v>EPA
           United Stcilos
           Environmental Protection
           Agency
           Office of Hatiiatiun programs
           Las Vegas Facility ^
           P.O. Box 15027
           Las Vegas NV 89114
'ZfH jtu/ ui- /'
August 1978
           Radiation
Radionuclide interactions
with Soil and Rock Media

Volume 1:
Processes Influencing Radio-
nuclide Mobility and Retention

Element Chemistry and
Geochemistry

Conclusions and Evaluation

-------
                                               EPA 520/6-78-007
                                               August 1978
                                               Volume 1 of 2
      RADIONUCLIDE INTERACTIONS WITH SOIL AND ROCK MEDIA

Volume 1:  Processes Influencing Radionuclide Mobility and Retention
           Element Chemistry and Geochemistry
           Conclusions and Evaluation
                              by
                          L. L. Ames
                          Dhanpat Rai
                           Battelle
                 Pacific Northwest Laboratories
                  Richland, Washington 99352
             Final Report for Contract 68-03-2514
                        Project Officer
                       Robert F. Kaufmann
                        Evaluation Branch
        Office of Radiation Programs-Las Vegas Facility
             U.S. Environmental Protection Agency
                    Las Vegas, Nevada 89114

-------
                                  PREFACE
     The Office of Radiation Programs carries out a national program designed
to evaluate the exposure of man to ionizing and nonionizing radiation, and to
promote the development of controls necessary to protect the public health
and safety and assure environmental quality.  A current area of great interest
and effort within the EPA is the development of environmental protection cri-
teria and standards for the management of radioactive wastes.  Accidental or
deliberate interactions of such wastes with various geologic media dictate
that the Agency have access to current knowledge concerning the occurrence and
mobility of selected radionuclides in the lithosphere.  Herein lies the purpose
and scope of the present study.
     Office of Radiation Programs technical reports allow comprehensive and
rapid publishing of the results of intramural and contract projects.  The
reports are distributed to groups who have known interests in this type of
information.  The study reported on herein -is expected to be of considerable
interest to the Department of Energy, the U.S. Geological Survey, the Nuclear
Regulatory Commission, counterpart organizations in foreign countries facing
similar nuclear-related issues, selected private consulting and environmental
groups, and factions within industry.  Ready availability of technical reports
to the scientific community as a whole and to the public is made possible by
distribution through the National Technical Information Service.
     Readers of this report are encouraged to inform the Office of Radiation
Programs of any omissions or errors by contacting the Director, Office of
Radiation Programs - Las Vegas Facility, U.S. Environmental Protection
Agency, Las Vegas, Nevada 89114.
                                         W. D. Rowe, Ph.D.
                                         Deputy Assistant Administrator
                                         for Radiation Programs

                                     111

-------
                                   CONTENTS

                                   Volume 1
EPA REVIEW NOTICE	, • .  11
PREFACE	-	111
ABSTRACT	  .  .  iv
TABLE OF CONTENTS	   v
LIST OF FIGURES	x11
LIST OF TABLES  	  .....  xv
SECTION 1 - INTRODUCTION	1-1
   METHODS AND CONSTRAINTS-	1-2
      Selection of Radionuclides  	  	 1-2
      Collection of Pertinent Literature	..1-4
   REFERENCES.	.  . 1-5

SECTION 2 - PROCESSES INFLUENCING RADIONUCLIDE MOBILITY
            AND RETENTION	 2-1
   NATURAL SOIL AND ROCK DISTRIBUTIONS	2-1
   SOLID PHAS£ AND SOLUTION SPECIES	 2-2
      Uncomplexed, Mononuclear and Polynuclear Solution Species .  .  .2-4
      Construction and Interpretation of the Diagrams  	 2-5
   ION EXCHANGE	2-6
      Ion Exchange Properties of Soil' and Rock Components ...... 2-8
   ANION EXCLUSION	2-13
   DIFFUSION	2-13
   REPLACEMENT REACTIONS 	 2-15
   PHYSICAL TRANSPORT AND FILTRATION 	 2-17
   SATURATION EFFECTS 	 2-19
   SPECIFIC RETENTION 	 2-25
   THE DISTRIBUTION COEFFICIENT OR Kd	 2-27
      Kd Relationship to Migration	2-30
      Field Determination of Kd Values	2-31

-------
                              CONTENTS  (contd)

   REFERENCES	2-33

SECTION 3 - ELEMENT CHEMISTRY AND GEOCHEMISTRY	3-1
   AMERICIUM	3-1
      Natural Soil and Rock Distributions	3-1
      Brief Chemistry	3-1
      Solid Phase and Solution Equilibria	 3-2
      Experimental Adsorption Results	 3-4
      Migration Results  	 .... 3-8
      Summary	3-10
      References.  .	.3-11
   ANTIMONY	3-12
      Natural Soil and Rock Distributions	3-12
      Brief Chemistry	3-13
      Solid Phase and Solution Equilibria	 3-14
      Experimental Adsorption Results	3-15
      Migration Results  	 3-22
      Summary.	 3-24
      References	3-25
   CERIUM	3-27
      Natural Soil and Rock Distributions	3-27
      Brief Chemistry	3-28
      Solid Phase and Solution Equilibria  	 3-28
      Experimental Adsorption Results	3-30
      Migration Results  	 3-32
      Summary	3-34
      References	3-35
   CESIUM .   .   .  '.	3-37
      Natural Soil and Rock Distributions	3-37
      Brief Chemistry .	3-38
      Solid Phase and Solution Equilibria  	 3-38
      Experimental Adsorption Results	3-38
      Migration Results  	 3-55
      Summary	3-62
      References	3-63
                                      vi

-------
                           CONTENTS (contd)

COBALT	3-69
   Natural Soil and Rock Distributions	3-69
   Brief Chemistry	 3-69
   Solid Phase and Solution Equilibria  	 3-70
   Experimental Adsorption Results	3-72
   Migration Results  	 3-74
   Summary.  	  .....  	 3-75
   References	3-76
                 -\
CURIUM	3-78
   Natural Soil and Rock Distributions	3-78
   Brief Chemistry	3-78
   Solid Phase and Solution Equilibria  	 3-79
   Experimental Adsorption Results	3-81
   Migration Results  	 3-82
   Summary.	 3-83
   References.	3-84
EUROPIUM  ...........	  .   .   . 3-85
   Natural Soil and Rock Distributions	. 3-85
   Brief Chemistry 	 3-85
   Solid Phase and Solution Equilibria  	 3-86
   Experimental Adsorption Results	 3-88
   Migration Results  	 3-91
   Summary	3-91
   References	3-92
IODINE	-	3-94
   Natural Soil and Rock Distributions  .  .   .	3-94
   Brief Chemistry	3-95
   Solid Phase and Soil  Solution Equilibria	3-96
   Experimental Adsorption Results	3-96
   Migration Results  	   	 3-97
   Summary	3-98
   References	3-99

                                   vii

-------
                         •  CONTENTS  (contd)

NEPTUNIUM .   .  :	3-101
   Natural Soil and Rock Distributions	3-101
   Brief Chemistry	3-101
   Solid Phase and Solution Equilibria   	  3-101
   Experimental Adsorption Results	3-103
   Migration Results  	   ....  3-106
   Summary	3-106
   References.  .   •	3-107
PLUTONIUM	3-108
   Natural Soil and Rock Distributions	3-108
   Brief Chemistry	3-108
   Solid Phase and Solution Equilibria   .  .  •	3-109
   Experimental Adsorption Results	3-115
   .Migration Results	  .  .	3-126
   Summary	3-132
   References."	."....  3.-133
PROMETHIUM	" .'  .   .	3-138
   Natural Soil and Rock Distributions	'3-138
   Brief Chemistry	3-138
   Solid Phase and Soil  Solution Equilibria	3-138
   Experimental Adsorption Results	3-139
   Migration Results  	  3-140
   Summary	3-140
   References	.  .   .	3-141
RADIUM	3-142
   Natural Soil and Rock Distributions	3-142
   Brief Chemistry	3-142
   Solid Phase and Solution Equilibria	"...  3-143
   Experimental Adsorption Results	  3-143
   Migration Results  	  3-145
   Summary	3-147
   References	3-147
                                  vi i i

-------
                             Volume 2 - Part A

EPA REVIEW NOTICE	    11
PREFACE   .   .   .   .	111
ABSTRACT	    1v
TABLE OF CONTENTS	     v
LIST OF ABBREVIATIONS	    vl
BIBLIOGRAPHIC CITATIONS (A THROUGH Me)	  -.   .   A-l

                             Volume 2 - Part B

EPA REVIEW NOTICE	    11
PREFACE	Ill
ABSTRACT	  .   .    1v
TABLE OF CONTENTS	     v
BIBLIOGRAPHIC CITATIONS (M THROUGH Z)   	   ...   M-l
ALPHABETICAL INDEX TO JUNIOR AUTHORS  	  .  	   Index-!
DOCUMENT INDEX	Index-121
GEOGRAPHIC INDEX   ......	Index-181
SUBJECT INDEX	Index-201
                                      xi

-------
                                   FIGURES


Number                                                                  Page

2-1   Types of column breakthrough curves	  .2-25
3-1   The relative stability of various americium solids in an
      oxidizing soil environment	3-3
3-2   The activity of various americium species in equilibrium
      with Am02(s) in em oxidizing soil environment  .	3-4
3-3   The activity of SbO  in equilibrium with various  antimony
      solids in an oxidizing soil  solution environment	3-14
3-4   The activity of various antimony species in equilibrium
      with Sb203(s) (cubic) with pNH4+ = 3.0 and pp- =  4.5	3-16
3-5   The activity of Ce   in equilibrium with phosphate
      levels from Variscite and Gibbsite (V and G), Dicalcium
      Phosphate Dihydrate (DCPD) and Octacalcium Phosphate
      (OCR).  ..'....	  .   .  .3-29
3-6   The activity of various cerium species in equilibrium
      with CeP04(s) in an oxidizing soil 'environment	3-30
3-7   The relative stability of cobalt solids in an oxidizing soil
      environment [p02(g) = 0.68 atm], pCa2+ = 2.5 and  phosphate
      levels in equilibrium with Variscite and Gibbsite (V and G),
      Dicalcium Phosphate Dihydrate (DCPD) and Octacalcium
      Phosphate (OCP)	3-71
3-8   The activity of various cobalt ions in equilibrium with
      CoC03(s) in an oxidizing soil environment	3-71
3-9   The activity of various curium ion species in equilibrium
      with Cm(OH)3(s)	3-80
3-10  The relative stability of various europium solids at
      pSOf- = 2.5	'	3-87
3-11  The .activity of various europium species in equilibrium
      with Eu(OH)3(s)	3-87
3-12  The influence of pH on the activity of europium in
      solution	3-89
3-13  The relative stability of various neptunium solids in
      an oxidizing soil  environment	3-102

-------
                           CONTENTS  (contd)

RUTHENIUM	3-150
   Natural Soil and Rock Distributions	3-150
   Brief Chemistry	3-151
   Solid Phase and Solution Equilibria   	 3-152
   Experimental Adsorption Results	3-154
   Migration Results  	 3-157
   Summary.	3-162
   References.	3-163
STRONTIUM  .......  	 3-166
   Natural Soil and Rock Distribution	3-166
   Brief Chemistry	3-166
   Solid Phase and Solution Equilibria   	 3-167
   Experimental Adsorption Results	3-169
   Migration Results  .	3-188
   Summary.	3-197
   References	3-198
TECHNETIUM.	.'  .. -	3-204
   Natural Soil and Rock Distribution.	3-204
   Brief Chemistry	 3-204
   Solid Phase and Solution Equilibria   	 3-205
   Experimental Adsorption Results	 3-205
   Migration Results  .	 3-207
   Summary	3-209
   References	3-210
THORIUM	3-211
   Natural Soil and Rock Distributions	 3-211
   Brief Chemistry	3-212
   Solid Phase and Solution Equilibria	  .  . 3-213
   Experimental Adsorption Results	3-215
   Migration Results  	 3-216
   Summary	3-217
   References.	3-218
                                    ix

-------
   TRITIUM	3-221
      Natural Soil and Rock Distributions	3-221
      Brief Chemistry	3-221
      Solid Phase and Solution Equilibria   	  3-222
      Experimental Adsorption Results	  3-222
      Migration Results   	  3-223
      Summary	3-226
      References	3-226
   URANIUM	3-228
      Natural Soil and Rock Distributions	  3-228
      Brief Chemistry	3-228
      'Solid Phase and Solution Equilibria	  3-230
      Experimental Adsorption Results.   .	  3-232
      Migration Results   	  3-235
      Summary	3-238
      References	3-239
   ZIRCONIUM	.   .	3-243
      Natural Soil and Rock Distributions   .  .   .	3-243
      Brief Chemistry	   .	  3-243
      Solid Phase and Solution Equilibria   	  3-245
      Experimental Adsorption Results	  3-247
      Migration Results   	  3-248
      Summary	  3-250
      References	3-251

SECTION 4 - CONCLUSIONS AND EVALUATION   	  4-1
   FACTORS INFLUENCING ADSORPTION OF RADIONUCLIDES ON
   GEOLOGIC MEDIA  	  4-2
   ADDITIONAL RESEARCH NEEDS	4-6
ACKNOWLEDGMENTS	  .  Ack-1

-------
                               FIGURES (contd)
Number                                                                  Page

3-14  The activity of various neptunium species in equilibrium with
      Np02(s) in an oxidizing soil environment ......... 3-103

3-15  The relative stability of various plutonium solids in an
      oxidizing environment Cp02(g) = 0.68 atm], pC02(q) =
      3.52 atm, pF- = 3.5 and phosphate levels in equilibrium
      with Variscite and Gibbsite (V&G) Dicalcium Phosphate
      Dihydrate (DCPD) and Octacalcium Phosphate (OCP) ...... 3-110

3-16  The relative stability of various plutonium solids in a
      very reducing environment Cp02(g) = 80 atm], pC02(q) =
      3.52 atm, pF° = 3.5 and phosphate levels in equilibrium
      with Variscite and Gibbsite (V&G) Dicalcium Phosphate
      Dihydrate (DCPD) and Octacalcium Phosphate (OCP) .....   : 3-110
3-17  The activity of various plutonium species in soil  solution
      in equilibrium with Pu02(s) at pH 8, pC02 - 3.52 atm, pCl" =
           - = 2.5 and pH2P04- = 5.0 ............ 3-112
3-18  The activity of various plutonium species in soil  solution
      in Pu02(s) in a mildly oxidizing soil  environment ......  3-113

3-19  The relative stability of various promethium solids .....  3-139

3-20  The relative stability of various ruthenium solids in an
      oxidizing soil environment  .............  3-153

3-21  The activity of various ruthenium species in equilibrium
      with RuOg (amorphous hydrate) in an oxidizing soil
      environment  ..................  3-153

3-22  The relative stability of various strontium solids at
      pCa2+ = pS042- = 2.5, pH4Si04 =3.1, and pC02(g) =1-52
      atm in equilibrium with Variscite and Gibbsite (V and G),
      Dicalcium Phosphate Dihydrate (DCPD) and Octacalcium
      Phosphate (OCP) ..................  3-168

3-23  The activities of various strontium species in equilibrium
      with SrC03(s) in the soil at pCl = pS042' =2.5, pNOa- =
      3.0, pF- = 4.5, pC02(g) » 1.52 atm and pH2P04" = 5.0 .....  3-169

3-24  The relative stability of various thorium solids in
      equilibrium with Variscite and Gibbsite (V&G) Dicalcium
      Phosphate Dihydrate (DCPD) and Octacalcium Phosphate (OCP).   .   .  3-213

3-25  Activity of various thorium species in equilibrium
      with Th02(s), pCl" = pNOs" = pS042' = 3.0, pF~. =
      4.5 and  HP0- = 5.0 ...............  3-214
3-26  The relative stability of various uranium solids in an
      oxidizing soil environment [p02(g) = 0.68 atm],  PC02(g)
      =3.52 atm, pK* = pNa+ = pNfy* = 3.0 and phosphate levels
      in equilibrium with Variscite and Gibbsite ........  3-231

3-27  Activity of various uranium species in equilibrium with
             in an oxidizing soil environment .........  3-232
                                     xi i i

-------
                               FIGURES (contd)

Number                                                                  Page

3-28  The stability of zirconium solids at 25°C in equilibrium
      with soil  silica (oH4Si04 =3.1)  	  3-245

3-29  The activity of various zirconium species in equilibrium
      with zircon [ZrSi04(s)] and soil solution silica (pH4$i04
      =3.1) at pCT = pS042- = 2,5,  pNOs" = 3.0 and pF~ = 4.5.   .   .   .  3-246
                                     xiv

-------
                                  TABLES
Number                                                                 Page
 1-1     LWR Plant Waste from Uranium Fuels Reprocessing 	  1-3
 2-1     Cation Exchange Capacities .  	  2-10
 2-2     Increase in Cation Exchange Capacities with Kaolinite
         Particle Size	'	2-10
 2-3     Increase in Cation Exchange Capacities for Some
         Smectites with Particle Size	2-11
 2-4     Cation Exchange Capacity in meq-100 g  Due to Grinding .   .   .  2-11
 2-5     Examples of Replacement Reactions   	  2-16
 2-6     Soil Capacities for Saturated and Unsatured Flow
         Conditions	2-20
 2-7     Dispersion Coefficients in the Unsaturated Zone of
         a Yellow Eolian Soil	2-23
 3-1     Americium Radionuclide Data.  .	3-2
 3-2     Americium Kd Values from Several  Organic Solutions ....  3-5
 3-3     Americium 50-Day Distribution 	  3-8
 3-4     Leaching of Americium Adsorbed on the  Soil by
         Organic Wastes 	  3-10
 3-5     Antimony Concentration in Igneous and  Sedimentary Rocks   .   .  3-13
 3-6     Antimony Radionuclide Data 	  3-13
 3-7     Radiochemical Analysis of Test Shot Debris Sampled
         about 200-ft Above the Shot Point at about 2 Years
         in Age	3-17
 3-8     Composition of Safford Copper Ore	3-17
                           •toe
 3-9     Solutions Used in    Sb Equilibrations with Glacioflu-
         viatile Sediments from Wells  	  3-18
 3-10    Sediments Used in Obtaining Equilibrium Distribution
         Coefficients with 125$b and the Solutions Given in
         Table 3-9	3-19
 3-11    Equilibrium Distribution Coefficient Values (ml/g)
         Between Seven Glaciofluviatile Sediments and the
         Solutions of Table 3-9 Containing 125Sb   	  3-19
                                    xv

-------
                              TABLES (contd)

Number                                                                 Page
         •joe
 3-12       Sb Analyses of Water and Soil Samples from a
         Seep Near Trench 7	'.   .  3-20
 3-13    Chemical  Compositions of Melt Glass and Primary
         Copper Ore.   .   .	3-21
 3-14    Composition  of Leaching Solution Influent  .   .   	  3-22
 3-15    Cerium Abundance in  Rocks   .   .  .  .   ._	3-27
 3-16    Stable Isotopes of Cerium	3-28
                       144
 3-17    Adsorption of    Ce  on Suspended Solids ........  3-34
 3-18    Cesium Content of Rocks	3-37
 3-19    Cesium Distribution  Coefficients for Several  Rocks,
         Minerals.and Soil  Types  	  3-40
 3-20 '   The Equilibrating Solution Composition  Plus Trace
         137cs That Was Utilized in All  of Berak's (1963)
         Kd Work With 24 Hour Equilibrations at  20°C	3-49
 3-21    Effect of Addition of Potassium to the  Influence on
         the Sorption of Cesium by Vermiculite from 0.5M NaN03
         Containing the Mass  Equivalent  of 2 uc  13'Cs/ml	3-49
 3-22    Effect of pH on 137Cs Adsorption	3-50
                                          1 "37
 3-23    Effect of NaNOs Concentration  on    Cs  Adsorption	3-50
 3-24    Summary of Cesium Kd Values	3-51
 3-25    Variation of Trace Cesium Kd Values with pH	  3-52
 3-26    Solution:Soil  Ratio  Effect-on Trace Cesium
         Equilibrium  Distribution Coefficients	3-53
 3-27    Cesium Equilibrium Distribution Coefficients  for Sand,
         Silt and  Clay Fractions of Burbank Loamy Fine Sand in
         0.5N NaCl	3-53
 3-28    Trace Cesium Adsorption by Three Hanford Soils from
         a 0.2N NaCl  Solution	3-54
 3-29    Cesium Kd Values Between Seven  Glaciofluviatile  Sediments
         and the Solutions  of Tables 3-9 and 3-10	3-55
 3-30    K Values  of  Cesium in the Saturated Subsoils  at
         Mol, Belgium	3-59
 3-31    Cesium Distribution  Coefficients and Relative Migration
         Rates in  Houthulst Clays at pH  3	3-59
 3-32    Properties of Natural Water Samples  	  3-61
 3-33    Adsorption of Radionuclides on  Suspended Solids   	  3-61
                                     xvi

