured for these minesoils are, in general,




less than the fluxes of C0_ measured by DeJong (1981) in grassland




and cultivated soils.  DeJong (1981) found CO.. fluxes ranging from


                   	-i r\         _n    —1         	c     _9    	*L

near 0.0 to -7 x 10    moles-cm  -sec   (-3 x 10   g-cm  -sec  ),




with fluxes normally near -2 x 10    moles-cm  -sec  .  The lower




CO  fluxes in the minesoils may be due to the lack of plant cover




and freshly deposited organic matter at the sites measured,




although no consistent pattern was obvious in the data.




     Undoubtedly, much of the variation found in the flux values




was due to the fact that we assumed a constant air-filled porosity -




tortuosity term in the diffusion calculation when, in fact, this




term would vary over time as the water content of the minesoils




changed (Troeh, et al., 1982).  The value we used was only approxi-




mate, thus, the magnitude of the flux terms shown in Table 6 are




also approximate.  However, the ratio of the C02 flow to 0_ flux is




an absolute comparison since the effect of diffusion path in the




spoil is identical for 0,, and C02 (Penman, 1940) and cancels out




when we take the ratio, r  = NCQ /NQ .  Table 6 lists the TN for the
                                 50

-------
average of the fluxes at the surface of each site.  The r  ranges from



-.10 at site 4 to -.46 at site 6 and is equal  to  -.31 for all  the



sites combined.  The lower values for rN occur at sites 3, 4 and



5 which are located on the upper portion of the mine site in the



predominantly sandstone overburden.  The differences between r  at



sites 3, .4 and 5 and 1, 2 and 6 are not significant however, because



of the great variability in the flux values.   These r  values indi-



cate that, on the average, from 2 to 10 times  as  much 0  is entering



the spoil material as C0_ is leaving.



     The source term Q was calculated for 02 and  CO,, for each layer



at the six sites.  Table 7 summarizes the results, showing the mean


                                —3    —1
values of Qn  and Q n  (moles-cm  -sec  ) and  the coefficients of

           °2      UU2

variation for these values.  The variation in  the Q terms was large



enough that no significant difference could be found between the



different layers, but several trends are obvious  from the data.



     The layers of greatest consumption of 0- are the surface layers



at almost every site.  The rate of 0« uptake in these layers is con-



sistently greater than for the deeper layers.   This finding is



consistent with the fact that plant roots and soil organisms are



most active in the surface layers and would be removing oxygen during



respiration.   The rate of 0_ uptake in the surface layer at the six


                        -12         -3    -1                   -1   —1
sites averaged -1.2 x 10    moles-cm  -sec   or 0.065 uliters-g  -hr



if a bulk density of 1.57 g/cm  is assumed.  This value is  much lower



than the values found by Stroo and Jencks (1982)  for 0~ uptake in



minesoils.  Even for barren minesoils they found  the lowest 0-  uptake
                                  51

-------
         Table  7,
Mean values and coefficients of variation of the source terms for 09 and CO
                   and  for  the ratio rr
                   correlation between Q _   and  Q
                                        C°2      °2
                           /Q^  at the individual depths at each site.  The

                                  r,  is also listed.
Ul





Site 1






Site 2






Site 3





Depth


cm
15
45
82
182
335
488
602
15
45
115
245
472
785
1098
15
45
105
228
380
532
v
X
moles
3
cm -sec
-3.6
3.6
3.1
- .92
- .55
- .073
.048
-5.5
-3.0
1.0
- .079
- .013
- .045
.027
-2.1
- .51
.003
- .072
.17
- .12
io12
c.v.


%
123
64
36
110
42
260
720
150
110
120
270
280
62
99
150
220
7900
140
92
-80
Qco2x
X
moles
3
cm -sec
1.5
-1.8
-1.4
.41
.23
.021
.13
2.5
1.2
- .38
.020
.010
.029
- .014
.35
- .004
.033
.059
- .064
.039
io12
C.V.


%
150
88
33
110
64
580
220
98
81
79
410
130
59
86
240
8000
350
71
56
92

X


—- —
-.38
-.48
-.49
-.50
-.47
-.15
-.29
,-.51
-.48
-.24
-.45
-.18
-.54
-.59
-.32
-.78
-.53
-.79
-.23
-.45
rq
c.v.


%
87
44
32
100
57
930
470
180
130
200
130
310
140
130
50
290
340
160
270
110
r


— —
-.85**
-.90**
-.63**
-.84**
-.47*
-.36
-.42
-.86**
-.85**
-.68**
-.77**
-.15
-.63**
-.49*
-.81**
-.88**
-.60**
-.71**
-.46*
-.58**

-------
Site 4






Site 5





Site 6


15
45
122
260
412
640
900
15
88
235
478
782
1088
75
228
342
.22
.082
-1.10
.12
.029
.003
- .010
-2.5
- .58
- .028
- .009
.007
- .014
- .090
- .052
- .20
1400
180
93
140
140
1500
580
160
130
327
310
270
120
110
90
78
.18
-.27
.28
-.027
-.017
.004
.006
.055
.13
.017
-.001
-.008
.011
.038
.030
.075
750
180
96
160
100
630
560
1620
160
170
1500
130
110
110
100
80
-.25
-.56
-.24
-.36
-.42
-.45
-.39
-.23
-.32
-.29
-.16
-.56
-.91
-.53
-.52
-.46
365
390
200
92
120
87
53
245
100
120
310
130
44
110
75
51
.22
-.77**
-.53*
-.59**
-.79**
-.69**
-.66**
-.30
-.63*
-.50
-.91**
-.81**
-.83**
-.86**
-.88**
-.88**

t
 Significant at the 5% protection  level.

it
 Significant at the 1% protection  level.

-------
rates to be 0.58 uliters-g  -hr   or about ten times  the values

found in this study.  This large difference may be due  to differ-

ences in the studied site, but is more likely due to  the different

techniques used to measure Q; a laboratory respirometer by Stroo and

Jencks (1982), and concentration gradients in this study.  Also,

since only estimated values for the porosity - tortuosity term were

used here to calculate Nn  and thus Q  , the absolute magnitude of
                        U2           °2
Q   is unknown.  Comparing the two data sets, the coefficient 4>/T

would have to be increased by a factor of 10 for ,the uptake rates to

agree.  Such an increase is certainly feasible but would result in

greater C«  flux rates for the minesoil than for the natural soils

studied by DeJong (1981).  Problems in measuring Yn  at shallow

depths, discussed in Part II, may also contribute to the difference,

as well as smearing of the 0« gradient as a result of atmosphere-

pressure fluctuations (Morth, et al., 1972).   Smearing of the gradi-

ents because of pressure fluctuations and deviation from strictly

vertical diffusion because of spatial variability in the porosity of

the minesoil may also account for the seemingly contradictory result

that the layers just below the surface act as 0  sources and CO

sinks (Table 7).

     Problems with using the data resulting from uncertainty in the

absolute magnitude of Qn  and Q _  is avoided by comparing the ratio
                       U2      CU2
of the fluxes r  = Q _ /Qn  and the correlation of the relative up-
               Q      2   2
take terms.   Table 7 shows the correlation coefficient between Q n
                                                                LU2
and 0-  and the ratio, r , for each layer.  In general,  the
                                 54

-------
correlation is strongest near the surface at each site indicating a



greater dependency of the changes in CCL and CL.  The rQ values are



extremely variable and range from -.15 to -.91.  Table 8 shows the



average values for the source term of 0~ and C0» and the r  values



for the entire depth at each site.  Again the values are extremely



variable which prevents any significant statistical comparison



between the sites.  However, there is a trend in the data for rQ



similar to that found for r.T where the r-. values for sites 3, 4 and
                           N            Q


5 are less than sites 1, 2 and 6.  The average rQ value for all the



sites was -0.30 which indicates that more than 3 times as much 0?



is being consumed as CO .







                            Conclusions





     Both the ratios of the diffusional flux of CO. to 0^ and the ratios



of the source terms for C0,j and €>„ showed considerable variability.



Site 6 showed considerably less variation than the other sites possibly



because it was the only site located in undistrubed bedrock where our



assumption of strict vertical diffusion was more valid.  The r  ranged



from -.10 to -.46 for the six sites and was equal to -.31 for all the



sites combined.  The rQ for the entire profile at each site ranged from



-.12 to -.46 and averaged -.30 for the combined site.  Both r  and r



indicate that from 2 to 10 times as much 0  is being consumed as CO-



with 3.3 times as much for all the sites combined.  These values lie



well within the range expected for carbonate neutralization of the acid



formed from pyrite oxidation and indicate that this mechanism accounts



for the bulk of the CO™ measured at this site although heterotrophic
                                 55

-------
Table 8.  Mean values and  coefficients of variation,  C.V., for the
          production rate  of  0~  and CO- and for their ratio, rn =
          Q   /Q  .   Rates are for the entire spoil profile at each
           CU ty   'y
          site averaged over  time.

Site



1
2
3
4
5
6
all
v]
X
moles
2
cm -sec
-.66
-1.9
-.47
-.24
-1.0
-.45
-.91

C.V.
%


210
130
140
480
170
37
180
Qco2x
X
moles
2
cm -sec
0.22
0.78
0.063
-.047
0.16
0.20
0.29
io10
C.V,


%
280
110
300
880
250
42
220
rQ



-.44
-.40
-.15
-.24
-.12
-.46
-.30
C.V.


%
350
180
550
450
230
33
330
                                 56

-------
respiration is undoubtedly still taking place.  The more negative ratios



at the sites located on the lower portions of the mine site may be as a



result of greater concentrations of carbonate in this portion of the



mine.  The more negative ratios may also be due to waters containing



acid formed in the upper reaches of the mine site flowing through the



lower sites and producing increased amounts of CCL for neutralization



reactions.  This may also explain the poor correlation between Q-  and



Q n  at the 488 and 602 cm layers at site 1, where Y    was very high.
 LU—                                                CUrt


The rn values indicate that a considerable amount of the acid produced



within the mine is being neutralized before it leaves the site.
                                 57

-------

-------
                            REFERENCES




1.  Bridge, B. J. and A. J. Rixon.  Oxygen Uptake and Respiration




    Quotient of Field Soil Cores in Relation to their Air-Filled




    Pore Space.  J. Soil Sci., 27:279-286, 1976.




2.  Bunt, J. S. and A. D. Rovira.  Microbial Studies of Some Sub-




    antartic Soils.  J. Soil Sci., 6:119-128, 1955.




3.  Cathles, L. M. and J. A. Apps.  A Model of the Dump Leaching Process




    that Incorporates Oxygen Balance, Heat Balance and Air Convection.




    Metall. Trans., 66:617-624, 1975.




4.  Clark, C. S.  The Oxidation of Coal Mine Pyrite.  Ph.D. Thesis,




    John Hopkins University, University Microfilms International,




    Ann Arbor, Michigan, 1965.




5.  DeJong, E.  Soil Aeration as Affected by Slope Position and




    Vegetation Cover.  Soil Sci., 131:34-43, 1981.




6.  Evans, D. D.  Gas Movement.  In:  Methods of Soil Analysis, C. A.




    Black, ed.  American Society of Agronomy, Madison, Wisconsin, 1965.




    Agronomy 9:319-330.




7.  Garrels, R. M. and C. L. Christ.  Solutions, Minerals and Equilibria.




    Freeman, Cooper and Company, San Francisco, California, 1965.




8.  Morth, A. H., E. E. Smith, and K. S. Shumate.  Pyritic Systems:  A




    Mathematical Model.  Environmental Protection Agency Technology




    Series 14010 EAH, Contract No. 14-12-589, Report No. 665771, U.S.




    Environmental Protection Agency, Washington, D.C., 1972.




9.  Ohio State University Research Foundation.  Acid Mine Drainage




    Formation and Abatement.  Water Pollution Control Research Series




    Program 14010 FPR, U.S. Environmental Protection Agency,




    Washington, D.C., 1971.
                                 58

-------
10.  Peaceman, D. W.  Fundamentals of Numerical Reservoir Simulation.




     Elsevier Scientific Publishing Company, New York, 1977.




11.  Pedersen, T. A., A. S. Rogowski, and R. Pennock, Jr.  Physical




     Characteristics of Some Minesoils.  Soil Sci. Soc. Am. J.,




     44:321-328, 1980.




12.  Penman, H. L.  Gas and Vapour Movements in the Soil.  II.  The




     Diffusion of Carbon Dioxide through Porous Solids.  J. Agr. Res.,




     30:570-581, 1940.




13.  Rixon, A. J. and B. J. Bridge.  Respiratory Quotient Arising from




     Microbial Activity in Relation to Matric Suction and Air-Filled




     Pore-Space of Soil.  Nature, 218:961-962, 1968.




14.  Rogowski, A. S.  Acid Generation within a Spoil Profile:  Preliminary




     Experimental Results.  In:  Seventh Symposium on Coal Mine Drainage




     Research, NCA/BCR Coal Conference and Expo IV, Oct. 18-20, Louisville,




     Kentucky, 1977.  pp. 25-40.




15.  Rogowski, A. S., H. B. Pionke, and B. E. Weinrich.  Some Physical and




     Chemical Aspects of Reclamation.  Reprint, paper presented at the




     North Atlantic Region ASAE 1982 Annual Meeting, 1982.




16.  Russell, E. J. and A. Appleyard.  The Atmosphere of the Soil:  Its




     Composition and  the Causes of Variation.  J. Agric. Sci., 7:1-48,




     1915.




17.  Stroo, H. F. and E. M. Jencks.  Enzyme Activity and Respiration  in




     Minesoils.  Soil Sci. Soc. Am. J., 46:548-553, 1982.




18.  Troeh, F. R., J. D. Jabro, and D. Kirkham.   Gaseous Diffusion




     Equations for Porous Materials.   Geoderma, 27:239-253, 1982.
                                 59

-------
                            SECTION 4
               A NUMERICAL MODEL OF ACID DRAINAGE
                 FROM RECLAIMED COAL- STRIPMINES
                          Introduction


     Acid drainage from reclaimed coal stripmines can be a severe

problem in the humid Eastern United States (Collier, et al., 1970).

The amount of pollutants leaving the reclaimed site can be reduced

by using proper-reclamation techniques when restoring the land

(Grim and Hill, 1974).  However, the development and evaluation of

best reclamation and management techniques is difficult because of

the great expense and time required to implement and assess the

value of a specific practice and because of the great variability

between even adjacent-stripmine sites.  The development of a mathe-

matical model that describes the acid production and leaching from

stripmined lands would facilitate the development of sound-

management practices by allowing assessment of each technique for

any given condition to be made in minutes rather than years.  Although

several models for pyrite weathering have been presented (Morth, et

al., 1972; Colvin, 1977? Cathles and Apps, 1977; Cathles, 1979) none

is completely  suitable for describing the process in reclaimed coal-

stripmines.  The model described by Morth, et al., (1972) is for

pyrite oxidation in deep mines and assumes that pyrite oxidation

takes place only on the surface of cracks and voids and not within

the rock matrix.  They assumed oxygen flux due to diffusion and
                                60

-------
atmospheric-pressure fluctuations.  Their model is reasonable for




deep mines but does not seem representative of pyrite oxidation




within mine spoil composed of coarse fragments.  Colvin (1977)




describes a model for pyrite oxidation of reclaimed stripmines where




diffusion serves as the mechanism of replacing oxygen with the




profile.  However, the model uses representative values of pyrite




oxidation rates and does not attempt to relate the oxygen consumption




rate to pyrite content and distribution within the spoil fragments




nor does it differentiate between pyrite oxidation mechanisms.




Cathles and Apps (1977) present a model for oxidation and leaching




of sulfide-bearing fragments where the pyrite distribution, degree




of weathering, and oxidation kinetics are considered.  However,  they




assumed air-convection to be the main mechanism of oxygen movement




and that pyrite was oxidized by ferric iron, where the ferric iron




was maintained at a constant concentration by bacterial activity.




While these conditions represent reasonable assumptions for the




very coarse, copper-waste dumps they modelled, they do not apply




well to reclaimed coal-stripmines where diffusion processes probably




dominate and the role of iron-oxidizing bacteria is unclear.   This




paper presents a parametric model of in situ oxidation of  pyrite




and subsequent leaching of the acid products from reclaimed coal-




stripmines,  based on oxygen diffusion, pyrite oxidation kinetics




and combined oxygen and ferric iron oxidation of pyrite, where the




activity of iron—oxidizing bacteria depends on the energy  available
                                 61

-------
from their substrate and the suitability of their environment for cell




viability and is calculated independently.








                        Description of Model






Basic Reactions.




     The chemistry of pyrite oxidation is extremely complex and only




partially understood.  However, the overall stoichiometry of the




reaction can be summarized by four equations (Singer and Stumm, 1968).




Two mechanisms for pyrite oxidation are possible (Ohio State University




Research Foundation, 1970).  One possibility is that oxygen can react




directly with pyrite to form sulfate and acid.





               FeS2 + 3.5 02 + H20 -»- Fe2 + 2S02~ + 2H+               [19]





Alternatively, ferric iron can replace oxygen as the direct oxidant.





            FeS2 + 14Fe3+ + 8H20 -> 15Fe2+ + 2S02~ + 16H+             [20]





In stripmine spoil, the only important source of ferric iron in Eq.




[20] is assumed to be the in situ oxidation of ferrous iron (Lau, et




al., 1970; Singer and Stumm, 1968);





               14Fe2+ + 3.5 02 + 14H+ -»• 14Fe3+ + 7^0                [21]





While the oxidation of ferrous iron is thermodynamically favorable,




the kinetics are extremely slow "at normal pH's of stripmine waters




(pH < 4; Singer and Stumm, 1970).  However, certain chemoautotrophic




bacteria (Thiobacillus ferrooxidans) are known to use the energy




released by Eq. [21] as their energy source and can significantly




increase the oxidation rate (Lundgren, 1975; Beck, 1960).  Thus, Eq.
                                  62

-------
[21] and consequently Eq. [20] are thought to be bacterially catalyzed




since ferric oxidation of pyrite can be significant only when bacteria




are active.




     As a final step in the pyrite oxidation process, the ferric iron




produced by  Eqs.  [19] and [20] may precipitate as a ferric hydroxide.




            Fe3+ + 3H20 •*• Fe(OH)3 + 3H+                         [22]




This reaction often takes place after the iron has been leached from




the stripmine site, with the iron hydroxide precipitation in surface




streams and  ponds.   Summing Eqs. [20] and [21] results in Eq. [19],




thus regardless of mechanism the result is the same; two moles of




acid are produced for every mole of pyrite and three and a half moles




of oxygen consumed.






Basic EOuations
     The reclaimed-mine environment in which Eqs. [19]-[21] and




possibly [22]  take place can be described as consisting primarily of




coarse (> 2mm)  fragments.  Pedersen, et al. (1980) and Ciolkosz, et




al. (1977) found coarse fragments composing 50-90% of the spoil




volume.  In this model, we consider all the pyrite oxidation to take




place within coarse fragments.  Since pyrite oxidation is a surface




reaction (Garrels and Thompson, 1960), the oxidant must diffuse from




the fragment surface through the fine pores of the fragment, which we




assume to be water saturated, to the pyrite mineral surface (Fig.




12).  Evidence indicates that the fine-grained or framboidal form




of pyrite is the most reactive (Clark, 1965 and Caruccio, 1973).
                                63

-------
   Liquid
   Pyrite
    Blebs
                                     ..          — Fe
                                     Y//////////////,
Figure 12.   Cross section of a  coarse fragment containing pyrite.
                              64

-------
The framboidal form may be the result of  the mineral being deposited




by microorganisms that were active in the original sediments  (Emrich




and Thompson, 1968).  Possibly for this reason, framboidal pyrite  is




evenly distributed within the rock formations in which it is  found




and the fragments derived from these strata during mining (Arora,




et al., 1978).  The rate at which pyrite  and oxidant react may be




controlled either by the chemical reaction rate or by the rate of




diffusion of oxidant into the fragment, or by both.  This process




can be described by a standard shrinking-core model (Levenspiel,




1972).  If we consider the fragments to be thin plates and the  oxida-




tion of pyrite to be first order with respect to oxidant concentration




and pyrite surface area, we can express the oxidation rate of pyrite




within a fragment by:





                      dX/dt
                              2tn(l-X)+tr
                                U       L*




where:




     X = fraction of pyrite remaining in fragment, (-)




    t  = time for complete oxidation of fragment if diffusion of




         oxidant is much slower than reaction ra.te, (sec)




    t  = time for complete oxidation of fragment if chemical




         oxidation is much slower than oxidant diffusion, (sec)




and




     t = time (sec)




     tn represents the total time required to oxidize all the pyrite




within a fragment when the diffusion of reactants and products




between the fragment surface and the pyrite grain is the rate-




controlling step.  t^ can be calculated from (Levenspiel, 1972):






                                65

-------
                         C ox



where:



   p   = molar density of pyrite within the fragment,



         (moles/cm )



     £ = one half thickness of fragment,  (cm)



     b = stoichiometric ratio between pyrite and oxidant



         consumption, (-)



    D  = effective diffusion coefficient for coupled oxidant-


                                       2
         product counter diffusion, (cm /sec)



   C   = concentration of oxidant at fragment surface,
    ox


         (moles/cm ).



     tp is the total time required for complete pyrite oxidation



within a fragment if the resupply rate of oxidant to the pyrite



surface is much faster than the chemical oxidation of pyrite.  t
                                                                L*


can be calculated from, (after Levenspiel, 1972; Cathles and Apps,



1975):


                             ppy

                     CC   bK.C  3d                              [25]
                            o OX



where:



    K  = first-order surface reaction rate constant for
     O


         pyrite oxidation per unit surface area of pyrite,



         (cm/sec);  i.e.,



         p   £ dx/dt = -bK_C
          py             S ox



     a  = surface area of  pyrite per unit volume of fragment,



         (cm  )
                                 66

-------
 and

      3 = effective thickness of  fragment within which pyrite

         is oxidized  (cm, see Appendix  B) .

      From Eq. [23] it is obvious that the oxidation  of pyrite will

 be independent of diffusion rates at initial  times,  (X =  1) ,  but

 may or may not become important  as the  pyrite 'is leached,  depending

 on the magnitudes of tc and t .  For t  »  t  ,  diffusion will never

 be an important rate-controlling step and the fragment will weather

 uniformly throughout its thickness.  For t_ »  tr, diffusion  is the

 rate-controlling step and a rim  depleted of pyrite will form  and

 slowly grow at the fragment surface.

      When both oxygen and ferric iron are present the  rate of pyrite

 oxidation is the sum of the rates of each oxidant acting alone (Ohio

 State University Research Foundation, 1970) and  their  reaction rates

 can be summed.
             tD(02)(l-X)+tc(02)
                                                                 [26]
Values for t_ and tn can be calculated from typical values for spoil
            U      D

and Eqs. [24] and [25].  For the values shown in Table 9, the
                                      Q           *> t           Q
calculated values are t^CL) = 27 x 10  sec, t_(Fe  ) = 35 x 10
                    Q               -31           r
sec, t (09) = 4 x 10  sec, and t (Fe  ) = .7 x 10  sec.  These
      Ll  Am                      \*t

calculated values indicate that we would expect diffusion to be the

rate-controlling step for both the oxygen and iron oxidants and

would expect that a leached rim, surrounding a relatively unweathered

core, would form in the fragment.  This agrees with the weathering
                                 67

-------
Table 9.  Coefficients used to calculatee typical values of  t

          and tn.
Parameter                                  Value





   p                       4.4 x 10~  moles cm"  (.13% pyrite  S)
    py

   £                       1 cm



   b(02)                   1/3.5



   b(Fe3+)                 1/14.



   D                       10    cm -sec
    c


   C  (0.)                 0.29 x 10~   moles-cm"  (.21 mole

    ox  2                    fraction)



   C  (Fe3+)               0.89 x 10~6  moles-cm"3 (50 mg/£)



   K (0»)                  83. x 10~  cm-sec"



   Ks(Fe3+)                4.4 x 10~6 cm-sec"1



   a                       217 cm"1



   3 (02)                  .0745 cm



   3 (Fe3+)                .0102 cm
                                 68

-------
pattern observed for sulfide-bearing fragments (Cathles, 1979;


Braum, et al.,  1974).



Oxygen Diffusion

     We assume  the main mechanism for resupply of 0- to the fragment


surface to be diffusion within the interfragment voids of the spoil


profile.  Atmospheric  pressure fluctuations (Morth, et al., 1972),


air pumping from rising and falling water»table levels, convection

due to temperature and compositional differences (Cathles, 1979)


and dissolved oxygen contained in percolating water may contribute


to the 09 flux  within  the spoil but diffusion should dominate,

especially where a fine-textured, surface soil layer is present


(Evans, 1965).   For one-dimensional diffusion of 0- into a r


spoil-bank at constant total pressure in which uptake of CL is


taking place and the water content remains constant over time,


the process can be described by:
                                       Do2
               d/dt     "  v    *  '"
Q0         [27]
  2
where:


       = air filled porosity, (-)
      A
                                     3
    Pn  = molar density 0-,  (moles/cm -air)


    Yn  = mole fraction of 0- in gas, (-)


     P  = total pressure,  (kPa)


     R  = universal gas constant, (8.31 kPa-cm /mole-K)


     T  = temperature, (K)


     T  = tortuosity of gas diffusion path, (-)


    Q   = 0  uptake rate,  (moles/cm )


     Z  = depth, cm


                                 69

-------
Dn  is the effective diffusion coefficient  of  0_  in an 0~,  CO-,



N2 atmosphere and is found from  the binary  diffusion coefficients



and the flux ratios, r. = N   /Nn  and r_ = N   /Nn   (Appendix A).
                      1    C02   02       2.    N2  02




                                                          -1
D

            D
                               _ V
             o, co
                                                                 [28]
              J ry )    O    " o » ^ v O





     For gaseous diffusion into spoil banks,  the bottom  of  the spoil




profile can be assumed to act as an effective diffusion  barrier




since below this boundary the air-filled porosity of  the bedrock




is significantly less than the spoil and commonly coincides  with




the water table.




     Qn  in Eq. [27] may be partitioned into  four separate processes

      U2
of CL consumption:





            Q02 = QPY + QAB + QCHOX + QHB




Qpy represents oxygen consumption by direct pyrite oxidation.  Q.




is the oxygen consumption rate of iron oxidation by autotrophic




bacteria as shown in Eq. [21].  Qr-rmv ^-s t^e rate of 0» uptake by
                                 CntJA                 £.
chemical oxidation of ferrous iron, and Q,TT1 represents the uptake
                                         HU


of oxygen due to respiration by plant roots or other soil organisms. •




The magnitude of each factor in Eq. [29] can be calculated separately.




     QpY. Q_Y can be calculated from Eq. [23], modified to account
                                  o

for the number of fragments per cm  of spoil, ( _ ^) and the

                                                •MTJ

                                                ^PY



stoichiometric ratio of pyrite consumption to 0? consumption, b (()-
                                 70

-------
                       -b(0 ) -1-       FFR FPY
            QPY = 	£	-1                 r301
                  2tD(02)(l-X)+t(,(02)   MWpy                     LJUJ




For a given fragment configuration and pyrite content, the magnitude



of QpY depends on the 0» concentration at the fragment surface and



the degree to which the fragment has weathered.



     Q  .  Q   represents the rate at which 00 is consumed through
      HB    HB                               /


respiration by plant roots and other soil organisms.  The magnitude



of Q   depends on type and extent of plant cover, time of the year
    HB


and the presence of biodegradable organic matter in the reclaimed



profile.  Since Q   does not affect pyrite oxidation other than
                 nJi


decreasing the available 0^ (Colvin, 1977), we have chosen a simple



temperature and 0- concentration dependent function to describe the



process.  0- uptake rates for bare (winter) and cropped (summer)



surface soils have been measured to range between 1.5 and 16.0 x



10    moles 0 -sec-cm  (soil) (Russell, 1973).  If we let the rate



equal zero at 0 C and a maximum value of 16.0 x 10    moles 0^-sec -

 -3      o
cm  at 30 C and assume that the observed values occurred at an 02



mole fraction of 0.17, but would drop to zero linearly as the 02



concentration goes to zero, we can represent Q _ as:


Q                  ?Y  T n          T > n r
             HB   ^ 0_  %BB       T > U 0

                                                                 [31]


                =0.0              T < 0°C



     Where Q    is -the base rate of 0  production by respiration and
            HBB                      2.
would be equal to 0.0 when no plants or biodegradable organic


                                                          Y
                                               -12             -1
matter are present in the soil and to -1.5 x 10    moles O.-sec
cm  - C.   for a well established agronomic soil.
                                 71

-------
     Q     .  The rate of chemical oxidation of ferrous  iron  is very
      CHOX


slow at low pH's and pH dependent throughout  the remaining pH range.



Singer and Stumm (1970) found that for pH > 4.5 the rate of  ferrous



iron oxidation at normal atmospheric pressure can be expressed by:



            d [Fe2+]/dt = - ^ [Fe2+] YQ  [H+]~2                 [32]





     with            ^ - 1.3 x 10~16 mole2-L~2-sec~1



where the brackets indicate concentration.  Below a pH  of 3.0, the



rate can be expressed as:



            drFe2+]/dt = -K2 [Fe2+] YQ                           [33]



                    K2 = 1.7 x 10~9 sec"1



For the entire pH range, Eqs. [32] and [33] can be combined  and  since



the oxygen uptake rate is equal to one fourth of the ferrous  iron



oxidation rate (Eq. [21]), we find the 0~ uptake rate per cubic  centi-



meter to be:
         A                       1                     — 9



                            = - 4W IFe'+J  V KiIH+J
                       chem                   2



where   is the water-filled porosity.



     Q  .   The rate at which chemoautotrophic bacteria can oxidize
ferrous iron, consuming 0^ in the process, is extremely variable



and depends on the size of the bacterial population and the condition



of the bacterial environment.  Expressing Q   in a manner similar to



chemical oxidation we have:


                -  *w  d [Fe2+],    *w
                               ,     w     f     ,                .  ,

             AB ~ 4000   dt    ~ ~ 4000 *B LFe  J  Y02           [35]


Where 1C is the factor that accounts for the bacterial "activity



the reclaimed spoil.  The lower limit for K^ is 0.0 when the auto-



trophs are not active.  An upper limit to 1C is set by assuming that
                                 72

-------
at maximum "activity," the bacteria are limited  by  the diffusion

rate of 0- to their surface.  If we assume  the bacteria are  attached

to the fragment surface within the water film surrounding the frag-

ment (Malouf and Prater, 1961), the maximum oxidation rate must

just equal the diffusion rate of 0? through the  film.
                           Y   =        /A£'                     [36]
            4003         o2         2

where A^ is the fragment surface area per milliliter spoil water,

D,, is the diffusion rate of 0- in water, where 0_ concentration is
 W                           2.                   L

expressed as mole fraction and 101 kPa (1 atmosphere) total pressure

is assumed, Yn  is the mole fraction of 0_ at the film surface (the
             °2                          ^
CL concentration at the fragment/bacterial surface is set equal to

0.0), and AX." is the film thickness.  Rearranging Eq. [36]


                 40°°
For A£' = 0.01 cm (ElBoushi, 1975) and Dy = 2.75 x lO'   moles 02~

cm— sec, the upper limit of K_ can be calculated for any pyrite and

fragment content.  Between the upper and lower limits, the magni-

tude of K_ is found from the bacterial "activity," which depends on

the energy available to the bacteria.