-------
                              TABLES (contd)

Number                                                                 Page
 3-34    Cobalt Concentrations in Igneous and Sedimentary
         Rocks in ppm	3-69
 3-35    Curium Radionuclide Data	3-79
 3-36    Curium 50-Day Distribution Computed from
         Sheppard et al	3-82
                       ?4?
 3-37    Kd Values for    Cm in Untreated Aiken Clay Loam ..... 3-83
 3-38    Europium Content of Igneous Rocks in ppm	3-85
 3-39    Europium Content of Sedimentary Rocks in ppm  	 3-85
 3-40    Europium Radionuclide Data	 3-85
 3-41    Variation of Trace Europium Kd Values on Mol Soils  .... 3-88
 3-42    Effect of pH and Europium Concentration on
         Europium Kd by Burbank Sand	3-90
 3-43    K Values of Europium in the Saturated Subsoils at
         Mol Belgium  .	  . 3-91
 3-44    Iodine Concentration in Volcanic Rocks  	 3-94
 3-45    Iodine in Sedimentary Rocks	3-95
 3-46    Range of Surface Soil Properties Used in the Methyl
         Iodide and Iodide Adsorptions  	 3-97
 3-47    Kd Values and Correlations Relating Soil Properties
         Iodide and Methyl  Iodide Adsorption  	 3-98
 3-48    Properties of Soil Samples  	 3-104
 3-49    Neptunium Kd (ml/g) as a Function of Soil  and Solution .   .  . 3-104
 3-50    Neptunium 50-Day Distribution Values Computed from
         Sheppard et al	3-105
 3-51    Plutonium Radionuclide Data 	 3-108
 3-52    Plutonium Adsorption Versus Particle Size   .  .   .'  .   .   .  .3-116
 3-53    Plutonium Kd Values for Quartz-of Various  Particle
         Sizes from A 0.001 N HN03" Plutonium Solution 	 3-116
 3-54    Plutonium Kd Values for Different Minerals of Coarse
         Clay Size from 0.001 N HN03-Plutonium Solution	3-117
 3-55    Rhodes'  Plutonium Kd Data as Reported by Evans	3-117
 3-56    Removal  of Plutonium from pH 7 Water by Several  Soil
         Minerals	3-120
 3-57    Percent Plutonium Leached by Extractants	3-125
 3-58    Plutonium Kd as a Function of .Oxidation State	3-125
                                     xv n

-------
                               TABLES (contd) -

Number                                                                  Page
 3-59  .   Kd- Values of 238Pu in Untreated Aiken Clay Loam	3-129
 3-60     Average Radium Content of Various Rock Types 	 3-142
 3-61     Radium Radionuclide Data	'.. 3-143
 3-62     Radium Kd Values from Lime-Neutralized Waste;
          1.25 g Exchanger/1 of Waste   ...'.'.	3-144
 3-63     Simulated.River Water Composition	.--3-144
 3-64     Radium Distribution Coefficients with the Solution
          of Table 3-63	3-145
 3-65     Abundances of Ruthenium in Igneous Rocks and
          Rock-Forming Minerals in ppm	:.....  .3-150
 3-66     Ruthenium Contents of Shales  .  .  .	3-151
 3-67     Atomic Percentages of Stable Ruthenium Isotopes 	 3-151
 3-68     Ruthenium Distribution Coefficients on a Hanford Soil
          as a Function of pH  .  .  :	3-154
 3-69     Equilibrium Distribution Coefficients Between Seven
          Glaciofluviatile Sediments and the Solutions in
          Table 3-9 Containing 106RU	  .  .  .3-156
 3-70     Properties of Natural Water Samples 	 3-160
 3-71     Adsorption of 106Ru on Suspended Solids in
          Natural  Waters 	 3-160
 3-72     Average Strontium Content of Igneous Rocks   .   ." .  .  .  .3-166
 3-73     Average Strontium Content in Sedimentary Rocks  -
          and Sediments	3-166
 3-74     Strontium Kd Values for Western Soils  	 3-171
 3-75     The Influence of Anions on the Strontium Kd in  the
          Hanford Composite Soil  	 3-171
 3-76     Summary of Strontium Kd Values	3-17-3
 3-77     Strontium Distribution Coefficients for Several Rock,
          Mineral  and Soil Types  .  .  .  :	3-174
 3-78     Microstrontium-Macrocalcium Discrimination of Some
          0.1 to 0.2 mm Rocks and Minerals	  .  .  .3-181
 3-79     Variation of Trace Strontium Kd Values with pH   	 3-182
 3-80     Composition of Burns, Mississippi, Montmorillonite .... 3-184
 3-81     Comparison of Experimental and Computed Strontium
          Kd Values for Burns Montmorillonite 	 3-184
                                    xvm

-------
                              TABLES (contd)

Number                                                                  Page
 3-82     Strontium Kd Values	  3-186
 3-83     Characteristics of the Soils Used in Strontium Kd
          Determinations 	   .3-187
 3-84     Trace Strontium Adsorption Characteristics from
          0.2N NaCl	,   .  3-187
 3-85     K Values of Strontium in the Saturated Subsoils
          at Mol, Belgium	  3-193
 3-86     Strontium Distribution Coefficients and Relative
          Migration Rates in Houthulst Clays at pH 3   .  .  .  • .   .   .3-194
 3-87     Technetium Content of Some Natural Materials .  .  .   .   .   .  3-204
 3-88     Range of Properties of 22 Surface Soils Used in the
          Pertechnetate Adsorption Studies 	   .  3-206
 3-89     Properties of Soils Used in the Technetium
          Adsorption Studies	3-206
 3-90     Properties of Soil Samples 	  3-207
 3-91     Technetium Kd as a Function of NaHCOi Concentration for
          a South Carolina Soil Characterized in Table 3-90  ....  3-207
 3-92     Properties of Bergland and Arveson Soils Used in
          the "Tc Extraction Work	3-208
 3-93     Extractability of 99Tc from Bergland and
          Arveson Soils  	  3-208
 3-94     Thorium Content of Common Rocks and Soils 	  3-211
 3-95     Thorium Radionuclide Data	3-212
 3-96     Abundances of Uranium in Natural Materials   	  3-229
 3-97     Nuclear Properties of Natural  Uranium Isotopes  	  3-230
 3-98     Uranium Radionuclide Data	>	3-230
 3-99     Uranium Kd Values	3-235
 3-100    Desorption of Uranium from Soils with Distilled Water
          and Carbonate Solutions	'	3-236
 3-101    The Average Zirconium Content of Rocks 	  3-243
 3-102    Natural Isotopes of Zirconium and Their Abundances  ....  3-244
 3-103    Zirconium-Niobium Kd Values as a Function of pH	3-247
 3-104    Adsorption of 95Zr and 95Nb on Suspended Solids
          in Natural Waters	  3-250
                                    xix

-------
                              TABLES (contd)


Number                                                                  Page

 4-1       Factors Reported to Effect Adsorption of Radioelements
          Over the pH Range of 4 to 9	4-3
 4-2      Predominant Solution Species of Elements in a  pH 4 to 9,
          p02 0.68 to 80, pC02 1.52 to 3.52,  pCl" = pNOs"  - pS042'  =
          3.0, pF~ 4.5 and pH2P04~ 5.0 Environment Without Organic
          Ligands  .   .	4-4
                                     xx

-------
                                 SECTION 1
                                INTRODUCTION

     The nuclear industry, like many other industries, produces wastes.  Part
of these wastes are radioactive and biologically hazardous without management.
In general, the hazard represented by the radioactive wastes decreases with
isolation and storage, due to radioactive decay with passing time.  However,
some of these wastes contain radioisotopes of americium, curium, iodine,
neptunium, plutonium, technetium and uranium with half-lives of many thousands
of years, so that these isotopes will persist.
     Selected radioactive wastes have been, and are being, disposed by land
burial.  Barriers around storage and disposal sites are generally constructed
to safeguard against radioactive contamination of the environment by wind
erosion or by trespassing of unauthorized personnel.  However, in the case of
accidental spills or breach of barriers, the radioactive elements can enter
the surrounding environment.  Water is one of the main transporting agents in
terrestrial environments and will  control dissemination of the radionuclides
below the ground surface.  The biological retention and mobility of the radio-
nuclides in geologic regimes will  be governed by the physical  and chemical
characteristics of the geohydrologic system.  The composition of the water in
the geohydrologic system is determined by the adsorption-desorption and
precipitation*dissolution reactions between the water and the solid phases
making up the soils, sediments and rocks.  As one of the principal controls
of the concentration of radionuclides in the geohydrologic system at a given
time, the interactions of the radionuclides with soils and rocks in an aqueous
system become of primary interest to understanding radionuclide migration and
retention.  As a consequence, the present work was undertaken to critically
review the literature available through 1976 concerned with radionuclide
interactions with soils and rocks.  The objectives of this study were to:
                                     1-1

-------
 1.  summarize and critically review the available information on radionuclide
     interactions with the geologic media, and
 2.  determine the deficiencies in the available data and outline future work
     required to correct these deficiencies.
     Several reviews already are available that include radionuclide inter-
actions with geologic materials (Gera, 1975;  Borg et al., 1976a; Dames and
Moore, 1976; Routson, 1973; Bensen, 1960).  However, these reviews do not
cover but a limited number of elements, refer to work done only in specific
geographical areas or their emphasis is not primarily the interactions between
geologic media and radionuclides.

METHODS AND CONSTRAINTS
     The fission and activation products that are present in nuclear process-
ing and other radioactive wastes and are most biologically hazardous were
selected for study.  For each selected element, pertinent available literature
on natural distribution in soils and rocks, thermodynamic data for solid and
solution species, chemistry of the element, radionuclide  adsorption-desorption,
radionuclide precipitation and dissolution and the mobility of the element in
laboratory and field situations was collected.  These data were organized and
used to critically review the present state of knowledge  of the interactions
of radionuclides with geologic media.
     Due to time (14 months) and level of effort (one man year) constraints,
the present study was limited in scope.  Soil and rock reactions with a
selected group of radionuclides were thoroughly reviewed.  Less effort was
spent on elemental chemistry and natural distribution. Available thermodynamic
data for various solids and solution species  used in this study were not
critically evaluated by the present writers.   However, the thermodynamic data
were checked for internal consistency.  Laboratory work,  field work and model-
ing were excluded from the study.
Selection of Radionuclides
     Based upon radionuclide concentrations in radioactive wastes  (Table 1-1),
19 potentially biologically hazardous elements were chosen for  critical  review
including americium, antimony, cerium, cesium, cobalt,  curium,  europium, iodine,

                                     1-2

-------
  TABLE 1-1.   LWR PLANT WASTE FROM  URANIUM  FUELS  REPROCESSING*
                  (SCHNEIDER AND  PLATT,  1974)
                        SELECTED ACTINIDES. curies/metric ton of U
Radlonucl Ide
225Ra
226Ra
229Th
23°Th
233U
234U
237NP
239Np
239Pu
24°Pu
"'Am
243Am
242Cn,
245Cn
Separation
3.48 x 10"8
1.97 x 10"8
3.49 x 10"8
2.04 x 10"5
2.31 x 10"7
3.77 x 10"3
3.40 x 10"'
1.76 x lO1
1.62 x 10°
2.37 x 10°
1.52 x 102
1.76 x 101
1.93 x 10*
2.76 x 10"'
1 Year
3.50 x 10"8
2.85 x 10"8
'3.50 x 10"8
2.05 x 10"5
1.84 x 10"6
3.95 x ID"3
3.40 x 10"'
1.76 x 101
1.62 x 10°
2.63 x 10°
1.52 x 102
1.76 x 101
4.08 x 103
2.76 x 10"'
SELECTED FISSION
3H
79Se
90Sr
93Zr
95Zr.
99Tc
103Ru
"KRu
107Pd
124Sb
125Sb
126Sn
129j
'34Cs
135Cs
137Cs
141Ce
'47Pm
148Pm
151 Sm
152Eu
">
6.92 x 102
3.98 x 10"'
9.72 x 10*
7.68 x 104
1.46 x 10°
2.77 x 10S
1.43 x 101
8.83 x 104
4.10 x 10S
1.10 x 10"'
8.53 x 101
7.99 x 103
5.46 x 10"'
3.74 x 10"2
2.15 x 105
4.96 x 10"'
1.07 x 105
S.64 x 104
7.71 x 10S
9.80 x 104
3.19 x 101
1.47 x 103
1.14 x 101
6.76 x 103
9.84 x 103
6.54 x 102
3.98 x 10"'
7.48 x 102
7.50 x 104
1.46 x 10°
5.65 x 103
1.43 x 101
1.48 x 102
2.06 x 105
1.10 x 10"'
1.26 x 10°
6.18 x 103
5.46 x 10"'
3.75 x 10"2
1.53 x 105
4.96 x 10"'
1.04 x 105
2.29 x 101
3.16 x 105
7.52 x 104
7.71 x 10"2
1.46 x 103
1.07 x 101
6.47 x 103
6.71 x 103
10 Years
4.22 x 10"8
1.09 x 10"7
4.22 x 10"8
2.09 x 10"5
1.49 x 10"5
6.64 x 10"3
3.41 x 10"'
1.75 x 101
1.62 x 10°
4.52 x 10°
1.55 x 102
1.75 x 101
3.16 x 10°
2.76. x 10"'
100 Years
7.31 x 10"7
1.11 x 10"6
7.31 x 10"7
3.42 x 10"5
1.47 x 10"4
2.56 x 10"-2
3.45 x 10"'
1.74 x 101
1.67 x 10°
8.90 x 10°
1.44 x 102
1.74 x 101
2.09 x 10°
2.74 x 10"'
1 ,000 Years
6.94 x 10"5
7.22 x 10"5
6.94 x 10"S
3.86 x 10"4
1.52 x 10"3
4.56 x 10"2
3.67 x 10"'
1.60 x 101
2.05 x 10°
8.25 x 10°
3.43 x 101
1.50 x 101
3.46 x 10"2
2.56 x 10"'
10,000 Years
5.54 x 10"3
2.91 x 10"3
5.54 x 10"3
3.74 x 10"3
1.56 x ID"2
4.45 x 10"2
3.74 x 10"'
7.10 x 10?-
4.02 x 10°
3.28 x 10°
1.31 x 10"'
7.10 x 1.0°
5.24 x ID"20
1.31 x 10"'
PRODUCTS, curies/metric ton of U
3?94 x 102
3.98 x 10"'
7.05 x 10"17
6.01 x 104
1.46 x 10°
3.40 x 10"'2
1.43 x 101
1.54 x 10"23
4.15 x 102
1.10 x 10"'
4.08 x 10"17
6.14 x 102
5.46 x 10*'
3.75 x 10"2
7.30 x 103
4.96 x 10"'
8.48 x 104
0.00
1.04 x 102
6.96 x 103
2.15 x 10"25
1.36 x 103
6.38 x 10°
4.38 x 103
2.14 x 102
2.47 x 10°
3.98 x 10*'.
0.00
6.53 x 103
1.46 x 10°
0.00
1.43 x 101
0.00
4.58 x 10"25
1.10 x 10"'
0.00
5.70 x 10"8
5.46 x 10"'
3.75 x 10*2
4.47 x 10"'°
4.96 x 10"'
1.06 x 104
0.00
0.00
3.19 x 10"7
0.00
6.46 x 102
3.53 x 10"2
8.89 x 101
2.32 x 10"'3
2.35 x 10"22
3.94 x 10"1
0.00.
1.50 x 10"6
1.46 f 10°
0.00
1.43 x 101
0.00
0.00
1.10 x 10"1
0.00
0.00
5.42 x 10"'
3.75 x 10"2
0.00
4.95 x 10"'
9.92 x 10"6
0.00
0.00
0.00
0.00
5.12 x 10"'
9.45 x 10"25
1.05 x 10"15
0.00
0.00
1.37 x 10"'
0.00
0.00
1.45 x 10°
0.00
1.39 x 101
0.00
0.00
1.10 x 10''
0.00
0.00
5.10 x 10"'
3.74 x 10"2
0.00
4.94 x 10"'
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
•Assumptions:  power 30 megawatts/metric ton, exposure 33,000 megawatt-days/metric ton,
            3.3X enriched U, flux 2.92 x 10'J N/cm'-sec, spent fuel reprocessed
            150 days after discharge, 0.5J fuel loss to waste.

                                           1-3

-------
neptunium, plutonium, promethium, radium, ruthenium, strontium, technetium,
thorium, tritium, uranium and zirconium.   This list includes most of the
elements that are of both immediate and long-term concern.   Cobalt is not
listed in Table 1-1 because it is not an  actinide or a fission product.   It
is included here because as   Co it is widely used in gamma irradiation
sources and occurs in other cobalt-containing materials that have been exposed
to a neutron flux.  Cobalt can often be found in industrial and medical  radio-
active wastes.  Radium, a uranium and thorium decay chain product, is listed
with the parent actinides for convenience in Table 1-1.  The selected radio-
nuclides, for example, also include the fission products and actinides calcu-
lated to be the most radiotoxic in the leachate from German high level waste-
borosilicate glasses (Scheffler et al., 1977).
Collection of Pertinent Literature
     Because of limited time and resources available for the literature  col-
lection and abstracting, an attempt was made to use already available biblio-
graphical data to critically review radionuclide interactions with geological
media.  However, it soon became apparent  that the available bibliographies
were not adequate (Environmental  Plutonium Data Base Group, 1972, 1973;
Martin et aT.7~T97~
-------
fauna.  The term soil is used in an engineering sense, and is not tied to any
specific depth such as the root zone.  A sediment has a different origin than
a soil, but is considered here to be composed of essentially the same mate-
rials.  A rock is a consolidated material  composed of several mineral phases
and may also contain organic matter.


REFERENCES

Ames, L. L., Dhanpat Rai and R. J. Seme.   1976.  A Review of Actinide-Sediment
Reactions with an Annotated Bibliography.   BNWL-1983.

Bensen, D. W.  1960.  Review of Soil Chemistry Research at Hanford.   HW-67201.

Borg, I. Y., R. Stone, H. B. Levy and L. D. Ramspott.  .1976a.  Information
Pertinent to the Migration of Radionuclides in Ground Water at the Nevada
Test Site.  Part 1:  Review and Analysis of Existing Information.  UCRL-52078,
Pt. 1.

Borg, I. Y., R. Stone, H. B. Levy and L. D. Ramspott.  1976b.  Information
Pertinent to the Migration of Radionuclides in Ground Water at the Nevada
Test Site.  Part 2:  Annotated Bibliography.  UCRL-52078, Pt. 2.

Dames and Moore.  1976.  Development of Monitoring Programs for ERDA Owned
Radioactive Low-Level Waste Burial Sites.

Environmental Plutonium Data Base Group.  1972.  Environmental  Aspects of
Plutonium.  A Selected, Annotated Bibliography.  ORNL-EIS-72-21.

Environmental Plutonium Data Base Group.  1973.  Environmental  Aspects of
Plutonium and Other Elements.  A Selected, Annotated Bibliography.   ORNL-EIS-
73-21 (Suppl. 1).

Faust, R. A., F. M. Martin, C. T. Sanders, and S. S. Talmage (comps. and eds.).
1975.  Environmental Aspects of the Transuranics:  A Selected,  Annotated
Bibliography.  ORNL-EIS-75-21-5.

Francis, C. W., S. S. Talmage and B. B. McMullin.  1975.  Radionuclide Move-
ment in Soils and Uptake by Plants.  A Selected, Annotated Bibliography.
ORNL-EIS-75-77.

Gera, F.  "1975.  GeochemicaT Behavior of Long-Lived Radioactive Wastes.
ORNL-TM-4481.

Martin, F. M., C. T. Sanders and S. S. Talmage (eds.).  1974.  Environmental
Aspects of the Transuranics.  A Selected,  Annotated Bibliography.  ORNL-EIS-
74-21 (Suppl. 3).

Routson, R. C.  1973.  A Review of Studies on Soil-Waste Relationships on the
Hanford Reservation from 1944 to 1967.  BNWL-1464.


                                     1-5

-------
Scheffler, K., U. Riege, K.  Louwrier, Hj.  Matzke,  I.  Ray and H.  Thiele.   1977.
Long Term Leaching of Silicate Systems:   Testing Procedure,  Actinides  Behavior
and Mechanism.  KFK 2456.  EUR 5509e, p.  29.
Schneider, K.
Alternatives.
2.D.22.
J. and A.  M. Platt (eds.).   1974.   High-Level  Waste Management
 BNWL-1900, Volume I, Appendix 2.D, pp.  2.D.7-2.D-8, 2.0.21-
                                     1-6

-------
                                  SECTION 2

                     PROCESSES INFLUENCING RADIONUCLIDE
                            MOBILITY AND RETENTION
     This section is concerned with a general discussion of some of the factors
that are expected to influence the radionuclide-geologic media interactions.
All of the factors discussed here may or may not influence the behavior of a
specific element.  The influence of these factors on various selected elements
is the subject of the next section.  The main purpose of this section is to:
 1)  briefly enumerate the factors and processes involved in radionuclide-
     geologic media interactions in the next section,
 2)  define various terms, and
 3)  outline-the reasons for developing, limitations and the method of construc-
     tion of various solid phase and solution species diagrams presented in
     the next sections.

NATURAL SOIL AND ROCK DISTRIBUTIONS
     It is important to determine the concentration of the stable isotopes
of a given element in the soils and rocks because the chemical behavior of
the stable isotopes would not differ from that of the radioisotopes of the
same element.   Hence, the presence of stable isotopes can affect the mobility
and retention of the radioisotopes.  For example, limestone contains an
                                                                  90
average of 610 ppm strontium.  Disposal or an accidental spill o'f   Sr-contain-
ing waste in a limestone area would very likely add much less strontium to the
environment than was already present as stable isotopes.  The stable isotopes
of strontium can be present in such abundance that they control the soil reac-
tions and mobility of the radioactive strontium.
                                     2-1

-------
     The average elementary abundances in rocks and soils are given in the
next section.  The actual elemental concentrations may range ..widely from the
average values given due to local conditions.  It is always necessary to
establish local base level concentrations to avoid disposal or storage of the
radioisotope of an element in environments with high concentrations of stable
isotopes of the same element.  Such an environment could enhance mobility of
the radionuclide.

SOLID PHASE AND SOLUTION SPECIES
     Among factors that influence radionuclide behavior in the environment,
the nature of the solid phases and solution species of the radionuclide is
one of the most important.  Through their precipitation and dissolution solid
phases would cause changes in solution concentration which, in turn, would
affect adsorbed ions.  The environmental  concentration of most radionuclides
is expected to be low so that discrete solid phases of the element may not
be present in the terrestrial environment.  In this case the element may be
distributed throughout the soil  and rock matrix and adsorption reactions may
control solution concentrations.  However, if the concentration of the element
in solution is such that a solid phase of the element can precipitate, the
knowledge of this solid phase and solution ions would help in predicting the
solution concentrations available for transport (Dutt, Shaffer, and Moore,
1972; Jurinak and Medrano, 1974; Rai and Franklin, 1973; Tanji and Doneen,
1966).  Granted the importance of the total solution concentrations of the
element, the nature of the predominant solution species are important since
they affect 1) adsorption through their charge; 2) adsorption because of
changes in the nature of the species due to alteration in solution properties
such as pH, Eh, competing ions and complexing ions; 3) movement through the
soil and rock matrix because of their physical size; and 4) plant uptake.
     Laboratory information on possible solid phase and solution species of
elements of interest in radioactive waste is almost nonexistent.  What little
information is available is very site and solution specific with such poorly
defined systems that no extrapolations can be made to other situations.   In order
to bridge this gap, thermodynamic data are generally used to construct Eh-pH
                                     2-2

-------
diagrams to predict stable, phases and solution species (Pourbaix, 1966;
Garrels and Christ, 1965).  This approach also has several  limitations, but
is the best method until extensive laboratory data become available.   Some of
the limitations of a theoretical approach are:
 1.  The reliability of the prediction is dependent upon the accuracy of the
     thermodynamic data which, in a large number of elements of concern in
     radioactive waste, contain a large error term.
 2.  The approach is limited to those compounds for which information is
     available at the present time.
 3.  This approach does not consider the kinetics of reactions.
 4.  The predictions apply only under equilibrium conditions.
 5.  The influence of organic ligands and colloids which may be very  important
     cannot be evaluated because the thermodynamic data are lacking.
     In order to develop solid phase-solution species diagrams,  accurate values
of the equilibrium constants of all the reactants and products are required.
The time and money available for this study allowed neither an extensive search
nor a critical evaluation or tabulation of the available thermodynamic data.
In a related DOE sponsored document, Rai and Seme (1978),  the equations and
thermodynamic data used to construct the solid phase-solution  species diagrams
are tabulated.  Therefore, equilibrium constants were selected from several
readily available sources, the most important of these sources being  Sillen  and
Martell (1964); NBS publications (Rossini et al., 1952; Wagman et al., 1968,
1969, 1971; Parker et al., 1971; Schumm, 1973), Latimer (1952);  Pourbaix (1966);
Cleveland (1970); Keller (1971); Burney and Harbour (1974); and  Baes  and Mesmer
(1976).  The thermodynamic data were used to develop equilibrium constants for
chemical equations that describe the behavior of various species.   Generally
these equations are used to develop Eh-pH diagrams (Pourbaix,  1966; Garrels  and
Christ, 1965) that provide the information regarding the stable  compounds and
solution species in the whole range of pH and Eh.  However, Eh-pH diagrams
can be used only for elements that exist in more than one oxidation state, and
these diagrams do not indicate the total amount of the element and the relative
amount of various species in solution under specified conditions.   Therefore,
activity-pH diagrams rather than Eh-pH diagrams were developed.   The  solid
                                    2-3

-------
phase-solution species diagrams of elements affected by oxidation-reduction
were developed at an assumed partial pressure of oxygen (0.2 atmospheres in
most cases).  However, the changes in solution concentration or the stability
of the solid phases that would occur due to the decrease in oxygen pressures
are also discussed.  This approach is similar to the one used by Lindsay (1972),
Baes and Mesmer (1976), and Rai and Serne (1977).
Uncomplexed, Mononuclear and Polynuclear Solution Species
     An element in solution can exist as free ions or as mononuclear and poly-
nuclear species (complexed).  The total  concentration of the element in solution
can be obtained by summing the concentrations of all the free and complexed
ions.   Mononuclear species are of the type ML, ML9, ML.,	ML , where M is
                                                 u    Oo     n
the metal ion and L is the complexing ligand such as C03~, HCOZ, F", NOl, Cl",
OH", CM", CgHcOy " (citrate) and C2H302~ (acetate).  Some examples of mononu-
clear species are CoOH+, Co(OH)2 , Co(OH)3" and Co(OH)4  .  Polynuclear species
are of the type MmLn, where m and n are integers.   Some examples of polynuclear
species are Co2OH3+ and Co4(OH)44+.
     In general, mononuclear species of the type ML maintain higher concentra-
tions in equilibrium with metal ions than the ML2	ML  species.   Hence, in
most cases mononuclear species of the type ML2	MLn do not contribute sig-
nificantly to total metal ion solution concentrations.   Hydroxide bridging is
apparently responsible for polynuclear species of M(II), M(III) and M(IV) ions
(Baes and Mesmer, 1976).  The polyanions of M(V) and M(VI) usually involve
oxygen bridging.  In most cases, the polynuclear species do not contribute
significantly to total metal ion solution concentrations.
     Polymerization of a radionuclide is a special case of hydrolysis and
precipitation, and is primarily a function of Eh,  pH, temperature, radionuclide
concentration and the presence or absence of complexing ligands.  For example:
 T)  Pu(IV) polymerizes but Pu(VI) and Pu(III) do not significantly polymerize
     (Cleveland, 1970),
 2)  at 10  M Pu(IV), 91% of the plutom'um was present as polymer at pH 8.5
     (Lindenbaum and Westfall, 1965), but true colloid was not present at
     pH values of <7.5 for 10"8M Pu(IV)  (Qrebenshchikova and Davydov, 1961),
     and

                                     2-4

-------
                                                       p
 3)  the presence of complexing ligands such as F", S04  , and oxidizing agents,
     reduce the polymerization of plutonium (Cleveland, 1970).
     The metal ions of interest in:this study that might be expected to poly-
merize include americium, antimony, cerium, cobalt, curium, europium, plutonium,
promethium, ruthenium, thorium, uranium and zirconium.  Usually the polymeric
                                       o           o
units are colloidal in size range (MO A to 10,000 A) and possess a surface
charge that tends to keep the units from coagulating and precipitating from
solution.  Rhodes (1957b) showed with laboratory studies of Kd versus solution
pH that 239Pu, 106Ru, 95Zr-95Nb and 144Ce Kd values all tended to diminish in
the pH region of 9 to 13, probably due to reduced surface change on the polymers
produced by the increasing amounts of NaOH added to reach the higher pH values,
resulting in reduced polymer adsorption by the soil.  Prediction of whether
or not colloidal size polymeric units will form during a precipitation reaction
and remain stable in solution is difficult because of the large number of sys-
tem variables that effect possible polymer formation.  However, the metal
cations with high charges (M  , M  ) show the greatest tendency to form
polymers.
     The stability constants that were available in the literature,and appli-
cable to the soil and rock environments selected, were utilized in this study
to estimate solid phases and solution species equilibria.   It is unfortunate
that reliable hydrolysis constants are not available for many of the elements
including plutonium.  Also missing are reliable formational constants for many
other complex ion species such as the nitrosyl ruthenium complexes.   In some
cases, there is a reasonable doubt as to the identity of some of the species,
including those formed with organic materials found in many soils and surface
waters.
Construction and Interpretation of the Diagrams
     The details of the method of constructing the diagrams presented in this
report can be found in Rai and Lindsay (1975) and Rai and Serne (1977).   In
these diagrams the pCOgCg), K , and pNa , etc., refer to the negative logarithm
of the activity (in moles/1 for solution species and atmospheres for gases).
The oxidation-reduction conditions are denoted by partial  pressures  of oxygen.
                                     2-5

-------
      In  the  solid phase diagrams, any compound that  lies below another com-
 pound at a given pH  is the more stable.  Thus, for any two solids at a specific
 pH, the  solid that maintains lower ionic activity is more stable.   In addition
 to stability, information regarding weathering and formation can also be
 obtained from these  diagrams.  If, in a given soil,  the soil solution composi-
 tion is  below the solid line, the soil is under-saturated with respect to
 the solid represented by the line, and the mineral will dissolve or weather.
 On the other hand, if a soil solution point is above the solid line, the soil
 is supersaturated with respect to the mineral represented by the line, and
 the mineral can precipitate.
     The solution species diagrams were developed assuming an oxidation-
 reduction environment, an equilibrium with a compound, and ionic activities
 of various cations (Ca  , Na+, NH/1") and anions (Cl~, HCCL, CO,  , NO,", F",
   n         • ,                  f                       J    J      J
 SO, ", and P0» ") that may be- present in the terrestrial environment.  As men-
 tioned earlier, the  organic ligands and colloids that may be present in the
 solutions were not considered.  The solution species diagrams depict anionic,
 cationic, and several neutral or uncharged species for which the data were
 available.

 ION EXCHANGE
     Ion exchange is a reversible interchange of ions between two compounds,
 one of which is insoluble in the medium in which the exchange is carried out
 (Amphlett, 1964).  It is one of the most common mechanisms responsible for
 radionuclide adsorption on geologic materials.

     If X" is a solid, negatively-charged medium,  an  exchange reaction  can
occur between two cations, -A  and B  as follows:
     X"A+ + B"1" -JX'B** A+<  .
The anion does not enter into the reaction in this case because the  solid
medium has a residual negative charge that is inherent in  the mineral  frame-
work.   In the case of clay minerals and zeolites,  the negative charge is
                               +3       +4
acquired by substitutions of Al    for Si    in the  mineral  lattice.
                                     2-6

-------
    .A selectivity quotient or constant can be experimentally derived from
contacting the exchange medium, based in cation A  , with a known concen-
tration of cation B .  For univalent-univalent cation exchanges, the
selectivity coefficient,

     KB _  [XV] [A+]
      A    [B+] [XV]. '

where the brackets denote solution and solid medium concentrations.  For a
univalent-divalent exchange, the reaction would be 2X~A+ + B   t(X") B   + 2A+,
with the rational selectivity coefficient,
      B              *2
        =
                   [A
      A   [B+2] [X V]2   '                      .
If the capacity of the solid medium is known in meq/g (m  ) and the initial
solution concentration in meq/ml (C ), a relationship between the selectivity

              B                                       +
coefficient, KjJ , and a distribution coefficient for B , KdB, can be derived.
Assuming that [A*] + [B*] = CQ and [xV] + [X~B+] = mQ then

      B . CX"B*3 CA*1 _ [XV] (Co - [B+])  =      . Co • [B+]    .
 .     A   [B+] [XV]   [B+]  (m0 -  [X'B+])          m0 -  [X'B]

 If  B   is  a  trace  component  ([B  ] much  less  than  CQ,  [X"B  ]  much  less  than
 m ),  then the  above  can be  written  as:
              P                m
     KB % K(JB _?_  and KdB £ KB £ ,
               0                0
or the Kd is inversely proportional to [A*] + [B+].  If Kd  were plotted
versus log [A*], a straight line would result with a slope of minus one.
                    +2
A divalent cation, A  , would yield a slope of minus two when plotted in
the above manner.  The same treatment applies equally well to anion exchange
reactions.
     One of the key parameters in the above treatment is the requirement that
the component for which the Kd value is determined be present in the system
                                      2-7

-------
in trace' quantities.  In practice, a trace quantity should be defined experi-
mentally by determining Kd values at several concentrations of the traced ion.
The Kd will remain constant, all other things being equal, over the traced ion
concentration range for which the Kd concept remains valid.
     An implied limitation can be found in the original exchange reaction,
 -++'-+    +
XA  + B  £ X B  + A ..  The expression requires that the solid medium, X  , be
based in cation A  before contacting the solution containing B .  A Kd value
that has been experimentally determined for cation B  without first. basing
the solid medium (soil, clay mineral, zeolite) in cation A , has been deter-
mined for an undefined system.  Also, because the dimensions of m  were meq/g,
and of C , meq/ml , the dimensions of the Kd value are ml/g.
     If the selectivity quotient is modified with the activity coefficients
for A and B on the solid medium, f.-and fg, and in the solution y. and yg, a
rational thermodynamic equilibrium constant, K, can be derived (Gaines and
Thomas, 1953):
                                           B

     The thermodynamic quantities are sometimes of value in establishing and
studying reaction mechanisms, but are limited in practical applications.  The
                                                             D
distribution coefficient (Kd) and the selectivity quotient (1C) where trace
concentrations are exceeded are experimentally determined values that are
characteristic of a given exchange system, and can be applied in several
practical situations.
Ion Exchange Properties of Soil and Rock Components
     As far as soils, sediments and altered rocks are concerned, ion exchange
means chiefly cation exchange.  The soil particles have an amphoteric charac-
ter, but in general they carry a net negative charge (Winklander, 1964).  For
cation exchange, the negative charge on the soil colloids increases with rising
pH.  The positive charge tends to increase with diminishing pH, giving rise to
an anion exchange capacity on the acidic side.  Chloride and sulphate ions,
for example, show no adsorption at pH 7 on kanditic or smectitic soil minerals
(Mattson, 1931).  Lowering of the pH activates basic groups,  R-OH, to R-OHL,
                                     2-8

-------
with the OH originating in clay mineral broken bonds and hydrous oxides of
iron, aluminum and manganese.  They take part in exchange reactions such as,
R-OH^Cl" + NO^ -" R-OH^ NO^ + Cl~.  However, the anion exchange capacities of
most soils are relatively minor compared to their cation exchange capacities
because the pH.of most soils is seldom as low as 4.0, and the number of sites
that can be activated to function as anion exchange sites usually are relatively
small as well.
     There are at least two types of anion exchange in clay minerals.  The
first involves replacement of the hydroxyl (OH") ions by other anions.
McAuliffe et al. (1947) showed that the OH ions of clays can enter into
exchange reactions by using clays with deuterium-tagged hydroxyls.   The replace-
ment of hydroxyls by fluoride ions also was reported (Dickman and Bray, 1941). '
     Another type of anion exchange involves the substitution of anions such
as phosphate, arsenate and borate on the edges of the (SiO,) (silica tetra-
hedron) sheets, perhaps growing as extensions of these sheets.   Other anions
such as sulfate, nitrate and chloride are not adsorbed because their geometries
are not comparable to the silica tetrahedra.
     A third, type of anion exchange site could occur as a positive site on
basal surfaces as a result of a local  electrostatic imbalance in the clay
framework such as an excess of aluminum in the octahedral position.
     In the cases of both OH ion replacement and anion exchange due to geo-
metrical similarities between the anion and silica tetrahedra,  exchange would
occur around clay mineral edges.  There would be very little exchange on basal
surfaces of expandable clay minerals.   Investigations of anion  exchange are
difficult because of the low pH requirement.   However, for clay minerals the
cation and anion exchange capacities should be about equal  at optimum pH
values for each.
     The anion exchange capacities of expanding clays were 12 to 31  meq/100 g
for smectites and 4 meq/100 g for vermiculite, while kandites varied from
6 to 20 meq/100 g (Hofmann et al., 1956).   Compare these anion  exchange
capacity data with those of Table 2-1  for cation exchange at pH 6 to 8.
     The cation exchange capacity ranges of some common clay minerals are
given in Table 2-1.  The ranges vary due to differences in crystallinity and

                                     2-9

-------
particle size.  In soils, rocks, weather rocks and sediments, cation  exchange
capacities are due mainly to the types and quantities of minerals  in  the  silt
(2 to 62 urn) and clay (<2 urn) size fractions.

         TABLE 2-1.  CATION EXCHANGE CAPACITIES  (CARROLL,. 1960,  1970)

   :                                     Cation Exchange Capacity,
                    Mineral             	meq/lOOg	
          .  Kaolinite                             3-15
            Halloysite (2H20) -                   5-10
            Halloysite (4H20)        '           40-50
            Smectites                            36-100
            Illite                               10-40
            Vermiculite                          100-150
            Chlorite                              4-47
            GlauconUe                           11-20
            Palygorskite                         20-30
            Allophane                        .     70
            Zeolites                             100-300
            Organic Matter in Soils.  '           130-350
            Feldspar                              <1
            Quartz                                <1

     There is a considerable variation in cation exchange capacity with particle
size shown by the non-expandable soil clays.  Kaolinite, for example,  shows  an
increase in exchange capacity as outlined in Table 2-2.  The expanding layer
silicates, on the other hand, show little change in cation exchange capacity
with particle size as shown for some smectites in Table 2-3.

      TABLE 2-2.' INCREASE IN CATION EXCHANGE CAPACITIES WITH KAOLINITE
                  PARTICLE SIZE (HARMON AND FRAULINI, 1940)

                      	Particle Size, urn	
   Cation Exchange       10-205-TOZ-41-0.50.5-0.250.25-0.10.1-0.05
   Capacity, meq/lOOg     2.4     2.6   3.6    3.8      3.9         5.4       9.5
                                    2-10

-------
         TABLE 2-3.  INCREASE IN CATION EXCHANGE CAPACITIES  FOR  SOME
                     SMECTITES WITH PARTICLE SIZE  (CALDWELL  AND
                     MARSHALL, 1942)

                                    	Particle Size,  um	
    Cation Exchange        Smectite    2-1    1-0.5   0.5-0.2   0.2-0.05   -0.05
    Capacity, meq/IOOg    Nontronite   60.8   61.0     64.35.70
                        Saponite     69.3   76.0     8.15       86.3      81.4
     An increase is to be expected with decreasing particle  size when  the
cation exchange capacity primarily occurs from broken  bonds  in minerals  such
as illite and kaolinite.  In the .case of expanding layer  silicates where most
of the capacity is on basal plane surfaces, particle size  has little effect
on cation exchange capacity.
     Excessive grinding of clay and nonclay minerals, causes  an increase  in
cation exchange capacity, probably a result of the same decrease in particle
size and number of broken bonds, as shown in Table 2-4.  A rubber-lined  ball
mill and polished agate balls were used for the grinding.