     K,,.  The primary concern of rhis model is the long term  (e.g.,

months) leaching of stripmined lands.  Since on this time scale

bacterial population fluctuations are comparatively rapid, we can

consider the bacterial "activity" to be in dynamic equilibrium with

the environment at all but initial times.  It is also desirable to

include the effects of the bacterial "activity" directly without
                                 73

-------
having to be concerned with the exact population size and metabolic
kinetics.  Most bacterial models are thus of little use to us since
they are concerned with population size or growth rate (Lacey and
Lawson, 1970; Schaitman, et al., 1969; Landesman, et al., 1966;
Lundgren, 1975).  Instead, we must devise a scheme that accounts
for the activity of chemoautotrophic bacteria directly.
     Treating the bacteria population as a single, idealized organ-
ism, we can represent the energy available to the population per unit
time, E , as being divided into two separate parts.  First, the
       A,
population must obtain sufficient energy in order to maintain its
current activity.  This maintenance energy, E , represents energy'
required to obtain nutrients from the environment, regenerate cell
components, and sustain metabolic porcesses.  E^ also includes the
energy required to shield each cell in the population from hostile
environmental conditions such as too low or too high of a pH or
high-salt concentrations.  If E  exceeds £1  then the remainder of
the energy available per unit time is put into population growth.
The total energy available to the population is partitioned
between maintenance needs and growth.

                       EA = SM + EG                             ™
An increase in cell numbers will only occur if E. > EL. and the cells
will cease to function or "die" and the population collapse when
E  > EL., where both E^ and E  have incorporated into them the
efficiency with which the specific organisms transfer energy from
external sources to internal uses.
                                 74

-------
     The energy requirement of an organism can be related  to  the
rate at which it consumes its energy substrate.  Arkesteyn  (1980)
found that Thiobacillus, ferrooxidans serves as the primary  auto-
troph in increasing the oxidation rate of pyrite.  These organisms
use the energy released from the ferrous-ferric oxidation reaction
as their only energy source.  He found that the sulfur-oxidizing
bacteria were incapable of oxidizing pyrite directly but could use
the sulfur released from chemical or T_. ferrooxidans induced
pyrite oxidation.  Therefore, in this model we'll assume that the
ferrous-ferric oxidation reaction is the sole energy source for
the pyrite-oxidizing autotrophs.  If we let RC represent the rate
at which the entire population oxidizes iron (moles of electrons/
unit time) and let AG represent the energy released in iron oxida-
tion (energy/mole of electrons) then:

                       EA = Rc AG - ^ + EG                     [391
which,  when the population is stable, (Ep - 0),  reduces to:
                                        (3
                       RC AG = EM                               [40]
Since we are assuming that the population can adjust itself rapidly
enough to always be in equilibrium with its environment, Eq. [40]
should always be true.
     We can combine the bacterially dependent terms in Eq.   [40]
by solving for AG:

                        AG = VRc • V                        I4l!
The magnitude of AG can be calculated if we know the iron ion
concentrations in the spoil solution.  AG can be found from:
                        AG = - FE                               [42]
                                 75

-------
where F is the Farady and E  (volts)  is  found  from the combined



Nernst equations for the exchange of one  electron in the  ferrous



iron oxidation and oxygen reduction half-cell reactions at  normal



pressure.
         0.46 - 8.6 x 10 5 T In
Y3[Fe3+]
                                [A3]
    >                                                  +        2+
where y-, y_ and y_ are the activity coefficients  for H  and Fe   and



Fe  , respectively.  Since AG is proportional  to  the driving force, E,



we will -consider the two interchangeable and  talk  of the energy as



though it were measured by the driving force, E  and thus let E.* = E



at constant cell population size.



     The value of E/J will vary in relation to the  environment of  the



organism, increasing when adverse conditions  are encountered.   This



can be represented by:

                              n

                   !£ = E'. / H  x.                             [44]
                    a.    mm .  ,   i                                J
                             i=l



where E'.  is the minimum value of E.' for ideal conditions and  x.
       mm                          rl                           i


represents the inhibiting effect caused by the ith inhibition agent.



x. varies from 1.0 for no inhibition to 0.0 for total inhibition



and subsequent cell "death" and population collapse.  For T.



ferrooxidans, we will consider  only 3 inhibition agents.   These are



defined as:



     x  H  Temperature dependence of cell viability.  T.



          ferrooxidans are mesophiles and based on work by



          Silverman and Lundgren (1959),  Landesmann, et al.
                                 76

-------
       (1966), Belly and Brock  (1974), Lundgren  (1975)


      and Malouf and Prater  (1961), we have assigned


      the peak cell activity  (x  =1.0) to be at 28°C,


      falling to zero at 4°C  (Ehrlich and Fox,  1967)


      and 55°C (Cathles and Apps, 1975).  A cubic


      equation was fit through these points such


      that


 x  =-1.23 x 10~5 T3 - 4.33 x 10~4 T2 + .0657 T -  .255       [45]


                                        4°C < T <  55°C


 x  =» 0.0                               otherwise


 Figure 13a shows, the calculated value of x  versus the


 normalized cell "activity" data from Belly and Brock


 (1974), Silvertnan and Lundgren  (1959) and Malouf  and


 Prater (1961).


x   = T. ferrooxidans are sensitive to pH.  At  low pH's, cell
 PH

      activity decreases probably due to acid attack of cell


      membranes.  At high pH's, activity also decreases


      perhaps due to a decrease in efficiency of metabolic


      processes that are tuned to low pH's.  From  the work


      by Silverman and Lundgren  (1959), Landesmann, et al.


       (1966), Schnaitman, et al. (1969), Ehrlich and Fox


       (1966) and Malouf and Prater (1961), the  maximum


      activity is between pH 2.5 and 4.0.  A quadratic


      expression was fit to the data given by Silverman


      and Lundgren (1959) and Schnaitman, et al. (1969) and


      is shown in Fig. 13b.  From this data:
                             77

-------
00
 to

 .8

 .6
-*

 .4

 .2

  0
                     a.
                      10   20   30  40o 50  60
                        Temperature °C
                             a.
1.0

 .8
                                            w .6
                                            O.
                                            X!
                                              .4
                           CM
                           O
                          X
            3
           pH
1.0

 .8


 .6

 .4

 .2


 0
                                                                            C.
             b.
  0  .05   .10  .15  .20  .25
    Oj Mole Fraction

            C.
           Figure 13.  Plots  of  the  inhibition factors (solid  lines).   a) x  versus  temperature ( C) ;

                       dashed line is  normalized data from Malouf  and Prater, 1961;  (•)  normalized data
                       from Silverman  and Lundgren, 1959; (±)  normalized data from Belly and Brock, 1974.

                       b)  x    versus pH;  (•)  normalized data  from  Silverman and Lundgren,  1959; (A)
                           pH
                       normalized data from Schnaitman et al.,  1969.   c) x   versus  oxygen mole fraction.
                                                                            U2

-------
     x   = -.348 pH2 + 2.26 pH - 2.66    1.54 < pH < 4.95        [46]



         = 0.0                           otherwise



     the maximum (x „ = 1.0) occurring at pH = 3.25.
                   pn


     Q  = Oxygen dependence of chemoautotrophic bacteria.  T.



          ferrooxidans are obligate aerobes whose activity



          ceases when 0_ is depleted..  At 25 C, Myerson  (1981)



          found their activity to decline when 0_ concentra-



          tions fell below 5% of saturated levels.  Thus, at



          normal atmospheric pressure:
               Xo2 = 1'°       Yo2
               Xo2 = Yo2'-01   Yo2
                                                                [47]
          The x   - 0  relation is shown in Fig. 13c






Thus, E' can then be written as:
the magnitude of E' varying with environmental conditions.  At this




point, we need to evaluate E'. .  No a priori relationships seem




obvious so instead we must turn to experimental data.




     We have found only one report in the literature with sufficient


                                       4

detail to evaluate E' .  .   Bailey (1968)  ran culture experiments with




pyrite solutions innoculated with iron-oxidizing bacteria.  From his




observations of ferric, ferrous and bacterial concentrations as a




function of time, we can evaluate E/..  Initially, his solutions showed
4

 Bailey,  J.  R.   1968.   Biological oxidation pyrite.  Unpublished M.S.

 Thesis,  The Ohio State University, Columbus, Ohio.
                                 79

-------
rapid bacterial growth  (doubling  time ^ 10 hr), but  as  the system



approached steady state conditions  the population  stabilized,  in



agreement with our model, and remained fairly  constant  thereafter



(AG = E£) .


                                            4
     Analysis of the results of Bailey 0.968)   are  shown in Tables



10 and 11.  The experimental results are divided into two  phases.



Phase 1 represents the growth phase of the bacterial population



when AG > EC,, while phase 2 represents the steady  state condition



when AG = E^.  To evaluate for E'.  , the data  from the  second  phase
           a.                    mm


of the experiment was used to calculate E and  EC, from Eq.  [43]  and



to find x , xn  and x   .  E'.  was then found by rearranging Eq.
         T   U —      pti    mxn


[48].  The average value for E".  from eight measurements  during



phase 2 was 0.173 volts (Table 10).  This value was  then used with



the data from phase 1.  Table 11 shows the analysis  of  the results



from five measurements during the intial growth phase.  The values



for EC. were---calculated by correcting the value for E'.  found in



Table 10 for x_, x.-  and x „.  Values for AG were found from Eq.
              T   02      pH


[43].  E' is the difference between AG and E.', in each  case E'>
        G                                   M                G


0.0 indicating that a surplus of energy is available for population



growth.   The last column in Table 11 shows the normalized  growth



rates of the bacteria.  Comparing the last two columns,  we  find



that the growth rate is roughly proportional to the  surplus energy,



E_ (especially after 224 hours) which is consistent with our simple
 I?


growth model.
                                 80

-------
Table 10.   Measured E., values and calculated x_, X- , x TT and
                     M                        I   U7   pn


           E .   values from Phase II of experiment by Bailey
            mm


           (1968).4

hr
593
639
688
758
833
880
929
1002
PH
3.10
3.05
2.95
2.85
2.80
2.80
2.80
2.70
XT
.80
.80
.80
.80
.80
.80
.80
.80
\
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
XPH
1.00
.996
.979
.954
.940
.940
.940
.905
**
.224
.231
.202
.230
.223
.209
.249
,240
E'.
mxn
.179
.184
.158
.175
. .168
.157
.187
.173
                                                    Ave. =  .173
                                 81

-------
00
           Table 11.  Measured and calculated coefficients for the bacterial "activity" model
                      from Phase I of experiment run by Bailey (1968) .

hr
224
362
412
439
459
pH
3.71
3.90
3.50
3.40
3.35
XT Xn XnH AG
1 U_ pH
.8QO 1.00 .935 .312
.800 1.00 .861 .301
.800 1.00 .987 .258
.800 1.00 1.00 .233
.800 1.00 1.00 .226
C ^
.231 .081
.251 .050
.219 .039
.216 .017
.216 .009
RR§
0.42
1.00
0.55
0.39
0.26

V
rl
\
is calculated

= EA - V
based on E . of 0.173.
min







            §
             RR  is  the measured  normalized growth rate of  the bacterial  population.

-------
     On the basis of this model of bacterial  growth, we  can  calcu-



late the ferric/ferrous ratio at any time and for any  conditions



by rearranging Eq. [43] and by using Eq.  [48]  to calculate Ej.
             2+     -v   L  J  n
           Fe       T3         2       T 8.8 x 10 "


                               2+
The actual oxidation rate of Fe   by bacteria must be such  that Eq.



[49] is satisfied.  That is the valueof K^ in Eq. [35] is adjusted



such that the ferric/ferrous ratio calculated in Eq. [49] is met.



     To extend the model of pyrite oxidation beyond steady-state



conditions of bacterial population "activity", allowances for popu-



lation growth and decline must also be included.  Population growth



is modeled by allowing the "activity" (K,,) to double every  12 hours



if conditions are favorable (e.g., AG > E/l) .  This compares to the


                                                         4
doubling times for population size found by Bailey (1968)   of



between 7.5 and 51 hours.   When environmental conditions deteriorate



in relation to sustaining population size, the population "activity"



is assumed to respond instantly and K^ decreases.





Chemidal Species



     Chemical species that are followed in the model include H ,


  2+    3+    2-
Fe  ,  Fe  , SO  , HSO, and the complexes of ferric iron.  Concentra-



tions for each are calculated using the activity coefficient for the



ion as calculated by the Davies equation (Davies, 1967, p.  60).
y. = exp
           - 0.5 z
- 0.31
                                                                [50]
                                 83

-------
which is satisfactory for ionic strengths less  than  about  0.3 moles/



kg.



     Ferric Precipitation.  The precipitation and complexation of



ferric iron must be considered.  One such step,  is the precipitation



of ferric hydroxide:



                  Fe3+ + 3H20 ->- Fe(OH)3  + 3H+                   [51]

                                       c



The change in standard free energy per mole of  electrons for this



reaction is AG  = -210.kJ and thus is spontaneous as written.  The
              K


equilibrium concentrations of Fe   can be calculated from  this



reaction, but the very low values for ferric concentration don't



agree well with measured values from acid—mine  drain-waters



(Nordstrom, et al., 1979), when crystaline forms of iron hydroxide



are used.  This is probably due to one of two factors, either the



kinetics of the above reaction have not been considered and may be



slow or the crystaline form of Fe(OH)_ may not be the controlling


              3+
species for Fe   solubility.  Amorphous Fe(OH)_ or some other



species may instead be the control (Langmuir and Whittemore, 1971).



     Fresh solutions can be significantly supersaturated with respect



to ferric hydroxide precipitates (Langmuir,  1971),  but this seems



to be primarily a function of nucleation kinetics.   In strip mines,



where nucleation surfaces are abundant, precipitation of ferric



hydroxide would not be a problem.   Analysis  of acid-mine drain-



waters has, in fact, shown this to be true (Nordstrom,  et al.,
                                 84

-------
 (1979), the waters seldom being supersaturated with respect to



Fe(OH),     :
      3 amor




          Fe(OH)3 anor + 3H+ •+ Fe3+ + 3H2



4.  Yet many measurements of total ferric iron concentrations exceed

                                   A
the calculated values (Bailey, 1968 ; Pionke, et al., 1980).  This



is primarily due to ferric complexes.  Ferric iron complexes strongly



with hydroxides and sulfates, which can raise the aqueous concentra-


          3+
tion of Fe   a hundred to a thousand times at moderate pH's (pH ^ 5).



Thus,  the complexed forms of ferric iron must be accounted for.



     Another reason we must take account of iron complexes is that



in most kinetic studies of pyrite oxidation or bacterial catalysis



of ferrous iron, ferric complexes have been ignored and values



reported as Fe   concentrations, are in reality, total aqueous


  3+    3+
Fe  ,  Fe_ .  Thus, the reaction kinetics are based on total dissolved



ferrous iron not just the Fe   ion.



     For calculating the pyrite oxidation rate in this model, the


  3+
Fe   concentration in solution is used in the rate expressions.
                                 85

-------
  3+
Fe   is found in one of two ways.  If the hydroxide solid is present




Eq. [52] is used to calculate the ferric iron concentration from




which the concentration of the complexes can be found.  The solid




iron hydroxide is then adjusted for changes in total dissolved  iron




minus pyrite dissolution.  If no solid hydroxide is present, the




dissolved iron species are adjusted so that the mass balance is




preserved and the species are in equilibrium with each other.   The




complexes considered in this model and their equilibrium constants




are shown in Table 12.






Removal of Reactio'n Products




     The dissolved products of the pyrite and bacterial-oxidation




reactions must be removed from the spoil profile.  Two possible




methods exist for H  removal; acid neutralization with a possible




increase in reserve acidity and leaching by deep percolation.  For




sulfate and iron, only leaching is assumed to remove the ions from




the profile.



     Acid Neutralization.  Acid produced by pyrite oxidation may be



neutralized or modified by the surrounding rock matrix or gangue



material (Rogowski,  et al., 1977).  These H  - gangue reactions




appear to be quite significant.  Lovell, et al., (1978) found that



in situ neutralization reactions were significant in all three



mine sites they studied,  effectively removing all of the acidity



in two of the sites.  Although not of  the same magnitude,  consider-




able neutralization and transformation to reserve acidity was




observed in a small column-leaching study (Pionke,  et al.,  1980).
                                86

-------
Table 12.   Iron complexes and log of the equilibrium constants
           for  their formation and for sulfuric acid dis-
           sociation and amorphous ferric hydroxide precipitation.

Reaction logK
H._0+Fe -*->-Fe(OH) +H - 2.94
2H-0+Fe -*-»-Fe(OH)9+2H - 5.70
2H?0+2Fe -^-HFe,, (OH)9 +2H - 5.22
3+ +
3H20+Fe «-»Fe(OH)- +3H -12.
aq
4H20+Fe3+^Fe(OH)~+ 4H+ -21.60
Fe +HSO.-«->-FeHSO, 0.60
4 4
4 4
3H++FeOOH+H2CK->-Fe3++3H 0 4 . 9
Reference
1
1
1
2
2
1
1
3

 References  refer  to:   1)  Sapieszko,  et  al.,  1977;  2)  Baes  and
 Mesmer,  1976;  and 3)  Nordstrom,  et  al.,  1979.
                               87

-------
Although not reported in the study, approximately 10% and 20%



of the H  formed from pyrite oxidation was consumed by neutraliza-



tion and transformation to reserve forms of acidity (mostly



aluminum), respectively (H. B. Pionke, personal communication).



Column studies by Braun, et al., (1974) have shown similar results.



     Carbonates may be very reactive with H  and are capable of



maintaining the water percolating through them at pH's near



neutrality (Freeze and Cherry, 1979).  The carbonate-H  reaction is



treated separately from other gangue-H  reactions because of this



large neutralization capacity and because of their use as liming



materials on reclaimed stripmines.   The reaction of carbonates



with H  is assumed to be controlled by the chemical kinetics.  The



rate of carbonate consumption per unit area of carbonate can be



described by (Wentzler,  1977).5



                           Qc = - Kc [H+]                       [53]


                                                      '  -1   -2
where 0  is the rate of  carbonate consumption (moles-sec  -cm  )
       c

                                              —6     2
and K  the first-order rate constant (2.9 x 10   L/cm -sec).
     c


Assuming the carbonate is in the form of distinct, pure fragments



that dissolve completely with time, we can describe the consumption



of carbonate mathematically in a manner similar to the consumption



of pyrite [Eq.  23] but without the diffusional control term (t ).



For spherical particles, Levenspiel (1972) gives the rate of change



of the fraction of carbonate sphere remaining, X , to  be:
 Wentzler,  T.  H.   1977.   A study of the interactions of limestone

 in acid solutions.   Unpublished M.S. Thesis,  The Pennsylvania State

 University,  University  Park,  Pa.
                                 88

-------
                       3  X 2/3  K [H+]

                     =  	°p  R C	                            [54]

                           c   s


 where p  is the molar  density  of the sphere and R  its original
       c                                          s


 radius.  The rate  of carbonate consumption and thus H  consumption



 can .then be calculated from Eq.  [54]  and the number of carbonate



 fragments per volume.




     Reactions between H  and  other  gangue minerals that  need to be



 considered include neutralization of  H   to form weak acids,  such



 as orthoclase reacting to  form kaolinite (Birkeland,  1974, p.  61).


                   +                 .           .        H*




   (orthoclase)                (kaolinite)



 or the exchange of H   for  other  cations  on clay colloids  and  oxides.



 Reserve acidity in the form of  aluminum  and aluminum hydroxides  may



 also serve as a sink for hydrogen.  In general,  the amount of  pH



modification caused by the gangue would  depend  on  the mineralogy



 and pH.   For this model a  simple empirical  relation was used  to



 partition the H  produced  into active solution-H   and combined



neutralized- or reserve—H .    The relation  has  the form:



              AH+ = AH*  (1.0 -  exp (G_(G -pH))  )                 [56]
                       K              15   A

        4-       +                                   +
where AH  and AIL  are  the actual increase  in free  H  and  the amount
        A.   f    J\


of H  produced by all  the remaining reactions respecively, and G.



and G,, are empirical constants.  G. has  the  physical  significance
     jj                            A


that when the solution pH equals G. all  the  H   produced is consumed
                                  A


by the gangue and the  pH remains constant.   G_  is  a  scaling factor
                                             D


which determines the rate at which the condition of  constant pH  is
                                 89

-------
 approached.  The shape of  the neutralization  curves used  in this




 study are shown in Fig. 14.




     Leaching.  Water percolating through  the reclaimed-spoil




 profile will flush the oxidation products  out of  the profile.




 Intuitively, it would seem unrealistic to  assume  that Darcy type




 flow governs the movement  of water in coarse-spoil  material.




 Indeed, ElBoushi (1975), measuring infiltration into coarse-stone




 rubble, and Rogowski and Weinrich (1981), modeling  infiltration




 into reconstructed-spoil profiles, have  demonstrated the  short-




 comings of Darcy's Law for describing water movement in coarse




 materials.  In addition, since it is assumed  in this model  that




 the water content of the spoil remains constant over time,  infil-




 tration studies like those quoted above are of limited help here.




 Considerable progress has been made in understanding and modeling




 the leaching of coarse materials (Rao, et al., 1980  a and b), but




 such treatments are beyond the intent of this model  where only a




 simple removal mechanism of solutes is desired.




     Instead, we first recognize that the model will be solved




 by finite-difference techniques and therefore divide  the spoil




 profile into a number of horizontal layers.  Within  each layer,




 the water is assumed to be completely mixed and homogeneous, which




 represents a major simplification of the leaching process.  In




 strongly structured soils,  it has been shown that infiltrating




water can penetrate very deeply in a short time with very little or




no interaction with the intervening soil layers (Quissenberry and
                               90

-------
                        NEUTRALIZATION  CURVE
          a
1.
                           2.      3.      4

                            EXPECTED PH
Figure 14.  Neutralization  curves as calculated from equation in


           text.  G_ equals 1.0 in both plots
                  o
                        G  equals  2.5 in
                         A
           plot 1 (•) and 2.8 in plot 2  (•).
                           91

-------
 Phillips, 1976;.Bouma and Dekker,  1978).   Incorporating  this

 phenomena into the leaching process, we  assume that  water  leaving

 any layer is partitioned to the underlying layers  in proportion to

 the inverse of the distance separating them,  1/d.  Water traveling

 between two layers does not interact with  the intervening  layers.

     For an n layer system, water  infiltrating the surface is

 partitioned among all layers proportional  to  the inverse of the

 depth to the layer (the distance is measured  from  the surface  to

 the center of the layer).  The fraction of  infiltrating  water  enter-

 ing any layer, i, is:
                                   n
                  fraction = 1/d./ I 1/d.                        [57]
                                1  3-1   J

 Water leaving any layer will also be partitioned proportional  to 1/d

 to the underlying layers.  The fraction entering any  layer j from

 layer i where layer 1 is at the surface is:
                                        n
             fraction. . = l/(d.-d.)/   £  l/(d -d.)              [58]
                     13       J   1  k=i+l     *  x

     The amount of water leaving any layer  is  equal to the amount

 entering,  which is just the sum of all the  fractional contributions

 from overlying layers, the 0 layer indicating  surface precipitation:

                   j-1                n
        outflow.  =  I   [l/(d.-d.)/   E  I/(d-d.)] outflow..    [59]
               J    i=0       3  X   k=i+l     k  1          X
      +    2-f
     H ,  Fe  , total sulfur, total dissolved  ferric iron, and  the

acid-neutralization products are carried with  the percolating water

and leached from  the mine profile.  The chemical load entering any

layer is  equal to the sum of water coming from each overylying layer
                                 92

-------
multiplied by the species concentration in the layer.  The amount of



each constituent leaving a layer is just the concentration of the



species in the layer multiplied by the outflow.



     An example of leaching as described by Eq. [58] is shown in



Fig. 15.   The figure shows the normalized concentration of a non-



interacting solute in the water leaving a 10-meter deep profile



that is divided into 20 layers.  The profile has a water-filled



pore volume of 0.13 and the water is infiltrating at a constant



rate of 50 cm/year.  The initial and boundary conditions for the



solute concentration are:



                C. = 1.0       1 < i < 20; t = 0.0

                 1                                              [60]

                C.  = 0.0      t > 0.0
                 in


Figure 15 shows a considerable smearing of the solute front, with



the concentration slowly tailing off to zero after more than 3



pore-volumes have passed through the profile.  The method of in-



verse weighting of water flow between layers mimics a system where



diffusion - dispersion processes are of considerable magnitude



(Rose, 1977).





Solution Method



     Equations [27], [28], [30], [31], [34], [35] and [49] serve as



the basis of a model for describing in situ pyrite-oxidation.



Equations [53], [56], [59] and the expressions  listed in Table 12 are



used to adjust the concentrations of each species and calculate the



leaching rate of the reaction products, where water is assumed to
                                 93

-------
       1.0
       0.8
                           PORE  VOLUMES
                         1              2
       0.6
   'EFF
       0.4
       0.2
       0.0
                        1000           2000
                                DAYS
                                           O	Q
                                             3000
Figure 15.
Example leaching curve showing the normalized concen-
tration versus water-filled pore volumes.  Profile was
10 meters deep, divided into 20 layers with a water-
filled pore volume of 0.13 and an infiltration rate of
50 cm/year.
                             94

-------
move downward only at a uniform annual rate.  For air movement in




the vertical direction only, Eq. [27] is a nonlinear, nonhomogeneous,




parabolic equation.  Equation [27]  can be solved using an implicit,




finite-difference technique in which the reclaimed-mine profile is




divided into N horizontal layers.




     The solution of these series of equations is very unstable due




to the complicated interdependence of the variables and to the non-




linear form of some of the equations (e.g., Eq. [49].  The solution




procedure involved estimating values for all of the variables at




time t + At, calculating the diffusion coefficients and source/sink




terms from Eq. [28], [30], [31], [34] and [35] and then solving Eq.




[27].  The calculated values for the 02, CO , Fe  , Fe   and H  con-




centration were then used to recalculate the diffusion coefficients




and source/sink terms and Eq. [27]  was solved again.  This procedure




was repeated until the values for 0», CO-, Fe.  , Fe   and H  con-




centrations had converged.  Values  for IL, and the leaching rates of




each chemical species were then updated for t + At and the entire




process repeated a maximum of 25 times or until all values had con-




verged.  The iterative scheme has proven to be stable and capable




of handling large (^ 6 months) time steps for all cases studied.




     A comprehensive testing of the model, comparing the model




results to measured values, was not possible due to the lack of a




sufficiently complete data set for  an active, oxidizing-leaching




pyrate body.  However, several simulations were run to determine the
                                 95

-------
interaction of several parameters in the model and to compare the



simulator results to known stripmine behavior.  For all simulations, a



reclaimed profile, 10 meters deep and divided into 20 equal horizontal



layers, was used.  Unless otherwise noted, each layer contained 75%



(wt/wt) coarse fragments (£ = 1 cm), containing a pyrite content of



0.25% (wt/wt) and a 20% water-filled porosity.  The bulk density of



each layer was 1800 kg/m  and a constant water saturation of 70% was



used.



     At the beginning of each simulation the profile air-space contained



0.21 mole fraction 0  and 0.0003 CO .  The spoil water contained 0.28


       2+                                       3+
mg/L Fe  ,4.8 mg/L sulfate, less than 1 mg/L Fe   and was at a pH of



5.0.  No soluble  ferric hydroxide was assumed to be present.  These



values for dissolved species may be unrealistically low for spoil water



in newly reclaimed mine-sites because of the large content of soluble



salts in spoil material  (Lovell, et al., 1978), but were used so that



the  formation of  oxidation products could be observed more easily.  After



the  pyrite started to oxidize, the water infiltrating the surface of the



profile was assumed to be at a pH of 5.0 and contain no dissolved iron,



with sulfate at a concentration of 4.8 mg/L.  In the simulations where



iron-oxidizing bacteria  are active, a 32 day "innoculation" period was



assumed.  During  these first 32 days, the bacteria were assumed not to



be active.  After this time a base "activity" for the bacteria,



represented by K = 10~   , was used.  That is,  the base rate of ferrous



oxidation catalyzed by bacteria was set to less than 0.1% of the



minimum chemical  oxidation rate of  ferrous iron.
                                    96

-------
                     Results and Discussion






     Two major aspects of the model were investigated.   The  signi-




ficance of iron-oxidizing bacteria was studied, since the way  in




which the model calculates their "activity" from  the energy  subtrate




and environmental conditions is unique.  Several  other simulations




were also performed to illustrate the flexibility of the model and




compare the model output to expected behavior.






Role of Iron Oxidizing Bacteria




     Since this model is unique in the way that the activity of




iron-oxidizing bacteria is accounted for, the effect of bacteria on




pyrite weathering as predicted by the model was examined.  In the




first simulation, Run 1, iron-oxidizing bacteria were assumed not to




be active and thus the only source of ferric iron was chemically




oxidized ferrous iron.  Figure 16 shows the fractional amount of




pyrite consumed within the entire profile from 0  days, when pyrite




oxidation begins, to 10,000 days or approximately 27 years.  The




pyrite is consumed slowly, with just over 22% consumed after 10,000




days, the consumption is due almost exclusively to direct 0,, reaction.




The ferric iron concentration never exceeds 0.5 ymoles/L at any time




in any layer.  The rate of pyrite oxidation due to ferric iron is




never greater than 0.07% of the direct 0., pyrite oxidation rate.




The rate of pyrite oxidation is slightly faster at initial times, but




is virtually constant (slope in Fig. 16 is constant) after 3000 days.




A similar trend is observed in Fig. 17, where the rate of total iron
                                97

-------
                                 86
(K
C
Tl
n
03
o
 O
 3
 o
 >-tl

•a
 O
 X
 H-
 o.
 H-
 N
 fD
 O.
 •3"
 H-
 3
m
3
 it
 O
 H-1
 fD
 CO
 c
 CO
 O
 b
  FRACTION  PYRITE

pop
ro       
        o  01
        >  O
        •<  o
        CP  o
 so
 CD
 3
 Cu

 pa
o
o
o
o
 p
 In
—r
p
CD
                                                C   C
                                                Z   2

                                                PO   —•

-------
                                        2

being leached from the profile  (moles/m -day)  is plotted  against  time



for the 0 to 10,000 day period.  A marked peak in  the leach  rate  is



observed at 2^.00 days (5.75 years), which then levels off slowly  to


                                                           2+
an almost constant value.  The average  concentration of Fe  within



the profile after 2J.OO days of pyrite oxidation is 0.005A moles/L.