      TABLE 2-4.  CATION EXCHANGE CAPACITY IN meq/100  g DUE  TO GRINDING
                  (KELLY AND JENNY, 1936)

                                         Grinding Time, Days
                             100 Mesh    ~2~     3        7
          Muscovite            10.5        -    76.0
          Biotite               3.0      62.0   72.5
          Kaolinite             8.0      57.5   70.4    100.5
          Montmorillonite     126          -    238.0

     The clay minerals are relatively sensitive to pH.  Chatterjee and Paul
(1942) and Mukherjee et al. (1942) have shown that so-called hydrogen  clays
are really hydrogen-aluminum or hydrogen-magnesium-aluminum  systems, the mag-
nesium and aluminum being removed from the clay framework  with resulting clay
damage.  Other things being equal, higher valence cations  more easily  replace
lower valence cations, and are more difficult to replace when already  present
on a clay or zeolite.  Hydrogen ions are an exception  because they behave
more like divalent or trivalent cations.  With ions of the same valence, the
                                    2-11

-------
smaller diameter hydrated cations are more tightly held than the larger diam-
eter hydrated cations.  The usual lyotropic replacement series on smectites,
for example, is Cs  > K  > Na  > Li  and Ba   > Sr   > Ca   > Mg  .   An excep-
tion arises when the cation just fits the size and charge requirements of the
anionic site, as does K  into the basal sites of illite.
     These cations can cause structural modifications in the clay or zeolite
structure that leads to their so-called "fixation" (Bray and DeTurk, 1939;
Wood and DeTurk, 1941; Nishita et al., 1956; Frysinger, I960; Tamura. and Jacobs,
1960; Tamura and Jacobs, 1961).
    •In contrast to clay minerals, soil organic matter has no well-defined
ability to bind exchangeable cations, although the exchange displacement fol-
lows the lyotropic order, as do most of the clays (Bartlett and Norman, 1938).
The cation exchange capacity increases with increasing pH values, arising from .
the increasing dissociation of phenols and carboxylic acids at higher pH levels
(Broadbent and Bradford,.1952).
     The cation exchange capacity of soil  organic substances are reported as
between 150 and 450 meq/100 g depending on the origin of the fractions.  Posner
(1966) showed-that the cation exchange capacities of humic acids extracted
from a red-brown soil under different conditions yielded substances  with rela-
tively similar cation exchange capacities.  Yields of humic acids varied from
2.7 to 12.0 g/kg soil and cation exchange capacities varied from 217 to 409
meq/100 g, with the averages at 7.0 g/kg soil for yield and 310 meq/100 g
capacity.   Cation exchange isotherms between humic acids and alkaline earth
and alkali metals were investigated in detail by Zadmard (1939).  The role of
carboxyl and hydroxyl groups in metal cation fixation by organic soil materials
was investigated by Schnitzer and Skinner (1965).  Specific blocking of the
functional groups indicated two reaction types:  1) a main reaction  with par-
ticipation of both carboxyl and phenolic groups, and 2) a lesser reaction
involving only carboxyl groups.  'No evidence was found for exchange  reactions
of the carbonyl groups.
                                    2-12

-------
ANION EXCLUSION
     Reactions between an inorganic anion  and a soil or rock exchanger can
involve both adsorption and repulsion.  Adsorption of anions is due to a
limited number of positive charges on soil minerals (anion exchange capacity).
Anion exchange and the capacities of various minerals are discussed in the
ion exchange section above.  Anion exclusion involves the repulsion of anions
by the predominantly negatively charged clay minerals and zeolites in soil
(Schofield, 1947; de Haan and Bolt, 1963; de Haan, 1964; de Haan, 1965).  The
higher the cation exchange capacity (negative charge) of the soil or rock,
the more pronounced the anion exclusion (Thomas and Swoboda, 1970).  The high
cation exchange capacity smectites exhibit anion exclusion to a greater degree
than do the kandites (kaolinite group).  Thus anions, such as chloride, iodide
and pertechnetate, can be repelled by the negative charges of a high cation
exchange capacity soil into the higher velocity soil  solution flow channels
and break through the soil column in less than one pore volume of displacing
solution (Thomas and Swaboda, 1970; Cassel et al., 1975).  The implication
for anionic radionuclides and anionic complexes of radionuclides is that their
displacement through high cation capacity soil  columns can take place at less
than one displacing solution volume, with a higher apparent migration velocity
than the displacing solution.

DIFFUSION
     One of the parameters of aqueous transport of ions is that of aqueous
diffusion.   The transport of matter in the absence of bulk flow is referred
to as the process of diffusion.   The flux of matter due to diffusion is pro-
portional to the concentration gradient and is  a molecular process.  In
general terms, the flux, J. , of component i in the z direction, is (Fick's
                   dfy    1Z
first law)  0.  = -D^-1- , where D is the proportionality constant and diffusion
            1Z   .  QZ                 n
coefficient, with the dimensions of cm /sec, J.  has  the dimensions-of moles/
  ?
cm /sec, n. is the concentration of material i  and the negative sign indicates
that flow of material i in the z direction is also in the direction of lower
i concentration.  In a binary electrolyte, at infinite dilution, the aqueous
diffusion coefficient is:
                                    2-13

-------
                               D  =
                               'o   (M-, + u2) F
where 'u, and Uo are the mobilities of the two ions (K  and Cl", for example,
as a binary electrolyte former), RT is the gas constant times the degrees
                                                                  o
Kelvin and F is a faraday.  The dimensions of D  are once again cm /sec.  The
ionic mobilities of many ions are known (Harned and Owen, 1958; Dean, 1973)
in aqueous solutions.  Scott et al .  (1974) defined an apparent or measured.
difusion coefficient (D.) and its relation to the aqueous diffusion coefficient
(D ) as (the change of concentration with time in the x direction)
                                 *r   D« ^2r
                                 9C    o 3 C
                           -      3t " R  3X2 '
where R is a retardation factor.
     The retardation factor,  R, was defined as

                                R = 1 + £  Kd
                                        £
where p  is the soil bulk density, e is the pore fraction (volume occupied by
the solution phase) and Kd is the equilibrium distribution coefficient with
dimensions of ml/g.  Note that the above expression is the same one that is
used' to evaluate the relationship between groundwater velocity and radionu-
clide migration velocity.  The same retardation factor serves in both cases.
The apparent diffusion coefficient is defined by DA = DQ/R.
                                            -8
     Relyea (1977) obtained D  values for 10" M plutonium nitrate in 0.01M
HNO,, 0.1M KNO, and 1M Ca(NO,), of 4.8 x 10~6 ± 2.5 x 10"6 cm2/sec, 7.8 x 10"5
           5   ?                  5            52
± 3.7 x 10   cm /sec and 3.0 x 10"  ± 2.0 x 10"  cm /sec, respectively.
These D  values may be compared to the limiting diffusion coefficient of KC1 ,
                        -52
for example, of 2.0 x 10   cm /sec.   Subsequent measurement of the plutonium
particle size in the 0.01M HNO, by means of the Stokes-Einstein equation
                              0                            o
showed that it had rapidly hydrolyzed to particles about 8 A in diameter at
25°C.  Increasing the nitric acid level to 0.1M reduced the effective radius of
                                   o
plutonium particles to less than 1 A.  The measured DQ from the Fuquay soil
supernate was 2.7 x 10"5 ± 1.2 x 10"5 cm2/sec at pH 2.0, 25°C and 8 x 10"10M
plutonium nitrate, and from the Burbank soil at pH 7.1 and plutonium 5 x 10"  M,

                                     2-14

-------
D  = 3.6 x 10"  ± 1.0 x 10~6 cm /sec.  The effective plutonium radius for the
 0                   o
Burbank sample was 6 A.  The D  values of various molecules are remarkably
similar in the absence of polymerization, or some other complicating factor.
The plutonium D» values obtained by Relyea for the Burbank sandy loam at
                              '             -in   7
e = 0.30, T = 21°C and 791 hours was 5 x 10    cnT/sec and at e = 0.14, T =
                              -11   2
21°C and 3600 hours was 8 x 10    cm /sec.  The above DA values can be com-
pared to the previously-given D  value for Burbank supernate to illustrate
the influence of the retardation factor (and Kd) on the mobility of high Kd
radionuclides.  If, on the other hand, the radionuclide Kd value is very low,
the apparent diffusion coefficient (D.) approaches the aqueous diffusion
coefficient (D ) in value.
     The simple expression, x" = ,/Vt (Jost, 1960) can be utilized to determine
the effect of D, the diffusion coefficient and t, time, on the distance (root
mean square distance) that a radionuclide will migrate.  The dimension of D
     2
is cm /sec and t also is in seconds with x in cm.  If D. is substituted for D
in Jost's expression above, it can be seen that the high Kd radionuclides such
as plutonium must have very high t values in order to diffuse a significant
distance.  However, the low Kd value radionuclides such as tritium or. techni-
tium, can diffuse distances of up to 25.cm/yr, due to D. values of close to
      -52
1  x 10   cm /sec.  Diffusion can be an important migration mechanism for the
low Kd radionuclides.  Its potential contribution to migration should be
examined for each radionuclide.

REPLACEMENT REACTIONS
     Replacement is a type of precipitation reaction in which one of the ions
of a final replacement product is contributed by a solid that is more soluble
in the contacting solution than is the final product.  Replacement reactions
can be divided into two kinds:  1) anion replacement reactions in which only
the anion in the solid is replaced by an anion in solution, with cations of
both solution and solid being common, and 2) cation replacement reactions in
which only the cation in the solid is replaced by the a cation in solution,
with anions of both solution and solid being common.  Some examples of anion
and cation replacement reactions are given in Table 2-5.  The principle govern-
ing replacement reactions, assuming that the solution-solid system obeys Raoult's
law, is a relatively simple solubility relationship.

                                    2-15

-------
                 TABLE  2-5.   EXAMPLES  OF  REPLACEMENT  REACTIONS
                             (AMES,  1961a;  AMES,  1961b)
Original . Ion in
Solid Solution
Reaction
Solid
Product
Anion Replacement
CaCO^ F"
0
CaCO, P0«
•J *T

CaC03 Ba+2
CaC03 Sr+2
CaS04 2H20 Ba+2
CaS04 2H20 Sr+2
CaC03 + 2F" ^ CaF2 + C03"2
5 CaC03 + 3P04"3' + OH" -^
Ca5(P04)3OH + 5 C03"2
Cation Replacement
CaC03 + Ba+2 t=; BaC03 + Ca+2
CaC03 + Sr ^ SrC03 + Ca+
CaS04'2H2.0 + Ba+2 f^ BaS04 + Ca+2
CaS04-2H20 + Sr 2 f^ SrS04 + Ca+2
CaF2
Ca5(P04)3OH .

BaC03
SrC03
BaS04
SrS04
                                     2H20
     The difference in solubility between. the original so'lid and replacement
product is the driving force for the reactions shown in Table 2-5.  The reac-
tion will cease as the ratio of replaced anion or cation to replacing anion or
cation approaches the same ratio as the activity products of the original solid
and solid product with the common anion or cation.  For example, if the
activity products are 1.5 x 10"9 for BaS04 and 3.6 x 10"5 for CaS04«2H20 during
the replacement reaction CaSO,,*2H90 + Ba+2  ^  BaSO, + Ca*2 + 2H,0,.the
                              *
                              *                           1.0       »
reaction rate approaches zero as the activity ratio of Ca   to Ba+2 approaches
3 6 x 10~5             4
— - Q, or 2.4 x 10 .  If excess calcium is added, the equilibrium can be
1.5 x 10-a
made to go to the left and form CaS04*2H20)  (gypsum).  When several solid
products are theoretically possible, only the least soluble will form.
     Replacement reactions involve the dissolution of the original solid at
an interface between solid product and original solid (Ames, 1963).  Bonds
between cation and anion in the original solid must actually be broken, the
ions diffuse into solution and recrystallize on the advancing solid product.
Ions must be able to diffuse through the solid product to the reaction face or
                                    2-16

-------
the replacement reaction soon ceases.  It is during the portion of the reaction
when ions are traveling between the more soluble original solid to the less
soluble final product that trace radionuclides able to satisfy the ion size
and charge requirements of the final product are included in the growing
reaction product (Ames, McHenry and Honstead, 1958; Ames, 1959; Ames, 1960).
                                        90
Replacement type reactions that removed   Sr from high salt wastes containing
0..01M P04~  were first reported by Rhodes (1957).  The presence of the phosphate
in the solution caused a reaction with the calcite in the soil to form apatite,
                      90
Ca5(P04)3OH, with the   Sr substituted for calcium in the apatite crystal
lattice.  Other cations that can be included in the apatite lattice during
the replacement reaction include promethium, uranyl ions and plutonium (Ames,
1960).   The cations that are removed into the crystallizing apatite are also
the cations found to be enriched in natural  apatites, including strontium,
uranium, the rare earths, thorium, manganese and barium (Deer, Howie and
Zussman, 1962).  The effects of phosphate concentration, calcite replacement
                          90
rate and pH on removal of   Sr from solution was given by Ames (1959).  Tamura
                                                                         90
(1962)  showed the effects of contact time and phosphate concentration on   Sr
removal by calcite replacement.   Belot and Gailledreau (1963) studied the
            90
kinetics of   Sr retention during the calcite-phosphate replacement reaction.
     One of the problems with the use of replacement type reactions for radio-
nuclide removal from waste solutions involves kinetics.  As mentioned above,
the reaction is diffusion-controlled.  As the thickness of solid product
becomes greater, the diffusion path becomes  longer and more tortuous, leading
to increasingly slower reaction rates.  If the solid product coating is too
thick,  reaction ceases and the radionuclides pass through the soil  without
removal by a replacement reaction.  In addition, it is usually necessary to
add a substantial amount of reacting ions such as phosphate to the waste solu-
tion before ground disposal, which can be costly and lead to a precipitate in
the solution prior to disposal.   Adjustment  of solution pH also may be required,
as in the case of the calcite-phosphate reaction.

PHYSICAL TRANSPORT AND FILTRATION
     Colloidal particles, due to their large size compared to solution species
and their changes in charge in response to pH, exhibit different migration and

                                    2-17

-------
retention behavior from solution species (Van Olphen, 1963; Grebenshchikova
and Davydov, 1965; Sheidina and Kovarskaya, 1970).  Colloidal radionuclide
particles that have been described in the literature are of two types.  These
are true colloids and pseudocolloids.  A true colloid is defined as composed
of a radionuclide only, while a pseudocolloid is a radionuclide adsorbed on
another colloidal size particle.  Several rare earths, zirconium, plutonium,
curium, antimony and americium could occur partially as true colloids, preci-
pitates or particulate suspensions (Rhodes, 1957; Bensen, 1960; Andelman and
RozzelT, 1970; Fukai and Murray, 1974; Price and Ames, 1975; Ames, 1976).  .The
downward movement of colloidal particles through soils would mainly depend
upon physical factors such as permeability, rate of water movement and
particle size.  This same dispersion and translocation of soil colloids occurs
under natural soil conditions (Drew, 1967; Gerasimov and Glazovskaya, 1960) and
may lead to problems with internal soil  drainage.  Ground disposal of low ionic
strength condensate wastes was reported by Routson (1973) to result in partial
plugging of the soil column and reduced infiltration rates because of disper-.
sion and filtration of upper soil  colloids.  Generally, the downward movement
would be limited because the colloids are subjected to filtration by the soils
and sediments (Routson, 1973; Price and Ames, 1975; Ames, 1976).  The lateral
or horizontal movement of colloidal particles depends upon their degree of
exposure to wind and surface water action (Hakonson et al., 1975).
     A recent report concerned with the long term leaching of silicate glasses
containing high level waste (Scheffler et al., 1977) serves as an example of
physical transport of several actinides.  The base glass composition used to
contain the radionuclides was Si02 = 50.5 wt%, Ti02 = 4.2 wt%, A1203 = 1.4 wt%,
B203 =13.6 wt%, CaO = 2.8 wt% and Na20 =27.5 wt%.  Two actinide-containing
glasses were made using the base glass formula.   One contained 78.43 wt% base
glass, 16.64 wt% fission product and corrosion product oxides, 4.90 wt% Am02
and 0.03 wt% Cn^.  The other glass contained 80.85 wt% base glass, 17.15 wt%
fission product and corrosion product oxides, 0.005 wt% Am02 and 2.00 wt% Pu02-
Leaching experiments with distilled water, 1M NaCl and a saturated brine solu-
tion were performed.  Actinide concentrations were on the order of 10" M in
the leachates after a period of initial  leaching, with rates of 10"  to 10" g/
  2
cm /day common.  An examination of the leachate to determine the form of the
actinides indicated that particles attached to the actinides actually dissolved
                                    2-18

-------
from the glass rather than the actinides alone, and that the particles were
                                                2
high silica polymers with an average size of 10   urn.  The actinides  (curium
americium, and plutonium) were bonded to the silica polymers in the glass
and remained so in the leachate.

SATURATION EFFECTS
     In migration of radionuclides, hydraulic conductivity is of primary
importance.  Hydraulic conduction is the ratio of the flux density (the volume
of water flowing through a cross-sectional area per unit time) to the hydraulic
gradient (the head drop per unit distance in the flow direction).  The reason
for the importance of hydraulic conductivity to radionuclide migration is the
requirement that a liquid (water) move the radionuclides either physically as
dispersed colloids or on dispersed soil colloids, or as ions or neutral com-
plexes in solution.  If the soil water does not move, then the radionuclides
in, or contacted by, the soil water do not move except by diffusion,  a rela-
tively slow process (Hajek, 1965; Hajek, 1966; Rancon, 1973).
     Coincidentally, the most important difference between saturated  and
unsaturated flow is the same hydraulic conductivity (Hillel, 1971).   When the
soil is saturated, nearly all pores are filled and conducting, and conductivity
is at a maximum.  As the soil becomes unsaturated, some of the pores  become
air-filled and the conductive cross-sectional area decreases.  In addition,
the first pores to empty under tension are the largest and most conductive,
and tortuosity is increased by these empty pores.  In unsorted soils  and sedi-
ments, the large pores that resulted in high conductivity at saturation become
barriers to liquid flow between smaller pores during unsaturated flow.  Hence,
the transition from saturated to unsaturated flow may result in a steep drop
in hydraulic conductivity of several orders of magnitude as the tension
increases from 0 to 1 bar.  At higher tensions, conductivity may be so low
that steep negative pressure gradients are required for any appreciable soil
water flow to occur.
     An interesting corollary of the pore size-conductivity relationship is
that at or near saturation, a sandy soil conducts water more rapidly  than a
clay soil with many micropores.  When the soils are unsaturated, however, many
of the micropores in the clay soil remain filled, and consequently, the hydraulic

                                    2-19

-------
conductivity in the clay soil does not decrease nearly as sharply as it does
in sandy soil under the same tension.   In fact, an occurrence of a sand layer
in a clay soil  can constitute a barrier to water flow through the clay soil
under unsaturated flow conditions (Hillel, 1971; Winograd, 1974).
     Most of the work on radionuclide adsorption by soils has been on nearly
saturated to saturated flow systems, mainly because saturation represents a
system parameter where hydraulic conductivity, and subsequently radionuclide
migration, is at a maximum value.  Because most of the laboratory radionuclide
adsorption work on soil columns is conducted at saturated flow conditions,
and the soil-water system is largely unsaturated during waste disposal opera-
tions, Knoll (1960) examined the possibility that the cation exchange capacity
of the soil under unsaturated flow conditions was not equal  to the soil capa-
city at saturated flow conditions.  His interest was in applying laboratory.
results to field situations.  Small  columns of'a fairly uniform, very fine
sand at constant temperature, packed bed density and tension were used with
                                          +2             +2           -2
an influent solution containing 1.0 g/1 Ca  ,0.1 mg/1 Sr   and 2 x 10
       Qfi
uCi/ml   Sr.  The results of the experiments are given in Table 2-6.  There
was apparently no difference in strontium capacity between saturated and
unsaturated flow conditions.

          TABLE 2-6.  SOIL CAPACITIES FOR SATURATED AND UNSATURATED
                      FLOW CONDITIONS (KNOLL, 1960)

 % Saturation    Average Flow Rate,  ml/hr    Strontium Capacity,  meg/100  g
      100                   1.3                       4.0 x  10"4
      100                   2.5                       4.4 x  l.O"4
      100                  10.3                       5.2 x  TO"4
      100                  11.0                       5.2 x  10"4
      100                  10.2                       4.6 x  10"4 '
       65                   2.1                        3.6 x  10"4
       65                   1.42                      6.0 x  10"4
       58                   0.22                      5.2 x  10"4
       48                   0.3                       5.4 x  10"4
       36                   0.2                       5.4 x  10"4
                                    2-20

-------
     However, it is worth noting that column flow rate differences between
100% saturation and 36% saturation are less than a factor of 100.  With use
of an aggregate of less uniform pore size, flow rate (hydraulic conductivity)
differences could have been much greater, the volume of soil contacted by the
aqueous phase under unsaturated flow conditions much smaller, and hence the
cation exchange capacity of soil might have differed considerably between
saturated and unsaturated flow conditions.  On the other hand, the clay frac-
tion of the soil constitutes the largest exchange capacity and smallest pore
sizes.  Since the smaller pores are involved in unsaturated flow, there may
be little obvious effect on the exchange capacity of the soil in desaturating
it.  If the columns are normalized by referring to a column pore volume rather
than a simple effluent volume, the difference in ion exchange capacities of
saturated and unsaturated flow columns tends to disappear.
     Nielsen and Biggar (1961, 1962a, 1962b) have shown that the drier the
soil, the greater the effluent volume required to reach a tracer effluent/
influent ratio (C/Co) of 1.0.  The desaturation eliminates the larger flow
channels and increases the stagnant water volumes which are difficult to dis-
place.  The relatively unmoving water volumes caused by desaturation are shown
as C/Co versus volume of effluent (ml)/volume of water in the column (ml).
The latter coordinate represented the effluent volume divided by the pore
volume used by the soil at that saturation.  Desaturating the soil  progres-
sively shifted the initial breakthrough to the left and increased skewness of
the breakthrough curves.  Aqueous diffusion is a relatively unimportant migra-
tion mechanism near saturation, but can become important as solution velocity
through the soil decreases at unsaturated flow conditions for low Kd
radionuclides.
     Nelson et al. (1962) conducted model field tests in disposal sites, one
24 in.2 and the other 6 In. .  The influent was a solution of 600 ppm Ca*
    QC
and   Sr.  The water table at the field site was 12 ft below the ground surface,
overlain by a fairly homogeneous fine sand and silt soil with 31% porosity.
Only a small part of the soil beneath the model disposal site was found to be
water saturated as measured by neutron moisture probe.   A section through the
disposal site was monitored to show moisture distribution to the groundwater
table.  Unsaturated flow conditions were typical of most of the section.

                                    2-21

-------
     Baetsle et al.  (1965) examined the breakthrough curves of   Co, added
to the influent as K^   Co (CN)g, through Mol white sand.  Percolation velocity
was about 3 cm/hr at soil saturation of 100%, 76%, 58% and 46%.  The data were
plotted as effluent volume in milliliters versus fraction of the influent   Co
concentration in the effluent at a given throughput volume.  The curves were
somewhat different in shape than those of Nielsen and Biggar (1961) because
percolation velocity was ten times faster through the   Co column, and the
columns were only about one-fifteenth as large as those used by Nielsen and
Biggar.  The differences resulted in a flattening of the terminal ends of the
  Co column breakthrough curves, and a very rapid appearance of the   Co tracer
in'the initial  column effluents.  Columns of yellow eolian sand were used to
       85
obtain   Sr breakthrough curves at 100%, 45% and 30% saturation.  These columns
tended to load in successive steps, indicating that the solution in the non-
saturated columns percolated through preferred channels.   The normalized
breakthrough curves also were given for the strontium work.  The normalization
consisted of plotting C/Co versus the volume of effluent in cm  divided by the
                                        3
volume of water in the soil  column in cm.   The unsaturated pore volume was
equal  to the fraction of soil  saturation times the soil  pore volume at 100%
                                                   134
saturation.'   Normalized curves also were given for    Cs loading on the yellow
eolian sand.   The stepwise loading was less pronounced but dispersion became
increasingly important as the soil  became less saturated.
     Several  hydrodynamic dispersion coefficients in the unsaturated zone were
determined and these value's  are given in Table 2-7.   The same method was used
to obtain dispersion coefficients for saturated flow.   After the pore volume
for a given saturation was determined on the column, a thin layer of    I con-
taminated sand was added and the elution curve of the radionuclide obtained at
the same saturation.   The elution curves counted at time t were then plotted
                                                P /Prt
on a cumulatively linear probability scale as S;c/Co  versus cm of column
length.  The curve on probability paper became  nearly a straight line.
Between the abscissa values  0.159 and 0.841, the standard deviation, a,  was
determined,  and using the relationship a = /2Dt, the dispersion coefficient,
D, was calculated..  Strontium adsorption had a smaller dispersion coefficient
in the less saturated columns.  The velocity of percolation through the soil
averaged 4 cm/day, or about  5 x 10"  cm/sec.  The authors concluded that

                                    2-22

-------
molecular diffusion played a major role  in radionuclide movement  as  soil
saturation became a very small value.  There  is considerable  controversy  in
the literature at present regarding the  extension of  the dispersion  concept
to a field scale.
        TABLE 2-7.  DISPERSION COEFFICIENTS  IN THE UNSATURATED  ZONE  OF
                    A YELLOW EOLIAN SOIL  (BAETSLE ET AL.,  1966)
    Radionuclide
       131r
    60
      Co(CN)
3.13x10
      -4
                                    Soil Saturation, %
100

-
75 55 45
Dispersion, cm^/sec
2.65xlO"4 l.lxlO"3 1.9xlO"3
2.20xlO"4 9xlO"4
35

2.5x10"3
30

-
6.10x10
      -4
7.4x10
     -4
9.5x10
3.3x10
-4
-6
1.2x10
               -3
                                                                  6x10
                                                                      -6
     Schwille et al.  (1967) reported some model experiments  on  fluid  flow  in
the transition zone from unsaturated to saturated  soils.  They  stress the
difficulties involved in analytically treating  liquid  flows,  even  in homogene-
ous media,-and suggest--that whenever possible field  tests should  be used to
investigate the migration of radioactive solutions.
     Routson and Serne  (1972), in a study involving  experimental  verification
of their one-dimensional PERCOL model, determined  the  breakthrough curves  and
85Sr capacity for unsaturated Tank Farm sandy loam soil  columns at 96%, 75%,
62% and 45% of saturation.  The column volumes  used  were adjusted for pore
volume at  that saturation.  The parameters are  given for each column  includ-
ing influent solution composition, soluble salts  in  the  initially dry soil
column, CaC03 content,  temperature, bulk density,  flow rate, soil volume,  soil
weight and column dimensions.  At 45% saturation,  the  flow rate was reduced  to
0.055 cm/hr.  There was no apparent reduction of  capacity as a  function of
saturation in these systems.  Obtaining curves  and capacities at  lower satura-
tions would have required nearly 1 year of experimental  time.   At the higher
tensions and lower  saturation, only a very limited amount of solution would
be transmitted per  unit area.  The fraction  of  the total radionuclides adsorbed
during ground disposal  of liquid wastes in the  very  low  saturation  zones was
considered to be negligible due to the small volumes of  solution  per  volume  of
soil.
                                    2-23

-------
     Rancon (1969, 1973) performed several field studies of water behavior in
unsaturated soils.  Simple methods were developed to measure in situ the move-
ments of the soil wetting front (deffned as 2% FLO by volume), the radionuclide
front and the .kinetics of soil drying.  Rancon stated that soil contamination
did not move when there was no transfer of liquid water by infiltration or
redistribution, at least over the time period of the study (3 months).  The
wetted front always constituted the maximum envelope of the contamination, and
all movement and redistribution of the radionculides took place within this
limit.  Iodine and strontium did not move in the soil column during soil drying.
by evapotranspiration.  However, if the water table is shallow, a steady flow
of liquid can occur from-a water source below to the evaporation sink above
(Hillel, 1971) as a function of the micrometeorological conditions (evaporation
rate) and water transmitting properties of the soil profile.  Such capillary
soil moisture transmission could transport radionuclides to the surface, just
as it has been known to transport salts to the surface (salinization).  Without
                       131
a carrier, part of the    I used became fixed in the soil during the drying out
phase of the evapotranspiration studies and did not move during a following
elution.  Tritium was'not retained in the soil under similar circumstances.
The protection afforded the groundwater by unsaturated soils was emphasized.
     Three breakthrough curves which might be encountered with the use of soil
columns are shown in Figure 2-1 (Nielsen and Biggar, 1962b).  In the first, no
interactions between solid, solvent and solute have occurred.  Piston flow
rarely, if ever, occurs in soil columns because pores of constant radii would
be required to attain it.  Ion exchange resin breakthrough curves approach
piston flow (Samuelson, 1963).  The second curve showing longitudinal disper-
sion effects can be seen with soil columns under the proper operating conditions
(where the solid is selective for the tracer ion and the column flow rate is very
slow, reducing hydrodynamic dispersion to a minimum).  This symmetrical break-
through curve (area A = area B) can be used to derive a Kd value where C/CQ =
0.5.  The third type of breakthrough curve in Figure 2-1 shows that the curve
is not symmetrical (area |A 7s area B).  This curve is typical of a soil composed
of equal-sized aggregates with intra- and inter-aggregate pore systems of
differing size, which gives an extremely wide range in pore velocity distribu-
tion that is bimodal at saturation.  In addition, the breakthrough curve can
                                    2-24

-------
                           1.0
                         y 0.5
                                         PISTONaOW
                           LO
                           0.5
                            0

                           1.0.
                        y  0.5
                        u
LONGITUDINAL
 DISPERSION
                                           WIDE RANGING
                                             VELOCITY
                                           DISTRIBUTION
                            "01     2     3     4
                                    PORE VOLUMES
               Figure 2-1.  Types of column  breakthrough curves
                            (Nielsen and Biggar,  1962b)
be displaced  to  the right by any chemical or  physical  process that retains  the
solute within  the  column.  Incomplete mixing  throughout the entire soil  solu-
tion moves the breakthrough curve to the  left of a piston flow curve.   In
reality, the  actual  radionuclide breakthrough curve usually represents  some
combination of the five types of curves discussed above.   Hence, column  break-
through data  can seldom be used to accurately determine Kd values.

SPECIFIC RETENTION
     Specific  retention, as defined by Meinzer (1933),. is the ratio between
the volume of  water that a soil or sediment will  retain against the pull of
gravity and its  packed volume from an initially saturated condition.  It is
                                     2-25

-------
usually given as a percentage and can be expressed as R = 100  (r/V), where R is
specific retention, r the volume of water retained against"the pull of gravity
and V the packed volume of the soil.or sediment.
     Disposal on a specific retention basis of selected liquid waste solutions
was practiced at Hanford from 1944 to about 1962.  As used at Hanford, specific
retention in practice represented the volume of liquid that could be discharged
to a disposal pit of known dimensions without leakage to the groundwater.  The
specific retention was expressed as a percent of the soil column volume as mea-
sured by the cross-sectional area of the disposal pit bottom and the height of
the soil column between water table and pit bottom.  Lateral spreading was
ignored to introduce a conservative element into the specific retention
calculation.
     A centrifugal technique for estimating soil specific retention properties
was reported by Bierschenk (1959).  Data were obtained indicating that the
natural moisture content of Hanford soils could be reproduced by draining a
1-cm thick soil sample from the same horizon at 1000 gravities for 1 hour.  The
average ratio of the natural field moisture content to. that of the centrifuged
samples ranged from 0.82 to 1.06.  The~speciTic""Tetention capacity of Hanford
soils varied from less than 1  to 2 vol%.   Usually the only reason for resorting
to specific retention was because normal  ground disposal methods were inadequate
(Haney, 1957).
     The problems associated with specific retention as a radionuclide waste
disposal technique include the fact that the forces holding the radionuclide
waste solution are relatively weak and that the radionuclides which are dis-
posed by this technique ofte.n remain in the retained solution.  Any water added
at the top of the soil column, from natural  precipitation or irrigation, would
lead to the migration of the retained waste solution into the groundwater.
Hence, specific retention as a waste disposal  technique would be limited to
arid areas where there is little downward migration of precipitation to the
groundwater and no chance of other water additions to the top of the soil
column.  In addition, large soil volumes are required, for specific retention
disposal because very little of the soil  column ion exchange capacity is
utilized.
                                    2-26

-------
THE DISTRIBUTION COEFFICIENT OR Kd
     One of the most useful concepts for dealing with radionuclide migration
and retention processes on geologic materials is that of the distribution coef-
ficient or Kd.  The application of the Kd concept to ion exchange reactions was
given above.  However, the distribution coefficient can be directly derived
from the same thermodynamic principles underlying ion exchange and other physi-
cochenvical reactions.  The more general derivation allows the application of
the Kd concept to a broad selection of radionuclide reactions with geologic
media.
     The distribution of a dilute solute (trace ion) between two immiscible
solvents (aqueous solution and solid medium), can be derived from y and y1,
the chemical potentials of the trace ion in- the solution and solid medium,
respectively.  At equilibrium, y = y1.  If both are ideal dilute solutions,
then y = y* + RT£nx and u1 = u1* + RHnx', and by definition, u* + RT2,nx =
y1* + RTfcnx1, where u* and y1* are standard free energies and x and x' are the
concentration of dilute solute in the solution and solid medium, respectively
(Denbigh, 1971).  Rearranging, RTJln ^-= - (u1* - u*).   Then ^- = Kd because  •
                                     A                        A
both y'* and y* are independent of composition.  The distribution coefficient,
(Kd), also is independent of the concentration of trace ion in the solution
   »
and on the soil.  The quantity y1* - y* is the standard free energy change,
AG*, for the reaction:

        Trace ion in solution -»• Trace ion on solid medium (soil, rock),
                              and RTAnK = -AG*,
which is the usual relationship for a chemical reaction.  If the solute (trace
ion) is very dilute, then the equilibrium mole fractions are proportional  to
the concentration, and Kd equals the concentration of trace ion on the solid
medium divided by concentration of trace ion in solution (times a correction
factor to yield a result in solution volume per weight solid medium).   The Kd
was originally proposed as a distribution law by Nernst in 1891 (Denbigh,  1971).
     A consideration of mechanisms is not involved in the derivation of the Kd
concept in the above generalized manner.  Removal onto the solid medium can be
due to a number of mechanisms.  However, it should be remembered that the trace
ion was considered to be in an ideal dilute solution both in the solution and

                                    2-27

-------
 on  the  soil.   The  trace  ion  could  conceivably  be  considered  in  an  ideal  dilute
 solution  if  it is  exchanged  onto the  soil  and  occupies  only  a very small  frac-
 tion  of the  soil ion  exchange  capacity.  The requirements'that  the Kd value
 remain  constant over  a concentration  range and that  the soil-trace ion behave
 ideally are  not compatible with precipitation  in  solution  as a  trace  ion  removal
 mechanism.   With precipitation as  a removal mechanism during Kd determinations,
 the soil  becomes an inert system constituent except  insomuch as it may affect
 the system pH  and  redox  environment.
      The  soil  can  be  considered a  passive  and  incidental part of the.removal
 process unless the trace ion can be shown  experimentally to  be  adsorbed on  the
 solid medium.   This can  be accomplished  by "blank" experiments  in  which the
 solution  containing the  tracer and no  soil  or  rock solids  is shaken along with
 the same  solution  containing the soil  or rock  solids.   If  there is evidence for
 trace radionuclide removal without the presence of soil  or rock solids in the
 system, the  Kd concept is not  applicable.                        •  ,
      Other practical  requirements  of a Kd  experiment include attainment of
.equilibrium  between the  solid  medium adsorbed  trace  ion-and  the trace ion con-
 tained  in the  solution.  Prior to  contact  of the  solid  medium and  trace ion,
 the solid medium should  be in  equilibrium  with the none-trace solution macroions.
 This  procedure should be followed  to ensure that  only the  trace ion is adsorbed,
 and that  only  in trace concentrations.   Trace  concentration  is  shown  experi-
 mentally  by  demonstrating that Kd  values remain constant over successive  trace
 ion dilutions.  This  procedure also demonstrates  that equilibrium  was attained
 using several  solution-solid contact times.  The  Kd  value  should refer to a
 single  ionic species, not a mixture of fission products.   The basic concept
 does  not  allow more than one ionic species  to  have a constant distribution
 between two  immiscible media.
                                                                             r
      The  Kd  is an  experimentally derived value that  is  associated  with the  cir-
 cumstances of  its  determination.   It cannot be extended to another chemical  sys-
 tem with  the same  solid  exchange medium.   Likewise,  the Kd from one exchange
 medium  in a  given  chemical system  cannot be used  for another solid exchange
 medium  in the  same chemical  system.
      An obstacle to Kd comparisons is  the  common  failure to adequately charac-
 terize  the solid exchange medium.  Even  with comparable solution compositions

                                    2-28

-------
and derived Kd values, the solid exchange materials can be completely different
mineralogically, physically and chemically.  A bulk chemical analysis, at least
a semiquantitative mineralogical and particle size analysis should be performed.
Very little of the literature Kd data include such characterizations of the
solid phase.  With rocks, the characterization work also is very important.
Adsorption of radionuclides on rock fissures often is higher than on fresh
surfaces because the fissure surfaces have been altered by earlier percolating
solutions to minerals such as zeolites or clays that are radionuclide selective
and do not occur on fresh rock surfaces.  Rancon (1972) suggested calculating
Kd based upon surface area rather than the mass of the solid in order to study
radionuclide adsorption by impermeable materials such as rocks.  This Kd is
defined as:

          i _       surface concentration              ml  solution
              residual concentration in solution   Cm2 Qf S(jrface area

Using a rock with a known surface area allows better comparisons than a known
weight because ion exchange is a surface area function for impermeable rocks.
However, once again, good characterization of the rock is required to allow
comparisons between rock types, or even within rock types.
     The radionuclides which can be expected to react with rocks and soils in
other ways than ion exchange include americium, cerium, curium, europium, pro-
methium, plutonium, tritium and zirconium, and sometimes antimony, uranium and
ruthenium.  Special care should be taken with the above radionuclides to assure
that the Kd values obtained are from trace concentrations and that the trace
ion is adsorbed on the solid medium.  Where the oxidation potential (Eh) is
important, such as in plutonium systems, means will have to be found for Kd
value determinations in controlled Eh-pH environments.
     Many techniques are used or under study for the determination of Kd values
including batch tests, low pressure column loading, axial- filtration and chan-
nel chromatography.  The last two techniques are in a development and proving
stage, and may or may not be comparable to Kd values determined by a batch
technique.  The standard Kd determination technique to which all other tech-
nique results are compared is the batch technique.  It is theoretically possible
to derive column Kd values comparable to batch Kd values.  However, difficulties
                                    2-29

-------
are usually encountered in practice because the soil or rock unique hydrologic
properties are superimposed on the radionuclide adsorption effects in the result-
ing breakthrough curve (Baetsle et al., 1966).  Consequently, the breakthrough
curve is not symmetrical  about the 50% breakthrough point and the Kd value as
taken at the 50% point is in error.  A batch Kd value is determined by agitat-
ing a weighed soil or rock sample with a known volume of chemically-characterized
solution until the radionuclide attains an equilibrium between solution and soil
or rock.  To obtain reproducible and comparable radionuclide distribution coef-
ficients, the following.experimental requirements should be followed:
 1.  mineralogically and  physically characterize representative rock or soil
     and chemically characterize the solution before use in Kd measurements,
 2.  all rocks should be  compared on a surface area rather than a weight basis,
 3.  dry a weighed sample of soil or rock at 105°C to obtain weight corrections
     for moisture content,
 4.  pre-equilibrate the  solids with nonradioactive waste constituents and
     discard solution,
 5.  add the radionuclide at several concentration levels and equilibrate versus
     time until  no significant changes are observed (precipitation of solid
     phases should be absent),
 6.  use triplicate samples for each equilibration,
 7.  a blank solution (containing no soil  or rock) should be run along with
     the samples at each  radionuclide concentration level,
 8.  the equilibrium solution composition should be determined including major
     cations, anions and  organics,
 9.  the pH should be measured before and after equilibrium, and
10.  a controlled redox potential is necessary where it has an effect on the
     radionuclide oxidation state to assure that only one species of radionu-
     clide is present in  the  equilibrating solution.
Kd Relationship to Migration
     The importance of an accurately determined Kd value cannot be overempha-
sized when it is used in  calculating the velocity of radionuclide movement in
geologic materials.  If K is equal to the groundwater velocity divided by the
                                              \(A rt
radionuclide migration velocity, then K = 1 + —*-, where:
                                    2-30

-------
  Kd = the distribution ccefficient =  (concentration of radionuclide on the
       solid/weight of solid)/(concentration in the solution/solution volume
       in the ml),
   p = bulk density of medium = weight of solid/volume of solid,, and
   e = void fraction = groundwater volume/volume of solid (Burkholder and
       Cloninger, 1977; Hiester and Vermeulen, 1953; Jackson et a!., in press).
       It may be seen that as the Kd value increases, the radionuclide migra-
       tion velocity must decrease a proportional amount, if the bulk density
       of the medium, the void fraction and groundwater velocity are assumed
       to be constant.
A doubling of the Kd value halves the radionuclide migration velocity in rela-
tion to the groundwater velocity.  Lester et al. (1975) with their computer
model of radionuclide migration have shown schematically the effect of Kd value
on radionuclide discharge rate at the outlet of a soil column for a three-
membered decay chain.  Impulse release was assumed and decay rates of all  three
radionuclides were neglected.  Differences in Kd'values for chain members sig-
nificantly reduced radionuclide discharge rates in comparison to the case
when the Kd values for all three radionuclides were the same, including zero.
Field Determination of Kd Values
     Using the above equation, a groundwater velocity in the field can be
determined between two wells, or between a disposal site and a river bank
spring with the use of tritium for a tracer.  Another radionuclide can be
added to the groundwater and the travel time over the same distance determined
for it.  If the bulk density and void fraction of the soil can be determined
as well, a field Kd can be derived.  Eliason (1967) has used the above method
to determine the travel times of groundwater and    I between a ground disposal
site and the Columbia River near Hanford N-Reactor.  The travel time for    I
was 1.28 times that of the waste water (79 days travel time for  H, 101 days
travel time for 131I).  An 131I Kd value of 0.06 ml/g was estimated by Hajek
(1968) from these data.  An advantage of this field method of Kd measurement
is that a resultant Kd value is obtained for soils and rocks which are seldom
homogeneous.  The laboratory Kd is useful for field applications only if the
field soil or rock are assumed to be homogeneous and represented by the
                                    2-31