The average pH of the spoil water is 1.99.  After 10,000  days of



weathering, the ferrous iron concentration and average pH of the



spoil water are 0.0032 moles/L and 2.22, respectively.  While the


  2+
Fe   concentrations seem to be well within the range of values found



for water samples removed from below a  reclaimed stripmine,  the pH



values appear to be about 0.5 of a pH unit lower than measured



values (Rogowski, et al., 1982).




     In the second simulation, Run 2, iron-oxidizing bacteria were



allowed to interact with the pyrite system as  predicted by the



model.  Figure 16 also shows the amount of pyrite consumed over time



for this simulation.   The consumption of pyrite is almost  identical



to the results of Run 1, where no bacteria are  active, the relative



increase in oxidation being less than 3%, which represents an abso-



lute increase in pyrite oxidation of 0.7% after 10,000 days.  The



plot for Run 2 in Fig. 17 shows the same behavior,  almost no differ-



ence is observed in the rate of total-iron removal for Run 1 and



Run 2.



     The lack of impact by the bacteria in this case is explained



by the fact that the pH of the spoil water drops rapidly below



2.6 for all layers after only 200 days and remains  between 1.8 and
                                99

-------
H
O
O
l.f
<£
1'°
UJ
io.e
O
o
x 0.6
z
o
£ 0.4
H
O no
1- U.2
0.0
(
i i i i i i i i l
o RUN 1
o RUN 2
-

*P*A*to^
cP ^cP^i
o
o
cf
D
IP
1 1 1 1 1 1 1 1 1








D 5000 10000
DAYS
             Figure 17.  The rate of leaching out  of the profile of all iron species versus time for Run 1

                         and Run 2.

-------
 2.3 for most layers for the remainder of the simulation.   Examina-
 tion of Fig. 13b shows that at these lower pH's, x _ drops to
                                                  pH
 between 0.2 and 0.7 indicating decreased "activity" of  the bacteria
 because of adverse H  activity and a decreased role for ferric iron -
 pyrite oxidation.  However, it is misleading to believe that the
 bacteria exert no influence under these conditions.  Figure 18 shows
 the rate of pyrite oxidation (moles/cm -sec) as a result of direct
 02 and ferric oxidation for Runs 1 and 2 after 5 years of  oxidation.
 Although the overall rate of pyrite oxidation is similar,  in Run 1,
 0- is the only important oxidizer while in Run 2 bacterially
 produced ferric iron begins to become important in the surface layer,
where the pH remains higher due to infiltration of water at pH 5.
 This increase in ferric iron concentration can also be seen in Fig.
 19 where the ferric/ferrous concentration ratio in the water leaving
 the profile is plotted against time.  The ratio is about 100 times
 greater during Run 2 than during Run 1.
     Two additional simulations were run under the same conditions,
but where interactions between the acid produced by pyrite weather-
ing and the rock matrix were assumed to occur.   Figure 20 shows the
fraction of pyrite consumed versus time for these two runs  compared
 to Run 2.   In Run 3, the gangue material is assumed to react with
H  so that the pH never falls below 2.5 (G  = 2.5).  In Run 4,  the
                                          A.
system is  maintained at a pH of 2.8 or above (G  = 2.8").  The H
                                               A
produced by pyrite oxidation, but not assumed to contribute to  the
                                101

-------
    0
    1
    2
    3
J  *
f  5
Q.
S  6
    7
    8
    9
   10
           Pyrite Oxidation  Rate x1014 (moles/cm3-sec)
          01               2              3
                    ^XX^XXXXXXXXXX
           vXXXXXXXXXXXXXXXXXV
           \XXXXXXXXXXV\X1
           \XXXXXXXX1
           ^OOsXXXXXXXXXXXX
          xxxxxxxxxxx-
           XXXXXXXXXXi
           XXXXXXX1
          XXXXX1
Figure 18.  Oxidation rate of pyrite for each layer after 5 years of
          oxidatio'n.  Fine cross-hatching is total oxidation rate
          for Run 1.  Open cross-hatching is rate of oxidation by
          direct oxygen reaction with pyrite in Run 2.  White cross-
          hatching on black is oxidation rate due to ferric iron
          reaction with pyrite in Run 2.  Below 1 meter, ferric
          oxidation is negligible.  Below 4 meters, oxidation rates
          of Run 1 and Run 2 are identical
                          102

-------
IU
UJ
1—
X
o
UJ 10~4
0
<
A lcf5
Ul
rO
UJ
U_
10~6
r ' ' ' ' ' L 00°°^
D oQo ao o -
"^ nO^^ ~
~o Qo° ~
a o°
~° oa"00
_ ° o RUN 2 -
_ —
— *~"
™~ D 1 I Kl 1
« r\ w mi rt O "™"
•-• f\ rt O O ^ "•"•
~ ' 0ooOo0000° I
Ooooooo° °
O o 1 1 1 1 1 1







10OO 3000 50OO 7000
DAYS
Figure 19.   The ratio of ferric iron to ferrous iron in the water
            leaving the profile for Run 1 and Run 2.
                             103

-------
H  activity,  is represented by the reserve acidity in the spoil



solution.


     Raising  the pH of the system has the effect of increasing



the x _ function value and thus increasing bacterial "activity."
     pH


In Run 3,  where the pH is maintained at 2.5 or higher, Fig. 20



shows a considerable increase in the amount of pyrite consumed over



time.  After  10,000 days, 30.5% of the pyrite is consumed for a



relative increase of 36% over the unbuffered, bacterial Run 2.



Most of this  increase occurs before 2,000 days, after which the rate



of pyrite consumption is very similar to Run 2 (plots are parallel).



When the system pH is maintained at 2.8 and above, 36% of the pyrite



is consumed after 10,000 days for a relative increase of 61% over



the unbuffered run.  Again, most of this increase over Run 2 occurs



before 2,000  days, but the rate of pyrite oxidation continues to



be greater than Run 2 or Run 3 as indicated by the continuously



diverging plots.


     To maintain the pH at 2.5 and 2.8, a considerable proportion of



the H* formed by Eqs. [18], [19] and [21] must be converted to


acidity by reactions with the gangue.  In Run 3, 39% of the acid



formed from pyrite oxidation is converted to acidity after 10,000



days of oxidation.  Again, it should be remembered that in this



model the acidity term represents both neutralization and conversion



to reserve forms of acidity such as Al  .  In Run A, 56% of the acid
                               104

-------
                  0
10000
Figure 20.   Fraction of pyrite  oxidized witliin  the  entire  profile  versus  time for Runs 2,  3, and A,

-------
from pyrite oxidation is converted to acidity in order to maintain


the pH at 2.8 or higher.  Although these percentages seem high,


they certainly fall within the range of observed behavior for strip-


mine lands (Lovell, et al., 1978).


     Figure 21 shows the rate of removal of total iron over time


for Run 3 and Run 4.  Run 2 is replotted on this figure for ease of


comparison.  For Run 3, where the pH is maintained at 2.5 or higher,


the peak in the total iron-leaching rate occurs at 1,750 days


(°u 5 years).   The rate of leaching is considerably greater than Run 2


throughout most of the simulation, being almost double when the maxi-


mum rate is observed.  Although greater in magnitude, the shape of


the rate vs.  time plot for Run 3 is very similar to Run 1 and Run 2,


Run A, however, shows a considerably different behavior.  The plot


for Run 4, where the pH is maintained at 2.8 or higher, shows a


tendency for  the rate to reach a plateau value of about 0.0065

       2
moles/m -sec  and then slowly oscillate around this value while the


plateau rate  slowly decreases with time.  This behavior is due to


two factors.   As 0_ penetrates a layer at a sufficient rate to


permit bacterial activity, the ferric iron concentration increases,

              *
which increases the overall rate of pyrite oxidation, which in


turn increases the total iron in solution and in the leachate.  This


process is what is occurring within the 250 to 300 cm layer between


3,300 and 4,000 days in Run 4.  A point is reached, however, where


the bacterially catalyzed-ferric iron concentration within the
                                106

-------
                                                5000
                                                DAYS
10000
Figure  21.   The  rate of leaching out  of  the  profile of all iron species versus time for Runs 2,  3,
            and  4.

-------
 layer .exceeds  the  ferric iron solubility  as  controlled by amorphous



 ferric hydrooxide  at pH 2.8.   The  ferric  iron then starts to  pre-



 cipitate,  removing iron from  solution  and from the leaching water.


                                                           3+
 The  rate of  leaching then starts to  decline  because the Fe   oxi-



 dant is maintained at a stable concentration while the pyrite



 concentration  decreases.   From Eq.  [26],  where t   is  greater  than



 t   (pyrite oxidation rate is  controlled by diffusion),  it can be
 W


 seen that  as X decreases  the  rate  of pyrite  oxidation decreases.



 This is what is occurring between  4,000 and  5,100  days,  where X,



 for  the layer  between 250 and 300  cm in which the  maximum rate of



 pyrite  oxidation is  taking place,  decreases  from 0.646  to 0.449.



 After 5,100 days the pyrite concentration in the layer  is suffi-



 ciently low that the oxidation rate and consumption of  0«  has



 decreased  to a point that increased levels.of  0- are  diffusing



 to the  next deepest  layer raising  that layer's  0«  concentration.



 The  process is then  repeated  in the deeper layer.   The  iron con-



 centration in  the deeper  layer  and the profile  as  a whole  and in



 the  leach water starts  to increase again.  Thus, the  oscillating



behavior of Run 4 in Fig.  21  is due to the division of  the profile



into distinct layers, while the constant value  of  the leaching rate



is controlled by the  precipitation of ferric  iron within  the profile



when the pH is maintained  at  2.8 or higher.   This  is  the first simu-



lation discussed where  iron precipitation is observed.



     A better indication  of the rate of pyrite  oxidation and acid



leaching for Run 4 compared to Run 3 is shown in Figure 22, where
                                108

-------
the total acidity being removed from the profile in. the leach water



versus time is plotted.  The total acidity is the sum of the con-



centrations in the leachate of H , one-half the acidity value



calculated from the gangue-H  interaction, and three times the ferric



iron values.  One-half of the acidity value is used in calculating



the total acidity in an attempt to partition the gangue-H  inter-



action between neutralization processes and the formation of reserve



acidity.  As can be seen in Fig. 22, the two plots are of similar



shape with the plot for Run 4 indicating an increase in the rate of



acid leaching of about 30% over Run 3 after 1,950 days of pyrite



oxidation.



     One other simulation was run to investigate the response of



the bacterial "activity" model.  Run 5 was conducted assuming no



gangue-H  interaction but where the x „ function [47] was lowered
                                     ptt


one-half pH unit so that the maximum, x   = 1.0,  occurred at pH
                                       pti


2.75.  This simulation was run because the mathematical form and



value of the coefficients in Eq. [48] represent only a first



approximation and are not absolute.  Differences in bacterial strains,



substrate type, and means of measuring bacterial activity may cause



discrepancies between the leaching studies cited in estimating the



x. values and bacterial activity within stripmine material.  Figure



23 shows the plots of pyrite consumed versus time for Run 5 and



Run 3.  The plots are of similar magnitude, although buffering the



pH at 2.5 or higher results in greater consumption of pyrite at
                                109

-------
OTI






to
3
0.
-C-
•







































(TO
C
ID
NJ
NJ
TOTAL ACIDITY |MEQ/M2-DAY)
H ,.
sT o 5 £
n
to
rr
(D
0
(-h
h-»
to
n_
•— '
H-
3
O
C
rr
O
t-h
rr
3-
m
Q
2 o £
H- > O
n> -< o
(ft O
o
fT
(D
•o
0
^T
ro
3
rr
H-
03
l^i

to
3
a.
to
D
rr
H-
<
to
o _
- 8
"^V- 1 1 1
~"***CH f\ -.
Or^P ri
°^% 5Qo 0

°°o ^\
o 1
/ oD*
0° 0°
0° rP°
/ °°
0 0°°
O rP
- s °
O Q
O n
8 8
8 8
fs
8
8 9
0 5
o n
Q Q
o 8 a o
O 0 «™
o Q *^ ^^
O n " ^ ^~
o 5 z z
o o
° 8 ^ w _
o 8
Q O
O n
Q U-
o fi
O 0
0 S
109 i

H) O
2 °
i
pa
c
3
CD

-------
         u
         Q.
            0.7
            0.6
            0.5
            0.4
         2  0.3
         o
         <
         or
             0.2
             0.1
            0.0
               0
                                                      o  RUN 3



                                                      a  RUN 5
             Qa
                                          j	I
5000


DAYS
                      J	I
10000
Figure 23.   Fraction of pyrite oxidized within entire profile versus  time  for Run 3 and Run 5,

-------
early times, but less consumption after 8,800 days or after about




28% of the pyrite is consumed.




     Runs 1-4 indicate that, at least for the parameters used, the




iron-oxidizing bacteria increase the pyrite oxidation rate only if




the pH is maintained above ambient levels of interactions with the




rock matrix.  Removal of H  by leaching alone (Run 2), was insuffi-




cient in preventing the pH in dropping to levels where the bacteri-




ally produced ferric iron only compensates for the loss in direct




0_ pyrite oxidation caused by the drop in 02 concentrations due to




bacterial consumption of 0--  The oxidation rate in Run 4, where




the pH is maintained at 2.8 or higher is controlled intermittently




by ferric solubility and bacterial "activity," with bacterial




"activity" controlling the oxidation rate until the solubility of




ferric iron is exceeded.  The solubility then controls the oxida-




tion rate until bacteria start to increase in the next deepest




layer because of increased 0? levels.






Model Sensitivity and Versatility




     Although it is beyond the scope of this paper to perform an




extensive sensitivity analysis of all the parameters in the model or




to model all possible management alternatives to reclaiming a given




mine profile, several possible combinations of parameters were




studied to demonstrate the versatility of the model.




     Effective Diffusion Coefficient.  Ultimately, the rate of pyrite




oxidation can exceed the rate at which 02 is supplied to the oxida-




tion zone.  In this model, 0- is assumed to be transported through




the mine profile by gas diffusion which is very sensitive to the





                                112

-------
air-filled porosity and tortuosity of  the flow path.   To  observe the

effect of varying the porosity and tortuosity on  the results,  two

simulations were run where the air-filled porosity from Runs 1-5

was doubled and the tortuosity was halved.  Examining  Eq.  [27],  it

can be seen that this combination results in increasing the effective

diffusion coefficient, D  ", by a factor of A.  In Run  6,  iron-
                         2
oxidizing bacteria are assumed not to be active within the profile,

analogous to Run 1.  For Run 7, bacteria are assumed to be active

as calculated by the model, with the spoil solution pH maintained

at a pH of 2.5 or higher (G.  = 2.5).  This is comparable  to Run  3.

     Figure 24 shows the plots for Run 6 and 7 for the consumption of

pyrite over time.  Run 3 is included for reference.  The  oxidation of

pyrite increases significantly with the increase in the effective

diffusion coefficient.  Run 6, without the bacteria, shows a signi-

ficant increase over Run 3, with 40% of the pyrite consumed after

10,000 days.  'When the bacteria are allowed to be active, Run 7

shows that a very large increase in the amount of pyrite  consumed

is obtained, with 70% of the pyrite consumed after 10,000 days.

This represents a 133% increase in pyrite oxidized over Run 3 and

a 75% increase over Run 6.   At the reduced air-filled porosity,

when bacteria were included at a buffered pH of 2.5 the amount of

pyrite oxidized increased by 24% over the case where no bacteria

were present (Run 3 compared to Run 1, Figs. 16 and 20).  Thus, at

the greater air-filled porosity, the effect of bacteria is increased

(75% increase in pyrite oxidation versus a 25% increase).  The
                                113

-------
00
Tl
i-l
01
O
rr
H-
O
3
O
b
   FRACTION  PYRITE

o      o       o
 O
 X
 N
 fB
 O.
3-
H-
3
T)
H
O
CO
c
CO
3
ro

i-h
O
i-t

33

3
CO

UJ
3
a*

-------
effect of increased pyrite consumption is  even  greater at  early

times.  After 500 days, 3 times as much pyrite  is consumed in Run  7

than Run 6.

     The effect of increased diffusion can also be seen in Fig. 25

where the rate of total iron leaching from the profile is plotted
                         *
against time.  The plot for Run 3 is also  shown for comparisons


with Runs 6 and 7.  The leach rate of iron shows a similar pattern

over time for the three runs, however, the magnitude is greater at

all times for Runs 6 and 7 compared to Run 3.  The leaching rate is

particularly greater in Run 7 where iron-oxidizing bacteria are

present.

     The effect of increasing the diffusion rate is perhaps better

shown in Fig. 26 and 27 where the 0? concentration with depth after

1,825 days (^ 5 years) of pyrite oxidation is plotted.  Figure 26

shows Runs 1 and 6 where bacteria are not  allowed to be active.  02

concentrations at the deeper depths are much higher for Run 6.

The shape of the curves reflect a process  where the consumption term

is proportional to the concentration (Kirkham and Powers, 1972).

This is in contrast to Fig.  27 where Runs  3 and 7 are plotted.

These 0« profiles show a near linear decrease with depth similar to

diffusion through a non-interacting zone to a layer of complete

0,, consumption.  0^ does not penetrate as  deeply in the profiles

where bacteria are active due to the concentration of pyrite oxida-

tion in the first few layers of high pyrite concentration rather

than slower oxidation rates distributed over several layers as is

the case when bacteria are not active.
                               115

-------
       CJ
        \3.0
        UJ
        O
        2
           2.0
z
o
a:
   1.0
           0.0
                                  1       1
          A
         A
                       ^/T-0.0216
              -   o

                          OA-0.0054
                                             1	1	1	T

                                               o RUN  3
                                               o RUN  6
                                               A RUN  7
                  J	I	I	L
                                             5000
                                             DAYS
                                                             I       I
                                                                       10000
Figure 25.   The  rate of leaching out of the profile of all  iron  species versus  time for Runs 3,
            6, and  7.

-------
      CL
      UJ
      Q
            0.0
       MOLE  FRACTION

0.05     0.10     0.15      0.20
          8
            o
          9 -
          10
                     T
         T
              o

              o

              o

             o
    o

    o

    o

    o
                                           o  o

                                      o     o

                                  O       o
              AFTER 5 YEARS

                 o  RUN  1

                 o  RUN  6
0.25
Figure 26.   The mole  fraction of oxygen within the profile versus

            depth  (m) after 5 years for Runs 1 and 6.
                             117

-------
                           MOLE  FRACTION













2
X
1—
Q.
LJ
Q








0
0


1


2.


3


4

5


6
7
8

9


10
.0 0.05 0.10 0.15 0.20 0.;
1 1 1 1 -
0 0
O D
-
o o
0 0
-
0 0
0 D
_
0 0
00
- -
CD
r AFTER 5 YEARS
33 o RUN 3

Q RUN 7
D
D
O
D
D
D
— —
D
D
I.I.
Figure 27.   The mole fraction  of oxygen within the profile versus
            depth (m)  after 5  years for Runs 3 and 7.
                            118

-------
     Pyrite Distribution.  Runs 1-7 were conducted with a uniform




pyrite distribution with depth to simplify the comparisons between




the runs.  Two simulations were performed to illustrate the depend-




ence of pyrite oxidation on pyrite distribution.  The two simulations




were run under conditions similar to Run 3 except that the initial




pyrite distributions were as shown in Fig. 23.  These distributions




are analogous to burying a "worst" pyrite layer at either 4.75




meters or 9.75 meters and replacing the best material or material




with the lowest pyrite content near the surface.




     Figure 29 shows the amount of pyrite consumed from 0 to 4,000




days for these two runs and Run 3.  Both Run 8, where the worst




layer is half-buried and Run 9 where the layer is completely buried,




show a marked reduction in pyrite oxidation.   However, Runs 8 and 9




show very little differences in pyrite oxidation, indicating that




for the combination of parameters used,  the position of the worst




layer is not as important as the position of  the first layer having




a sufficient pyrite content to support an oxidation rate that




effectively consumes all of the oxygen.   This is best illustrated




in the 0« versus depth plots for these two runs after 3,650 days




(10 years).  Figure 30 shows these 0- profiles.  In each case, 02




does not penetrate significantly beyond the first layer having a




pyrite content gerater than 0.25% (wt/wt), which is sufficient to




consume most of the 0--
                                 119

-------
KJ
O
                               .01
.02
    Fract ion   Pyr ite

0       .01      .02      0
01
.02      .03
V
1
2
3
J 4
f 5
ex
» 6
Q
/
8
9
in
|
1
§
|
|
|

-------

o-
rr
H-
0
3

«•

?3
C

O3
X-N
cr
c
H-
(B
a.
o
rr
lj
Ln
<§-

C
3
O.

3
VO
^
buried
O
rr

vo
ui
-
>"x



00
r»
NJ
vO
T!
11
01
n
rr
H-
0
3

O
It!
-o

H-
rr
re
O
H-
D.
H-
N
re
D-
s:
rr
H-
3

re
D
H-
re
•o
o
i-h
H-
I—1
re
1
en
c
CD
rr
H-
3
re
i-n
O
?o
C
3
FRACTION PYRITE
POP
o *_ ro
P











O
O
0





g 8
K 0
01 o




GJ
8
O






O

BL. ! i '
Wt>
fi t>
Q. t>
S >
H. »
M >
S 0
%. >
B *> ooc>

ft ^ CD ao m
6 \ c c <
?, % m m z
8 p o o
8 * — — 2
jr» ^ ^^ ^^ C/^
0 *->! ^J H
§ t> 01 01
B > 2 S
fi &
8 &
8 p
8 p
8 *
8 . *
8 >
8 *
8 *
B *
8 p
ff »

JD C^
r? >
% >
HD >
O t>
1 I 1 1
o
0
re
3
rr
if

-------
              0.0
              MOLE  FRACTION
       0.05     0.10     0.15      0.20
        X   5
        D.
        Lu)
        Q
             8
            10
CD
CD
n>
x
                                                at
                                             DO
                                         00
                                      o o
                                  0 0
                Q O

            O   O

         0   O

     O   O

 O    O

 O  O

 DO
                                 AFTER 10  YEARS
                                 o BURIED (4.75 M)
                                 o BURIED (9.75M)
0.25
figure  30.   The mole fraction of oxygen within the profile versus
            depth (m)  after 10 years for Run 8 (buried  at 4.75  m)
            and Run 9  (buried at 9.75m).
                            122

-------
 Inhibition of Bacteria


     Comparing Run 1 to Run 3 or Run 4,  it  is apparent  that,  at  least


 under some circumstances, a considerable reduction  in the  oxidation of


 pyrite over time can be achieved if a system can be shifted from


 bacterially catalyzed Fe   oxidation of  pyrite to direct  02 oxidation.


 Indeed, techniques for inhibiting bacterial activity have been pro-


 posed as a means of reducing acid drainage from reclaimed stripmines


 (Kleinmann and Erickson, 1981).  If bacteria can be inhibited from


 the beginning of pyrite oxidation, a reduction of the magnitude


 represented by the difference between Run 3 or 4 and Run  7 and Run 6


may be possible.  Amending the stripmine spoil after oxidation has


 started and bacteria have been established in order to inhibit


bacterial activity and reduce acid drainage has also been suggested


 (Kleinman and Erickson,  1981).  To simulate the situation in which


iron-oxidizing bacteria  are inhibited after they have become


established,  a run was made identical to Run 7 but where  the


bacterial activity was assumed to cease after approximately two


years of oxidation.  This represents the extreme case where the


addition of a bacteriacide is 100% effective in destroying the
      •
bacteria throughout the  profile.   Figure 31 shows the rate of


leaching of total iron over time for this simulation along with the


results  of Run 7 for comparison.   Although the bacterial activity


ceases  after  700 days, allowing the oxidation system to shift from


Fe  -dominant to O^-dominant,  no difference in the leaching rate of
                                123

-------
K3
                  1
                  CM1
                   y
                   o
                   2
                   ^ 2
                   x
z
o
ir
                   t
                   O
                                    D
                                        00


                                    DO       O
                cP

                O
               a
              o
              o
              o
             o
                              O
                 KB=O.O
                             e
                            8
                           a
                       o
                 1000
                 a  UNINHIBITED
                   a
                                             INHIBITED
                                                    I
2000


DAYS
3000
                                                                                     O  O
4000
           Figure 31.  The rate of  leaching out of the  profile of all iron species versus  time for Run  7  (no


                      inhibition)  and Run 10  (complete inhibition after 2 years).

-------
iron is observed until 250 days later.  From 950 days onward,  the



plots continue to diverge such that a large reduction is observed



after 3,000 days, or 2,300 days (6.3) years after bacterial inhibi-



tion.  The system demonstrates an apparent sluggishness in response



to the decreased oxidation rate.  One reason is that not until after



day 800 does the ferric iron concentration fall below levels allow-



ing the direct 0- oxidation mechanism to become dominant.  Another



reason is that at the assumed leaching-rate of 50 cm/year, 2.16



years are required to remove one water-filled pore volume from the



10-meter deep profile.  Thus, any change in oxidation rate will not



be fully expressed in the leach water until months and perhaps years



later.





Carbon DioxideGeneration



     Several simulations were made to examine the way in which the



model handled carbon dioxide production and flux within the mine



profile.  Four simulations were run for a period of five years;



Run 11,  a control where no source of carbon dioxide was assumed,



Run 12,  where heterotrophic respiration was assumed to be occurring



in the top 50 cm of the profile, Run 13, where respiration was



assumed to be occurring in the bottom layer of the profile only,



and Run 14, where darbonates were assumed present in the bottom



layer only.  Run 12 is analogous to a mine profile with a well


                                                    -12
established vegetative cover, where Q__ = - 0.3 x 10    moles
                                     DD
                                125

-------
0 -cm  -sec  - C   in Eq .  [31].  Run 13 represents  the case where




heterotrophs are active deep within the mine profile, utilizing




buried organic matter or digestible coal.  (}„„.., Eor Run 13 was
also set to -0.3 x 10    moles 0_-cm  -sec  - C   .  Run 14 was  run




to simulate carbonate neutralization of acid .  Only one layer was




assumed to contain carbonate at a content of 0.083% (wt/wt) .  The




low carbonate content in the profile was used because the H -




carbonate reaction rate as predicted by the model was very rapid




and dominated the pyrite oxidation reaction even at very low-




carbonate contents.  Although rapid reactivity of carbonates in




acid spoil has been observed (Geidel and Caruccio, 1977), we be-




lieved the neutralization rate used in this model to be unrealistic-




ally fast mainly because it was based on work with freshly exposed




limestone (Wentzler, 1977)  and ignored diffusion controlled kinetics




and possible build-up of iron coatings on the carbonate surface




(Geidel and Caruccio, 1977) both of which would reduce the neutraliza-




tion rate.  Even at the low-carbonate content used here, the pH of




the bottom layer remained above 5.0 compared to approximately 2.5 for




the other layers.




     In every case the production of CO,, reduced the oxidation of




pyrite.  After five years 73, 85 and 88% of the pyrite consumed in




the control was consumed in Runs 12, 13 and 14,  respectively.  The




reduction in pyrite oxidation is due primarily to the decreased




oxygen concentration in the spoil profile.  The reduction in 0_




concentrations in Run 12 was due mainly to 0  consumption in the
                                126

-------
surface layers.   In contrast, reductions in 0- concentrations in



Runs 13 and 14 were primarily due to increased resistance to oxygen



diffusion in the profile because of the counter-current C0_



diffusion up through the profile (see Appendix A) .   In addition to



reducing pyrite  oxidation because of C02 production, Run 14 showed



reduced levels of acid being flushed from the profile because of



carbonate neutralization..  Although Run 14 had a greater amount of



pyrite oxided than Run 12, the acid removed from the profile after



5 years was the  same;  73% of the control value • for  Huns 12 and



14 compared to 80% for Run 13.



     Table 13 shows the flux ratio, r  = N_n /Nn ,  for each layer
                                     K    LU-  U „


in Runs 12-14.  Even for these simple examples it can be seen that



the behavior of  r  with depth is fairly complicated.  Values of r



are both positive and  negative in each run and may not show a uni-



form increasing  or decreasing pattern with depth.  For example,



Run 12 shows a minimum r  value below the surface layer at a depth



of 275 cm.  This may,  in part, explain some of the difficulties



encountered with the interpretation of measured fluxes in Part



III.
                                 127

-------
Table 13.  Ratio of the flux of
C02 to 0  for Run 12 (hetero-
           trophs active in layer 1 only), Run 13  (heterotrophs
           active in layer 20 only), and Run 14  (acid neutraliza
           tion by carbonates in layer 20).

Layer
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Depth
(cm)
25f
75
125
175
225
275
325
375
425
475
525
575
625
675
725
775
825
875
925
975

Run 12
-.47
.16
.12
.10
.093
.089
.090
.096
.12
.21
.57
.76
.83
.86
.90
.94
.99
1.0
1.1
1.2
VNo2
Run 13
.0003
.0002
-.0009
-.0016
-.0018
-.0022
-.0028
-.0037
-.0053
-.010
-.028
-.074
-.10
-.14
-.18
-.24
-.33
-.45
-.61
-.72

Run 14
.0003
.0003
.0003
.0004
.0001
-.0013
-.0026'
-.0033
-.0048
-.0091
-.027
-.082
-.13
-.18
-.26
-.38
-.57
-.88
-1.4
-1.8

'Depth to middle of layer.
                                128

-------

-------
                              REFERENCES




1.  Arkesteyn, G. J. M. W.  Pyrite Oxidation in Acid Sulfate Soils:




    The Role of Microorganisms.  Plant and Soil, 54:119-134, 1980.




2.  Arora, H. S., J. B. Dixon, and L. R. Hossner.  Pyrite Morphology




    in Lignitic Coal and Associated Strata of East Texas.  Soil Sci.,




    125:151-159, 1978.




3.  Baes, C. F., Jr. and R. E. Mesmer.  The Hydrolysis of Cations.




    John Wiley and Sons, New York, 1976.




4.  Beck, J. V.  A Ferrous-Ion-Oxidizing Bacteria.  I.  Isolation and




    Some General Physiological Characteristics.  J. Bacteriol. ,




    79:502-509, 1960.




5.  Belly, R. T. and T. D. Brock.  Ecology of Iron-Oxidizing Bacteria




    in Pyritic Materials Associated with Coal.  J. Bacteriol. ,




    117:726-732, 1974.