-------
laboratory soil sample.  However, the laboratory Kd measurement is relatively
rapidly accomplished and is inexpensive.  The field Kd measurement has the dis-
advantage of being time-consuming and expensive.  Another disadvantage of the
field. Kd measurements is the difficulty of accurately determining a composite
void fraction and bulk density for the field soil, sediment or rock.
     An additional method of determining a Kd value for a waste disposal site
in the field involves a knowledge of the volume of waste disposal to the ground,
the radionuclide concentration per unit volume, the volume and bulk density of
the soil and the average radionuclide concentration per unit weight of soil.
Essentially this field method is a much larger version of a laboratory Kd
determination.  A major problem with this type of field Kd determination is
that the saturated flow directly beneath the disposal site soon grades into
unsaturated flow., with much lower volumes of waste transmitted through the
unsaturated flow area.  Usually an arbitrary decision is required to select a
wetted zone for the average soil radionuclide loading value.  An exception
would be a radionuclide that is adsorbed very close to the disposal site.  Thus,
the radionuclides with high Kd values present the least problems with this type
of field Kd determination.
     An example of the latter type of field Kd is given by Schmalz (1972) who
obtained auger samples down to 53 cm in TRA disposal pond bottom sediments at
INEL, Idaho.  These samples were analyzed for    Cs,   Sr,   Co and    Ce in
three intervals of 0 to 15.3 cm, 15.3 to 30.5 cm, and 30.5 to 45.9 cm.  The
analytical results, were used to estimate the quantities of each of the above
radionuclides in the three measured zones and these gradations extrapolated to
15.3 m to.obtain a total radionuclide concentration in that volume of sediments.
Knowing the volume of sediment and its estimated average bulk density, the radio-
nuclide contents of the sediments, and the volume and radionuclide content of
the waste disposed to the TRA pond, Kd values were estimated.   The laboratory
Kd value for    Cs was 285 ml/g, for example, while the field Kd value was
600 ml/g.  Considering the assumptions as to homogeneity of sediments underly-
                           137
ing the TRA pond and their    Cs content, perhaps the field and laboratory Kd
values are not as far apart as the numbers would seem to indicate.
     One of the problems associated with the second type of field Kd determina-
tion is that usually the influent to the. waste disposal site is not constant in

                                    2-32

-------
either composition or volume, and disposal records are often too generalized to

determine waste solution compositional and volumetric variations.  Redox poten-
tials and final pH values are seldom known and are difficult to measure in field
systems.  The experimental requirements for a reproducible laboratory Kd value
can seldom be followed for a field Kd determination.


REFERENCES

Ames, L. L., Jr.  1959.  The Genesis of Carbonate Apatites.  Economic Geology.
54:829-841.

Ames, L. L., Jr.  1960.  Some Cation Substitutions During the Formation of
Phosphorite from Calcite.  Economic Geology.  55:354-362.

Ames, L. L., Jr.  1961a.  Anion Metasomatic Replacement Reactions.   Economic
Geology.  56:521-532.

Ames, L. L., Jr.  1961b.  Cation Metasomatic Replacement Reactions.  Economic
Geology.  56:1017-1024.

Ames, L. L., Jr.  1963.  Kinetics of a Replacement Reaction.   Economic Geology.
58:1229-1236.

Ames, L. L.   1976.  Actinide Occurrences in Sediments Followina Ground Disposal
of Acid Wastes at 216-Z-9.  BNWL-2117, pp. 39-71.

Ames, L. L., Jr., J. R. McHenry and J. F.  Honstead.  1958.  The Removal  of
Strontium from Wastes by a Calcite-Phosphate Mechanism.  A/CONF. 15/P/395.

Amphlett, C. B.  1964.  Inorganic Ion Exchangers.  Elsevier Publishing Co.,
New York.

Andelman, J. B., and T. C. Rozzell.  .1970.  Plutonium in the Water Environment.
1.  Characteristics of Aqueous Plutonium,   p. 118-137.  IN:  Radionuclides in
the Environment.  Adv. in Chem. Ser. No. 93.  The Am. Chem. Soc., Washington,
D.C.

Baes, C. F., Jr. and R. F. Mesmer.  1976.   The Hydrolysis of Cations.  John
Wiley and Sons, New York.

Baetsle, L., W. F. Maes, J. Souffriau and P. I.  Staner.  1966.   Migration de
Radio-Elements Dans le Sol. EUR 2481.f.

Bartlett, J. B. and A. G. Norman.  1938.  Changes in the Lignin of Some Plant
Material as a Result of Decomposition.  Soil Sci. Soc. Amer.  Proc.   3:210-216.

Belot, Y. and C. Gailledreau.  1963.  Kinetics of Strontium-90 Retention by a
Calcite-Phosphate Extrapolation Assay.  _INj  Proc. of an Intern. Symp. on the
Retention and Migration of Radioactive Ions in Soils.  Saclay,  France.  Oct.
16-18, 1962.

                                    2-33

-------
Bensen, D. W.  1960.  Review of Soil Chemistry Research at Hanford.  HW-67201.

Bierschenk, W. H.  1959.  Techniques for Estimating the Specific Retention
Properties of Hanford Soils.  HW-61644 Rev.

Bray, R. H. and E. E. DeTurk.  1939.  The Release of Potassium from Nonreplace-
able Forms in Illinois Soils.  Soil Sci. Soc. Amer. Proc.  3:101-106.

Broadbent, F. E.  and G.  R. Bradford.  1952.  Cation-Exchange Groupings in the
Soil Organic Fractions.   Soil Science.  74:447-457.

Burkholder, H. C. and M. 0. Cloninger.  1977.  The Reconcentration Phenomenon
of Radionuclide Chain Migration.  BMWL-SA-5786.

Burney, G. A. and R. M.  Harbour.  1974.  Radiochemistry of Neptunium.  NAS-NRC
Nuclear Sci. Ser. NAS-NS-3060.

Caldwell, 0. G.  and C. E. Marshall.  1942.  A Study of Some Chemical and Physi-
cal Properties of the Clay Minerals Nontronite, Attapulgite and Saponite.  Mis-
souri Univ. Agr.  Expt. Sta. Res. Bull.  354.

Carroll, D.  1960.  Ion  Exchange in Clays and Other Minerals.  Geol. Soc.
Amer. Bull.  70:749-780.

Carroll, D.  1970.  Rock Weathering.  Plenum Press, NY.

Cassel, D. K., M. Th. Van Grenuchten and P. J. Wierenga.  1975.  Predicting
Anion Movement in Disturbed and Undisturbed Soils.  Soil Sci. Soc. Am.  Proc.
39(6):1015-1019.

Chatterjee, B. and M. Paul.  1942.  Interaction Between Hydrogen Clays and
Neutral Salts.  Indian Journ. Agr. Sci.  12:113-120.

Cleveland, J. M.   1970.   The Chemistry of Plutonium.  Gordon and Breach Science
Publishers, New York.

Dean, J. A., Editor.  1973.  Lange's Handbook of Chemistry.  McGraw-Hill, NY.
p. 6-30.

Deer, W. A., R.  A. Howie and J. Zussman.  1962.  Rock-Forming Minerals.
Volume 5, Non-Silicates.  Longmans, Green, London,  pp. 323-338.

de Haan, F. A. M.  1965.  The Interaction of Certain Inorganic Anions with
Clays and Soils.   Agr. Res. Reports No. 655, 167 pp.

de Haan, F. A. M.  1964.  The Negative Adsorption of Anions (Anion Exclusion
in Systems with Interacting Double Layers.  Journ. Phys. Chem.  68:2970.

de Haan, F. A. M. and G. H. Bolt.   1963.  Determination of Anion Adsorption by
Clays.   Soil Sci. Soc. Am. Proc.  27:178.

Denbigh, K. . 1971.  The  Principles of Chemical Equilibrium, 3rd Edition,  Cam-
bridge University Press, London, pp. 256-257.

                                    2-34

-------
.Dickman,  S.  R.  and  R.  H.  Bray.   1941.   Replacement  of  Adsorbed  Phosphate from
 Kaolinite by Fluoride.   Soil  Science.   52:263-275.

 Drew,  J.  V.  (ed).   1967.   Selected  Papers  in  Soil Formation  and Classification.
 Soil Science Society of  America,  Inc.   Madison,  Wisconsin.

 Dutt,  G.  R.,  M.  J.  Shaffer and  W. J. Moore.   1972.   Computer Simulation  Model
 of  Biophysic -  Chemical  Processes in Soils.   Univ.  of  Arizona,  Tucson.   Agric.
 Exp. Stn.  Tech.  Bull.  196.

 Eliason,  J.  R.   1967.  Field  Evaluation of Ground Disposal of Reactor  Effluent  -
 13Q1-N Crib.  BNWL-CC-1032.

 Frysinger, G. R.  1960.   Cation Exchange Behavior of Vermiculite-Biotite Mix-
 tures. .Clays and Clay Minerals.  9:116-121.

 Fukai, R.  and C. N. Murray.   1974.  Adsorption and  Desorption of Plutonium  and
 Americium in  Freshwater  - Sediment  and  Seawater  - Sediment Systems.  IN_:  Activi-
 ties of the  International  Laboratory of Marine Radioactivity.   IAEA-163,
 pp. 96-103.

 Gaines, G. L. and H. C.  Thomas.   1953.  Adsorption  Studies on Clay Minerals.
 II.  A Formulation  of  the Thermodynamics of Exchange Adsorption.  Journ.  Chem.
 Phys.  21:714-718.

 Garrels,  R.  M.  and  C.  L.  Christ.  1965.  Solutions,  Minerals  and Equilibria.
 Harper and Row.. New York. "pp.  213-216.

 Gerasimov, I. .P. and M.  A.  Glazovskaya.  1960.   Fundamentals  of Soil Science
 and Soil  Geography.  Israel Program for Scientific  Translations,  Jerusalem.

 Grebenshchikova, V. I. and Y. P. Davydov.  1961.  A Study of the State of
 Pii(IV) in  Dilute Solutions of Nitric Acid.  Radiokhimiya.  3:155-164.

 Hajek, B.  F.  1965.  Some Adsorption, Diffusion, and Dispersion  Characteristics
 of Anions  in  Soils.  BNWL-36, pp. 4.1-4.4.

 Hajek, B.  F.  1966.  Plutonium  and  Americium  Mobility  in Soils.   BNWL-CC-925.

 Hajek, B.  F.  1968.  Waste Disposal to  the Ground at 100-N.   BNWL-CC-1670.

 Hakonson,  T.  E., J. W. Nyhan and W. D.  Purtymum.  1975.  Accumulation and Trans-
 port of Soil  Plutonium in Liquid Waste  Discharae Areas at Los Alamos.  IAEA-SM-
 199/99.

 Haney, W.  A.  1957.  Disposal of High Cobalt-60  Scavenged Wastes.  HW-48862.

 Harmon, C. G. and F. Fraulini.   1940.   Properties of Kaclinite  as a Function of
 Its Particle Size.  Jour.  Amer. Ceram.  Soc.   23:252-258.

 Harned, H. S. and B. B.  Owen.   1958.  The Physical  Chemistry  of Electrolytic
 Solutions.   Reinhold Publishing Corp.,  NY.

                                    2-35

-------
 Hiester, N.  K. and T. Vermeulen.   1953.  Saturation Performance of  Ion-Exchange
 and Adsorption Columns.  Chem.  Eng. Prog.  48:505-516.

 Hillel, D.  1971.  Soil and Water.  Physical Principles and Processes. Academic
 Press, New York.

 Hofmann, U., A. Weiss, G. Koch, A. Mehler and A. Scholz.  1956.   Intracrystal-
 line Swelling, Cation Exchange  and Anion Exchange of Minerals of  the Montmoril-
 lonite Group and of Kaolinite.  Nat!. Acad. Sci. Pub!.  456.  pp. 273-287.

 Jackson, R.  E., W. F. Merritt,  D.  R. Champs, J. Gulens and K. J.  Inch.   In
 Press.  The  Distribution Coefficient as a Geochemical Measure of  the Mobility
 of Contaminants in a Ground Water  Flow System.  Paper given at a  Conference on
 the Use of Nuclear Techniques in Water Pollution Studies, 6-9 December  1976,
 Cracow, Poland, under IAEA Sponsorship.

 Jost, W.  1960.  Diffusion in Solids, Liquids and Gases.  3rd Edition.  Academic
 Press, New York, NY.

 Jurinak, J.  J. and J. Santillan - Medrano.  1974.  The Chemistry  and Transport
 of Lead and  Cadmium in Soils.   Utah State Univ., Logan Agric. Exp.  Stn.  Res.
 Rep. 18.

 Keller, C.    1971.  The Chemistry of the Transuranium Elements.  Vol. 3.  Kern-
 chemie in Einzeldarstellungen.  Verlag Chemie/GmbH.

 Kelley, W.  P. and H. Jenny.. 1936.  The Relation of Crystal Structure to Base-
 Exchange and Its Bearing on Base-Exchange in Soils.  Soil Science.  41:367-382.

 Knoll, K. C.  1960.  Adsorption of Strontium by Soils Under Saturated and
 Unsaturated  Flow Conditions.  HW-67830.

 Latimer, W. M.  1952.  Oxidation Potentials.  Second Edition.  Prentice-Hall,
 Englewood Cliffs, NJ.

 Lester, D.  H., G. Jansen and H. C. Burkholder.  1975.  Migration  of Radionu-
 clide Chains Through an Adsorbing Medium.  IN:  Adsorption and Ion  Exchange.
 AIChE symposium Series.  71:202-213.

 Lindenbaum, A. and W. Westfall.  1965.  Colloidal Properties of Plutonium.
 Int. J. Applied Radiat. Isotop.  16:545.

 Lindsay, W.  L.  1972.  Inorganic Phase Equilibria of Micronutrients in  Soils.
 IN:  Micronutrients in Agriculture, pp. 41-57.  Soil Sci. Soc. Amer. Inc.
 Madison, WI.

 Mattson, S.  1931.  The Laws of Soil Colloidal Behavior.   IV.  Amphoteric
 Behavior.  Soil Science.  32:343-365.

McAuliffe, C.  D.,  M. S. Hall, L. A. Dean,  and S. B.  Hendricks.  1947.  Exchange
 Reactions Between Phosphates and Soils.  Soil  Sci.  Soc.  Amer. Proc.   12:119-123.


                                    2-36

-------
Meinzer, 0. E.  1933.  Outline of Ground-Water Hydrology with Definitions.
U.S.G.S. Water Supply Paper 494.

Mukherjee, J. N'. , B. Chatterjae, and P. C. Goswami.  1942.  Limiting Exchange
of Aluminum Ions from Hydrogen Clays on the Addition of Neutral Salts.  Jour.
Indian Chem. Soc.  19:400-407.

Nelson, J. L., W. A. Haney, K. C. Knoll and D. W.  Bensen.  1962.  Laboratory
Pilot and field Scale Studies of Fission Product Migration Rates in Soil.
HW-SA-2744.

Nielsen, D. R. and J. W. Biggar.  1961.  Miscible  Displacement in Soils:  I.
Experimental Information.  Soil Sci. Soc. Amer. Proc.  25:1-5.

Nielsen, D. R. and J. W. Biggar.  1962a.  Miscible Displacement:  II.  Behavior
of Tracers.  Soil Sci. Soc. Amer. Proc.  26:125-128.

Nielsen, D. R. and J. W. Biggar.  1962b.  Miscible Displacement:  III.   Theoreti-
cal Considerations.  Soil Sci. Soc. Amer. Proc.  26:216-221.

Nishita, H., B. W. Kowalesky, A. J. Steen, .and K.  H. Larson.  1956.   Fixation
and Extractability of Fission Products Contaminating Various Soils and Clavs.
I.  Sr90, Y91, Ru105, CsT37 and CeT44.  Soil Science.  81:317-326.

Parker, V. B., D. D. Wagman, and W. H. Evans.  1971.  Selected Values of Chemi-
cal Thermodynamic Properties.  Tables for the Alkaline Earth Elements (Elements
92 through 97) in the Standard Order of Arrangement.  NBS Technical  Note 270-6.

Posner, A. M.-  1966.  The Humic Acid Extracted by  Various Reagents from a Sail.
Part 1.  Yield, Inorganic Components and Titration Curves.  Jour.  Soil  Sci.
17:65-78.

Pourbaix, M.  1966.  Atlas of Electrochemical Equilibria in Aqueous  Solutions.
Pergamon Press, Oxford, England.

Price, S. A. and L. L. Ames.  1975.  Characterization of Actinide-Bearing Sedi-
ments Underlying Liquid Waste Disposal Facilities at Hanford.   ARH-SA-232.

Rai, D. and W. T. Franklin.  1973.  Program for Computing Equilibrium Composi-
tion of CaC03 and CaS04 Systems from Irrigation Water Compositions.   Council of
U.S. Univ. for Soil and Water Develop, in Arid and Sub-Humid Areas.   Water
Manage. Tech. Rep. 29.  Colo. State Univ., Fort Collins.

Rai, D.'and W. L. Lindsay.  1975.  A Thermodynamic Model for Predicting the
Formation, Stability and Weathering of Common Soil Minerals.  Soil  Sci. Soc.
Amer. Proc.  39:991-996.

Rai, D. and R. J. Serne.  1977.  Plutonium Activities in Soil  Solutions and the
Stability and Formation of Selected Plutonium Minerals.   J. Environ. Qua!.,
6:89-95.

Rai, D. and R. J. Serne.  1978.  Predicted Solid Phases and Solution Species
of Elements Generally Present in Radioactive Wastes.  DOE Sponsored Document
for the National Waste Terminal Storage Program, in publication.

                                     2-37

-------
Rancon, D.  1969.  Contamination des Sols.  Evolution de L'humidite Dans les
Sols Non Satures Apres Rejet Sur Une Surface Limitee.  CEA-R-36350).

Rancon, D.  1972.  Practical Utilization of the Distribution Coefficient for
the Measurement of the Radioactive Contamination of Minerals in Rocks, Soil
and Subterranean Water.  ANL-TRANS-931.

Rancon, D.  1973.  Contamination des Sols.  Comportement dss Ions Radioactifs
Dans les Sols Non Satures Apres Rejets Dreau Sur Une Surface Limitee.  CEA-R-
3635(2).

Rancon, D.  1973.  The Behavior in Underground Environments of Uranium and
Thorium Discharged by the Nuclear Industry.  IN:  Environmental Behavior of
Radionuclides Released in the Nuclear Industry.  IAEA-SM-172/55, pp.  323-346
(in French).

Relyea, J. F.  1977.  The Behavior of Plutonium in Soil.  Ph.D. Thesis.  Uni-
versity of Arkansas, Fayetteville, AR.

Rhodes, D. W.  1957a.  The Adsorption of Plutonium by Soil.  Soil Science.
84:465-471.

Rhodes, D. W.  1957b.  The Effect of pH on the Uptake of Radioactive Isotopes
from Solution by a Soil.   Soil  Sci.  Soc. Amer.  Proc!  21:389-392.

Rossini, F. D., D. D. Wagman, W. H.  Evans, S.  Levine and I. Jaffe.  1952.
Selected Values of Chemical Thermpdynamic Properties.  NBS Circular 500,
Washington, D.C.

Routson, R. C.  1973.  A Review of Studies on  Soil-Waste Relationships on the
Hanford Reservation from 1944 to 1967.  BNWL-1464.

Routson, R. C. and R. J.  Serne.  1972.  Experimental Support Studies  for the
PERCOL and Transport Models.  BNWL-1719.

Samuelsori, 0.  1963.  Ion Exchange Separations  in Analytical Chemistry.  John
Wiley and Sons, NY.  pp.  97-112.

Scheffler, K., U. Riege,  K. Louwrier, Hj. Matzke, I. Ray and H. Thiele.  1977.
Long" Term Leaching of Silicate  Systems:   Testing Procedure, Actinides Behavior
and Mechanism.  KFK 2456.  EUR  5509e.