6.  Birkeland, P. W.  Pedology, Weathering and Gemorphological




    Research.  Oxford University Press, New York, 1974.




7.  Bouma, J. and L. W. Dekker.  A Case Study on Infiltration Into Dry




    Clay Soil.  I.  Morphological Observations.  Geoderma 20:27-40,




    1978.




8.  Braun, R. L., A. E. Lewis, and M. E. Wadsworth.  In-Place Leaching




    of Primary Sulfide Ores:  Laboratory Leaching Data and Kinetics




    Model.  In:  Solution Mining Symposium 1974, American Institute of




    Mining, Mettalurgical and Petroleum Engineers, Inc., New York,




    1974.  pp. 295-323.




9.  Cathles, L. M.  Predictive Capabilities of a Finite Difference




    Model of Copper Leaching in Low Grade Industrial Sulfide Waste




    Dumps.  Mathematical Geol., 11:175-191, 1979.
                                  129

-------
10.  Cathles, L. M. and J. A. Apps.  A Model of the Dump Leaching




     Process that Incorporates Oxygen Balance, Heat Balance and




     Air Convection.  Metal. Trans., 6B:617-624, 1975.




11.  Ciolkosz, E. J., R. L. Cunningham, G. W. Peterson, and R. Pennock,




     Jr.  Characteristic Interpretations and Uses of Pennsylvania




     Minesoils.  Pa. Agric. Exp. Stn. Prog. Rep., 1977.




12.  Clark, C. S.  The Oxidation of Coal Mine Pyrite.  Ph.D. Thesis,




     John Hopkins, University, University Microfilms, Inc., Ann Arbor,




     Michigan.  No. 65-10-278, 1965.




13.  Collier, W. R., R. J. Pickering, and J. J. Musser.  Influences of




     Strip Mining on the Hydrologic Environment of Parts of Beaver




     Creek Basin, Kentucky, 1955-66.  U.S. Geological Survey




     Professional Paper 427-C, 1970.




14.  Davis, C. W.  Electrochemistry.  William Clowes and Sons., Limited,




     London, 1967.




15.  Ehrlich, H. C. and S. J. Fox.  Environmental Effects on Bacterial




     Copper Extraction from Lower Grade Copper Sulfide Ores.   Biotech.




     and Bioeng., 9:471-485, 1967.




16.  ElBoushi, I. M.  Amount of Water Needed to Initiate Flow  in Rubbly




     Rock Particles.  J. Hydro1. , 27:275-284, 1975.




17.  Emrich, G. H.  and E.  G. Thompson.  Some Characteristics of




     Drainage  from  Deep Bituminous Mines in Pennsylvania.   In:




     Proceedings Second Symposium on Coal Mine Drainage Research,




     Mellon Institute, Pittsburgh, Pennsylvania, May 14, 1968.




18.  Evans, D. D.   Gas Movement.  In:  Methods of Soil Analysis, C. A.




     Black, ed.  American  Society of Agronomy, Madison, Wisconsin.




     Agronomy,  9:319-330,  1965.







                                  130

-------
19.  Freeze, R. A. and J. A. Cherry.  Groundwater.  Prentice-Hall, Inc.,




     Englewood Cliffs, New Jersey, 1979.




20.  Garrels, R. M. and M. E. Thompson.  Oxidation of Pyrite by Iron




     Sulfate Solutions.  Am. J. Sci., Bradley Vol. 258-A:57-67, 1960.




21.  Geidel, G. and F. T. Caruccio.  Time as a Factor in Acid Mine




     Drainage Pollution.  In:  Proceedings Seventh Symposium on Coal




     Mine Drainage Research, Louisville, Kentucky, Oct. 18-20,




     National Coal Association, Washington, D.C., 1977.




22.  Grim, E. C. and R. D. Hill,  Environmental Protection in Surface




     Mining of Coal.  EPA-670/2-74-093, U.S. Environmental Protection




     Technology Series, U.S. Environmental Protection Agency, Concinnati,




     Ohio, 1974.  pp. 1-277.




23.  Kleinmann, R. L. P. and P. M. Erickson.  Field Evaluation of a




     Bactericidal Treatment to Control Acid Drainage.  In:  Symposium




     on Surface Mining,  Hydrology, Sedimentation and Reclamation,




     Lexington, Kentucky, December 7-11, University of Kentucky,




     Lexington, Kentucky, 1981.




24.  Lacey, D. T. and F. Lawson.  Kinetics of the Liquid-Phase Oxidation




     of Acid Ferrous Sulfate by the Bacterium Thiobacillus ferrooxidans.




     Biotech, and Bioeng., 12:29-50, 1970.




25.  Landesman, J., D. W. Duncan, and C. C. Walden.  Iron Oxidation by




     Washed Cell Suspensions of the Chemoautotroph Thiobacillus




     ferrooxidans.  Can. J. Microbial. 12:25-33, 1966.




26.  Langmuir, D.  Particle Size Effect on the Reaction Geothite =




     Hematite + Water.  Am. J. Sci., 271:147-156, 1971.
                                  131

-------
27.  Langmuir,  D. and D.  0. Whittemore.  Variations in the Stability of




     Precipitated Ferric  Oxyhydroxides.  In:  Nonequilibrium Systems in




     Natural Water Chemistry, R. F. Gould, ed.  Advan. in Chem., Series




     106, Amer. Chem. Soc., 1971.  pp. 209-234.




28.  Lau, C. M., K. S. Shumate,  and E. E. Smith.  The Role of Bacteria




     in Pyrite Oxidation  Kinetics.  In:  Third Symposium on Coal Mine




     Drainage Research, May 19-20, Mellon Institute, Pittsburgh,




     Pennsylvania, 1970.




29.  Levenspiel, 0.  Chemical Reaction Engineering.  John Wiley and




     Sons, Inc., New York, 1972.




30.  Lovell, H. L. , R. R. Parizek, D. Forsberg, M. Martin, D. Richardson,




     and J. Thompson.  Environmental Control Survey of Selected




     Pennsylvania  Strip Mining Sites.  Final Report to Agronne National




     Laboratories, #31-109-38-3497, 1978.




31.  Lundgren, D.  G.  Microbiological Problems in Strip Mine Areas:




     Relationship  to the Metabolism of Thiobacillus ferrooxidans.




     Ohio J. Sci., 75:280-287, 1975.




32.  Malouf, E. E. and J. D. Prater.  Role of Bacteria in the Alteration




     of Sulfide Minerals.  J. Metals 1961:353-356, 1961.




33.  Morth, A. H., E. E.  Smith, and K. S. Shumate.  Pyritic Systems:  A




     Mathematical  Model.   Environmental Protection Agency Technology




     Series 14010  EAH, Contract No. 14-12-589, Report No. 665771, U.S.




     Environmental Protection Agency, Washington, B.C., 1972.




34.  Myerson, A.  S.  Oxygen Mass  Transfer Requirements During  the Growth




     of Thiobacillus ferrooxidans  on Iron Pyrite.  Biotech, and Bioeng.,




     23:1413-1416, 1981.
                                  132

-------
35.   Nordstrom, D. K., E. A. Jenne, and J. W. Ball.  Redox Equilibria of




     Iron in Acid Mine Waters.  In:  Chemical Modeling in Aqueous




     Systems, E. A. Jenne, ed.  American Chemical Society, Washington,




     D.C., 1979.




36.   Ohio State University Research Foundation.  Sulfide to Sulfate




     Reaction Mechanism.  Water Pollution Control Research Series




     14010, FPS 02/70, Federal Water Pollution Control Administration,




     Washington, D.C., 1970.




37.   Pedersen, T. A., A. S. Rogowski, and R. Pennock, Jr.  Physical




     Characteristics of Some Minesoils.  Soil Sci. Soc. Am. J.,




     44:321-328, 1980.




38.   Pionke, H. B., A. S. Rogowski, and R. J. DeAngelis.  Controlling the




     Rate of Acid Loss from Strip Mine Spoil.  J. Environ. Qual.,




     9:694-699, 1980.




39.   Pionke, H. B. , A. S. Rogowski, and C. A. Montgomery.  Percolate




     Quality of Strip Mine Spoil.  Trans. ASAE, 23:621-628, 1980.




40.   Quisenberry, V. L. and R. E. Phillips.  Percolation of Surface-




     Applied Water in the Field.  Soil Sci. Soc. Am. J., 40:484-489,




     1976.




41.   Raats, P. A. C.  Unstable Wetting Fronts in Uniform and Non-uniform




     Soils.  Soil Sci. Soc. Am. Proc., 37:681-685, 1973.




42.   Rao, P. S. C., R. E. Jessup, D. E. Rolston, J. M. Davidson, and




     D. P. Kilcrease.  Experimental and Mathematical Description of




     Non-absorbed Solute Transfer by Diffusion in Spherical




     Aggregates.  Soil Sci. Soc. Am. J., 44:684-688, 1980a.
                                  133

-------
43.  Rao, P. S. C., D. E. Rolston, R. E. Jessup, and J. M. Davidson.




     Solute Transport in Aggregated Porous Media:  Theoretical and




     Experimental Evaluation.  Soil Sci. Soc. Am. J., 44:1139-1146,




     1980b.




44.  Rogowski, A. S., H. B. Pionke, and J. G. Broyan.  Modeling the




     Impact of Strip Mining and Reclamation Processes on Quality and




     Quantity of Water in Mined Areas:  A Review.  J. Environ. Qual.,




     6:237-244, 1977.




45.  Rogowski, A. S., H. B. Pionke, and B. E. Weinrich.  Some Physical




     and Chemical Aspects of Reclamation.  Preprint, paper presented




     at the North Atlantic Region ASAE 1982 Annual Meeting, 1982.




46.  Rogowski, A. S. and B. E. Weinrich.  Modeling Water Flux on




     Strip-Mined Land.  Trans. ASAE, 24:935-940, 1981.




47.  Rose, D. A.  Hydrodynamic Dispersion in Porous Materials.  Soil




     Sci., 123:277-283, 1977.




48.  Russell, E. W.  Soil Conditions and Plant Growth.  Longmans,




     London, 1973.




49.  Sapieszko, R. S., R. C. Patel, and E. Matijevic.  Ferric Hydrous




     Oxide Soils.  2.  Thermodynamics of Aqueous Hydroxo and Sulfate




     Ferric Complexes.  J. Phys. Chem. , 81:1061-1068,  1977.




50.  Silverman, M. P. and D. G. Lundgren.  Studies of  Chemoautotrophic




     Iron Bacterium, Ferrobacillus  ferrooxidans.  II.  Manometric




     Studies.  J. Bacteriol.,  78:326-331, 1959.




51.  Singer, P.  C. and W. Stumm.  Acidic Mine  Drainage:  The Rate




     Determining  Step.   Science, 167:1121-1123,  1970.
                                   134

-------
                 APPENDIX A
APPLICABILITY OF PICK'S LAW TO GAS DIFFUSION
                     135

-------

-------
                          Introduction





     The exchange of gases between the soil and  the atmosphere is of



primary concern to soil scientists.  Gas movement  in the soil can be



caused by both convection and diffusion.  Convection results when a



difference in total gas pressure exists between  two locations.



Diffusion will occur even if the total pressure  is uniform as long as



there is a spatial difference in the chemical potential, normally



represented by differences in concentration or partial pressure, of



the components of a gas mixture (Kittel, 1969).  While both processes



are important in soils, diffusion is considered  to dominate during



normal gaseous exchange (Evans, 1965).



     The fundamental equation used to describe one dimensional



diffusion in soil is Pick's First Law.



                    q± = - a DF  dCi/dz                         [Al]

                               i



where q. is the mass transfer rate of component  i per unit area


   -2 -1
(ML  T  ), D   is the Fickian diffusion coefficient for component i


  2 -1       1                           -3
(L T  ), C. is the concentration of i (ML  ), z  is distance and a is



the correction term that accounts for the air-filled porosity and



tortuosity of the diffusion path in soils (Troeh, et al.,  1982).



Since in the remainder of this paper the magnitude of a is inconse-



quential as long as its magnitude is independent of gas type



(Penman, 1940b), we will let it equal 1.0 and drop it from the



remaining equations.
                                 136

-------
     Alternatively,  Eq.  [Al]  can be written in terms of mole fractions




by invoking the equation of state for ideal gases and Dalton's Law to




give:





                    Ni = -V   fedVdz                       [A2]





where now N.  is the  molar flux,  P is the total gas pressure, R the




universal gas constant,  T absolute temperature,  Y. the mole fraction




of component  i and t is  time.   While useful in soil science, it must




be remembered that Fick's Law when applied to gases is strictly an




empirical relation,  borrowed  from studies with solutes and  shown to




agree well with observed diffusional processes in air (Kirkham and




Powers,  1972, p.  429;  Penman, 1940a and 1940b).   The form of Eq.




[A2], the flux of a  gas  being equal to its gradient multiplied by a




single coefficient,  is very misleading and requires closer




examination.




     The equations for gaseous  diffusion were developed by  Stefan




(in Wilke, 1950)  and Maxwell  (1952) among others.   The equations



are based on  the  statistical  mechanics description of ideal gases




developed by  Gibbs and Boltzman (Kittel, 1969).   These equations



have been extensively developed  in the literature and found to




agree very closely with  experiments on gas-diffusion (Hirschfelder,




et al.,  1964; Fairbanks  and Wilke, 1950).  Since the interest  in




gas-diffusion in  soils has increased in recent years (Smith,  1977;




Bakker and Kidding,  1970;  DeJong,  et al., 1979;and others)  and Fick's




Law is used to describe  the process in most soils literature (see
                                  137

-------
Wood and Greenwood, 1971 for an exception)  it would  seem advisable

to compare Pick's Law  (Eq.  [A2]) to  the  Stefan-Maxwell  diffusion

equations in order to  gauge its validity.   This paper will examine

several simple cases of diffusion of interest in  soil science.   It

is not our intention   to develop the Stefan-Maxwell  equations  in

depth, nor to fully present the' theory underlying them.   For  this,

the reader is directed to the comprehensive works by Chapman and

Cowling (1939) or Hirschfelder, et al.  (1964), or to the excellent

review by Marrero and Mason (1972).   We  shall, for the  remainder of

the paper, consider only"diffusion in the gas phase, ignoring  gas-

liquid interactions, convection, turbulence or other mass flow

phenomena, and for simplicity will consider only  steady-state

diffusion.



           Stefan-Maxwell Equations  for  Gas Diffusion


     Chapman and Cowling (1939, p. 244)  have shown that  for dif-

fusion, the average velocity of one  gas  in  relation  to  a second  gas

is represented by:

                   n          ^(.•njn^)   n±n (m -nO
          v,-v,= - —— D^ r  	r-	 + 	J	J	
                                 9r         p,           3r
                                           3T/3r ]               1A31
where the terms are defined at the end of this paper.  For our pur-

poses, it is only important to note that the difference in average

velocities of two gases (v.-v.) depends on four separate terms on the

right side of Eq. [A3], which may be thought of as the chemical

                                 138

-------
potential for the gas pair (Kittel, 1969, p. 215).  The  first  term




represents ordinary or concentration diffusion, which we are con-




cerned with in this paper*  The second term represents diffusion




caused by differences in total pressure but should not be confused




with mass flow.  As an example of pressure diffusion, lighter  ele-




ments such as helium will concentrate in the upper portions of the




atmosphere due to diffusion driven by pressure differences in  the




atmosphere.  The gases respond to the pressure gradient  created by




gravity and not to gravity directly.  The third term represents a




diffusional flux created by an external force, such as a flow of




charged particles in an electric field giving rise to an electric




current.  The final term represents thermal diffusion.   In thermal




diffusion, a temperature difference will cause diffusion within a




gas mixture, with the lighter components concentrating near the




warmer zone.  If the temperature gradient is maintained, the gas




components will continue to separate until ordinary diffusion




exactly balances thermal diffusion.  Of the four terms,  only con-




centration or ordinary diffusion is important under normal circum-




stances in soil and plant processes, although thermal diffusion is




used in commercial refining techniques for separating isotopes  of




some gases (Jones and Furry,  1946; Vasaru, et al., 1969).




     For ordinary diffusion in one dimension of one gas, i,  through




another gas, j,  where total pressure and temperature are constant,




Eq. [A3] reduces to:
                                  139

-------
                           2
                         n       d(n./n )

               v.-v.  --- 2— D.. - ^— —                     [A4]
                i   j      n^   ij    dz



              nT
Dividing by - D..,  Eq. [A4]  can be rearranged such that:

            ninj   XJ



               v  n n -v  n n        d(n n )
Dividing both sides  of  Eq.  [A5] by Avagadro's number, MQ, convert



the number densities to molar densities.  Remembering that N. = v.



n./M ,  Eq. [A5] becomes:




              NiWi =   i d
-------
mixture and not just i and j.   However, this difference is slight

(within the experimental error of measurement) and can be ignored

(Marrero and Mason, 1972).

     In general, Eq. [AS] represents n-1 independent equations for
                   n
n unknowns (since  Z  Y. = 1.0) and as such the solution is indet-
                  i-1  1
erminate.   However, under certain simplifying assumptions, solutions

can be found for the series of equations.  We will examine several

of these situations applicable to soil and plant science.



                        Binary Gas System


     For the simplest case of  diffusion in a two-component system

(i and j), Eq..[A8] yields only one unique equation, identical to

Eq. [A7],  since Y.  + Y.  = 1.0.  Eq. [A7] can be rearranged inl:o a

form similar to Pick's Law (Eq. [A2]).  By letting r..  equal the

negative value of the ratio of the fluxes, r.. = - N./N., and sub-

stituting 1-Y. for  Y.,  we can  rearrange Eq. [A7] to obtain:


                                   ddz
                X     a

Comparing Eqs.  [A9]  and [A2]  we find that:


               DF.  ' l-(l-r..)Y.

The Fickian diffusion coefficient depends not only on D.., but also

on the flux ratio and mole ratio of component i.  Two special

circumstances immediately follow from Eq. [A10] .
                                 141

-------
Tracer Diffusion

     In cases where Y. is present only in trace amounts (i.e., Y. =

0.0), Eq. [A10]  for the Fickian diffusion coefficient reduces to:

                            D   = D                              [All]
                              i     J

independent of flux ratios and concentration.

Closed System

     A second possibility for binary gas diffusion is diffusion in a

closed system where, since the total pressure remains constant, N. =

- N , r   = 1.0 and, from Eq. [A10], D_  again just equals D...
   J   Ji                             Fi                    ^

Open System

     A more general situation is the steady state diffusion of gases

through  an inert layer where the concentrations of the gases are

held constant at one end because of an infinite source/sink for the

gases (the atmosphere) and the concentrations are fixed at the other

end by consumption or production processes.  An example would be

oxygen diffusing into a soil ped and oxidizing ferrous to ferric

iron where oxygen and nitrogen comprise the ped atmosphere.  For an

open system Eqs. [A9] and [A10]  cannot be solved without a_ priori

knowledge of r...  FigureAL illustrates the variation of D   with
              J1                                           i
changes in r..  and Y. as calculated by Eq. [A10].  The reduced
                                *
Fickian diffusion coefficient, D_  = D_ /D..  is plotted for several
                                Fi    Fi  1J
values of r.. and for 0.0 < Y. < 1.0.  As discussed for equi-molar,

counter-current  diffusion in a closed system and for tracer diffusion,

D    is constant for all values  of Y. when r.. =1.0 and for all
                                  142

-------
            2.0
Figure Al.   Values of the reduced Fickian diffusion coefficient, D
                                                                   F. '
                                                                    i
            in a binary gas mixture plotted against the gas mole


            fraction of component i.  Curve labels are the values  of


            the flux ratio, r.., for which the curve was calculated.
                                   143

-------
values of r.. when Y. =0.0.  D   diverged rapidly  however  as  Y.
           31       i          F±             *                   i

increases, decreasing as r.. becomes less  (reduced  resistance to


diffusion) and increasing as r.. increases.   In  the extreme  it  can


be seen from Eq.  [A10] that D   can become negative when  the two
                              i
gases are diffusing in the same direction  (r.. < 0.0),  that  is,


gas i would diffuse against its gradient.  This  behavior  has been


measured by Duncan and Toor (1962).
                  Three Component Gas Systems



     The soil atmosphere can normally be characterized as a three-


component  (ternary) system.  In a ternary system composed of gases


i, j and k, where Y. + Y. + Y  = 1, Eq. [A8] yields two unique
                   1    J    *•

equations:


                 „           N.Y.-N.Y.   N.Y-N.Y.
                                 ij          ik
                             N.Y.-N Y.   N Y, -N, Y.
                      J          ij         Jk


Again, the general solution of these equations is impossible without


further simplifying assumptions.



Tracer Diffusion


     If we let component i represent a gas at trace concentration, so


that Y. + Y,  = 1, then Eq. [A12] which describes the diffusion of i,
      J    ^

becomes :
                                 144

-------
                                  y     Y
                                         ik
or upon solving for N,:


                      - D..D.
                     j ik4k ij



which is identical in form to Fick's Law with the Fickian  diffusion


coefficient dependent only on D.., D..  and the mole fractions;'Y.
and Yk.
                       D..D
               n
               D
                F.   Y.D.n
                 i    j ik  k ij
In general, for tracer diffusion in a n-component gas mixture, the


tracer will diffuse according to Fick's Law where the diffusion


coefficient is equal to:


                      n        n       n

               D   =  H  D  /  I   (Y ( n D  )                    [A17]
                 i   j=l  1J  j=l   J k=l lk
                                          or j
and is independent of the diffusional fluxes of the other components.



One Gas Stagnant


     Eqs. [A12] and [A13] can also be simplified if we assume one


of the gases is stagnant.  This is analogous to quasi-steady state


respiration in a soil where oxygen and carbon dioxide are being


exchanged with the atmosphere and nitrogen, the third gas component,


(k), is stagnant.  In this case with N  = 0.0, Eqs. [A12] and [A13]
                                      &

become:
                                  145

-------
      P/RT dY±/dz = N±
      P/RT dY./dz = N.
                             Y.

                                       Y.
                                  - N.
                       Y.     Y.
                        i  ,    I
                             D..     D

                                             Y.
                                  - N.
                                                                [A18]
                                                                [A19]
These equations can be solved for the molar fluxes N. and    if Y ,



Y  and Y,  are known.  In the case of counter-current, equi-molar
 j      k

diffusion, where N  » - N.  (r.± = 1.0),  Toor (1957) has solved these



equations exactly and found that the solution must simultaneously



satisfy Eqs.  [A20] and [A21].
                K
         Ni = - Nj = DT=5T
                 J    jk  ik
                                        .
                                        ln
D..
:r±i) (Y.  - Y.
D.,     i,    10
 jk     1     2
                              D

                              -
                                   (Y   - Y  )  = ln(Y  /Y

                               ik    Jl    32        ^2
                                                           [A20]
                                                          [A21]
The second subscript on Y refers to the points 1 and 2 separated by



the distance z.   These equations can be used to describe diffusion



in soils where the source and sink terms for CL (i)  and CCL  (j)  are



determined by plant root or soil organism respiration and the res-



piration quotient,  CO- produced/O- consumed, is near 1.0 (e.g.,  r   =



1.0).  However,  it must be remembered that this is only a good



approximation for respiration in soils during quasi-steady state



conditions because if Y. or Y.  is changing rapidly,  N,  4= 0.0 and
                       i     3                        K


the simplification is not justified.



     In general,  we can solve Eqs. [A18] and [A19] only if the ratio



r.. is known.  Substituting r..N. for N. in Eq. [A18]  and solving
                                  146

-------
for N. and substituting N./r  . for N. and  solving  for  N  in Eq „
     i                   3  31      i                   3


[A19] yields
          N  =
           i   D..Y.-H)...D.lrT.
                ik j  13 k  31 ik i

          N
           3   D.^D.VD  Y/r-~  T
           J    jk i  13 k  jk j  31
which are in the form of Fick's Law with diffusion coefficients



                           D..D
               D
                F.   D..Y.+D..Y.+T..D..Y.
                 i    ik 3  13 k  31 ik i





                   = 	' i.1—1^	

                F.   DMY.+D..Y.+D..Y./r..
                 i    jk i  13 k  jk 3  31
These equations demonstrate that the apparent diffusion coefficient



in Fick's Law depends not only on the binary diffusion coefficients



but also on the gas composition and the flux ratio, r^ .  As am



example, D  can be calculated for 0? from Eq. [A24] and for CCL from



Eq. [A25] in a ternary atmosphere where i = CL, j = CO., k = N. and
N   =0.0.  For this case Eqs. [A24] and [A25] become:

  2
            °2    02'N2XC02  02,C02"N2 rC02, 02°2ty2
                                                                 [A26]
                                      +D
Values for the binary diffusion coefficients can be calculated from



Table Al (next section).  For a temperature of 20 C and a pressure of



101 kPa these values are 0.159 x 10~A , 0.202 x 10~4 and 0.159 :K 10~4



(m2/sec)for D        D              ^ respectively.  The Flckian



                                  147

-------
diffusion coefficients can  then be calculated  if values  for Yn ,  Y n ,

                                                              U2   L 2

Y>T  and r_.  are known.  For  example, with Yn  =0.15, Y    = 0.06,
 N2      C02                                02           C02


Y,T  =0.79 and r.,^  _ = 0.5,  D_   and D_    are calculated  to be
 N,             CO,,0-        Fn       F
  2        ,      2.  L      _, 2^2         ")

0.210 x 10   and 0.150 x 10  m /sec, respectively.   These two points



are indicated on Fig. A2 and  A3 where the values of  D    and D


                                                      °2       C°2
are plotted for various flux  ratios.  In both  figures Y   is held

                                                         2

constant at 0.790 while Y   varies from 0.0 to 0.21.  Carbon dioxide



comprises the remainer of the atmosphere (0.21 < Y    <  0.0).
                                                  co2


     The apparent diffusion coefficients for both 02 and CO- vary



little for r .  n  near 1.0.  As the flux ratio deviates further
            (JUn , \Jn


from 1.0, the 09 diffusion  coefficient diverges rapidly  as  Y_

                                                             U2

increases (Fig. A2).  D  for  C00 is less variable but can deviate
                       r       2.


markedly from its value at  r  _  n  = 1.0 at low values of the  flux

                            CU2'U2

ratio and high C0_ (low 02) concentrations (Fig. A3).  Values  for



the flux ratio in soils have  been reported ranging between  0.6  and



4.0 (Bunt and Rovira,  1955; Rixon and Bridge,  1968; Bridge  and



Rixon, 1976).  Although none  of these authors  reported the  0_ and



C0_ mole fractions at which these ratios were  found, extreme values



of 0.0 to 0.21 for 0- and 0.0 to 0.09 for CO-  have been measured



(Russell and Appleyard, 1915).  Using the values of 0.05 for Y
                                                              co2


and 0.15 for Yn  and a flux ratio of 2.0, we can see from Figs. A2



and A3 that deviations in the fickian diffusion coefficients of 15%



for 0« and 9% for C0»  from  the values at r _  n  =1.0 are  possible.
     Z              2.                      CU2,U2


Figures A2 and A3 illustrate  the potential hazard of applying values



of the Fickian diffusion coefficient from one  situation  to another
                                 148

-------
      Q-20-
         .14
Figure A2.  Values of the Fickian diffusion coefficient, D  , for 0_
                                            2
            in an 0--C02-N2 atmosphere (in m /sec) .  The N« ;nole
            fraction is held constant at 0.790, while the 0_ mole
            fraction, plotted on the abscissa, is  varied between 0.0
            and 0.21.  C0« comprises the balance of the atmosphere.
            Values are for 20°C and 101 kPa with the N  flux being
            set to 0.0.  Curve labels are the values of the flux
ratio r
text.
                      , O
                            Point represents value calculated in
                                149

-------
                                              .20    .25
Figure A3.   Values for the Fickian diffusion coefficient, D ,  for
                                                2
            C0_ in an 0 -CO -N_ atmosphere (in m /sec).   The N? mole
            fraction is held constant at 0.790, while the 0~ mole
            fraction, plotted on the abscissa, is varied between 0.0
            and 0.21.  CO  comprises the remainder of the atmosphere.
            Values are for 20 C and 101 kPa with the N«  flux being
            set to 0.0.  Curve labels are the values of  the flux
            ratio r n    .  Point represents value calculated in
                   C°2'°2
            text.
                                  150

-------
where  the flux ratio and gas composition may not be  the  same.   This


may be, at least in part, the reason why flux calculations  based


concentration gradients do not compare well to other  techniques for


measuring flux (DeJong, et al., 1979).


     Equation fA8] also applies to diffusion in four-component  or


greater gas mixtures.  02> C02> N2 and Ar or 02> C02, N2, Ar and H20


vapor  atmospheres would be mixtures of interest in soils.   However,


as in  the ternary case, unless some simplifying assumptions can be


made,  the exact solution for diffusion in these mixtures is not


possible.




                  Binary Diffusion Coefficient
     D.. is the proportionality coefficient relating the flux of ±


and j to their gradient in a binary gas system.  Conceptually D..


can be thought of as a macroscopic averaging of the interactions


or collisions between components i and j.  The greater the number of


collisions, the more diffusion is impeded and the smaller D.. must


be.  For this reason, D..  must be inversely proportional to density,


and thus pressure since the number of collisions will increase


linearly with pressure at  constant temperature.  Thus:




                        Dij = Dij/P                            [A28]

       *
where D.. is the temperature dependent, pressure independent binary


diffusion coefficient.  However, any decrease in D..  due to a


pressure increase will be  exactly balanced by an increase in the
                                  151

-------
number  of molecules  or  flux  carriers  per volume of  gas  (increased


concentration,  C) .   This  can be  seen  by substituting  Eq.  [A28]  into


Eq.  [A29] for the flux  in a  binary  system:




                Ni =  - l-(l-r.')Y. fe  dVd2                     tA29J



the  pressure terms drop out  and  the diffusion  equation  is pressure
 independent when expressed in mole fractions:


                                                                [A30]

When working with the concentration form of the diffusion equation


(Eq.  [Al]), the pressure dependence of the diffusion coefficient


must be retained:
where P. is the total pressure at which D.. is known and P2 is' the


pressure at which diffusion is taking place.


     Relations for calculating D.. for any gas pair, based on


classical statistical mechanics were independently developed by


Chapman and by Enskog (Chapman and Cowling, 1939).  The solution

                                       *
involves successive approximations to D...  The approximations con-


verge rapidly, so that the first approximation, which is temperature


dependent but independent of gas composition and at most the second


approximation, which introduces slight compositional variation,  is
                                 152

-------
                            *
sufficient for calculating D.. to within at most several percent


error (Marrero and Mason, 1972).


     Values for D..  (more commonly D..), can be found in numerous


papers in the literature (see Marrero and Mason, 1972), with some


recent values for carbon dioxide, nitrous oxide, ethylene and


ethane presented by Pritchard and Currie (1982).  Marrero and Mason


(1972) have compiled an extensive tabulation of binary diffusion


coefficients taken from the literature.  Based on  the Chapman-


Enskog equations and extensive experimental results, they have com-


piled semi-empirical relations for the  first approximation of over


70 gas pairs.  Diffusion coefficients for gas pairs of interest in


soil research are tabulated in Table Al.  In all cases shown, the

                         *
second approximation to D.., which is based on the compositional


ratio of component i to component j, represents a correction to the


first approximation of less than 1.0% at normal temperatures for


the entire compositional range.  This correction is within the;


experimental error of the measured values and can be neglected for


all the gas pairs listed.  The equation used by Marrero and Mason

                      *
(1972) to calculate  D.. is:



               ln(D*.) = ln(a) + b ln(T) - c/T.                [A32]



They computed the constants a, b and c  by matching Eq. [A32], whose


form is suggested by the theoretical work of Chapman-Enskog, to a


compilation of experimental data for each gas pair.  The temperature
                                  153

-------
                                         * t
Table Al.  Coefficients for calculating D..  for  gas pairs of
           interest in soil research (after Marrero and Mason,
           1972).

Gas
Couple

Ar-CH,
Ar-N2
Ar-02
Ar-air
Ar-C02
CH4"N2
CV°2
CH4-air
N2-02
N2-H20
N2-C02
02-H20
o2-co2
Air-H20
Air-C02
H20-C02
C02-N20
a x 107
KPa-m
0
0
0
0
1
1
1
1
1
0
3
0
1
0
2
9
0
b

2/sec-Kb
.792
.913
.987
.926
.76
.01
.68
.04
.14
.188
.18
.191
.58
.189
.73
.33
.284
1
1
1
1
1
1
1
1
1
2
1
2
1
2
1
1
1
.785
.752
.736
.749
.646
.750
.695
.747
.724
.072
.570
.072
.661
.072
.590
.500
.866
0
0
0
0
89
0
44
0
0
0
113
0
61
0
102
307
0
c
K
.0
.0
.0
.0
.1
.0
.2
.0
.0
.0
.6
.0
.3
.0
.1
.9
.0
T range

307-10~4
244-104
243-104
244-104
276-1800
298-104
294-104
298-104
285-104
282-373
288-1800
282-450
287-1083
282-450
280-1800
296-1640
195-550
Uncertainty?
limits
%
3
2
3
3
3
3
3
3
3
4
2
7
3
5
3
10
3

 ln(D..) = ln(a)  4- b ln(T)  - c/T,  where T is temperature, (K) .

?    1J          *
 Uncertainty in D.. term at lower  end of listed temperature range.
                                  154

-------
range over which these coefficients were fit and the uncertainty

           *
limits in D.. at the lower end of the temperature range are also


given in Table Al.

                                *
     An example calculation of D.. using values in Table Al is as


follows.  The pressure-independent, diffusion coefficient for the


gas pair 0»-C07 is calculated from Eq. [A32] with the coefficients


from Table Al of a = 1.58 x 10  , b = 1.661 and c = 61.3  At a


temperature of 293 K:



     ln(D*  _n ) « ln(1.58 x 10~7) + 1.661 ln(293) - 61.3/293.   [A33]
         °2'C°2



        D*   0   = 0.00160 m2-KPa-sec~1



or at atmospheric pressure, from Eq. [A28]



                    Dn  _n  = 0.00160/101.
                     °2'C°2
                                                                 [A34]

                            = 0.159 x 10~A m2/sec


The diffusion coefficients are highly sensitive to temperature with


a 30% variation possible under normal, seasonal soil temperature


fluctuations.  For example, the binary diffusion coefficient for the


0,,-CO  gas pair is equal to 0.136 x 10~  at 0°C (273 K) and 0.165 x


10~4 at 30°C (303 K) or D0  rf,  is 22% greater at 30°C than 0°C.
                         °2'C°2
                                  155

-------
                             SUMMARY


     The comparison of the Stefan-Maxwell equations and Fick's Law

for diffusion of gas through vapor shows that only under certain,

special circumstances is the diffusion coefficient for Fick's Law,

D  , a constant independent of the mole fraction of i and the
  i
diffusion flux of other gases.  These special cases are the diffusion

of a trace amount of component i and equi-molar, counter-current

diffusion in a binary gas mixture, or equi-molar, counter-current

diffusion of two gases in a ternary system with the third gas

stagnant.  In general, D   is dependent on the binary diffusion
                         i
coefficients, the composition of the gas mixture, and the diffusional

flux ratios of the gas components.  Measured values of D   can only
                                                         i
be accurately extrapolated to other circumstances when all the above

conditions are similar.  In an CL, C02 and N» atmosphere where N? is

stagnant, variations on the order of 10% from the tracer value of

D  for 00 and C0_ are possible with variations in the mole fraction,
 c      /       /

or flux ratio.  The diffusion coefficients are most sensitive to

changes in the flux ratio for ratios less than 0.8 or greater than

1.2.  The temperature and pressure dependent nature of the diffusion

coefficient must also be considered.
                                 156

-------

-------
                              REFERENCES




 1.   Bakker,  J.  W.  and A.  P.  Ridding.   The Influence of Soil Structure




     and Air  Content on Gas Diffusion in Soils.   Neth. J. Agric. Sci.,




     18:37-48,  1970.




 2.   Bridge,  B.  J.  and A.  J.  Rixon.  Oxygen Uptake and Respiratory




     Quotient of Field Soil Cores in Relation to their Air-Filled




     Pore Space.  J. Soil Sci., 27:279-286, 1976.




 3.   Bunt, J. S. and A. D. Rovira.  Microbial Studies of Some Sub-




     anarctic Soils.  J. Soil Sci., 6:119-128, 1955.




 4.   Chapman, S. and T. G. Cowling.  The Mathematical Theory of




     Nonuniform Gases.  Cambridge University Press, London, 1939.




 5.   Curtiss, C. F. and J. 0. Hirschfelder.  Transport Properties of




     Multicomponent Gas Mixtures.  J.  Chem. Physics, 17:550-555, 1949.




 6.   DeJong,  E., R. E. Redmann, and E. A. Ripley.  A Comparison of




     Methods  to Measure Soil Respiration.  Soil Sci., 127:299-306,




     1979.




 7.   Duncan,  J.  B.  and H. C.  Toor.  An Experimental Study of Three




     Component Gas Diffusion.  Am. Inst. Chem. Eng., 8:38-41, 1962.




 8.   Evans, D.  D.  Gas Movement.  In:  Methods of Soil Analysis,




     C. A. Black, ed.  American Society of Agronomy, Madison,




     Wisconsin,  1965.  Agronomy 9:319-330.




 9.   Fairbanks,  D.  F. and C.  R. Wilke.  Diffusion Coefficients in




     Multicomponent Gas Mixtures.  Indust. Eng. Chem., 42:471-475,




     1950.




10.   Hirschfelder,  J. 0., C.  F. Curtiss, and R. B. Bird.  Molecular




     Theory of Gases and Liquids.  John Wiley and Sons,  Inc., New




     York, 1964.
                                  157

-------
11.  Jones, R.  C.  and W.  H.  Furry.   The Separation of Isotopes by




     Thermal Diffusion.   Rev.  Modern Phys.,  18:151-224,  1946.




12.  Kirkham, D. and W.  L.  Powers.   Advanced Soil Physics.   Wiley-




     Interscience, New York, 1972.




13.  Kittel, C.  Thermal Physics.   John Wiley and Sons,  Inc.,  New York,




     1969.




14.  Marrero, T. R. and E.  A.  Mason.  Gaseous Diffusion Coefficients.




     J. Phys. Chem. Ref.  Data, 1:3-118, 1972.




15.  Maxwell, J. C.  Scientific Papers, II.   Dover Publications, Inc.,




     New York,  1952.




16.  Penman, H. L.  Gas and Vapour Movements in the Soil.   I.   The




     Diffusion of Vapours through Porous Solids.  J. Agr.  Res.,




     30:437-462, 1940a.




17.  Penman, H. L.  Gas and Vapour Movements in the Soil.   II.  The




     Diffusion of Carbon Dioxide through Porous Solids.   J. Agr.




     Res., 30:570-581, 1940b.




18.  Pritchard, D. T. and J. A. Currie.  Diffusion Coefficients of




     Carbon Dioxide, Nitrous Oxide, Ethylene and Ethane in Air and




     their Measurement.  J. Soil Sci., 33:175-184, 1982.




19.  Rixon, A.  J. and B. J. Bridge.  Respiratory Quotient Arising from




     Microbial Activity in Relation to Matric Suction and Air-Filled




     Pore-Space of Soil.  Nature,  218:961-962, 1968.




20.  Russell, E. J. and A. Appleyard.  The Atmosphere of the Soil.  Its




     Composition and the Causes of Variation.  J. Agric. Sci., 7:1-48,




     1915.




21.  Troeh, F.  R., J. D. Jabro, and D. Kirkham.  Gaseous Diffusion




     Equations for Porous Materials.  Geoderma, 27:239-253, 1982.
                                  158

-------
22.  Smith, K.  A.   Soil Aeration.   Soil Sci.,  123:284-291,  1977.




23.  Toor, H. L.  Diffusion in Three-Component Gas Mixtures.   Appl.




     Ind. Chem. Eng. J., 3:198-207, 1957.




24.  Vasaru, G., G. Miller, G. Reinhold, and T. Fodor.   The Thermal




     Diffusion Column.  Veb Deutscher Verlag der Wissenschaften,




     Berlin, 1969.




25.  Wilke, C.  R.   Diffusional Properties of Multicomponent Gases.




     Chem. Eng. Prog., 46:95-104,  1950.




26.  Wood, J. T. and D. J. Greenwood.  Distribution of  Carbon Dioxide




     and Oxygen in the Gas Phase of Aerobic Soils.  J.  Soil Sci.,




     22:281-288, 1971.
                                  159

-------

-------
             APPENDIX B
PARAMETER VALUES USED IN THE PYRITE
          OXIDATION MODEL
                   160

-------

-------
D
 c


     The coupled counter-current diffusion coefficient for the



diffusion of reactants and products within the coarse fragments was



assigned a value comparable to measure diffusion coefficients in


                                              —7   2
soil.  Cathles (1979) used a value of 1.6 x 10   cm /sec for D  in
                                                              c


stone fragments.  Elgawhary   et al. (1970) and Palmer and Blanchar



(1980) measured values of about 1.8 x 10   and from 0.3 - 1.0 x 10


  2
cm /sec for zinc diffusion and for potassium diffusion in soils,



respectively.  A value from the lower range was used for this model


                         -7   2
and was equal to 1.0 x 10   cm /sec.
K
 s



     Numerous workers have investigated pyrite oxidation rates,



unfortunately most have ignored the fact that pyrite oxidation is a



surface reaction, which makes estimation of K  difficult.  McKay and



Halpern (1958) and Braley (1954)  showed pyrite oxidation rate is



proportional to pyrite surface area.  Lorenz and Tarpley (1964)



studied the reaction rate for three different pyrite sizes under



a variety of oxidizing conditions.   Clark (1965) combined their



results with those of Braley (1954), (which gave surface area of



pyrite fragments versus particle  size as measured by screen sieving),



to show the relationship between  reaction rate,  oxidant concentration



and pyrite surface area.  Making  surface area determinations for



irregularly shaped,  framboidal pyrite grains is  extremely difficult.
                                  161

-------
Extrapolating these results from one set of samples to another must




be done with extreme caution due to the great variety of pyrite




grain shapes and sizes.  However, we can assume that the most finely




ground pyrite will tend to have the most uniform fragment shape and




use these values to calculate K  from the data presented by Clark
                               5

                                       —9
(1965).  This gives a value of 1.8 x 10   cm/sec for K  when oxygen
                                                      s



is the oxidizer (curve 1, Clark, 1965, Figure 12, p. 113), and 8.6 x


   /                                Q                         O_i_

10   cm/sec (curve 3) and 8.1 x 10   cm/sec (curve 4) when Fe   is




the oxidizer.




     Singer and Stumm (1968) in studying the reduction rate of Fe




by pyrite found that the reaction is first-order with respect to


  3+
Fe   concentration, but conversely, they found the reaction rate




constant was inversely dependent on the Fe   concentration.  They




had no explanation for this apparent contradiction, but from the




data it appears that their assumption of non-limiting pyrite surface



area was in error.  This can be seen in their Figures 5-18 where



doubling the surface area at fixed Fe   concentrations caused a



doubling of the reaction rate.  If available surface area was not



limiting an increase would not be expected.  Extension of the experi-


                                       3+

ments to greater pyrite surface area-Fe   concentration ratios should




verify this relationship.  Using the Singer and Stumm (1968) values




for the reaction rate constant and the surface area information




presented by Clark (1965), we calculate K  values for Fe   oxidation
                                         s



to be 4.4 x 10   cm/sec when available pyrite surface area is least
                                  162

-------
limiting and 10.0 x 10   cm/sec when most limiting.  The former



value is preferred in view of the argument above.



     In a study by the Ohio State University Research Foundation



(1970) reaction rate and surface area determinations were made for


       3+
both Fe   and 02 oxidants of primary sulphur ball pyrite.  They

                                                               _g

found the maximum value of K  for 0~ reduction, to be 12.0 x 10
                            S      £


cm/sec when dissolved 0,, concentrations were the least and a


                            —8
minimum K  value of 4.4 x 10   cm/sec when dissolved 0- concentra-
         s                                            t


tions were the greatest.  These values indicate a slight coupling



effect of available surface area with 0~ concentrations.  They also


                   3+                                      —6
found K  for the Fe   system to equal a maximum of 4.4 x 10


                         3+                                   —8
cm/sec for the minimum Fe   concentration studied and 7.5 x 10


                         3+
cm/sec for the highest Fe   concentration.



     Pionke, et al. (1980) measured pyrite oxidation rates for fresh


                                                    2-
shale fragments, exposed to 20% 0_, to be 0.45 mg SO,  per hour.



They did not specify pyrite grain size or surface area, but measured



the pyrite concentration in the shale to be 3.8% by weight.  From



the work of Caruccio (1973),  we can assume that the active-pyrite



grains in their samples were on the order of 0.2 microns in size.



Using this approximation and assuming the grains to be spherical,

                                                               	Q

we can calculate the K  value for their material to be 8.3 x 10.
                      S


cm/sec,  in excellent agreement with the other values.  K  values
                                                        S

                                         3+              3+
for the 0« system,  K (0,),  and for the Fe   system, K (Fe  ),  are
         ^          s  £                             s

                                                          —8
summarized in Table Bl.  In this model a value of 8.3 x 10
                                 163

-------
                           3+
Table Bl.  K  values for Fe   and  0-  system
Source
                                         K (cm/hr)
                                          s
     Fe
                                        3+
Clark (1965)
Singer and Stumm
  (1968)
OSURF (1970)
Pionke, et al.
  (1980)
                                             oxidant
8.6 x 10
                                         -8
                                8.1 x  10
                                         -8
4.4 x 10
      ,-8
        -6
                               10. x 10
4.4 x 10
                                         -6
                                 7.5 x  10
                                         —8
18. x 10
                             -8
12. x 10
                             -8
                     4.4 x  10
                              -8
                     8.3 x  10
                              -8
                                  164

-------
cm/sec was used for K  (0_) and a value of 4.4 x 10   cm/sec  for
                     s  ^

     3+
K  (Fe  ) was used.
 s
     The surface area of pyrite per unit volume of fragment was



calculated from the pyrite content of a fragment.  Donaldson, et al.



(1977) measured the surface area for sandstone to be approximately


  4   2
10  cm /g.  This surface area was then modified by the pyrite content



and the pyrite and fragment densities.  In particular:





               a = 104 . pFR . (FPY PFr/Ppy)2/3                [Bl]





where p   is the fragment density (g/cm ), p   is the pyrite density



(5.0 g/cm ), and FPY is the fraction of pyrite in a fragment (g/g).



For a fragment having a 0.002 pyrite fraction and a bulk density of


        3                 23
2.1 g/cm , ct equals 187 cm /cm .
     The thickness of the fragment in which pyrite is being oxidized



can be calculated in a manner similar to that used by Cathles (1979).



First, we assume that the oxidant is diffusing through a leached rim-



of the fragment, where no uptake is occurring, to a zone where pyrite



is oxidizing at a rate proportional to the oxidant concentration.





               DC d2C(x)/dx2 - a KsC(x)  =0                    [B2]
or:
                        2   ° Ks
                        ^ - --=-D(x)  - 0                      [B3]
                                 165

-------
where C is concentration of the oxidant, a  the  surface  area of  pyrite



per unit volume, D  the diffusion coefficient of  the oxidant  in the



fragments, K  the first-order reaction rate constant and  x  is depth



into the fragment.  The solution to Eq.  [B3] is of  the  form





          C(x) = C,  exp(a K /D )1/2 + C0 exp I-(a K /D  )1/2J    [B4]
                  J.        a  C        ^            S  C




Eq. [B4] can be solved subject to the boundary  conditions:
          dC(x)/dx =0.0          x = I, one-half thickness     [B5]



                                         of fragment



          C(x) = C^ = C^        x = r = Ri                   [B6]



where i," is the depth of the leached rim and R = 1.0 - fraction



pyrite remaining/initial pyrite fraction.  Taking the derivative



of Eq. [B4] with respect to x and using Eq. [B5] gives:



               0.0 = b C^ exp(b£) - b GZ exp(-b£)               [B7]


                             1/2
where            b = (ctK /D ) '
                        s  o


Rearranging Eq. [B7]:



          CL = C2 exp(-b£)/exp(b«.) = C2 exp(-2b£)               [B8]



From Eqs. [B4] and [B6] we find:



          CD«, • cn exp(bR£) + C0 exp(-bR£,)                      [B9]
           K.X    X             /



or rearranging:



          CL = (CR£ - C2 exp(bR£ ) / exp(bR£)}                  [BIO]



Eqs.  [B8] and [BIO] can be used to find the values of C  and  C^,



giving:
                                 166

-------
             -      CR£exp(-2b£)


          Cl = exp (bR£-2b£)+exp (-bR£) =  CR£ Cl
             __   _


           2 = exp (bR£-2b£)+exp (-bR£) =  CR£  C2
     The reaction, rate, R,  , for  the fragment  can be expressed for



a unit area and depth, dx,as:



                    R,  = a K  C(x) dx                          [B13J
                     dx      s




The rate per unit surface area for the entire  fragment is:
          R^, = /  a KsC(x) dx = /£ ex KsC(x)  dx                 [B14]





Since no reaction is taking place in the leached  rim,  0 <  x < R£.



Substituting Eq. [B4] for C(x) and Eqs.  [Bll]  and [B12]  for C  and



C_, Eq. [B14] becomes




     ^ = Q Ks fRi [CR£ Cl exP(bx) + CR£ C2 exP(~bx>3  dx       [B15]




which upon integration becomes:





               R  = a KC/b  c  [exp(bi) -  exp(bR£)]
                  - C, [exp(-b£) - exp(-bR£)]l
                     Z                       j
                                                                [B16J
     For the pyrite oxidation model, we want to replace  the  expression



for R,^ with one of the form




                       RT  = a Ks Cx=o  3                       [B17]




Combining Eqs. [B16] and [B17] and solving for 3 gives:
                                 167

-------
               C

              7 -  JC. I«cp(b£) - exp(bRJl)]  -  C,  [exp(-b£)

               x=a
                 8                                              [B18]
                                                x=o



     ffithin the depleted rim there is no uptake of oxidant and  the



C(x) vs x curve will be linear.  The flux of oxidant, Q0 , can be



expressed as:






               *T = Q* = -Dc dC(x) = -Dc  %5=IT              [B19]



and




               C
               -s±- = 1.0 - R&/D  Q /C
               C                co  x=o
                x=o

                                                                [B20]

                    - 1.0 - R£/D
                                c  T




Substituting Eq. [B17] into Eq. [B20] gives






               CL./C    = 1.0 - R&/D a K  3                     [B21]
                R£  x=o             c   s                       l    J



which when substituted into Eq. [B18] gives:



               3 = C  (1.0 - R2./D  a K  3)/b                    [B22]
                                 C    S



which can be rearranged



               3 = C /(b + R£/D  a K  C-)                       [B23]
                               c    s


Eq. [B23] was then used to calculate 9 as a function of oxidant



concentration, reaction rate, diffusion rate and degree of pyrite



weathering.
                                  168

-------
FFR

     The fraction of fragments composing the spoil, FFR, was assumed


to 0.75 (wt./wt.)  This compares with values measured by Pedersen,

et al. (1980) of from 0.5 to 0.9.



FPY and p
	 py

     The average pyrite content of the spoil composing a reclaimed


stripmine was measured by Rogowski (1977) to be approximately 0.001

pyritic S by weight or 0.002 pyrite by weight.  In this model we

assume all the pyrite to be contained within the coarse fragments

which comprise 75% of the total spoil weight.  Thus, we set FPY

equal to .002/.75 or 0.0025.  For a bulk density of 1.5 and a

fragment fraction of 0.75, this value of FPY corresponds to a molar

density of pyrite of p   = 0.0025 • 0.75 •  1.5/120. = 23 x 10~6
                      py

moles pyrite/cm .
PB> *A and
     The bulk density for minesoils was measured by Pedersen, et

                                     3
al. (1980) to average about 1.57 g/cm , which corresponds to a

                                                   3
total porosity of 0.4 (particle density = 2.62 g/cm ).  We have


assumed a percent water saturation of 75% in this model based on


measured values (Rogowski, et al., 1982).  With these values for


total porosity and water content, a good estimate for the air-

                               33            33
filled porosity,  ., is 0.12 cm /cm  and 0.28 cm /cm  for  .
                  A                                        Vr
                                  169

-------
Unfortunately, a mistake was made in running the simulations where


  was set to 0.06 cm /cm  instead.  This corresponds to a bulk
 A
                   3
density of 1.7 g/cm , which is still within the range measured by


Pedersen, et al. (1980).
                  2
     Colvin (1977) measured the diffusion coefficient for 07 in


shale spoil.  He found the ratio of the diffusion coefficient in


spoil to the coefficient in air to be 0.013 at an air-filled


porosity of 0.13.   For these conditions T can be calculated to be


T = 0.13/.013 = 10, which was the base value in the model.
                                 170

-------
                              REFERENCES




1.  Braley, S. A.  Summary Report on Commonwealth of Pennsylvania,




    Department of Health Industrial Fellowship,  Nos. 1,  2,  3,  4,




    5, 6 and 7.  Mellon Institute,  Fellowship No. 326B,  1954.




2.  Caruccio, F.  T.  Characterization of Strip-Mine Drainage by Pyrite




    Grain Size and Chemical Quality of Existing  Groundwater.  In:




    Ecology and Reclamation of Devastated Land,  R. J. Hutnik and G.




    Davis, eds.  Gordon and Breach, New York, 1973.  pp. 193-226.




3.  Cathles, L. M.  Predictive Capabilities of a Finite  Difference




    Model of Copper Leaching in Low Grade Industrial Sulfide Waste




    Dumps.  Mathematical Geol., 11:175-191, 1979.




4.  Clark, C. S.   The Oxidation of Coal Mine Pyrite.  Ph.D.  Thesis,




    John Hopkins University, University Microfilms, Inc., Ann




    Arbor, Michigan.  No. 65-10-278, 1965.




5.  Donaldson, E. C., R. F. Kendall, and B. A. Baker. Surface-Area




    Measurement of Geologic Materials.  Soc. Petrol. Eng. J.,




    15:111-116, 1975.




6.  Elgawhary, S. M., W. L. Lindsay, and W. D. Kemper.  Effect of EDTA




    on the Self-Diffusion of Zinc in Aqueous Solution and in Soil.




    Soil Sci. Soc. Am. J., 44:925-929, 1970.




7.  Lorentz, W. L. and E. L. Tarpley.  Oxidation of Coal Mine Pyrites.




    Report of Invest. No. 6247, U.S. Bureau of Mines, 1963.




8.  Mckay, 0. R.  and J. Halpern.  A Kinetic Study of the Oxidation of




    Pyrite in Aqueous Suspension.  Trans. Met. Soc., AIME 212:301-309,




    1958.
                                  171

-------
 9.  Ohio State University Research Foundation.   Sulfide to  Sulfate




     Reaction Mechanism.   Water Pollution Control Research Series




     14010 FPS 02/70,  Federal Water Pollution Control  Administration,




     Washington, D.C., 1970.




10.  Palmer,  C. J. and R.  W.  Blanchar.   Prediction of  Diffusion




     Coefficients from the Electrical Conductance of Soil.  Soil




     Sci. Soc. Am. J., 44:925-929,  1980.




11.  Pedersen, T. A.,  A.  S. Rogowski, and R.  Pennock,  Jr. Physical




     Characteristics of Some Minesoils.   Soil Sci. Soc.  Am.  J.,




     44:321-328, 1980.




12.  Pionke,  H. B., A. S.  Rogowski, and  C. A. Montgomery. Percolate




     Quality of Strip Mine Spoil.   Trans. ASAE,  23:621-628,  1980.




13.  Rogowski, A. S.  Acid Generation within  a Spoil Profile:




     Preliminary Experimental Results.   In:  Seventh Symposium on




     Coal Mine Drainage Research,  NCA/BCR Coal Conference and  Expo




     IV, Oct. 18-20, Louisville, Kentucky, 1977.  pp.  25-40.




14.  Rogowski, A. S.,  H.  B. Pionke, and  B. E. Weinrich.   Some  Physical




     and Chemical Aspects of Reclamation.  Preprint, paper presented




     at the North Atlantic Region ASAE 1982 Annual Meeting,  1982.




15.  Singer,  P. C. and W. Stumm.  Acidic Mine Drainage:   The Rate




     Determining Step.  Science, 167:1121-1123,  1970.
                                  172

-------
              APPENDIX C
COMPUTER LISTING AND EXAMPLE OUTPUT OF
       ACID MINE DRAINAGE MODEL
                173

-------

-------
             Computer Listing of Acid  Mine Drainage Model
  1.     //MNWXXXXX      JOB      (DXJ79)
  2.     // EXEC PGM-UMSG,PARM-(INTERACT, 'TO D1J JOB RUNNING')
  3.     //DELETE DD VOL-REF-MEN.P65440.DlJ.LIB,
  4.     //  DSN-MEN.P65440.D1J.DMNLS8,
  5.     //  DISP-(OLD,DELETE)
  b.     /*
  7.     // EXEC FWCG.PARM-NOSOURCE
  8.     /'JOBPARM FULLSKIPS,UCS-TN,FORMS-16,V-NORMAL
  9.     //SYSIN DD *
 10.           IMPLICIT REAL'S (A-H.O-Z)
 11.           INTEGER O.OPRINT
 12.           REAL'S NA,NB,NAI,NBI,KB(50>,KBMAX,NC2,NC(50),NCI,NCIO,LSMAX,
 13.           1KBO,KB1,KB2,KBO(50),KBA(50),KBH(50),KBU50),LS(50).LSO(50)
 14.           REAL** KSP,K11,K12,K13,K14>K22,K.ML,KMHL,KA2,KOX1,KOX2,INFILR,
 15.           IL,KSO,KSF,DCO,DCF,KLS,MWLS,DC02
 16.           DIMENSION X(50), YA(50),YB<50),NA(50),H2C03(50) ,OH2C03(50),
 17.           1NB(50),R(50),T(50),PHI(50),FFR(50),RHOB(50),TE(50),TO(50),
 18.           2FPY(50),RHOFR(50),UA(50),UB(50),UAN(50),EA(50),EB(50),
 19.           3DX(50),DA(50),DB(50),FA(50),FB(50),TA<50),1AO(50),YBO(50),
 20.           4PYO(50),QA(50),qB(50),PY(50),YAX(50),YBX(50),YAH(50),YAL(50),
 21.           5PYX(50),NAI(50),NBI(50),UAO(50),UBO(50),U(50),RR(50),
 22.           6ALPHA(50),DELTAO(50),DXFO(50),DXF(50),TB(50),C03FL(50),TFE3C(50),
 23.           7AFE3T(50),FE2(50),FE3(50),PYFC50),PY02(50),TFE3CX(50),
 24.           8ST(50),FE2X(50),YAXO(50),YBXOC50),GA(50),GB(50 ) ,
 25.           9FFEO(50),WC(50),010(50),OH(50),OE3(50),OEOH(50 ) ,OEOH2(50),
 26.           10EOH4(50),OE20H2(50),OES04(50),OEHS04(50),OEOOB(50),OS04(50)
 27.           2,H(50),FEOOH(56),OEOR3(50),OST(50),OE2(50),OHBOX(50),HBACT(50),
 28.           30E20XOO) ,OHS04(50),G1(50) .DELTAFC50) ,HC(50) ,AHLS(50) ,AHLSX(50),
 29.           40E2B(50),OBC(50),FE3FL(50),FE2FL(50),HFL(50),DX2(50),
 30.           5STFLC50),HCFL(50),CAFL(50),CA(50),OCA(50),RLS(50),FLS(50)
 31.           EXTERNAL DO,DC02,DAB,DBC,DAC,FLDX
 32.           COMMON/ADEL/KSO,DCO,KSF,DCF.L
 33.           1/ACHEMI/FE31,H1,TFE3C,K.SP,KA2,K11,K12,K13,K14 , K22 , KML , KMHL , MCHEMI
 34.           I/ACHCAL/OIO,OEOH,OEOH2,OEOH3,OEOB4,OE20H2,OES04,
 35.           20EHS04,0£OOH,OS04,OHS04,ST1FE2,OR>OE3,OE2,
 36.           30ST,CA,OCA,H2C03,OH2C03,I/AHADJ/HC,OHC,HCFL,GA,GB/AFLUX1/H,FE3,X,
 37.           4DX,DT,INFILR,N
 38.           EQUIVALENCE (FFEO.OE2B)
 39.           CALL ERRSET(208,256,-1,1)
 40.           CALL TIMREM(IREMO)
 41.           READ, IDUMP
 42.
 43.
 44.
 45.
 46.
 47 .
 48.
 49.
 50.
 51.
 52.
 53.
 54.
 55.
 56.
 57.
 58.
 59.
 60.
 61.
 62.
 63.
 64.
*5.
 66.
 67.
 68.
 69.
 70.
 71.
                                     174
c
c
c



















c
c
c
c

SET IN CONSTANTS

MCHEMI — 10
MM-MSMEEP-M-0
RHOPY-5.0
PI-3. 14159265
KSP-79433.0
KA2-.073
Kll». 00116
K12-2.0E-06
K13-1.0E-12
K14-2.5E-22
K22-6.0E-06
KML-84.0
KMHL-4.0
KOX1-1.3E-19
KOX2-1.7E-12
BO-1./3.5
BF-1./14.
EMIN-. 173
IFCIDUMP.EQ. 1 ) THEN