Schmalz, B: L.  1972.  Radionuclide Distribution in Soil  Mantle of the Litho-
sphere as a Consequence of Waste Disposal at the National Reactor Testing
Station.  IDO-10049.

Schofield, R. K.   1947.  Calculation of Surface Areas from Measurement of
Negative Adsorption.  Nature 160:408.

Schnitzer, M. and S. I. M.  Skinner.   1965.  Organo-Metallic Interactions in
Soils:  4. Carboxyl and Hydroxyl Groups  in Organic Matter and Metal  Retention.
Soil Science.  99:273-284.
                                     2-38

-------
Schwille, F., W. Lippok and D. Weisflog.  1967.  Model Experiments on Fluid  Flow
in the Transition Zone from Unsaturated to.Saturated Soil.  IAEA-SM-93/11,
pp. 151-159.

Scott, H. D., R. E. Phillips and R. F. Paetzold.  1974.  Diffusion of Herbicides
in the Adsorbed Phase.  Soil. Sci. Soc. Am. Proc.  38:558-562.

Schumrn, R. H., D. D. Uagman, S. M. Bailey, W. H. Evans, and V. B. Parker.  1973.
Selected Values of Chemical Thermodynamic Properties.  Tables for the Lanthanide
(Rare Earth) Elements (Elements 62 through 76) in the Standard Order of Arrange-
ment.   NBS Technical Note 270-7.

Schwille, F., W. Lippok and .0. Weisflog.  1967.  Model Experiments on Fluid  Flow-
in the Transition Zone from Unsaturated to Saturated Soil.  IAEA-SM-93/11,
pp. 151-159.

Sheidina, L. D., and E.  N. Kovarskaya.  1970.  Colloidal State of Pu(IV)  in
Aqueous Solutions.   Sov. Radio Chem.  12:229-233.

Sillen, L. G. and A. E.  Martell.  1964.  Stability Constants of Metal-Ion Com-
plexes.  Special Publication No. 17.  The Chemical Society, London.

Tamura, T. and D. G. Jacobs.  1960.  Structural Implications in Casium Sorption.
Health Physics.  2:391-398.

Tamura, T. and D. G. Jacobs.  1961.  Improving Cesium Selectivity of Bentonite
by Heat Treatment.   Health Physics.  5:149-154.

Tamura, T.  1962.  Strontium Reactions with Minerals.'  TID-7628, pp. 187-197.

Tanji, K. K. and L. 0. Doneen.  1966.  A Computer Technique for Prediction of
CaC03 Precipitation in HC03 Salt Solution.  Soil Sci. Soc. Amer. Proc.
30:53-55.

Thomas, G. W. and A. R.  Swoboda.  1970.  Anion Exclusion Effects on Chloride
Movement in Soils.   Soil Sci.  110(3):163-166.

Van Olphen, H.  1963.  An Introduction to Clay Colloid Chemistry.  Interscience
Publishers, New York.

Wagman, D. 0., W. H. Evans, V. B.  Parker, I. Halow, S. M. Bailey, and
R. H.  Schumm.  1968.  Selected Values of Chemical Thermodynamic Properties.
Tables for the First Thirty-Four Elements in the Standard Ord.er of Arrangement.
NBS Technical Note 270-3.

Wagman, D. D., W. H. Evans, V. B.  Parker, I. Halow, S. M. Bailey, and
R. H.  Schumm.  1969.  Selected Values of Thermodynamic Properties.  Tables for
Elements 35 Through 53 in the Standard Order of Arrangement.  NBS Technical
Note 270-4.
                                    2-39

-------
Wagman, D. D., W. H'. Evans, V. B. Parker, I. Halow, S. M. Bailey, R. H. Schumm,
and K. L. Churney.  1971.  Selected Values of Chemical Thermodynamic Properties.
Tables for Elements 54 Through 61 in the Standard Order of Arrangement.  NBS
Technical Note 270-5.

Winklander, L.  1964.  Cation and Anion Exchange Phenomena.  IN:  Chemistry
of Soils, Chap. IV, ACS Monograph,  pp. 163-205.

Winograd, I. J.  1974.  Radioactive Waste Storage in the Arid Zone.  EOS, Trans.
Am. Geophys.  Union.  55:884-394.

Wood, L.  K. and E. E. DeTurk.  1941.  The Chemical Effect of Soluble Potassium
Salts in Some Illinois Soils..  Soil Sci. Soc. Amer. Proc.  5:152-161.

Zadmard,  H.  1939.  Zur Kenntnis der Kolloidchemischen Eigenschaften des Humus.
Kolloid-Beihefte.  49:315-364.
                                    2-40

-------
                                 SECTION 3
                    ELEMENT CHEMISTRY AND GEOCHEMISTRY

     The chemistry and geochemistry of the 19 elements chosen for review are
examined in the following section from the standpoint of radionuclide
migration and retention by geologic materials.  Some of the important radio-
isotopes normally found as fission products or neutron activation products
are enumerated.  Adequate chemical data were available only for the elements
with naturally-occurring stable isotopes.  The radiochemistry of all of the
isotopes of an element is nearly identical in respect to their reactions
with geologic materials, with possible minor exceptions in  H and  H.  A
summary of what is presently known (1977) relative to the radionuclide-
geologic media interactions for the selected elements is given.  It will be
noted that the data presented for cesium and strontium adsorption on soils
and rocks are inordinately large.  This circumstance is due more to historical
accident than to an unusually active experimental interest in cesium and
strontium adsorption.  In the past, many of the low to intermediate level
wastes were disposed to ground.  Two of the most hazardous radionuclides
contained in these wastes were cesium and strontium, and hence the early
interest in examining their migration and retention in soils.
AMERICIUM
Natural Soil and Rock Distributions
     The element americium has not been reported to occur naturally in soils
or rocks.

Brief Chemistry
     Americium has not been reported in nature bat is formed in high alpha
                                       238         241   8- 241
densities from uranium by the reaction    U(a,n) -»•    Pu *     Am, or from
                                     3-1

-------
high burnup piutorn urn having a high 241Pu  content by the reaction  239Pu
(n,Y)  - 240Pu(n,Y) -241Pu  5- 241Am (Keller,  1971).

     The americium radionuclide data of importance to radioactive wastes are
given in Table 3-1.  It is found in the following four oxidation states:
Am(III), Am(IV), Am(V) and Am(VI).  The most stable state of americium is the
trivalent.  The higher oxidation states are strong oxidizing agents and are
therefore stable only in media that contain no oxidizable substances (Coleman,
1963).  The ionic radius of Am+3 is 1.07 A (Ahrens, 1952).

                 TABLE 3-1.   AMERICIUM RADIONUCLIDE DATA
                             (KELLER, 1971)
Isotope
241Am
242mAm
242Am -
243Am
Ha If -Life
458
152
16
7370
years
years
hours
years
Decay Moi
a
IT, a
0-, EC
a
     The stability constants and-complex formation of Am(III) are probably
more well known and studied than any of the other trivalent actinides.  The
tendency to complex ion formation is a function of factors such as ionic size
and charge.  In order of decreasing complexing tendency, the actinide series
is generally M  >M02  >M  >M02  .  For univalent anions, complexing ability
with Am(III) follows the order  for F~>N03~>C1~>C104~, and for bivalent anions,
CO,~2>oxalate~2>SO,,~2 (Keller,  1971).
  0                *
Solid Phase and Solution Equilibria
     Information regarding the  possible solid phases of americium in soil and
rock environments  is lacking.   However, the relative solubilities of several
americium solids based upon estimated thermodynamic data (Latimer, 1952;
Keller, 1971) are given in Figure 3-1 as a function of pH.  These are oxides
and hydrous oxides of americium contained in an oxidizing environment (pO« -
0.68 atm).  The solubility of the solids decreases rapidly in the direction
of increasing pH.  As the environment becomes less oxidizing, the activity
                                    3-2

-------
of Am   associated  with  each  of the solid phases [except Am(OH)3] increases
rapidly.  At  PC^g)  =  3-58  atm' for example,  the curve for Am   activity
from Am(OH)4(s)  increases by  two log units.
                -4
             e
             <
             81
               -10
               -12
               -14
                                                       """AnHOHL
                             AmO-lOH..
V-. (pO, 8.68)
                        4567
                    9      10
          Figure 3-1.  The relative stability  of  various  americium
                       solids  in an oxidizing  soil  environment
                       [p02(gj = 0.68 atm]
     The activity of various  americium  species  in  solution in equilibrium with
AmCL(s) in an oxidizing environment  (pO«  =  0.68 atm)  is  given in Figure 3-2.
  •                                                     2+            ?+
The thermodynamic data for all the species  except  AmOH   and  AmHgPO^"   were
selected from Si 11 en and Martell  (1964).  The AmOH2+  and AmHgPO^2*  data were
selected from Keller (1971).   In  the normal  soil pH range,  the complex
              2+          +
species Am(OH)   and AmSO,,  would control the total activity  of americium,
     ' 3+
with Am   of somewhat lesser  importance.  In the environmental  conditions
chosen, chloride, nitrate and  phosphorus  species were relatively unimportant
in terms of the total americium present in  solution.   However it should be
                                                                     2+
kept in mind that in a relatively high chloride environment,  the AmCl   , for
example, could become the principal americium species in solution.   With an
                                    3-3

-------
increase in .pCL (more reducing environment) and the concentration of anions:
the curves given in Figure 3-2 would move upwards.  However, the relative
position of most of the isotherms would not change.  The americium species
all of cationic in the normal soil pH range of from 4 to 8.  It is to be
expected, therefore, that ion exchange could be an important americium
removal mechanism.
             £  -12
Figure 3-2.   The activity of various americium species in equilibrium
             with Am02(s) in an oxidizing soil environment Cp02(g) -
                             - =     "        =                 =
             0.68 atm
Experimental Adsorption Results
pCT = 3.0 and
                                                                   = 5.0.
     The adsorption of americium by soils has been studied by several investv
gators.  Hajek and Knoll (1966) and Knoll (1965, 1969) described several
americium adsorption experiments on Hanford wastes with emphasis on the
effect of organics.  An acidic, high salt (5.4M N03) waste spiked with
organics typically found in waste streams (lard oil, carbon tetrachloride,
DBBP and  TBP) that contained americium showed very rapid breakthrough when
                                  >,
percolated through a Hanford sand.^i The americium Kd was less than 1 ml/g at
a soil-solution pH as high as 3.  Plutonium in these solutions had a Kd of
                                    3-4

-------
2 to 3 ml/g.  If the wastes were neutralized and the resultant supernatant was
mixed with 20% by volume organics, the Kd increased to 41 ml/g.  A water leach
of the neutralized sludge mobilized some americium which, upon contact with
soil, yielded an americium Kd of 500 ml/g.  In all cases, the americium Kd was
less than the plutonium Kd.  In other experiments Knoll loaded columns of soil
with tap water spiked with americium.  Various organic solutions were then per-
colated through the columns and the quantity of americium leached recorded.
One hundred and thirty column volumes'of TBP(20%)-CC14 (80%) leached 10% of
the adsorbed americium versus 5% of the plutonium.  DBBP (30%)-CC14(70%) after
160 column volumes .lea.ched 15% of the americium and 40% of the plutonium.
D2EHPA-TBP in normal paraffin hydrocarbons after 30 column volumes leached 80%
of the americium and 30% of the plutonium.  Hydroxyacetic acid at 0.25M released
100% of the americium and 55% of the plutonium after 10 column volumes.   At a
lower concentration, 0.025M, hydroxyacetic acid removed 50% of the americium
and 40% of the plutonium after 70 column volumes.  When the americium and plu-
tonium were added to the. organic solution and then percolated through soil
columns, the Kd values listed in Table 3-2 were observed, assuming a column bulk
density of 1.5 g/ml.  There were significant differences between the two cases;
americium contacting soil  then being leached with organic solutions and americium
added directly to organic solutions and contacting soils.
         TABLE 3-2.  AMERICIUM Kd VALUES FROM SEVERAL ORGANIC SOLUTIONS
                     TBP = TRIBUTYL PHOSPHATE, DBBP = DIBUTYLBUTYLPHOS-
                     PHORATE, D2EHPA = Di-(2ETHYLHEXYL) PHOSPHORIC ACID
                     (KNOLL,  1965)

                                                 Americium Kd,
                       Organic Solution              ml/q
                  TBP(20%)-CC14(80%)                 1.6
                  DBBP(30%)-CC14(70%)                0.6
                  D2EHPA-TBP in Hydrocarbons         0
     Routson et al. (1975) used batch adsorption experiments (10 g soil to
25 ml solution) and 24-hour equilibrations to determine the Kd for americium with
desert sand and an Eastern sandy clay.  The Kd was determined as a function
of solution calcium and sodium concentration at an initial  pH of 2.5 to 3.1,
                                     3-5

-------
and for the sandy clay, decreased as the calcium concentration  increased.
At 0.002M Ca the americium Kd was 67 ml/g; at 0.20M Ca the Kd was 1 ml/c
For the Burbank sand the Kd was greater than 1200 for all concentrations of
calcium.  The americium Kd for the sandy clay also was a function of the
sodium concentration in solution; 280 at 0.015M Na and 1.6 at 3.00M Na.
Although a final pH was not given, the Burbank sand is known to be alkaline,
suggesting that americium adsorption may be sensitive to pH.
     Van Dalen et al. (1975) determined the americium Kd for Dutch subsoils  '
from 90% saturated NaCl solutions at pH 7 to 8.  Samples consisting of mainly
                                       4
illite and kaolinite had a Kd of J5_x 10 ; for a river sand the Kd was 400.
Gypsum bearing and clay bearing sandstones had intermediate Kd values.  The
pH dependence on sorption processes for americium between pH 5-8 was minor.
     Fried et al.  (1974a, 1974b) determined the adsorption of americium onto
a basalt.  Addition of salts to the solution lowered the adsorption of
              +3       +4
americium.  La   and Zr   cations lowered the americium adsorption much more
       +?       +
than Sr ~ and Na  suggesting ion exchange as the mechanism controlling
adsorption of americium.     '
     In Fried et al. (1977) the distribution of americium at a 32-year-old
disposal site in tuff, a very porous, soft volcanic rock, was investigated
by coring the tuff at the site to 14 ft.  Counting of wafers at different
depths down the tuff core showed the americium to be concentrated mainly at
a peak at about 9 ft from the disposal site bottom.   A core of the same tuff,
but uncontaminated, was used to duplicate the field results without much suc-
cess.  A near-surface americium peak was obtained rather than the peak at
9 ft found in the field.
     Fried (in press) also  studied the movement of    Am through Los  Alamos
tuff by using water to elute americium already adsorbed on the tuff.   The rela-
tive migration ratio for americium was 100 urn/meter  of water flow.-  Upper limit
for the relative migration ratio was 500 urn/meter of water flow.  Over 95%
of the americium was found very close to the point of deposition.  Surface
adsorption coefficients were derived for americium on National Reactor Testing
                                                         2
Station (now INEL) basalt in which the surface area in cm  was used rather
than the basalt weight in the Kd calculation.   In these experiments, rock
discs of known surface area were immersed in a solution of 1 x 10" M

                                    3-6

-------
Small aliquots of the solution were removed and counted at  12-hr  intervals
until a constant count was reached.  The Kd value for americium was  24_._4_±
12 ml/cm  for INEL basalt.  The values for Kd decreased rapidly as Na+,  Ca   ,
Sr   , La   and Zr 4 ions were added to the americium equilibrating solution
in increasing quantities.
     Sheppard et al. (1976) reported distribution between 12 characterized
                                                       241
soils from various locations and a solution containing    Am(III).   The  loca-
tions included Muscatine, Illinois; Rich!and, Washington; Sarnwell,  South
Carolina.; and Idaho Falls, Idaho.  Equilibrations were for  up to 4 months
with sampling of the solution at intervals.  The distribution results are
given in Table 3-3.  The distribution values ranged from 45,500 ml/g, for an
Idaho soil to 125 ml/g for a Washington soil.  The distribution values were
not constant with time, but steadily increased 1% to 86%/month.  With Burbank
soil, there was a 13% decrease/month.  No change was detected for another
Washington soil (Hanford).  It was postulated that such adsorption behavior
was consistent with the Am(III) being in the form of charged colloids or
hydrolysis products including Am  , Am(OH)+", Am(OH)2+ and Amn(CH)op and
removal' by cation exchange and adsorption.   It was reported that the removal
of soil organics and raising the system temperature did not appreciably affect
the rate of approach to equilibrium or the distribution values.  Analysis of
the soil cation exchange capacity data indicated a high correlation  (r =
0.96) between soil cation exchange capacity and clay content.  Correlation
does not appear as marked between soil clay content and americium distribution.
     Glover et al. (1977) also studied americium sorption by soils.   Seventeen
characterized soils from around the United States were equilibrated for 48 hr
                                            -10           -8
with distilled water solutions containing 10"  M Am and 10" M Am.   Eight
replicates were run per soiVat each americium concentration.  Results were
expressed as Kd values.  Kd values ranged from 82 to 10,000 on the different
                                        1Q     	1?	
soils.   Generally, the Kd values for 10   M and 10  M Am were the same, or
nearly so, with some exceptions.  Regression analyses were performed on the
americium sorption data to determine whether or not significant relationships
existed between americium sorption and soil physical or chemical  characteris-
tics.  Initial results indicated that a direct relationship existed between
soil cation exchange capacity, clay content and americium Kd value.
                                    3-7

-------
         TABLE 3-3.   AMERICIUM  50-OAY DISTRIBUTION.   (COMPLETED FROM
                      THE DATA OF  SHEPPARD ET AL.,  1976)

                              Am  Distribution,         Monthly Change
             Soil  Identity       	ml XT	       in Distribution Ratio
             Muscatine               4,830
             silt loam
             Burbank       •            714
             loamy sand
             Ritzville                 971                  +1"
             silt loam
             Fuquay sand,               476                  +32:;
             0-5 cm depth
             Fuquay sand,               417                  4.32?;
             5-15 cm depth
             Fuquay .sand,          '249           ' •      +27"
             15-50 cm depth
             Hanford A                 125                   0"

             Hanford B                 833                  4-34:.

             Idaho A                 3,920                  +30?:.

             Idaho B                43,500                  +36r,

             Idaho C                37,000                  +81S

             Idaho 0                10,900                  +53?;
     Polzer and Miner  (1977) relate americium adsorption data  on the same
soils  to  differences  in the effective positive  charge of americium (Korotkin,
1972)  as  influenced by system pH.   Kd values for  americium with  several soils
from around the United States are given in basic  and acidic systems.
Migration Results
Field  Studies-
     Fowler and Essington (1974) detected a possible difference  in solubility
between americium and  plutontum fallout in soils.   Americium may be more solu-
ble than  plutonium and may become the radionuclide of prime concern because
                                       3-8

-------
it has a faster migration rate in soils.  Several instances of decreasing  .
Pu/Am ratios with depth in the soil profile could be explained by differential
Plutonium and americium solubilities.
     Laboratory Studies--
     Hajek (1966) used contaminated soil samples from the Hanford 216-Z-9
covered trench upper 60 cm to determine the leach rate of americium.  Small
columns of 5 and 7.5 g of soil were leached with groundwater (composition
                                                           o
not given) and IN NaN03 at leach rates of less than 4 ml/cm /day.  Effluents
were collected at intervals and plotted as percent americium eluted versus
column volumes of leaching solution throughput.  About 7.5% of the americium
was leached with 8 column volumes of groundwater, after which the leach
rate became zero.   In the other column, about 6.5% of the americium was
eluted with 20 column volumes of.groundwater, and americium was still being
slowly leached.  With the IN NaNO-, leaching, the americium leaching curve
level off at 33% americium removed from the column.   The author considered
this to indicate that at least part of the americium was exchangeable.
     Cline (1968) demonstrated that soil pH was directly related to amenciurn
migration.  Americium nitrate was applied to two soil columns and 254 cm of
irrigation apqlied as an influent.  In the acid soil (pH 4.5) the americium
was retained in the top 5 cm with 98% in the top 1  cm.   Americium penetrated
to 20 cm in the basic soil and only 76% remained in the top one centimeter
of the column.  A larger fraction of the americium was  Am   at 4.5 than at
7.5.
     Knoll (1969) studied the problem of disposal of organic wastes to the
ground and the affect of this disposal on the adsorption or stripping of
radiorrucTides from the soil  column.   The organic wastes studied included
dibutyl butylphosphonate (D88P), di-(2-ethylexyl) phosphoric acid (D2EHPA), a
mixture of Lard Oil  and CC1, (Fab Oil), methylisobutylkatone (Hexone), a com-
mercial  mixture of straight'chain hydrocarbons including decane, undecane,
dodecane, tridecane and tetradecane (NPS) and tributyl  phosphate (TBP).   The
effect of organics on soil ion exchange properties  was  studied for an alka-
line, Burbank sandy loam soil type by saturating the soil  in a 10 cm column
with each of the above organics and loading with    Cs  traced water.  The
                                    3-9

-------
cation exchange properties of the soil were not affected by the prior con-
tact with organics, at least in terms of the    Cs capacity.  In other columns,
the americium was loaded on the column and then contacted with the organic
waste.  The americium in the effluent was monitored.  The results of leaching
with organic wastes is shown in Table 3-4.  The soil adsorption of americium
carried by organics was investigated by passing the americiurn-contaminated
organics through the same soil.  The 50% column breakthrough of americium
occurred at 9 column volumes of TBP-CCl^ (20 to 80%), at 3..5 column volumes
of DBBP-CCl^ (20 to 80%) and the soil .adsorbed none of the americium contained
in 0.4M D2EHPA-0.2M TBP in NPH.  .