INITIALIZE VALDES IF T-0.
(IDUMP-1 )

C
C
c



















c
c
c
c

-------
  72.            READ31,OPRINT,MBAT,1.,RADIUS,Y,Z,
  73.           1DCO.KSO, DCF,KSF,KLS,P,RC,El,ET,KT,INFILR,MVLS,RHOLFi!
  74.            INFILR-INFILR/(365.*86400.)
  75.            PR-P/RC
  76.            DX(1)-1.
  77.            X(l) —.5
  78.            IP(MT.Eq.l)  THEN
  79.            DDT-DT-86400.»2.
  80.            ELSE
  81.            DDT-DT-MT
  82.            END IF
  83.            READ35,DEPTH,ML,VINC.NPL
  84.            N-NL*liPl.+ l
  85.            IF(S.EQ.l) READ31.N
  86.            READ36,X(2),YAd ) ,YBd),Td) ,PHI(1),WC(1 ) ,FFR(I ) ,
  87.           1RHOBO ),FPY(1 ) ,FFEO(1 ) , RHOFRd ) , TE ( 1 ) , GA C 1 ) , GB ( 1 },HBACT(1 ) ,
  88.           2RLS(1),FLS(1)
  89.            DO 30 1-2,N
  90.            IF(X(2).GT.O.O)  X(I)-X(I-I)+X(2)
  91.            IF(YA(1).NE.-1.) YA(I)-YAd-I)
  92.            IFmU).NE.-l.) fB(I)-YBd-l)
  93.            IF(FFRd ).NE.-1.)  FFR(I)-FFRd)
  94.            IF(RBOB(1).NE.-1.)  RHOB(I)-RHOB(1)
  95.            IF(FPY(1 ).NE.-1.)  FPYd)-FPY(l)
  96.            IF(FFEO(1KNE.-l.)  FFEO(I )-FFEO(1)
  97.            IF(RBOFR(1J.NE.-l.) RHOFRd )-8HOFR(l )
  98.            IF(Td).NE.-l.)  TCI)-T(l)
  99.            IF(PHI(1).NE.-1.)  PHKI)-PHI(l)
100.            IF(WCd).NE.~l.) VC(I)-WCd)
101.            IF(TE(I).HE.-25.)  TE(I)-TEO)
102.            IF(GAd).NE.-l.) GA(I)-GAd)
103.            IF(GB(1 ).NE.-1.) CB(I)'GBd)
104.            IFCRLSd ).NE.-1. )  KLS(I)-RLSd)
105.            IFtFLSd ) .NE.-l. )  FLS(I)-FLSd )
106.            IFCHBACTd ).NE. 1.)  HBACT( I )-HBACT ( 1 )
107.            IFCI.KE.21)  FLSd)-0.
108.      30     CONTINUE
109.      31     FORMAT(I2,/,I2,/,13(G10.0,/) , I 10,3(/,G10.0))
110.      35     FORMAT(G10.0,/,I2,/,G10.0,/,I2)
111.      36     FORMAT(GIO.O)
112.            IF(X(2).EQ.-1. )  READ,(X(I),I-1,N)
113.            IF(X(2).EQ.O.) THEN
114.            NI-NPL
115.            DO 32 1-2,NL
116.            NI-NI+NPL*VINC**(1-1)
117.      32     CONT1K0E
118.            DI-DEPTH/NI
119.            11-2
120.            DO 34 J-l.NL
121.            DO 33 1-1 ,NP1.
122.            DX(II)-D1
123.            X
-------
151.
152.
153.
154.
155.
156.
157.
156.
159.
160.
161.
162.
163.
164.
165.
166.
167.
168.
169.
170.
172 .
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
184.
185.
186.
187.
188.
189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
201 .
202.
203.
204.
205.
206.
207.
208.
209.
210.
211 .
212.
213.
21*.
215.
216.
217.
218.
219.
220.
221.
222.
223.
224.
225.
40
1/,10X,'FFR- ' .F6.3.8X,'RHOB-  ',F6.3,9X,'FPY- ',F7.4,/,1 OX,
2'FFEO- ',F6.3,7X,'RHOFR-  ',F6.3,8X,'TE-  ' , F6.3,/,1 OX,
3'GA- ',F8.4,7X,'GB-  '.F6.3.11X,'HBACT- ',F6.3/,1 OX,'RLS-  '.
4F8.4,6X,'FLS-  ',F8.4,8X,'RHOLFR-  ',F8.4,/,1 OX,'L-  '.F6.3)
 DO 40    J-l.N
 I-N-J+1
 RLSd)-2.*MULS*lCl.S/(RLSd)*RHOLFR)
 FLSd)-FLSd)*RHOBd)/(2.*MVLS)
 H2C03d)-DC02(TEd))*YB(I)
 TO(I)-TEd)-TEd)+273.
 ALPHAd)-(FPYd)*RHOFRd)/RHOPY)**(2./3. )*RHOFR(I )*10000.
 Zl-WC(I);  22-PHId)
 WC(I)-Z2*Z1
 PHId)-Z2*(l .-Zl)
 Td)-PHId)/T(I)
 OHBOXd)-OE2Bd)-OE20Xd)-UBd)"0.
 HCd)-1.0-20
 KSA(I)-KZO(I)-KBd)-O.D-15
 LSOd)-LSd)-PYOd)-UAOd)-DXFOd)-PYd)-1.0;PYF(I)-PY02(I)-1.0-25
 IF(FPYd) .EQ.0.0)  PY(I)-0.0
 IF (RLS d)*FLSd). EQ.0.0)  LSO d )-LS (I )-0 . 0
 YAXOd)-YAXd)-YAOd)-YA(I)
 YBXOd)-YBXd)-YBOd)-YB(I)
 CALL DELCALC ALPHA d) , PY ( I ) , DELTAOd ) , DELTAF (I ) )
 NAId)-NA(I)-1.0;  NBI d )-NB d ) — 1 .
 CAd)-OE2(I)-FE2(I)-5.00-07
 STCD-5.0D-05
 FEOOHd)-1000.«FFEO(I)*RHOBd)/C89.«UC(I))
 H(I>-1.00-05
 CALL CHEMI(-1,H d),FE 3 d),FEOOH d),AFE3T(I),G1d))
 Hd)-OH(I); FE3d)-OE3d)
 UANd)--DCO*RHOBd)*FFRd ) *DO (TE (I ) ) * ALPHA C I ) *KSO / ( 3 2 . *RHOFR d ) *L)
 UAd)-YA(I)*UANd)*DELTAO(I)/(ALPHAd)*DELTAOd)*KSO*
1L*(1 .-PY(D)-f-DCO)
 U(I)-UAd)/YAd)
 DXFCD —120.* BF*DCF* ALPHA ( I) *KSF/( RHOFR d )«FPY(I)*L)
 R(D—0.05
 CONTINUE

 LSOO )-LS(l )-PY(l )-UA(l )-DXF(l)-OAOd)-UAK( 1)-FEOOH(1 )-0.0
 OCA(1)-CA(1)-ST(1)-FE2(1)-HC(1)-FE3(1)-1.OE-15
C
C
C
C
ELSE
READ(43,202)
1TZ1,TZ2,TZ3,1
READ(43,201)
READ(43,201)
READ(43,201)
READ(43,201 )
READ<43, 201 )
READ(43,20I )
READ(43,201 )
READ(43,201)
READ(43,20j )
READ(43,201 )
READ(43,201 )
READ(43,201 )
READ(43,201 )
READ(43,201)
READ(43,201)
READ(43,201 )
READ(43,201 )
READ(43,201)
READ(43,201)
READ(43,201)
READ(43,201)
READ(43,201 )
READ(43,201 )
READC43.201 )
READ(43,201 )
C
RETRIEVE DUMPED VALUES C
CIDDMP-0) C
C
OPRINT , 0, N , NBAT ,MBAT ,MT ,E 1 , PR, DT , DDT , TT , ET , TUT ,
rZ4,TZ5,TZ6,TZ7,TZ8,KSO,DCO,KSF,DCF,L, IKFILR.TFEO
(PY(I),I"1,N)
(PYOd).l-l .N)
(TECI ) ,1-1 ,N)
(TOd) ,1-1 ,NT)
CX(I),I-1,N)
(DXd),I-l,N)
(Td),l-l,N)
(UANd ) ,1-1 ,N)
(UA(I),I-1,K)'
(UAOd).I-J ,N)
(ALPHAd) ,1-1 ,N)
(PHId) ,1-1 ,N)
( WC (I ) , I - 1 , N )
(RHOBd).I-l ,N)
(RHOFRCI) ,1-1, K)
(FFRd ) , 1-1 ,N)
(FPY(I) ,1-1, K)
(FEOOHd) ,1-1 ,N)
(ST(I),I-1,K)
(FE3d),I-l ,N)
(FE2(I) ,1-1, N)
(Hd).I-l.N)
(DXFd ) , 1-1 , K)
                                        176

-------
226.            READ(43,201)  (DXF0(1 ) , I-1, )O
227.            READ(43,201)  (OE2B(I ) . I - 1,K)
228.            READ(43,201)  (OE20X(I),1-1,N)
229.            READ(43,201)  (KB(I),I-1,N)
230.            READ(43,201)  (KBOCI ) ,1-1,N)
231.            READ(43,201)  (010(1),I-l,N)
232.            READ(43,201)  (HC(I),1-1.N)
233.            READ<43,201)  (HBACT(I).I-l ,N)
23*.            READ(43,201)  (OHBOXd) .1-1 ,N)
235.            READ(43,201)  (Rd>.I-l.N)
236.            READC43.201)  (OE3(I) .1-1,N)
237.            READ(43,201)  (OE2(I),I-1,N)
238.            READ<43,201)  (OH(I),1-1,N)
239.            READ(43,20I)  (YAO(I ) ,I-1 ,N)
240.            READ(43,201)  (YBO(I ) ,I-1 ,N)
241.            READ(43,201)  (GA(I),1-1,N)
242.            READ(43,201)  (GB(I ) ,I-1,N)
243.            READ(43,2Q1)  (CA(I),1-1,N)
244.            READ(43,201)  (LS(I ) , I-1 ,N)
245.            READ(43,201)  (LSOd),I-l,N)
246.            READ(43,201)  (RLS(I),I-1,N)
247.            READ(43,201)  (FLS(I),I-l,N)
248.            READ(43,201)  (H2C03(I),I-1,N)
249.            READ(43,201)  (OH2C03(I),I-1,N)
250.            FPY(1)-0.0; PYF(1)-PY02(1)-l.OD-25;  E1-.00001
251.            DO 38 I-l.N
252.            CALL DELCAL( ALPHA d),PYd),DELTAO(I),DELTAF(I))
253.            CALL CHEMIU,H(1),FE3(I) ,FEOOH(I) ,A.FE3T(I) ,Z)
254.            KBACI)-KB(I)
255.            NAI(I)-NA(I)-1.0;  NB1(I)-NB(I)"-!.
256.            GKD-1.0
257.      38    CONTINUE
258.            IF(l.EQ.Z) THEN;  TT=TT-DT/86400.
259.            DT-D0T-DT*2.
260.            TT-TT+DT/86400.  ;  END  IF
261.            END IF
262.            HCFLCD-0.
263.            EB(1)-EA(1)-1.0
264.            EA(1)-DB(1)-0.
265.            PYX(1)-PYO(l)-PY(l)
266.            B(l)-NA(l)-NBC1)-0.0
267.            OHCU)-HC(l)
268.            KBAd)-KB(l)
269.            QA(l)«YAX(l)-YA(l)
270.            QB(l)«YBX(l)-TB(l)
271.            NAI(N)-NB1(N)-TB(N)-TA(N}-0.
272.             Z-FLUX(O)
273.      c***«»*********«**********««*******«****«****«*******************C
274.      C                                                                  C
275.      C     CONSTANT  PART OF  TRANSMISSIBILITIES  ARE CALCULATED          C
276.      C                                                                  C
277.      c*»***«**««*******»*******•»««*»**•»*«**«**»***•*«»****«***«.*«**c
278.      400   CALL TIMREM(IREM)
279.            1F(2.*IREM-IREMO.LT.50) GO  TO 200
280.            IF(TT.GT.1850. )  GO TO  200
281.            IREMO-IREM
282.            PRINT789
283.      789   FORMAT('l')
284.            DO 70 I-l.N
285.            TEMP-TE(I)
286.            TA(I)-Td)*DAB(TEHP)«DAC(TEMP)/TE«P
287.            TBd )-Td)*DAB(TEMP)*DBC(TEMP)/TS>lP
288.      70    CONTINUE

290.      C                                                                  C
291.      C     CALCULATE  EXTRAPOLATED  VALUES AND LEACH RATES               C
292.      C                                                                  C
293.      C****************************************************»****«****««c
294.            YAXOd )-YAX(l )-YA(l);  YBXO ( 1 ) -YBX ( 1 )-YB (1 )
295.            DO 50 1-2,N
296.            ASX-LSd) + (LSd)-LSOd))
297.            LSOd)-LS(l)
298.            LSd)-DMAXl(ASX,O.OD+00)
299.            YAXd)-YAd)-KYAd)-YAOd))
300.            YBXd)-YBd) + (YB(I)-YBO(I))
301.            IF(YAXd)-LT.O. )  YAXCI)-YA(I)/2.  ; IF ( YBX (I )-LT. 0. )  YBX (I )• YB < I)/2 .
302.            IF(O.GE.l)  THEN
303.            Z3-H(I) + (Hd)-OHd»
304.            Zl-FE3d)-HFE3d)-OE3(D)
305.            Z2-FE2(I) + (FE2d )-OE2(I ))
30b.            OE3d)-FE3(I)
                                        177

-------
307.           OE2U )-FE2(I )
3oe.           mzi.cT.o.)  THEN;  FE3(i)-zi; ELSE; FE3(i)-FE3d)/2.;  END IF
309.           IF; OBC2-DEC(TE2 )
                                        178

-------
390.            NC2-(NC1-NC10 ) r. .-t-NCU'
391.            NCIO-NCI
392.            IF(M.CT.O) NC2-(NC2+NC(I))/2.
393.            NCd)-NC2
394.            2-1.; MKM-0
395.            WHILE U.GT.1.OD-1i. AND . MMM .LT.2 2)
396.            Zl-NA(I);  Z2-NBCI)
397.            NAd)-T2«DAB2*DAC2/(YB2*(DAC2-DAB2)-'SA2*DA»2-H>AB2><'(22«YA2/
398.           1(T2*DAB2) + NC2*YA2/(T2*DAC2)-PR/TE2»((YA3-YA2)«XXH-(YA2-YA1)»XX2))
399.            NB(I)-T2*DAB2*DBC2/(YA2*(DBC2-DAB2)-YB2*DAB2+DAB2)*(NA(I)«YB2/
400.           1(T2*DAB2)+NC2«YB2/(T2*DBC2)-PR/TE2*((YB3-YB2)*XX1+(Y82-YB1)*XX2))
401.            Z"DABS(NA(I)-Z1 )-t-DABS ( NB ( I )-Z 2 )
402.            MMM-MMW+1
403.            END WHILE
404.      74     IF(NAd).EQ.O.O)  THEN
405.            RRCD-R2 — 1.0D-H6
406.            IF(NBd).EO.O.O)  RR( I )-R( I )-R2-PI
407.            DA2-1.
408.            ELSE
409.            R2-RR(I)-NBd)/NA(I)
410.            RMAX-DMAX 1(RMAX,1.-DABS(R(I)/R2)}
411.            Rd)-R2-(R.Rd)-»-Rd))/2.
412.            DA2-YB2*(DAC2-DAB2)-YA2*(R2*DAC2+DAB2)+DAB2
413.            END IF
414.            IF(R2.EQ.O.O)  THEN
415.            DB2-1.
416.            ELSEIF(R2.EQ.PI)  THEN
417.            DB2-1.
418.            ELSE
419.            IF(R2.EQ.-1 .OD-H6)  THEN
420.            Z-0.0
421.            ELSE
422.            Z-DBC2/R2
423.            END IF
424.            DB2-YA2*(DBC2-DAB2)-YB2*(Z+DAB2)+DAB2
425.            END IF
426.            FAd)-4.»PR*TA3«TA2/(TA3*DA2+TA2*DA3)
427.            FB(I)-4.«PR*TB3*TB2/(TB3*DB2+TB2«DE3)
428.            DX2(I)-(X2-X1)*(X3-X2)+(X3-X2)**2
429.      80     CONTINUE
430.            Z-0.0
431.            D085J-:,N
432.            I-N+2-J
433.            DA(I)-FA(I-1)/DX2(I)
434.            DB(I)-FB(I-1)/DX2(I)
435.            FA(I)-FA(I)/DX2(I)
436.            FB(I)-FB(I)/DX2(I)
437.            FE21-FE2U)
438.            Hl-H(I)
439.            IF(NB(I) .EO.0.0)  FB(I)-DBd)
440.            Ud)-UAN(I)«DELTAO(I)/(ALPHA(I)*DELTAO(I)*KSO*
441.           1L*(l.-PY(I))+DCO)
442.            IF(O.EQ.O) UAO(I)-U(I)«YA(I)
443.            X1-DMIN1(0.OD+00,2.*HBACT(I)*(TE(l)-273.»
444.            EAd)--(DA(I>+FA(I)-(0(l)-«CCI)/4-*FE21 * ( KB { I )-i-KOXl
445.           l*Hl**(-2)+KOX2) + Xl ) )-PHI(I)*PR/(TE(I)*DT)
446.            EB(I)--(DB(I)+FBd)+DC02(TE(I))/CDT*1000.)}
447.            QAd)--PHId)*FR*YAO(I)/(TO(I)*DT)
448.            OB(I)-X)*YA(I)-((OH2C03(I)+C03FL(I))/ 1 000 . +(LSO(I)-LS(I))»FLS(I))
449.           1/DT
450.            Z-1 -+Z+DABS(QB(I))+YB(I )/TE(I )-YBOCI )/TOd )
451.            IF(Z.EQ.O.O) THEN
452.            QB(I)-YBd)
453.            EB(I)-1.0
454.            DB(I)-FB(I)-FBd-l)«0.
455.            ELSE
456.            EB(I)-EB'-E".A» "•:"".tv.v j>(Ay« A".* ^ -?Y'! AX "'•'": A>«F T>'/v" C. r

-------
87
120
C
C
C
CONTINUE
UT-0.
DO 115 J-l.N
I-N+l-J
PYRITE CONSUMPTION IS CALCULATED

C
C
C
 475.            IF(YAX(J).LT.O.) THEN;  YAX{J)-YA(J> ;  YA(J)-YA(J)/2 .
 476.            PRINT.J,YAX(J),'-YAX' ,YA(J) , '-YA'; END IF
 477.            IF(DB(J).EQ.O.) THEN
 478.            YBX(J)-YBX(J-1)«DEXP((NA(J)/T(J)+NA(J-l)/TfJ-l))«(X(J)-X(J-l))/
 479.           1(2.*DAB(TE(J))«PR/TE(J)))
 480.            END  IF
 481.            YAX(J)-YA(J)-(YA(J)+YAX(J))/2. ; YBX ( J )-YB(J )-(YB(J)+YBX(J))/2.
 482.
 483.
 484.
 485.
 486.
 487.
 488.
 489.
 490.
 491.            IFd.EQ. 1)  GO TO 115
 492.            TOE3C-OEOH(I)+OEOH2(I)+OEOH3(I)+OEOH4(I) + 2.*OE20H2U)+OEHS04(I)
 493.           l-t-OES04(I)-K)EOOH(I)
 4 94 .            AMAX-DMAX1(AMAX,DABS(YAX(I)-YAXO( I ) ) )
 495.            BMAX-DMAX1(BMAX,DABS(YBX(I)-YBXO(I)))
 496.            YA2-YA1-YA(I)
 497.            YBl-YB(I)
 498.            YAXOCI)-YAX(I); YBXO ( I )-YBX ( I )
 499.            IF(FPY(I).EQ.O.) THEN;  Z3-Z4-0.0; GO TO  111;  END IF
 500.            Z3-DT*U(I)*YA1«BO*120./(RHOB(I)*FFR(I)  '
 501.           1*FPY(I))
 502.            Z4-DT*DXF(I)*AFE3T(I)*FE3(I)*DELTAF(I)/(L*D£LTAF(I)«ALPHA(I )
 503.           1*KSF»O .-PY(I))+DCF)
 504.            PY(I)-PYO(I )+Z3+Z4
 505.            IF(PY£I).LT.O.O) THEN
 506.       .     PY(I)-PYO(I)/2.
 507.            Z3--Z3«PY(I)/(2.*(Z3+Z4))
 508.            Z4—(PY(I)+Z3)/2.
 509.            END  IF
 510.            IF(O.NE.O)CALL  DELCAL(ALPHA(I),PYO(I ) , DELTAO(I),DELTAF(I))
 511.      Ill    IF(O.EQ.O)  THEN
 512.            PYU)-PYO(I)
 513.            UAO(I)-0(I)»YA1
 5U.            END  IF
 515.
 516.
 517.
 518.
 519.
 520.            IF(O.EQ.O)  THEN
 521.            AFE21-AFE22-DFE20X-AS04-0.
 522.            ELSE
 523.            FE32-FE31-FE3U)
 524.            FE32-FE31
 525.            H2-R1-HCI)
 526.            KB2--1.
 527.            KBl-KB(l)
 528.            AFE21—Z3»1000.*RHOB(I ) *FFR ( I ) *FPY ( I ) / ( 1 20. *WC( I ) )
 529.            AFE22— 1 5 . *Z4 * 1 000 . *RHOB ( I ) *FFR ( I) *FPY ( I ) / ( 1 20. *WC( I ) )
 530.            FE21-FE2(I)
 531.            FE22-FE2X(I)
 532.            TC-(TO(I)-HE(I))/2.-Z73.
 533.            XTE— 1 .23D-05*TC*«3-4.33D-04«TC**2 + 0.0657*TC-0.255
 534.            IFCH2.LE.O.) PRINT,I,E2, '-H2 ' , X
 535.            PH—DLOG10(H2)
 536.            XPH--.348*PH**2+2.26*PH-2.66
 537.            IF(XPH.LE.O.O) THEN; Z2 5-KB2-Z5-KB(I)-0.0; GO TO 803; END IF
 538.            X02-YA2/.01
 539.            IF
-------
476.            PRINT,J,YAX(J), '-YAX' ,YA(J), '-KA';  END IF
477.            IF(DB(J).EQ.O.) THEN
478.            YBX(J)-YBX(J-1>*DEXP«NA(J)/T(J)+NA(J-1 ) /T(J-1))«(X(J)-X(J-l))/
479.           1(2.«DAB(TE(J))*PR/TE(J)))
480.            END  IF
481.            YAX(J)-YA(J)-(YA(J)+YAX(J))/2. ;  YBX ( J )-YB ( J )- ( YB ( J )+YBX( J ) ) t'l .
482.     87     CONTINUE
483.            UT-0.
484.     120    DO  115  J-l.N
485.            I-N+l-J
486.     c******«*****»************«***************»*********************«C
487.     C                                                                   C
488.     C       PYRITE  CONSUMPTION IS CALCULATED                           C
489.     C                                                                   C
490.     c********•«**»*****«*******•******«**•**»*****•*««**«****»«**««**c
491.            IFd.EQ.1)  CO TO 115
492.            TOE3C-OEOHd)-K>EOH2d)+OEOH3d)+OEOH4d )-f2.*OE20H2(I)+OEHS04(I)
493.           l-K>ES04d)+OEOOHd)
494.            AMAX-DMAXl(AMAX,DA.BS(YAXd)-YAXOd)) )
495.            BMAX-DMAX1(BMAX,DABS(YBX(I)-YBXO(I)))
496.            YA2-YAl-YAd)
497.            YBl-YB(I)
498.            YAXOd)-YAXd ); YBXO(I)-YBX(I)
499.            IF(FPY(I).EQ.O.) THEN; Z3-Z4-0.0; GO  TO 111; END  IF
500.            Z3-DT*Ud)*YAl*BO*120./(RBOBd)*FFRd)
501.           1*FPY(D)
502.            Z4-DT*DXFd)*AFE3Td)*FE3d)*DELTAFd)/(L*DELTAFd)*ALPHAd)
503.           l*KSF*d.-PYd))+DCF)
504.            PYd)-PYOd)+Z3+Z4
505.            IF(PYd) .LT.0.0) THEN
506.            PYd)-PYOd)/2.
507.            Z3 — Z3*PY(I)/(2.*(Z3+Z4))
508.            Z4 —(PY(I)+z3)/2.
509.            END  IF
510.            IF(O.NE.O)CALL DELCAL ( ALPHA ( I ) , PYO ( I ) , DELTAO ( I ) , DELTAF ( I) )
511.     Ill    IF(O.EQ.O)  THEN
512.            PYU)-PYO(I)
513.            UAO(I)-U(I)*YA1
514.            END  IF
515.
516.
517.
518.
519.
520.            IF(O.EQ.O)  THEN
521.            AFE21-AFE22-DFE20X-AS04-0.
522.            ELSE
523.            FE32-FE31-FE3CI)
524.       .     FE32-FE31
525.            H2-H1-HCI)
526.            KB2 — 1.
527.            KBl-KBd)
528.            AFE21—Z3*1000.*RHOB(I)*FFR(I)*FPYd)/(120.«UCCI))
529.            AFE22 — 15.*Z4*1000.*RHOBd)*FFRd)*FPYd')/d20.*WCd))
530.            FE21-FE2(I)
531.            FE22-FE2XCI)
532.            TC-(TOd)+TEd))/2.-273.
533.            XTE--1.23D-05*TC**3-4.33D-0'*TC**2+0.0657*TC-0.255
534.            IF(H2.LE.O.) PRINT,I,H2,'-H2',K
535.            PH--DLOG10CH2)
536.            XPH--.348«PH**2+2.26*PK-2.66
537.            IF(XPH.LE.O.O) THEN;  Z2 5-KB2-Z5-KB(I)-0.0; GO TO 803;  END IF
538.            XC2-YA2/.01
539.            IF(X02.GT.1.) X02-1.0
540.            EM-EMIN/(XTE*X02«XPH)
541.            IFUTE.LE.O. -OR.XPH.LE.O.O.OR.X02.LE.O.) TREK
542.            PRINT,l.XTE,'-XT',X02,'-X02',XPH,'-XPH',YA2,'-YA'
543.            KB(I)-KB1/100.
544.            GO  TO  803
545.            END  IF
546.            IFCEM.GT. 1. 18) THEN;  KBd)-0.0;  CO TO 803; END  IF
547.            Z25-Z5-G1(I)**(-4)*H2*YA2*«.25*DEXP(( .46-EM)/(8.6D-05*TE(I ) ) )
548.            Z18-Z5-(FE31-Z5«FE22)/(1.+Z5)
549.            Z5-FE22+Z5
550.            KB2-KBd)-(0£2d) + AFE21*YA2/YAl+AFE22*d . -Z 1 8 /FE3 1 )+FE 2FL CI )
551.           1-Z5)/(Z5*ET*YA2*1000.)
552.            IF(KBd). LT.0.0) KB ( I) - . 0 1 *DSQRT ( KB 1 ) *3 . 3D-08
553.     803    Z11-2.*DT/8640G.
554.            Z11-DMIN1(2.0D+01,Z11)
555.            KB(I)-DM1N1 (KB(I).KBO{I)*2.*«211  '.
55b.            Z16-4.*WC(I)*2.75D-l]/f.O)**r*FE225
C
C KB IS CALCULATED
C
C
C
C
                                        181

-------
557.
558.
559.
560.
561.
562.
563.
564.
565.
566.
567.
568.
569.
570.
571.
572.
573.
574.
575.
576.
577.
578.
579.
580.
581.
582.
583.
584.
585.
586.
587.
588. '
589.
590.
591.
592.
593.
594.
595.
596.
597.
598.
599.
600.
601.
602.
603.
604.
605.
606.
607.
608.
609.
610.
611.
612.
613.
614.
615.
616.
617.
618.
619.
620.
621.
622.
623.
624.
625.
626.
627.
628.
629.
630.
631.
632.
633.
634.
635.
636.
KBd )-DMlNl (KB(I) ,Z16)
KBAd)-KB(I)
KB(I)-KB1
IF(KBA(I)*YAl*FE21.LT.1.0D-22) KBA(I)-0.0




IF(K.BA(I)«KB(I).NE.O.) THEN; KBMAX-DMAX1 ( KBMAX, DABS (KB ( I )-KBA (I ) )
1/KBCD)
ELSE; JCBMAX-DMAX1 (KBMAX, DABS ( KBA( I) ) ) ; END IF
IF(KBAd).EQ.KB2) FERMAX-DMAX1 ( FERMAX . DABS ( Z25-FE3 1 /FE22 > /Z25 )
C C
C FE3.FE2 AND H ARE CALCULATED C
C C
FE2Xd)-(OE2d)+AFE21+AFE22+FE2FL(I)
1 >/{r.+DT*1000.«YAl*(KOXl*Hl**(-2)
2-t-KOX2+KBd)))
WHILE(FE2X(I) .LT.O. )
FE2FL
1F(H(I) .GT. AST) THEN -
STL-AST
ELSE; STH-AST; END IF
AST-(STH+STL)/2.
IF(TT.LT.O) PRINT,I,M,IST,STH,'-HH',H(I),'-H',AST,
1 '-AST' ,STL, '-HL' ,HFL(I) , '-HFL'
IST-IST+1
IFdST.CT.20) THEN
DFE2B-DFE2B/2.
KB(I)-KB(I)/2.
END IF
END WHILE
END IF
C
C S04 IS CALCULATED
C
AS04--2.*(Z3+Z4)*1000.*RHOB(I)*FFR(I)*FPY(I)/
1(120. *WC(I))
ST(I)-OST(I )+AS04+STFL(I)
IF(ST(I ) .LT.O. ) THEN











/
)





















182

-------
637.
63E.
639.
640.
641.
642.
643.
644.
645.
646.
647.
648.
649.
650.
651.
652.
653.
654.
655.
656.
657.
658.
659.
660.
661.
662.
663.
664.
665.
666.
667.
668.
669.
670.
671 .
672.
673.
674.
675.
676.
677.
678.
679.
680.
681.
682.
683.
684,.
685.
686.
687.
688.
689.
690.
691.
692.
693.
69-.
695.
696.
697.
696.
699.
700.
701.
702.
703.
704.
705.
706.
707.
708.
709.
710.
711.
712.
713.
~T 1 i.
















c
c
c
























c
c

c








317

114













115

C
C
C
1ST-)
STH-2.*OST(I)+AS04
STL-0.0
AST-(STH+STL)/2.
WHILE (DABS 
-------
715.            1F(«.GT.50)  PR I NT 2,M,AMAX,BMAX,PYMAX,HMAX,FEMAX.FERMAX
716.      :     FORMATC/.1X,'  M-  ',13,'  AMAX-' ,D11.3, ' BMAX-',Dl1.3,
717.           1' PYMAX-' ,D11.3,'  HMAX-',011.3.' FE3MAX-' ,D 1 1.3,
718.           2* FERMAX-'.D11.3)
719.            IF{OPRINT.GE.3)  THEN
720.            DO 900 1-1,K
721.      900   PRINT12,I,X(I),YA(I),YB(I),NACI).NB(I),UA(I).UB(I),R2,PYU>
722.            END IF
723.            IF(M.LT.35)  THEN
724.            IF(AMAX.CT.10.«E1) CO TO 500
725.            IF(BMAX.GT.10.*E1) GO TO 500
726.            IF(PYMAX.GT.El)  GO TO 500
727.            IF(HMAX.CT.£1*1000.) GO  TO 500
728.            IF(FEMAX.GT.E1*1000.) GO TO 500
729.            END IF
730.      999   FORMAT(1X,I2,2X,6D11.4)
731.            MSWEEP-MSWEEP+1
732.      901   IFCOPRINT.CE.1 .OR.MSWEEP.GT.20) PRINT902,MSWEEP,M,RMAX,KBMAX,
733.           1FLMAX.LSMAX.FCMAX
734.      902   FORMAT(37X,'AFTER  SWEEP',14,5X,'AND' ,I 4 , '  ITERATIONS',/
735.           1,1 OX, 'RMAX-' ,D11.3,8X, 'KBMAX-' ,D11.3,8X, 'FLMAX-' ,D11.3,8X,
736.           2'LSMAX-',D11.3,'FCMAX-',011.3)
737.            IF(OPRINT.GT.1)  PRINT 2,M,AMAX,BMAX,PYMAX,HMAX,FEMAX,FERMAX
738.            IFCMSWEEP.LT.25)   THEN;  M-0
739.            IFCOPRINT.EQ.4)  PRINT785,(I,YA(I),1-1,N)
740.            IF(OPRINT.EQ.4)  PRINT785,(I,YAX(I),I-1,N)
741.            IF(OPRINT.EQ.4)  PRINT788,(I,KB(I),1-1,N)
742.            IF(OPRINT.EQ.3)  PRINT788,(I,KBA(I),I-1,N)
743.            DO 786 1-2,N
744.            FE2(I)-(FE2X(I)+FE2(I))/2.
745.            YA(I)-(YA(I)+YAX(I))/2.; YB(I )-
-------
 796.            Z8-FHJX(7)
 797.            DO 124 J-2.N
 798.            I-N-M-J
 799.            NC1-YBX(I)-PHI(I)*PR/TE(I)«DX(I)/DT*(YAO(I)-YA(I)+YBO(I)-YBCI)  .
 800.           1+YAOU-M )-TA(I + l )+YBO(I+l )-YB(I-H ))/2.+NCI
 801.      124   CONTINUE
 802.             DO 125  I-l.H
 803.            H2C03(I)-DC02(TE(I))*YB(I)
 804.            IF(I.NE.N)  THEN
 805.            Dl-DX(I)
 806.            D3-DX(I+1)
 807.            D2-D1+D3
 808.            T2-(T(I)*D3+T(I-H7*D1)/D2
 809.            TE2-(TE(I)*D3+TEU+1 )*D1)/D2
 810.            DAB2-DAB(TE2)
 811.            DAC2-DAC(TE2)
 812.            DBC2-DBCCTE2)
 813.            YA3-YACI+1)
 B14.            YAl-YA(I)
 815.            X3-XU + 1)
 816.            Xl-X(I)
 817.            YB3-YB(I+1)
 818.            YBl-YB(I)
 819.            YA2-(D1*YA3+D3*YA1 )/D2
 820.            YB2-(D1*YB3+D3*YB1)/D2
 821.            Z10-1
 822.            HHILECZ10.GT. l.OD-U)
 823.            Zll-NAI(I); Z12-NBKI)
 824.            NAI(I)-T2«I>AB2*DAC2/(YB2*(DAC2-DAB2)-YA2*DAB2+DAB2)*(Z12*
 825.           1YA2/(T2«DAB2)+YBX(I)*YA2/(T2«DAC2)-PR/TE2*(YA3-YA1 )/(X3-Xl ))
 826.            NBI(I)-T2*DAB2*DBC2/(YA2*(DBC2-DAB2)-YB2*DAB2+DAB2)*(NAI(I)*
 827.           1YB2/(T2*DAB2)-I-YBXCI)*YB2/(T2*DBC2)-PR/TE2*(YB3-YB1 )/(X3-Xl ))
 828.            Z10"DABS(NAI(I)-Z11)+DABS(NBI(I)-Z12)
 829.            END WHILE
 830.            END IF
 831.            R(I)-NB(I)/NA(I)
 832.            U3-U(I)*YA(I)
 833.            Ul-03
 834.            UAO(I)-U3
 835.            U3--U1*BO*2.
 836.            U3-03/DXU)
 837.            U4-DMIN1(O.OD+00,2.*YA(I>*HBACT(I)*(TE(I)
 838.           1-273.))
 839.            Dl-X(I)-<-DX(I)/2.
 840.            OE20XCD — FE2(I)«YA(I)*(KOXl*H(I)**(-2 )+KOX2)
 841.            YAX(I)-U2-1./A.*VC(I)*0£20X(I)
 842.            U2-U2+1./4.«WC(I)«(-KB(I ) *YA(I ) *FE2(I))
 843.            Z-DMAX1(Z,DABS(?Y(I)-P10(I)))
 844.      .      FE20X-YAX(I)*4.*DT
 845.            FE2B--KB(I)*YA(I)«FE2CI)
 846.            TFE2B-FE2B*DT««C(I)
 847.            OE2B(I)-FE2B
 848.            DXFO(I)-DXF(I)*AFE3T(I)»FE3(I)*DELTAF(I ) /(L*DELTAF(I)'ALPHA(I)"KSF
 849.           1*(1.-PY(I))+DCF)
 850.            IF(DABS(NB(D) . LT . l.OD-20) NB(I)-0.0
 851.      123    PRINT12,I,X(I),YA(I),YB(I),NA(I),NB(I),HI,U2,U4,FE20X,PYCI)
 852.            IF(MBAT.EQ.l) WRITE(NBAT,851 )  I,X(I),YA(I),YB(I),NA(I),NB(I),UJ ,
 853.           IZll.PY(I)
 854.            PRINT13,D1,NAI(I),NBI(I),PYF(I)/PY02(I),TFE2B
 855.           1,1.5(1)
 856.            IFCMBAT.EQ.1) URITECNBAT,13) D1,NAI(I),DELTAO(I)
 857.      126    FORMAT(1X,I3,8(1X,D10.4))
 858.      125    CONTINUE
 859.      12     FORMAT(1X,I4,1X.F8.1,2F10.4,3X,2(D10.4,3X),4(E!0.3,2X),2X,F&.3)
 860.      13     FORMAT(6X.F8.1,20X,2(3X,D10.4),25X,2(2X,D10.3),4X,F6.3)
861.      851    FORMAT(IX,14,1X,F8.1,2F10.4,3X,2(D10.4,3X),/,2(D10.4,3X),
 862.           1F7.4)
 863.            PRINT15
864.            PRINT215
865.            DZ7-Z7-TFE-0.0
866.            DO  128 1*1,K
867.            Z7-Z7-KPY02(I)+PYF{I))*FPY(I)
868.            IF(I.CT.l) DZ7-DZ7+FPY(I)*DX(I)
869.            IF(I.NE.l) CALL  CHEMI(1 ,H(I ) , FE3(I),FEOOH(I),AFE3T(I),PYX WRITE(NBAT,16 ) I , PH,FE2CI ) , FE3(I),FEOOH
-------
876.           10S04(I),010(1).FE20X.TFE2E
877.           FRINT16,I,PH,FE2(I),FE3d),FEOOHd).HCCI).OS04(I),010(1),
878.           1H2C03(I),CA(I),Z11
879.           IFCTT.GE.32.) KBOCI)-DMAXI(KB(I),KBA{I),1.OD-15)
880.           IF(KBd>+KBOd).LT.1.0D-12) GO TO  128
881.           IF(KBd).NE.O.O) THEN;  XI-DABS(KB( I )-KBO(I ) )/KB(I )
882.           ELSEIF(KBOd).EQ.l.00-15)  THEN;  Xl-0.;  ELSE; Xl-100.; END IF
883.           KBMAX-DMAX1(DABS(KB(I)-KBO(I))/KBO(I),KBMAX. XI )
884.           KBMAX-KBMAX
885.     128   CONTINUE
886.           Z7-86400.*Z7/(DT*DZ7)
887.     15    FORMATC1 ' ,1X, 'LAYER',4X, ' PH ' , 7X , ' FE2 ' , 9X, ' FE3 ' , 7X, ' FEOOH ' ,
888.           17X,'ACIDITY',7X,'S04',6X,'IONIC  STR',5X,'H2C03',
889.           28X,'CA',7X,'TFE3CMPLX')
890.     215   FORMAT(17X,'	MOLES/L',
89 ! .           j '	 , 7X, '-'./)
892.     16    FORMAT(1X,I4,IX,F8.2,3X,9(D9.3,3X))
893.           IF(OPRINT.EQ.3)  PRINT?
894.     7     FORMATC5X, 'B',6X,'S04' ,5X, 'HS04' ,4X, 'FE3' ,5X, 'FEOH',
895.           13X,'FEOH2',3X,'FEOH3',3X,'FEOH4',
896.           22X,'FE20H2',3X,'FES04',2X,'FEHS04')
897.           TFED-TFE-TFEO
898.           TFEO-TFE
899.           TFE-(TFED+(Z2+Z3)*DT/a64000.-DT*INFILR*(FE3(l)+FE2(l)))/TFE
900.           IP(TT.EQ.O.O) TFE-0.0
901.           DO  127 I-l.N
902.           FH--DLOG10(B
-------
 956.
 957.
 958.
 959.
 960.
 961.
 962.
 963.
 964.
 965.
 966.
 967.
 968.
 969.
 970.
 971.
 972.
 973.
 974.
 975.
 976.
 977.
 978.
 979.
 980.
 981.
 982.
 983.
 984.
 985.
 986.
 987.
 988.
 989.
 990.
 991.
 992.
 993.
 994.
 995.
 996.
 997 .
 998.
 999.
1000.
1001 .
1002.
1003.
1004 .
1005.
1006.
1007.
1008.
1009.
1010.
101 1.
1013.
1014.
1015.
101 6.
1017.
1016.
1019.
1020.
1021.
1022.
1023.
1024.
1025.
1026.
1027.
1028.
1029.
1030.
1031.
1032.
1033.
1034.
1035.
      END IF
      DT-DMINl(DT,1.5768P+07j
      END IF
140   IF(TT.LE.32.)  THEN
      IHTT+DT/86400. .GT.32.) DDT-DT-864 00 . *2 .
      END IF
      TT-TT+DT/86400.
      END IF
      MSWEEP-MM-M-0
      IF(TT.GE.ET) GO  TO 200
      GO TO 400
c
c
c
c

DATA IS DUMPED IF TIME
LIMIT IS EXCEEDED

C
C
C
C
200   PRINT,'       "'"EXECUTION EXCEEDS  TIME ALLOWED, JOE DUMPED*'***'
      WRITE(42,202)  OPRINT,0,N,NBAT,MBAT,MT,E1,PR,DT,DDT,TT,ET,TUT,
     lTZl,TZ2.TZ3,TZ4,TZ5,TZ6,TZ7,TZ8,KSO,DCO,K.Sr,DCF,L,INFILR,TFEO
      WRITE(42,201 )
      WRITE(42,201)
      WRITE(42,201)  (PY(I).I-l,N)
      WR1TE(42,201 )  (?10(I),1-1,N)
      URITEC42.201)  (TE(I),1-1,N)
      WRITE(42,201)  (TO(I),I-1,N)
      WR1TE(42,201)  (X(I),I-1,JO
      WRITEC42.201)  CDX(I),I-1,N)
      WRITE(42,201)  (T(I),I-1,N)
      WRITE(42,201)  (UAN(I),1-1,N)
      WRITE(42,201)  (UA(I),1-1,N)
      WRITE(42,201 )  (I)AO(I) ,1-1,N)
      WRITE(42,201)  (ALPHA(I),I-1,N)
      WRITE(42,201 )  (FKI(I) ,1-1 ,N)
      WRITE(42,201)  (UC(I),1-1,N)
      WR1TE(42,201 )  (RHOB(I),I-1,N}
      WRITE(42,201 )  (RHOFRCI ) ,1-1 ,N)
      WRITE(4:,201 )  (FFRCI),1-1,N)
      •»RITE(42,201 )  (FPY(I) ,1-1 , K )
      WRITEC42,201)  (FEOOH(I),1-1,N)
      «RITE(42,201)  (ST(I),1-1,N)
      WRITE(42, 201 )  (FE3CD , 1-1 ,N)
      WRITE(42,201)  (FE2(I),I-1,N)
      WRITE(42,201)  (H(I),1-1,N)
      WRITE(42,201)  (DXFCI),1-1,N)
      WRITEC42.201)   ,1-1 ,N)
      WRITE(42,201)  (CA(I),1-1,N)
      WRITE(42,201)  (LS(! ) ,1-1,N)
      BRITE(42,201)  (LSOCi;,1-1,N)
      WRITE(42,201)  (RLS(I),1-1,N)
      WRITE(42,201)  (FLS(I),1-1,N)
      WKITE(42,201)  (H2C03(I),1-1,N)
      WRITE<42,20n  (OH2C03U) ,1-1 ,N)
201   FORMAT(IX, 2D28. 1 8)
202   FORMAT(1X,515,110,2(1X,F10.8),/,lX,4D17.10,/,5(lX,D17.10),/,
     11X,4D18.10,2E13.6,/,1X,4E13.6,D17.10)
      STOP
      END
                          D.O. FUNCTION
               DO*YA  - GRAMS 02 PER  KL
                                         187

-------
 1036.
 1037.
 1038.
 1039.
 1040.
 1041.
 1042.
 1043.
 1044.
 1045.
 1046.
 1047.
 1048.
 1049.
 1050.
 1051.
 1052.
 1053.
 1054.
 1055.
 1056.
 1057.
 1058.
 1059.
 1060.
 1061.
 1062.
 1063.
 1064.
 1065.
 1066.
 1067.
 1068.
 1069.
 1070.
 1071.
 1072.
 1073.
 1074.
 1075.
 1076.
 1077.
 1078.
 1079.
 1080.
 1081.
 1082.
 1083.
 1084.
 1085.
 1086.
 1087.
 1088.
 1089.
 1090.
 1091.
 1092.
 1093.
 1094.
 1095.
 1096.
 1097.
 1098.
 1099.
 1100.
 1101.
 1102.
 1103.
 1104.
 1105.
 1106.
 1107.
 1108.
 1109.
 1110.
1111.
1112.
1112.
1114.
1115.
 1116.
 1117.
  DOUBLE PRECISION  FUNCTION DIM 1 ,)
  REAL'S T
  DO-.032*10."(2237.8/T-15.803+.018117«T)
  RETURN
  END
c
c
c
c





c
c
c
c
c
c





c
c
c
c
c
c





c
c
c
c
c
c




c
c
c
c
c

D.C02. FUNCTION
DC02-HENRY 'S LAW TEMP. DEPEND. COEFFICIENT

FUNCTION DC02(T)
REAL'S T
DC02-. 03388
RETURN
END

FUNCTION DAB

THIS FUNCTION CALCULATES THE TEMPERATURE DEPENDANT
02-C02 DIFFUSION COEFFICIENT.

DOUBLE PRECISION FUNCTION DAB(T)
IMPLICIT REAL'S (A-H.P-Z)
DAB-1 . 56D-05*T**1 . 661 /DEXPC61 . 3/T)
RETURN
END

FUNCTION DAC

THIS FUNCTION CALCULATES THE TEMPERATURE DEPENDANT
02-N2 DIFFUSION COEFFICIENT.

DOUBLE PRECISION FUNCTION DAC(T)
IMPLICIT REAL'S (A-H.P-Z)
DAC-1.13D-05*T"1.724
RETURN
END

FUNCTION DBC

THIS FUNCTION CALCULATES THE TEMPERATURE DEPENDANT
N2-C02 DIFFUSION COEFFICIENT.

DOUBLE PRECISION FUNCTION DBC(T)
IMPLICIT REAL'S (A-H.P-Z)
DBC-3. 15D-05'T"1 .57/DEXP(113.6/T)
RETURN-
FUNCTION FLUX
CALCULATES LEACHING RATE FOR H , HC , FE3 , FE2 , ST
AKD IS BASED ON CONCENTRATIONS AT OLD TIME STEP ONLY.

C
C
C
C





C
c
c
c
c
c





c
c
c
c
c
c





c
c
c
c
c
c




c
c
c
c
c
 DOUBLE  PRECISION FUNCTION FLUX(MO)
 IMPLICIT  REAL'S (A-H.O-Z)
 REAL«4  INFILR
 DIMENSION HCFL(50),OH(50),OE3(50),OE2(50),OST(50),OHC(50),S(50),
1X(50),DX(50),A(50,50),OEOH(50),OEOH2(50),OEOH3(50),OEOH4(50),
20E20H2(50),OES04(50),OEHS04(50),OS04(50),OHS04(50),010(50),
30EOOH(50),ST(50),FE2(50),H(50),FE3(50),HC(50),CA(50),CB(50),
4CA(50),OCA(50),H2C03(50),OH2C03(50)
 COMMON/ACHCAL/OIO,OEOH,OEOH2,OEOH3,OEOH4,OE20H2,OES04,
20EHS04,OEOOH,OS04,OHS04,ST,FE2,OH.OE3,OE2,
30ST,CA,OCA,H2C03,OH2C03,I/AHADJ/HC,OHC,HCFL,GA,GB
4/AFLUXl/H,FE3,X,DX,DT,INFILR,N
 IF(MO.GT.O)  CO  TO 25
 NN-N+1
 H(NN)-F£3(NN)-FE2(NN)-OH(NN)-OE3(NN)-OE2(NN)-OST(NN)-OHC(NN)-0.
 HC(NN)-OEOH(NN1-OEOH2(NN)-OEOH3(NN)-OEOH4(NN)-OE2OH2(NN)-0.
 CA(NN)-OCA(NN)-ST(NN)-OEHS04(NN)-OES04(NK)-OS04(NK>-OHS04(NN)-0.
 H2C03(NN)-OH2C03(NN)-0.
 X(NN)-X(N)->-DX(N)/:.
 DO 10  II-2.N
                                         188

-------
                                                                                                                                                                                                                u5c»^c^U'^ww^o*cn^*^tn^u^jwo^oo^o*u>^u>lo»—  o ^  cr>
OD
TVS0
~ -* •*. o
!_, (^ KJ 1
1 1 O to
H NJ <_.
*"' *» "
31 -^s
KJ 4
O 3J

\_/ (-1
-- 1
CJ *—
ti ^

O
'£
)*
O

to
0
o

*".
f-.
\
V
•f
r>
» 4










to z n i z t-> 2; i i a;
o * t-t '-"•N :z o o itt-i^^rto
O Z M M » 3J2JMMI
O M (I^WOM ll^tJ

^ » O — » » O
»-* on »-* 3: z
*-» > > ^ no
o ii 3: (i
*_- O *-* CJ
v_* PI «^* pi
* 25 * X
O O O O
n :r 31 a:
> » n *
•^ o — • o
MO M 30
,_x >. vv n


i i
•— • fcj
\-^ -^f

n n
HJ M















tJ O
O "I -^
H 3!

^— • y-.
* M
•-v O
O
t-


^
4-
0
11
C/l
0

/-^
t-4
v_^


















35 1 » 0
1 1 1
iroS
* CO
"< 0
~- is
4 t
U 0
1 PI
•- x

^ 0

in ^
H <-
•^ l


i_j 4.
^ O
•4- ti
t> w
*-> o
f-
•^
1
l->












»j ?; » *
00 * H4 ^ ^
t J O '"N II
-^ + *

ro *t)

4 <-
M -w>
M ^-v

^ PI
if ro
O <~*
Pt U
to |
/-N »-•
M ^^
•^ 4
O
M
N)

!_,
1
'-
,_,
4.
O
PJ
N>

t-i
I


4.
n
H

0 K>
ro o 4 -^
tO -s^
10 pi
t-* U>
M ^-*
4 -*
o 4
PI O
in u>
£> M

t— < -->
•w »
4 O
o n
M i>>
t/l x-»
O M

'~x 4
M O
^ w
o
3:
f^
H
"^
O
PJ
O



1-4


0
PI
O
5

1 1 1 •— to
4 » 4- >-
n o ~
>-» tii *»)
c- pi
t Ct
*- 1
4- ^
O 4
n o
NJ PJ

33 ^
Kl (-<

C. *-
t v>


4 o
O f*l
Pt w

t/> t-t
O 1
t- •->
U 4
1 O
N- PI
^ o

o -->

tn (
o »-

^-> 4-
t_ o
1 PI
•- o
^ ffi
S^SMMS^^^^^!^

^ 4 *
O Hto OC1OOOO
33 M« OOOOOO
^> *
H 0 _)HHHHH
•^ 3! OOOOOO
5 V S^^JSSS
^* ^
P3 O


v^ t_<
1

^,
4.


Ci
t


^rf-
4.
tc
c,
(

*^t
4
n








1 h^ yO f^>
H Z H •
O --. M
O H •
1
5C
? 3
W M
^ H
•""
>
71!


f
r*

B
?4



„
25

^^
w
f
"jj*
i— "

2










                                                                                                                                                                                                                                      Mir)o>-ar'O>wMaon(
                                                                                                                                                                                                                                         CJSSr4     O  »  [/I  7!  |     ^^l      5^ 35  M (/) ''>         35  I
                                                                                                                                                                                                                                         >tot~»OKJ5^»~|t-Hp1C^KiNJHl
                                                                                                                                                                                                                                      O|Mt—fOM«       n*-^u*-  •   »o«   *-4»    n«   OO*-'-'
                                                                                                                                                                                                                                         —  25-       *«1O     ZU     35O      S5r.I-.XM         Z
                                                                                                                                                                                                                                         O  PI '^-f  I             P1?S1»       llpl^-(-i»   |f-|pl>
                                                                                                                                                                                                                                                 I  t_,                i  (sj ^-     ^-  ^-     -^-  (  35  M  0      I
                                                                                                                                                                                                                                                >-                 ^«^_x*
-------
1200.
1201 .
1202,
1203.
1204.
1205.
1206.
1207.
1208.
1209.
1210.
1211.
1212.
1213.
1214.
1215.
1216.
1217.
1218.
1219.
1220.
1221.
1222.
1223.
1224.
1225.
1226.
1227.
1228.
1229.
1230.
1231.
1232.
1233.
1234.
1235.
1236.
1237.
1238.
1239.
1240.
1241 .
1242.
1243.
1244.
1245.
1246.
1247.
1248.
1249.
1250.
1251.
1252.
1253.
1254.
1255.
1256.
1257.
1258.
1259.
1260.
1261.
1262.
1263.
1264 .
1265.
1266.
1267.
1268.
1269.
1270.
1271.
1272.
1273.
1274.
1275.
1276.
1277.
1278.
1279.
1280.
1281 .
S2EI.
      Y-2.*H2C03(1 )/CH2C03(I)+OH2C03(l ))«OH2C03O)
100   IF(I.EQ.NN)  GO  TO 110
      FLUX-(CI-Y*TI)*INFILR*DT/DX(I)
      RETURN
110   FLUX-CI*INF11.R*864000.
      RETURN
      EKD
c
C SUBROUTINE FOR CALCULATING DELTA
C
C
C
C
1

10
      SUBROUTINE  DELCAL(A,PY,DELTA,DELTB)
      IMPLICIT  REAL*8  (A-Z)
      REAL*4 KSO,KSF,DCO,DCF,L
      COMMON/ADEL/KSO,DCO,KSF,DCF,L
      PYI-1.0
      Y-DSQRT(KSO«A/DCO)
      R-(1.-PY/PYI)*L
      C-(DEXP(-Y«R)-DEXP(Y*(R-2.*L)))/(DEXP(Y*(R-2.'L))+D£XP(-Y*R))
      DELTA-C/(Y*<1.+R«RSO*C/(DCO«Y)))
      Y-DSQRT(KSF«A/DCF)
      R-(1.-PY/PYI)*L
      C-(DEXP(-Y*R)-DEXP(Y*(R-2.*L)))/(DEXP(Y*(R-2.*L))+DEXP(-Y*R))
      DELTB-C/(Y*(1-+R*KSF«C/(DCF«Y)))
      RETURN
      END
     ••*******************»****»****».»*«**.****«**********«***»*C
                                                                   C
                         THOMAS'S ALGORITHM   .                     C
                                                                   C
     l**********»************»******»»*t*+»*»*t******************c
      SUBROUTINE  THOMAS(A,B,C,D,N,X)
      DOUBLE PRECISION  A(N),B(N),C(N),D(N),X(N),W(50),G(50),DEN(50)
      DO 1 1-2, K
      IF(I.NE.N)  W(I)-C(I )/DEN(I)
      G(I)-(DU)-AU)*Ca-l))/DEN(I)
      CONTINUE
      X(N)-GCN)
      3t(N-l)-G(N-l )-W(N-l)«X(N)
      N-N-1
      IF(N.CT.l)  GO  TO  10
      RETURK
      END
                        SUBROUTINE HADJ
               THIS  ROUTINE ADJUSTS THE H PRODUCED  TO
      ACCOUNT FOR  NEUTRALIZATION BY THE GANGUE MATERIAL.
      SUBROUTINE HADJ(AH,OH,I,H)
      IMPLICIT REAL'S  (A-H,0-Z)
      DIMENSION HC(50),OHC(50),HCFL(50),GA ( 50),GB(50)
      COMMON/AHAEJ/HC,OHC,HCFL,GA,CB
      MMM-0
      IFCAH.LE.O..OR.GACI).LT.-9.) TREK
      H-OH+AH
      HC(I)«OHC(I)+HCFL(I)
      WHILE(HC(I).LT.O.)
      HCFL(I)-HCFL(I)/2.
      HC(I)-HC(I)-HCFL(I)
      MMM-MMM+1; IF(MMM.CT.10)  THEN; PRINT,'
      GO TO 732
      END IF
      ZND WHILE
732   RETURN
      END IF
      «-!; Zl-1.0
      AL-OH
      AC-DM INI (1 .OD+0!**(-GA(I» ,OH+AH)
      AM-(AL+AC)/2.-OH
      WHILE(DABS(Zl-AM).CT..0001*AM)
      Zl-AM
      APH—(DLOC10(OHHDLOC10(OH+AM))/2.
      AAH-AH*(1.-DEXP(CBU ) * < CA (I )-APH)»
      IFtAV.-LT. AAH)  THEN
                                                 *«*ERROR***1120'
                                       190

-------
1283.
1284.
1285.
1286.
1287.
1288.
1289.
1290.
1291.
1292.
1293.
1294.
1295.
1296.
1297.
1298.
1299.
1300.
1301.
1302.
1303.
1304.
1305.
1306.
1307.
1308.
1309.
1310.
1311.
1312.
1313.
1314.
1315.
1316.
1317.
1318.
1319.
1320.
1321.
1322.
1323.
1324.
1325.
1326.
1327.
1328.
1329.
1330.
1331 .
1332.
1333.
1334.
1335.
1336.
1337.
1338.
1339.
1340.
1341.
1342.
1343.
1344.
1345.
1346.
1347.
1348.
1349.
1350.
1351.
1352.
1353.
1354.
1355.
1356.
1357.
1358.
1359.
1360.
1361.
1362.
1363.
1364.
!365.
733
20
25
 ELSi.
 AC-AM+Oh
 END IF
 AM-( AL+AG)/2.-OH
 M-H-fl
 IF(M.CT.50)  PRINT,I,OH,'-H' ,AAH,'-AAH' , '
 IF(M.GT.SO)  STOP
 END WHILE
 H-OH-t-AM
 HC(l)-OHC(I) + AH-AM-t-HCFL(I )
 WHILE(HC(I).LT.O.)
 HCFLd)-HCFL(I)/2.
 HC(I)-HC(I)-HCFL(I)
 MMM-MMM-H ;  IF(MMM.GT.SO)  THEN; PRINT,'
 GO TO 733;  END IF
 END WHILE
 RETURN-
 END
                                                  HC IS  NOT  CONVERGING*
                                                    ***ERROR***1132'
c
c
c
c
c
c
c
c

SUBROUTINE CHEMI

THIS ROUTINE CALCULATES THE IONIC STRENGTH, ACTIVITY
COEFFICIENTS, AND CONCENTRATIONS OF THE FERRIC SPECIES FOR
ANT GIVEN PR, ST, TFE3, AND TFE2. IT RETURNS THE ACTIVITY
COEFFICIENTS AND THE TOTAL IRON(III) ACTIVITY.

.

i
1
t
(
(
I
 SUBROUTINE  CHEMI{MO,H,FE3,FEOOH,AFE3T,DELIO)
 IMPLICIT  REAL*8 (A-H.O-Z)
 REAL*8 lO.DELFE(SO)
 REAL*4 KSP.K11,Kl2,Kl3,Kl4,K22,KML,KMHL,KA2
 DIMENSION OH(50),010(50),OE3(50),OEOH(50),OEOH2(50),HC(50),
10EOH4(50),OE20H2(50),OES04(50),OEMS 04(50),0504(50),OE2(50),
20EOH3(50) ,OHS04(50) , OE-OOH ( 50 ) , OST ( 5 0) ,ST(50),FE2(50),TFE3C(50),
3HCFL(50),OHC(50).GA(50),GB(50),CA(50),OCA(50),H2C03(50),OH2C03(50)
 COMMON /ACHEMI/FE31 , HH 1 , TFE3C , K.S? , KA2 , K 11 , Kl 2 , Kl 3 , Kl 4 , K2 2 , KML , KMHL ,
1MCHEMI/ACHCAL/010,OEOH,OEOH2,OEOH3,OEOH4,OE20H2,OES04,
20EHS04,OEOOH,OS04,OHS04,ST,FE2,OH,OE3,OE2,OST,CA,OCA
3,H2C03,OH2C03,I/AHADJ/HC,OHC,HCFL,GA,GB
 M-l; X-0.0
 IF(MO.LE.O)  THEN
 DELFE(I)-0.
 OEOOH(1)«FEOOfi
 IO-0.5«H+1.5*ST(I)
 FEOH-FEOH2-FEOH3-FEOH4-FE20H2-FES04-FEHS04-0.
 ELSEIF(MO.GT.1) THEN
 X2-Z1-FE3RI-FE3
 HR-H
 CALL HADJ(HR,OH(I),I,H)
 H-HH1; FE3-X2-FE31
 10-010(1);  FEOH-OEOH(I);  FEOH2-OEOH2(I )
 FEOH3-OEOH3(I); FEOH4-OEOH4(I ) ;  FE2OH2-OE20H2(I)
 FESO4-OESO4(I); FEHS04-OEHS04(I); OFEOOH-OEOOH(I)+FEOOH
 ELSE
 R2-FE3; X-IO-OIO(I)
 END IF
 G1-10.**(-.5*(DSQRT(IO)/(1 .4-DSORT(IO))-.3»10))
 G2-G1**4
 C3-G1««9
 G4-G1**16
 GO TO 25
 10-0.25*(H-fflS04-'-F£OH2+FEOH4+FES04+4.0*(FE2(I) + S04->-FEOH+FEHSOJi)
H-8.0*HC(I) + 9.0*FE3-H6.0*FE20H2)-HO/2.
 IFdO.GT. 120. )  PRINT,I,'-  I ' , M, '-M ' , ST ( I ) , ' - ST ' , FE2 (I ) , '-FEI '
 IF(IO.GT.120.)PRIKT2,H,PH,S04,HS04,FE3,FEOH,FEOH2,FEOH3,FEOHi,
1FE20H2,FES04,FEHS04,FEOOK
 IFdO.GT. 120.)  PRINT,FE3R1, ' -FE3RI ' , FE3R , '  -FE3R',HR,'  -HR',
ICELFEd) , '-DELFE' , FE3E , ' -FE3E ' , H6 , ' -H6 ' , HC (I ) , ' -HC ' , 10 , ' - I 0 '
 IF(M.CT.IS)  THEN; IO-X;  END  IF
 G1-10.**(-.5*(DSQRT(IO)/(1.+DSQRT(IO))-.3*10))
 C2-G1**4
 G3-G1**9
 G4-G1**16
 1F(MO.EQ.4)  THEN
 Z1-FE3
 H6-H-DKAX1(H,OH(I)/100.)
 FE3E-KSP/C3*(G1*H)**3
 FE3R-FE3R:-G3«FE3E*(KH / ( H*G 1 *G2 ) + K 1 2 / ( H** 2 *G ! ** 3 )-t-K 1 3 '
1(E*«3*C1**3)-HC)4/(H**4«G1**5)+G3*FE3E*K22/(G4«H**2«G1**2)
2*K«1,*SO4«C2/G!+KMHL*HS04«C1 /G2)+OEOH(l)^-
3+OEOH4(I)+OE20K2(I)*OIS04(I)+OEHS04(I)
                                        191

-------
1306.
1367.
1368.
1369.
1370.
1371.
1372.
1373.
1374.
1375.
1376.
1377.
1378.
1379.
1380.
1381 .
1382.
1383.
1384.
1385.
1386.
1387.
1388.
1389.
1390.
1391.
1392.
1393.
1394.
1395.
1396.
1397.
1398.
1399.
1400.
1401.
1402.
1403.
1404.
1405.
1406.
1407.
1408.
1409.
1410.
1411.
1412.
1413.
1414.
1415.
1416.
1417.
141 8.
1419.
1420.
1421.
1422.
1423.
1424.
1425.
1426.
1427.
1428.
1429.
1430.
1431.
1432.
1433.
1434.
1435.
1436.
1437.
1438.
1439.
1440.
1441.
1442.
1443.
I 444.
1445.
88
27
28
 IFCFEUOH.GT. Cl. )  THE:.
 FE3-FE3E
 ELSE
 FEOOH-0.0
 FE3-(FE3III+0£OOH(I)+OEOH(I)+OEOH2(I>+OEOH3 +
10E20H2+K13/(H**3*Gl**3)+K14/(H*«4*Gl**5>+C3*FE3«K22
3/(C4*H**2*Gl**2)-HCML*S04*G2/Gl-t-KMHL«HS04*Gl/G2))
 END IF
 IF(FE3.LE.O.O)  FE3-DMAX1(O.OD+00,FE3R+DELFE(I))
 H1-DMAX1(OH(I)/10..H); HA-10.;  MMMM-1 ;  IF(K.EQ.l)  KMMM-1
 WHILE(DABS(HA-H).GT..001«H.AND.MMMM.LT.2)
 MMMM-MMMM+1;  IFU.EQ.32)  FRINT , MMMK , HA , ' -HA ' ,H . ' -H '
 HA-H
 AH-HR+3.*(FEOOH-OEOOH(I))-(OEOH(I)+2.*(OEOH2(I)+OE20H2(I))
H.3.*OEOH3(I)+4.*OEOH4(I)-t-OHS04(I)+OEHS04(I)) + FE3*G3*
2(K11/(H1*G1«G2)+2.«K12/(H1*H1*G1**3)+3.*K13/(H1**3*C1**3)+4.«K14/
3{H1**4«G1««5)+2.*K22*G3«FE3/(G4*H1**2*C1**2)+KMHL«HS04»G1/G2)
4+HS04
 CALL  HADJ(AH,OH(I),I,H)
 H2-DMAX1(OH(I)/100.,H)
 IF ( H . EQ . OH ( I ) / 1 00 . ) HC ( D-OHC ( I ) / 1 0.
 END WHILE
 IF(M.EQ.9)THEN
 IFCH2.GT.H1)  THEN
 HH-H2; HL-H1
 ELSE
 HH-H1; HL-H2
 B-(HL-HiH)/2.
 END IF
 ELSEIF(M.GT.9)  THEN; IF(H.LT.Hl)  THEN;HH-H1
 ELSE; HL-H1;  END IF
 H-(HL+HH)/2.
 1FCI.GT.35)  PRINT88,HR,FE3R1,FE3R,FE3,FE3E,FEOOH,B,HL,HR,H2,IO
 ELSEIF(M.LT.9.)  THEN
 H-(Hl+H2)/2.
 IFC1.GT.33)  PRINT,I,El,'-HI', P. 2, '-H2'
 END IF
 FORMAT(1X,F10.5,' -HR  '.F10.5,'
1' -FE3  ',F10.5,' -FE3E  ',F10.5,
2F1C.5,'  -HL  '.F10.5,'  -HH '.FJ0.5,'
 GO TO 27
 END IF
 IF(MO.LE.O)  THEN
 X2-Z1-FE3-.2*(H*G1)**3/G3
 FEOOH-OEOOH(I)-FE3-FEOH-FEOH2-FEOH3-FEOH4-2.*FE20H2-FES04-FEHS04
 IF(FEOOH.LT.O.O) THEN
 FEOOH-0.0
 END IF
 END IF
 IF(FE3.LT.O.O)  THEN
 PRINT28.I.M
 FORMATC//,35X, 'WARNING' ,/, 10X, 'FE3 .LT. 0.0;    TOTAL COMPLEXES
1EXCEED  FEOOH',/,5X,'PROGRAM WAS EXECUTING  LAYER',14,',  ITERATION'
214,'  IN  SUBPROG CHEMI')
 PRINT,FE3RI, ' -FE3RI',FE3R, '  -FE3R',HR,' -HR',
IDELFE(I), '-DELFE' ,FE3E, '-FE3E' ,H6, '-H6'
 PRINT2,H,PH,S04,HS04,FE3,FEOH,FEOH2,FEOH3,FEOH4,FE20H2,
1FES04.FEHS04,FEOOH
 STOP
 END IF
 S04-ST(I)/(l.+K«KA2*G2-t-KML«FE3*C3*G2/Cl+K>1HL*H«FE3*KA2*Cl*C3)
 HS04-H*S04*KA2»G2
 FEOF-FE3«K11*G3/(G1*G2*H)
 FEOH2-FE3«K12"G3/(G1««3*H«*2)
 IF(FEOH.GT.1000.)PRINT.FE3,'-FE3',FEOH,'-FEOH',R,G1,G2,G3,'1354'
 FEOH3-FE3*K13*G3/(H**3*G1**3)
 FEOH4-FE3«K14«C3/(H**4*G1**5)
 FE20H2-(FE3/H)**2*K22*(G3/G1)*«2/G4
 FES04-FE3*KML*S04*G2*G3/G1
 FEHS04-FE3«KHHL*HS04»G1«G3/G2
 IFCMO.GT.l) THEN
 DELFE(I)-FEOH+FEOH2+FEOH3+FEOH4+2.*FE20H2+FEHS04+FES04-
10EOH(I>-OEOH2(1)-OEOH3(I)-OEOH4(I)-2.*OE20H2(I)-OEHS04(I>-
20ES04(I )
 END IF
 IF(DABS(FE3-X2).LE. -001«FE3) THEN-
 IP (DABS(IO-X)flO.LE..01) GO TO 30; EKD  IF
 X-IO
 X2-FE3
•FE3RI '.F10.5,' -FE3R  '.F10.5,
 -FEOOH ' ,/,IX,FlO.5, '  -H ',
    •H2 '.FlO.5,'-  10  ',/)
                                        192

-------
1447.
1448.
1449.
1450.
1451.
1452.
1453.
1454.
1455.
1456.
1457.
1458.
1459.
1460.
1461.
1462.
1463.
1464.
1465.
1466.
1467.
1468.
1469.
1470.
1471.
1472.
1473.
1474.
1475.
1476.
1477.
1478.
1479.
1480.
1481.
1462.
1483.
1484.
1485.
1486.
1487.
1488.
1489.
1490.
1491.
1492.
1493.
1494.
1495.
1496.
1497 .
1498.
1499.
1500.
1501.
1502.
1503.
1504.
1505.
1506.
1507.
1508.
1509.
1510.
1511.
1512.
1513.
1514.
1515.
1516.
1517.
1518.
1519.
1520.
1521 .
1522.
1523.
1524.
152S.
      M-M+1
      CO TO  20
30    AFE3T-((G1*(FEOH2+FEOH4+FES04) + G2 *(FEOH+FEHS04)+2.*G4*FE20H2
     1+FEOH3)/(FE3*G3)+G3)/1000.
      TFE3C(I)-FEOH+FEOH2+FEOH3+FEOH4+2.*FE20H2+FEHS04+FESO4+FEOOH
      IF(MO.LE.l)  THEN-
      DELI 0-FE3+FE2(I)+FEOOH+FEOH+FEOH2+FEQH3+FEOHA+2.«FE2OH2+
     1FES04+FEHS04
      OEOH(I)-FEOH;  OEOH2(I)-FEOH2; OEOH3(I)-FEOH3
      OEOH4(t)-FEOH4; OE20H2 ( I )-FE20H2 ;  OEOOH ( D-FEOOH ;  OES04 ( I )-FES04
      OEHS04(I)-FEHS04;OS04(I)-S04; OHS04(I )-HS04; OST(I)-ST(I)
      OHC(I)-HCd);  OCAd)-CA(I); OH2C03 ( I )-H2C03 ( I )
      IF(MO.EQ.-l) THEN; OH(I)-H; OE2(I)-FE2(I ) ;  OE3(I)-FE3;  END IF
      ELSE;  DELIO-G1
      END IF
      OIOCD-IO
      IF(I.EQ.32)  GO  TO 60
      IF(MCREMI)  40,50,60
40    RETURN
50    PRINT1,I,M,10
1     FORMAT{/, 10X,'LAYER ',14,', I ITERATIONS -  ',12,531,
     1'IONIC  STRENGTH - ',F6.4)
      RETURN
60    PRINT1,M,IO
      PH —(DLOG10(G1*H))
      PRINT2,H,PH,S04,HS04,FE3,FEOH,FEOH2,FEOH3,FEOH4,FE20H2,
     1FES04,FEHS04,FEOOE
2     FORMAT(7X,'H',8X,'PH',7X,'S04',7X,'HS04',6X,'FE3',7X,'FEOH',/,
     13X,D8.2,3X,F6.3,3X,4(D8.2,2X),//,5X, 'FEOH2' ,5X, 'FEOH3',5X, '1-EOH4'
     24X,'FE20K2',5X,'FES04',4X,'FEHS04',4X,'FEOOH',/,3X,7(D8.2,2S))
      PRINT3,FE3,AFE3T*FE3
3     FORMAT(/,5X, 'FERRIC IRON', 87.,'TOTAL FERRIC',
     1' IRON  ACTIVITY',/.6X.D10.4,6X.D10.4)
     - RETURN
      END
//DATA.FT51F001  DD UNIT-BAT,FILES-SDXDT1
//DATA.FT52F001  DD UNIT-BAT,FILES-SDXDT2
//DATA.FT53F001  DD UNIT-BAT,FILES-SDXDT3
//DATA.FT54F001  DD UNIT-BAT,FILES-SDXD74
                 DD UNIT-BAT,FILES-SDXDT5
                 DD VOL-REF-MEN.P65440.D1J.LIB,
//  DSN-MEN.P65440.D1J-DMNLS8,
//  DCB-(RECF«-FB,LR£CL-100,BLKSIZE-3100) ,
//  SPACE-(TRK, (3,5),RLSE),DISP-(NEW,KEEP )
//DATA.INPUT  DD  *
                                     IDUMP
                                     OPRINT
                                     MEAT
                                     L
                                     RADIOS
                                     T DIMENSION
                                     Z DIMENSION
                                     DCO
                                     KSO
                                     DCF
                                     KSF
                                     KLS
                                     P
                                     RG
                                   El
                                     ET
         1                           MT
                                     INFILR (CM/YR)
                                     MWLS
                                     RHOLFR
                                     DEPTH
                                     NL
                                     VINC
                                     NPL
                                     X2
                                     YA
//DATA.FT55F001
//DATA.FT42F001
 1
 0
 0
2.
4.
100.
100.
.0000001
.000000083
.0000001
.000004*
.00000291
1.
82.3
.0000!
 10300.
50.
 50.
 3.
1000.
20
1.
 1
0.
.21
.0003
10.
.313
.7
.75
1.80
0.0025
.0
2.1
                                     T
                                     PHK.180, .216)
                                     WC
                                     FFP.
                                     RHOBC 1 .67,1 .80)
                                     FPV
                                     FFEC
                                     RHOF?
                                      193

-------
1526.
1527.
1528.
1529.
1530.
1531.
1532.
1533.
1534.
1535.
1536.
1537.
1538.
1539.
1540.
1541.
1542.
1543.
1544.
15.
2.50000000
1.11000
0.0
1.0
.00083
0.0 0.
.00025
.02325
.0025

0 0.0
.00025
.0025 •
.0025 .
TE
GA(0.
GB(1.
HBACT
RLS
FLS
.00025 .00025
.00025 .00025 .00025
.0025 .0025 .0025
0025 .0025 .0025 .0025
0.0 -0.30-12 0.0 0.0 0.0
0.0  0.0  0.0  0.0  0.0
0.0  0.0  0.00.00.0
//DATA.FT43F001 DD DSN-MEN.P65440.01J.DMNLS7,
//  DISP-(OLD,KEEP),VOL-REF-MEN.P65440.D1J. LIB,
//  DCB-(RECFM-FB,LRZCL-10C,BLKSIZE-3100),
//  SPACE-(TRK,(17,3),RLS£)
/*
// EXEC PGM-UKSG,PARM-(INTERACT,'TO D1J JOB  DONE')
                                     194

-------
Example Output from Run 4

RMAX-

RHAX-

RMAX-

RMAX-

RMAX-
AHAX-








0

0

0

n

0
0.
I
I






.8710-03

.7870-03

. 1 15D-0?

.61fln-03

,R83n-m
162D-0'' BMAX
1 KB -
4 KB •
7 KB -
10 KB -
13 K II "
- 16 KB -
- 19 KB -










-
0
0
f)
n
0
0
n
AFTER
KRMAX-
AFTF.R
KBMAX-
AFTER
KBMAX-
AFTK.R
KBI1AX-
AFTF.R
KBMAX-
n. 1 1 '.n-i 7
.0
. 10910-07
,f>2?nn-o;
.n
.0
.n
. n
SWEEP 21
0,5170+00
SWBKP 22
0.277D400
SWEEP 23
0.226IUOO
SWEEP 24
n.!391MOO
SWEEP 25
o. ifl3n+on
PYMAX- O.I







2
5
•' 8
- 1 1
- 1'.
- 17
- 20
AND 5 ITERATIONS
FLMAX- 0. 1150-01
AND 4 ITERATIONS
FLMAX- 0.fi45D-02
AND 1 ITERATIONS
FLMAX- 0.7980-02
AND 3 ITERATIONS
FLMAX- 0.2690-02
AND 1 ITERATIONS
Fl.MAX- 0. 6690-02
74D-05 UMAX- 0.441D-04
KB - 0.99840-08 1
KB - 0. 1639D-07 I
KB • 0.7992D-07 1
KB - 0.0 I
KB - 0.0 I
KB - 0.0 I
KB - 0.0 I











LSMAX-

LSMAX-

LSHAX-

LSHAX-

I.SHAX-

0.0 FCMAX-

0.0 FCMAX-

0.0 FCHAX-

0.0 FCMAX-

0.0 FCMAX-
FE3MAX- 0. 3000-03 FERMAX- 0.
3
6
9
- 12
- 15
- 18
- 21
KB - 0
KB - 0
KB - 0
KB - 0
KB - 0
KB - 0
KB - 0
.77600-08
. 281 ID-07
.U08D-08
.0
.0
.0
.0

0. 2810-01

O.U5D-01

0.697D-02

0.397D-02

0.825D-03
1370*00








-------
        AFTER    6166.0 DAYS
SOLUTION WAS FOUND IN  155 ITERATIONS
LAYER
1

2

3

4

.•i

6

7

8

9

to

1 1

12

13

I'l

15

16

17

18

19

20

21

DEPTH
-0.5
0.0
25.0
50.0
75.0
100.0
125.0
150.0
175.0
200.0
225.0
250.0
275.0
300.0
325.0
350.0
375.0
400.0
125.0
450.0
4 7 5 '. 0
500.0
525.0
550.0
575.0
600.0
625.0
650.0
675.0
700,0
725.0
750.0
775.0
800.0
825.0
850.0
875.0
900.0
925.0
950.0
975.0
1000.0
02
0.2100

0. 1 754

0.1411

0. 1081

0.0777

0.0500

0.0264

0.0091

0.0068

0.0053

0.0041

0.0032

0.0024

0.0019

0.0014

0.001 1

0.0009

0.0007

0.0006

0.0005

0.0005

C02
0.0003

0.0003

0.0003

0.0003

0.0003

0,0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

0.0003

N02
0.6720D-10
0.67150-10
0. 53620-10
0. 31850-10
0.3050D-10
0. 29240-10
0.27770-10
0.26390-10
0.2472D-IO
0.2315D-10
0.2114D-10
0. 19220-10
0. 16440*10
0.1376D-10
0.7729D-
0. 1812D-
0. 14990-
0.1I8ID-
0. 10590-
0.9330D-12
0. 83330-12
0. 73070-12
0.6503D-12
0.56760-12
0.5033D-I2
0.4373D-I2
0. 38630-12
0.3340D-12
0.29370-12
0. 25240-12
0.2205D-12
0. 18780-12
0.1623D-12
0. 13630-12
0. 1 1550-12
0.94400-13
0. 76990-13
0. 59350-13
0.4406D-13
0.2863D-13
0. 1 'i 3 4 0-13
0.0
NC02
0.2024D-I 3
0.2023D-13
0. 16150-1 3
0.95930-14
0.91870-14
0.8806B-I4
0.8365D-14
0. 79490-14
0. 7450D-I4
0. 69770-14
0.637ID-I4
0. 5794D-I4
0. 49550-14
0.414HO- 4
0.23290- 4
0.54380- 5
0.44990- 5
0.3540B- 5
0.31771)- 5
0.27990- 5
0.250ID- 5
0.21930- 5
0.19530- 5
0.17040- 5
0.15120- 5
0.13130- 5
0.11610- 5
0. 10030-15
0.88250-16
0. 7582D-I6
0. 66250-16
0.56430-16
0. 48780-16
0. 40950-16
0. 34720-16
0. 28370-16
0.23140-16
0. 17840-16
0. 1324D-16
0. 86070-17
0.43100-1 7
0.0
II02I-Y
0.0

-0.2350-13

-0. 1640-13

-0. I 190-13

-0.857D-14

-0.584D-I4

-0.353D-I4

-0.205D-14

-0. 5670-14

-0.4920-14

-0. 4020-14

-0. 3240-14

-0.2590-14

-0.2050-14

-0. 1620-14

-0. 1290-14

-0. 1030-14

-0.8340-15

-0.6980-15

-0.612D-15

-0. 5700-15

U02FF.OX
-0.

-0.

-0.

-0.

-0.

-0.

-0.

-0.

-0.

-0.

-0.

-0.
-
-0.

-0.

-0.

-0.

-0.

-0.

-0.

-0.

-0.

8610-26

210D-I3

3600-13

4500-13

5600-13

727D-13

106D-12

238D-12

6930-14

1770-17

126D-17

91fll)-l8

681D-1B

5100-18

3860-18

295D-18

2290-18

1820-18

1500-18

130D-18

1210-18

U02HB/OR
0.0
0. 1000+01
0.0
0.810D+00
0.0
0.208D+01
0.0
0.3580+01
0.0
0. 6210+01
0.0
O.II9D+02
0.0
0.288D+02
0.0
0. 1150+03
0.0
0.176D+OI
0.0
0.31 ID+00
0.0
0.271D+00
•0.0
0. 2570+00
0.0
0. 2780+00
0.0
0.302D+00
0.0
0.336D+00
0.0
0. 3790+00
0.0
0.429D+00
0.0
0.482D+00
0.0
0.5300+00
0.0
0.560D+00
0.0
0.559D+00
CHOX/BACTOX
-0. 3810-18
0.0
-0.1660-09
-0.928D-06
-0.359D-09
-0. 159D-05
-0. 3200-09
-0.199D-05
-0.265D-09
-0. 2480-05
-0.200D-09
-0.322P-05
-0. 1320-09
-0.46BD-05
-0.230D-09
-0. 105D-04
-0. 122D-09
-0.307D-06
-0.784D-10
0.0
-0.556D-10
0.0
-0.4060-10
0.0
-0.3010-10
0.0
-0.226D-10
0.0
-0. 1 710-10
0.0
-0. 130D-10
0.0
-0. 101D-10
0.0
-0,8070-1 1
0.0
-0.665D-1 I
0.0
-0.577D-1I
0.0
-0. 533D-1 1
0.0
XPY/XLS
0.0
0.0
0.345
0.0
0.231
0.0
0.184
0.0
0.186
0.0
0.235
0.0
0.343
0.0
0.641
0.0
0.957
0.0
0.969
0.0
0.975
0.0
0.979
0.0
0.982
0.0
0.985
0.0
0.987
0.0
0.988
0.0
0.9B9
0.0
0.990
0.0
0.991
0.0
0.991
0.0
0.992
0.0

-------
LAYER





















T
T
T
T
T
T
T
T
T
T
T
T
1
2
3
4
5
6
7
8
9
10
1 1
12
13
14
15
16
17
18
19
20
21












PH
5.
2.
2.
2.
2.
2.
2.
2.
2,
2.
2.
2.
2.
2.
2.
2 .
2.
2.
2.
2.
2.













00
91
BO
an
BO
BO
80
80
80
80
80
80
80
60
81
81
81
82
82
82
82








0.


0.

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0








FE2
. 1000-14
.3810-03
. 1040-02
.1210-02
. 1400-02
. 1640-02
.2040-02
. 1040-01
.7360-02
.6090-02
.5570-02
.5260-02
.5070-02
.4920-02
.482D-02
.4730-02
.4670-02
.4620-02
.4580-02
.4560-02
.*Mn-ft?







11
20540-02



II
1 1060+02
FE3 FF.OOII ACIDITY S04 IONIC STR
0. 1000-14 0.0 0.10011-14 0.1000-14 0.105D-03
0.2530-03 0. 2830-01 0.4900-09 0.1920-02 0.6810-02
0.7000-03 0.651D-OI 0.9520-03 0.415D-02 0.1890-01
0.8860-03 0,6!OD-01 0.3970-02 0.6310-02 0.3630-01
0.1050-02 0.5480-01 0.7370-02 0. 8700-02 0.5580-01
0.1230-02 0. 3840-01 0.1170-01 0.1170-01 0.802O-01
0. 1450-02 0.2700-01 0. 1840-01 0.1670-01 0.1180+00
0.1670-02 0.0 0. 2950-01 0.301D-01 0.2080+00
0.1120-04 0.0 0.2000-01 0.1890-01 0.1330+00
0.1380-05 0.0 0.1700-01 0.1610-01 0.1130+00
0.8620-06 0.0 0.157D-01 0.1490-01 0.1050+00
0.6050-06 0.0 0.1500-01 0.1430-01 0. 9980-01
0.4510-06 0.0 0.1450-01 O.I38D-01 0.9670-01
0.3520-06 0.0 0.1420-01 0.1350-01 0.9460-01
0.2830-06 0.0 O.I40D-OI 0. 1330-01 0.9310-01
0.2350-06 0.0 0.1390-01 0.132H-01 0.9210-01
0.2000-06 0.0 0.1380-01 0.1310-01 0.913D-01
0.1730-06 0.0 0.1370-01 0.130D-01 0.9070-01
0.1530-06 0.0 0.136n-01 0.1290-01 0.9030-01
0.1380-06 0.0 0.1360-01 0.1290-01 0. 90011-01
n. 1260-06 o.o n. iif.o-oi 0.1290-01 o. 8980-01
INCRF.lir.NTAL FRACTIONAL IRON MARS BAt.ANCF,
-.26440-03
FRACTIONAL PYRITF, CONSUMPTION PKR M*«2 PER DAY
-.29600-04
TOTAL FRACTIONAL PYRITE CONSUMPTION PER M'*2
-.25630+00
HOLES ENTERING THE WATER TABLE PF.R H**2 PER DAY
ACIDITY FE3 FE2 TOTAL SULFATE
0.18530-01 0.53550-04 0.6IJ6n-o? n^i'tOR-O! 0
TOTAL MOLES TO ENTKR THE WATER TABLE PER M**2
ACIOITY FE3 FE2 TOTAL SULPATR
0.14B5l)f03 0.27630 + 00 0.1**SDl-02 0.12770+03 0
II2C03
0.1020-04
0. 1020-04
0. 1020-04
0. 102D-04
0. 10211-04
0. 102D-04
0. 1020-04
0, 1020-04
0. 1020-04
0. 1020-04
0. 1020-04
0. 1020-04
0. 1020-04
0, 1020-04
0. 1020-04
0.1020-04
0. 1020-04
0. 1020-04
0. 1020-04
0. I02D-04
0.1020-04







CALCIUM
.2900i;-i2

CALCIUM
.62940-03

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0








0 .


0.
CA
. 1000-14
. 1000-14
. 1000-14
, 1000-14
. 1000-14
. 1010-14
. 1090-14
. 1460-14
.2570-14
.5150-14
. 1010-13
.1830-13
.3040-13
.4660-1 3
.6650-1 3
.893D-13
. 1 140-12
. 1380-12
.1610-12
.18111-12
.2020-12







H2C03
i 39iu-u»

1I2C03
85850-01
TFE3CMPH
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
n.
0.












216D+05
2680+01
2080+01
2260+01
2410+01
2570+01
2780+01
3190+01
2850+01
2750+01
271D+01
2690+01
2680+01
2690+01
270D+OI
2720+01
2730+01
2750+01
2770+01
2790+01
28011 + 01













-------
                                  -TECHNICAL REPORT DATA
                            /Please read Instructions on the rcitrsc before completing)
                               '	~   "                      |3. RECIPIENT'
1  REPORT NO
                                      kCCESSIOf
4. TITLE AND SUBTITLE
  Atmosphere and temperature within a reclaimed coal-
  stripmine and a numerical simulation of acid mine
  drainage from stripmined lands	
                                                           5. REPORT DATE
                           6. PERFORMING ORGANIZATION CODE
7. AUTMOR(S)
  D.  B. Jaynes, A. S. Rogowski,  and H.  B.  Pionke
                           8. PERFORMING ORGANIZATION


                                         3
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  Northeast Watershed Research  Center
  USDA-ARS, 110 Research Building A
  University Park, Pennsylvania 16802
                                                           10. PROGRAM ELEMENT NO.
                            11. CONTRACT/GRANT NO.

                               EPA-IAG-D5-E763
 12. SPONSORING AGENCY NAME AND ADDRESS
     U.S. Environmental Protection Agency
     Office of Research  & Development
     Office of Energy, Minerals &  Industry
     Washinaton, B.C.   20460    	
                            13. TYPE OF REPORT AND PERIOD COVERED
                             Interim 9/1/75-8/31/80	
                            14. SPONSORING AGENCY CODE

                                  EPA-ORD
15. SUPPLEMENTARY NOTES
     This project is part of the EPA-planned and coordinated Federal Interagency
     Energy/Environment R&D Program.           	_^^_____	
16. ABSTRACT
      Oxygen, 62, carbon  dioxide,  €62 and temperature were  measured with depth along a
 transect of an acid,  reclaimed,  coal stripmine over a  two  year period.  Spoil-atmosphere
 Q£ concentrations  decrease  with  depth but approach zero  only in a small portion of the
 transect.  Most of the mine remains well oxygenated (02  >  10% by volume) down to 12-
 meters depth.  C02 concentrations ranged from near atmospheric levels to greater than
 15%.  At some locations,  especially within 2 meters of the surface, variations in 02 an<
 C02 are correlated with  changes  in the spoil temperature.   Spoil temperatures in layers
 below 3 meters remain in a  range conducive to iron-oxidizing, bacterial activity year
 around.  Flux ratios  of  CO- and  02 and the source/sink rates of the two gases indicate
 that carbonate neutralization of the acid produced by  pyrite oxidation is the dominant
 source of C02>
      In a second phase of the study, a numerical model describing the production and
 removal of acid and acid by-products from reclaimed coal-stripmines is presented.  Both
 direct oxygen and  bacterially catalyzed pyrite oxidation is considered.  The pyrite
 oxidation rate is  assumed to be  controlled by first-order, solid-liquid kinetics and
 simple diffusion of oxidant into reactive, coarse, stone fragments.  Oxygen supply into
 the reclaimed profile is considered to be governed by  one-dimensional, gas diffusion.
            (Circle One or More)
KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                                                                         c.  COSATI Field/Group
                                  Energy Conversion
                                  Physical Chemistry
                                  [norganic Chemistr
                                  ^Organic Chemistry
                                  Chemical Engineering
                                          6F   8A    8F

                                          8H  IDA   10B

                                          7B   7C   13B
13. DISTRIBUTION STATEMENT
                                              19. SECURITY CLASS (Tins Report/
                                                                          21. NO. OF PAGES
                                              20. SECURITY CLASS (Thispage)
                                                                         22. PRICE
EPA Form 222O-1 (9-73)

-------