            TABLE 3-4.  LEACHING OF AMERICIUM ADSORBED Of! THE
                        SOIL BY ORGANIC WASTES (KNOLL, 1969)

                                             Americium       Column
       	Organic Haste	      Removal, %      Volumes
       D8BP-CC14 (30 to 70%)                    15             80
       Fab Oil                                    1             30
       0.4M D2EHPA-0.2M TBP in NPH              80             50
       0.25M Hydroxyacetic Acid, pH 3.7        100              5
       0.125M Hydroxyacetic Acid, pH 3.8        55             10
       0.025M Hydroxyacetic Acid, pH 3.8        50 ,           70

Summary
     Americium adsorption (Kd values) on various soils and rocks have been
correlated with 1) cation exchange capacity of soils (Sheppard,  Kittrick  and
Hart, 1976; Glover and Miner, 1977), and 2) the concentration of the competing
cations (Na*, Ca  , Sr  , La   and Zr  ) so that the adsorption  of americium
decreases with an increase in concentration of competing cations (Fried,
Friedman and Quarter-man, 1974; Fried, Friedman and Weeber, 1974; Routson,
Jansen and Robinson, 1975; Sheppard, Kittrick  and Hart, 1976; Fried, 1977).
Although none of the workers have tried to distinguish the contributions of
polymer adsorption, cation exchange, hydrolysis and precipitation to the Kd
value, the laboratory data indicate that americium removal from  solution may
be predominantly via an ion exchange mechanism.  This conclusion is supported

                                    3-10

-------
by the predictions based upon thermodynamic data.  The americium species

are all cationic in the normal soil pH range from pH 4 to pH 8  (Figure 3r2).

References

Ahrens, L. H.  1952.  The Use of lonization Potentials., Part I.  Ionic Radii
of the Elements.  Geochim. et Cosniochim. Acta.  2:155.
Cline, J. F.  1968.  Uptake of ~  Am and L^?u by Plants.  BNWL-714,
pp. 8.24-8.25..

Coleman, J. C.  1963.  The Kinetics of the Disproportionate of Americium (V).
Inorg. Chem. 2:53.

Fowler, E. B. and E. H. Essington.  1974.  Soils Element Activities October.
1972 - September, 1973.  IN:  The Dynamics of Plutonium in Desert Environ-
ments.  P. B. Dunnaway and M. G. White (eds). NVO-142, pp. 7-16.

Fried, S. (in press) The Retention of Plutonium and Americium by Rock.  To
appear in Science.
Fried, S., A. M. Friedman, J. Mines, G. Schmitz, and M. Wheeler.  1977.
Distribution of Plutonium and Americium at an Old Los Alamos Waste Disposal
Site.  IN:  Proc. Sym. on Dynamics of Transuranics in Terrestrial and Aquatic
Environment.  Gatlinburg, TN.  October 1975.

Fried, S., A. M. Friedman, and L. A. Quarterman.  1974a.  Annual Report on
Project AN0115A, Fiscal Year 1974.  ANL-8115.

Fried", S., A. M. Friedman, and R. Weeber.  1974b.  Studies of the Behavior
of Plutonium and Americium in the Lithosphere.  ANL-8096.  pp. 10-11.

Glover. P. A., F. J. Miner, and W. L. Polzer.  1977.  Plutonium and Americium
Behavior in the Soil/Water Environment.  I. Sorption of Plutonium and Ameri-
cium by Soils..  BNWL-2117, pp. 235-246.

Hajek, B. F.  1966.  Plutonium and Americium Mobility in Soils.  BNWL-CC-925.

Hajek, B. F. and K. C. Knoll.  1966.  Disposal Characteristics of Plutonium
and Americium in a High Salt Acid Waste.   BNWL-CC-649.

Keller, C.  1971.  The Chemistry of the Transuranium Elements.  Vol. 3.
Kernchemie in Einzeldarstellungen.  Verlag Chemie/GmbH.

Knoll, K. C.  1965.  Reaction of High Salt Aqueous Plus Orqanic Waste With
Soil.  BNWL-CC-313.

Knoll, K. C.  1969.  Reactions of Organic Wastes in Soil.  BNWL-860.

Korotkin, Yu.S.  1972.  Study of Transplutonium Element Hydrolysis.  II.
Hydrolysis of Americium (III) in the Presence of Ions with Positive and
Negative Hydrat-ion Energy.  JIWR-P-6403.

-------
Latimer, W. M.  1952.  Oxidation Potentials.  Second Edition.  Prentice-Hall,
Englewood Cliffs, NJ.

Polzer, W.  L. and F. J. Miner.  1977.  Plutonium and Americium Behavior in
the Soil/Water Environment.   II.  The Effect of Selected Chemical and Physical
Characteristics of Aqueous Plutonium and Americium on Their Sorption by Soils.
BNWL-2117.
                                                                   go
Routson, R. C., G. Jansen, and A.  V. Robinson.   1975.  Sorption of  JTc,
237Np and 24lAm on Two Subsoils from Differing Weathering Intensity Areas.
BNWL-1889.

Sheppard, J.  C., J.  A. Kittrick, and T.  L. Hart-  1976.  Determination of
Distribution  Ratios  and Diffusion  Coefficients of Neptunium, Americium, and
Curium in Soil-Aquatic Environments.  RLOT2221-T-12-2.

Sillen, L.  G. and A. E. Marten.  1964.   Stability Constants of Metal-Ion
Complexes.   Special  Publication No.  17.   The Chemical Society, London.

Van Dalen,  A., F. DeWitte, and J.  Wiskstra.  1975.  Distribution Coefficients
for Some Radionuclides Between Saline Water and Clays,  Sandstones and Other
Samples from  the Dutch Subsoil.  Reactor Centrum Nederland.   75-109.
ANTIMONY

Natural Soil and Rock Distributions

     The average concentrations of antimony found in igneous and sedimentary
rocks are given in Table 3-5.   Hawkes (1954) reported on the antimony content
of the soils of the Nyeba lead-zinc district of Nigeria.  Soils that were
30 to 120 meters from ore bodies contained from 1 to 5 ppm Sb.   Ward and
Lakin (1954) reported 2.3 to 9.5 ppm Sb in Idaho and North Carolina soils.


                                   3-12

-------
Boyle (1965) reported that soils from the Galena Hill area, Yukon, Canada,
contained on the average 1.0 ppm Sb.  Little is known about the behavior of
antimony during rock weathering.  Antimony may accumulate in shales by adsorp-
tion on clay minerals and hydrous oxides such as Mn and Fe precipitates.
Antimony may be concentrated in soils relative to igneous rocks, also as an
adsorbate.
    TABLE 3-5.   ANTIMONY CONCENTRATION IN IGNEOUS AND SEDIMENTARY ROCKS
                (TUREKIAN AND WEDEPOHL, 1961)
                                       ppm
Igneous
Ultramafic
0.1
Basaltic
0.2'
Granitic
0.2
Syenite
0.1
Shale
1.5
Sedimentary
Sandstone
0.01
Carbonate
0.2
Brief Chemistry
                                                         1 pi
     The stable isotopes of antimony found in nature are    Sb (57.25%) and
123
   Sb (42.75%).  The radionuclides of interest in waste disposal  are given
in Table 3-6.
                  TABLE 3-6.  ANTIMONY RADIONUCLIDE DATA
                              (WEAST,
Isotope
124Sb
125Sb
126msb
126Sb
Half-Life
60.3 days
2.7 years
19 minutes
12.5 days
Decay Mode
3"
8-
IT, B-
6-
     Antimony may exist in (-III), (0), (III) and (V) oxidation states, readily
losing its 2(s) and 3(p) electron's.  Ionic radii are Sb"3 2.45 A, Sb+3 0.76 A
      +5      °
a'nd Sb   0.62 A (Ahrens, 1952).  The important antimony minerals include native
antimony (Sb), stibnite (Sb2S3), jamesonite (2PbS'Sb2S3), boulangerite
(5PbS'2Sb2S3), kermesite (Sb^O) and senarmontite (Sb203).  Stibnite is
the most common antimony mineral.
     Antimony, in the form of Sb  , forms complexes such as Sb(QH)2+,
Sb(OH) ~  SbCl.OH", SbCU"2 and SbS "3 (Sillen and Martell, 1964).   The exis-
      H       J   +3    °  +5      J                                 .     J.-3
tence of simple Sb   and Sb   is unlikely.  Further, Sb is amphoteric.   Sb  ,

                                    3-13

-------
for example, reacts as Sb20, + 3H20 + 20H - 2Sb(OH)4"  in bases,  and in  acids,


Sb00^ •*• H90 + 2H* -»• 2Sb(OH)?+.   Sb*" dissolves  in strong acids  forming  com-
  C. •$    d.                 £•

plexes such as SbCl,~, and in strong bases forms complex antimonides such as
r                  o

Sb(OH)c" (Latimer, 1952).   SbCU dissolves in a limited amount  of water to
      o       •                 j

give a clear solution which, upon dilution, precipitates insoluble oxochlo-


rides such as SbOCl and SbALCl- (Cotton and Wilkinson, 1962).   Izrael  et al.

                      125
(1969) mentioned that    Sb was amphoteric, with a solubility minimum at


pH 7.0.



Solid Phase and Solution Equilibria



     Thermodynamic data for Sb203 (Sillen and Martell,  1964) and Sb^Og,


Sb(OH)3(s), Sb204(s), Sb205, Sb2S3, SbCl3, and SbF3 (Wagman et  al., 1968) com-


pounds were selected and plotted in Figure 3-3 to determine the relative sta-


bility of the antimony solid phases.  This figure represents a  highly oxidiz-

                         0 68
ing environment (00 = 10     ).  Under these conditions SbCl3,  SbF3, and Sb^


are very soluble and therefore are not shown.
    Figure 3-3.  The activity of SbO  in equilibrium with various antimony

                solids in an oxidizing soil solution environment Cp02(g)

                = 0.68 a tan].         _

-------
                                              0 68
     Under these oxidizing conditions (02 = 10     atm) the minerals in
increasing order of stability throughout the pH range can be arranged as
follows:  orthorhombic Sb20v cubic Sb^O-j, orthorhombic Sb/L, cubic Sb^L,
Sb(OH)3, Sb204, and Sb^Oj..  Solids containing antimony in a +5 state
(Sb20- and Sb20j and the Sb^S., would change their relative positions with
changes in the oxidation-reduction conditions, whereas the other solids will
retain their relative positions.  Sb2Qc and Sb2Ch will become less stable
with an increase in reducing conditions, while Sb2S., will become the most
stable solid phase in extremely reducing conditions.  Sb^S- is very soluble
in an oxidizing environment because of the oxidation of sulfur.to sulfate,
which is the stable solution species (Garrels and Christ, 1965).   Under
extremely reducing conditions and total  sulfur levels of approximately 10" M
                                                      ?_
where the dominant sulfur species is no longer the SO." , the most stable
mineral will be Sb2S3.
     In oxidation-reduction conditions corresponding to oxygen pressures of
<10" " atmospheres, Sb(OH)3 would be the most stable solid.
     The relative activity of various solution species in equilibrium with
Sb203(s) is given in Figure 3-4.  If equilibrium with Sb(OH)3(s)  is assumed,
the activity of all the species would be approximately 2.9 log units lower
with the lines still parallel to the lines given in Figure 3-4.   Thermodynamic
data for Sb(OH)0  were obtained from Baes and Mesmer (1976)  and  all the other
data were selected from Wagman et al. (1968).   In the environmental pH range
of interest (4 to 8), it can be seen that HSb02° and Sb(QH)3° are the species
that would govern the total activity of antimony present in  solution.   The
other species [SbOF°, Sb(OH)/, SbO+, Sb(OH) ~ or Sb09~ and  NH.SbC,,0] do not
                            Cm               *T        £        ^    £
contribute significantly to the total antimony present in the solution and
hence can be ignored.  Since the predominant solution species [HSb02°,
Sb(OH)3°] are neutral, ion exchange as an antimony adsorption mechanism
is not expected to be important.
Experimental Adsorption Results
     Essington and Nork (1969) used two different models to  predict radio-
activity migration through basalt and Rainier Mesa tuffs.   Many  Kd measure-
                              125
ments were made, and a Kd for    Sb of 1.4 ml/g was used in  the  modeling work.

-------
     Figure 3-4.   The activity of various antimony species in equilibrium
                  with Sb203(s) (cubic) with pNH4+ = 3.0 and pF- = 4.5.
                  Oxidation-reduction changes do not affect the concentra-
                  tion of species reported in this figure.

125Sb was reported concentrated in Mediterranean bottom sediments by
Giacoletto and Triulzi (1970), and 124Sb was suspected by Osterberg et al.
(1966) to be present in precipitates collected from treated Columbia River
water.  124Sb is not a fission product, but an activation product resulting
from the former operation of once-through- cooling of the Hanford reactors.
     Izrael and Rovinskii (1970) measured the chemical states of fission
                                          IOC
products in crater water and reported the    Sb was 85% anionic, 8.5%
cationic and 6.5% colloidal.  Arnold and Grouse (1971) have determined the
radioactive contamination of copper recovered from ore fractured with
nuclear explosives.  Recovery was simulated  by leaching test shot rubble of
the radionuclide content given  in Table 3-7.  One-hundred gram  rubble samples
were  leached for 6 days with 200 ml of synthetic  copper ore  leach liquor
(2g Cu+2,  Ig Fe+3, 3g  Fe"2/liter as sulfates, with H2S04  to maintain desired
pH of  1.0,  2.0 and 3.6).  The  125Sb leached  from  the  rubble was  about 12%  at
pH 1.0, 4%  at pH 2.0 and too  low to permit  radionanalysis at pH 3.6.  The
gross  composition of  the rubble has not  been identified.
                                    3-16

-------
           TABLE 3-7.  RADIOCHEMICAL ANALYSIS OF TEST SHOT DEBRIS
                       SAMPLED ABOUT 20Q-FT ABOVE THE SHOT POINT
                       AT ABOUT 2 YEARS IN AGE (ARNOLD AND GROUSE,
                       1971)
               Radlonuclide     Concentration, dpm/g x 104
                  144Ce                    <0.03
                  9°Sr                      Q.2
                  106Ru                     1.4
                  60Co                    Absent
               Gross Gamma                  1.0

                                            125
     Adsorption of radionuclides, including    Sb, by low-grade copper ora
in Safford, Arizona, was also determined.   The chemical  analysis of this
ore is given in Table 3-8.

               TABLE 3-8.  COMPOSITION Of  SAFFORD COPPER ORE
                           (ARNOLD AND CROUSE,  1971)
                      Concentration,                  Concentration,
          Element          wt% _    Element         wt% _
            Cu              1.27             Ti        ,    0.56
            A1              7.31             K             4.03
            Ca              0.006            Na             1.98
            Mg              0.73             F             0.10
            Fe              3.71             C03            0.28
            Mn              0.002            IS             0.62
            Ni              0.011            SiO,         .  58.5
     Five ml of copper ore leach liquor, of the same composition as listed
above, were'adjusted in pH with H2S04 and the  25Sb added.   The solution
was contacted 16 hr with 1 g of minus 20 mesh Safford ore.   At pH 1.5,  the
distribution coefficient obtained was about 40 m'!/g, at pH  2.1, about 50 ml/g
and at pH 2.7, about 45 ml/g.  Arnold and Crouse (1971) state, however, that

                                   3-17

-------
with other adsorption tests on Safford ere, and with  solutions  containing
125
   Sb prepared by leaching radioactive debris, antimony was not as  strongly
adsorbed as in the above test.  These results suggest that the  chemical  forms
of antimony in the two tests were not the same.
     R. J. Serne of Battelle, Pacific Northwest Laboratories  (?ML),  1973,
equilibrated several sediments with simulated neutralized high-level  waste
solutions containing   3Sb.  The sediments and solutions are  described  in
Tables 3-9 and 3-10 and equilibrium distribution values (Kd's)  are  given in
Table 3-11.  Some Ca(OH)2 sludge developed in solutions III and IV  during
makeup.  This occurred because of the relatively high calcium nitrate content
                                                         IOC
and final pH (12.0).  This sludge was removed before  the    Sb  was  added,
                                                  125
and the solutions were used for Kd measurements.      Sb levels  were measured
immediately before use in the Kd measurements so that sludge  removal  did not
affect Kd values.
   TABLE 3-9.  SOLUTIONS USED IN 125Sb EQUILIBRATIONS WITH GLACIOFLUVIATILE
               SEDIMENTS FROM WELLS.  SEDIMENTS WERE  PRE-EQUILI3RATED WITH
               THESE SOLUTIONS MINUS THE 125sb ADDED  LATER AS A
               MICROCONSTITUENT
                         Molarity
                                 Constituent
                                              Solution
1.0
0.002
0.65
0.025

4.0
0.002
0.65
0.025

1.0
3.5
0.65
0.025

4.0
3.5
0.65
O.C25
0.6
1.18
0.79
0.002
0.15
0.25
O.C1
SjSG3
ca(:;G3),
:m,:;o3 "
pH=12
NaNO.
Ca(N03)2
'
-------
            TABLE 3-10.
SEDIMENTS USED IN OBTAINING EQUILIBRIUM
DISTRIBUTION COEFFICIENTS WITH 125$b AND
THE SOLUTIONS GIVEN IN TABLE 3-9.  CAPAC-
ITIES WERE DETERMINED ON SEDIMENTS MINUS '
THE >2mm MATERIALS
                       wt%
             Cation  Exchange
CaCOs g/IOOg
Sediment >2mm Removed Capacity, mea/lGQq Sediment
1,
2,
3,
4,
5,
5,
7,









silt
gravelly sand
sand
silty sand
caliche
silty sand
gravel
TABLE 3-11.

Sediments
1
2
3
4
5
6
7
0.0 6.96 13.2
39.1 6.73 1.24
20.1 6.16 1.30
5.0 5.95 2.60
7.88 37.00
5.6 5.28 7.65 '
58.9 6.73 0.95
EQUILIBRIUM DISTRIBUTION COEFFICIENT VALUES
(ml/g) BETWEEN SEVEN GLACIOFLUVIATILE SEDIMENTS
AND THE SOLUTIONS OF TABLE 3-9 CONTAINING ^25sb.
EACH VALUE IS AN AVERAGE OF THREE EQUILIBRATIONS
Solution
I II III IV V
0.036 0.548 79.7 122.8 0.081
0.0 0.483 81.4 71.1 0.045
0.0 0.491 55.4 109.0 0.055
1.27 2.09 83.7 43.6 0.228
0.73 0.56 15.8 14.7 0.140
0.65 0.40 15.3 24.1 .0.140
1.28 1.19 17.6 24.0 0.180
     The relatively large 125Sb Kd values obtained on the sediments with
solutions III and IV are generally up to three orders of magnitude higher
than the Kd values measured with the other solutions.  The high calcium
solutions also are solutions III and IV, suggesting that precipitation of
calcite has scavenged or coprecipitated a portion of the antimony.  The
solutions were all quite basic in pH(12), enabling them to absorb C02 from
the air and precipitate Ca^.
                                   3-19

-------
 •  '              125
     Movement of    Sb from trenches and seepage pits in the Melton and
Bethel Valleys, Oak Ridge, Tennessee, to White Oak Creek was investigated
by Ouguid (1975, 1976).  Typical results are given in Table 3-12.  Rela-
tively minor amounts of the    Sb are found in the groundwater or seepage
                                                            125
water compared to the amounts located on the soil.  Similar    Sb values
                                                                           _2
for burial ground 4 in Melton Valley showed a range of from 1.0 to 1.3 x 10
dpm/ml in the groundwater and seepage water.  This also represents a rela-
tively minor antimony movement from the soil into the groundwater.  Computed
antimony Kd values ranged from 10  to 10  ml/g.

           TABLE 3-12.  125Sb ANALYSES OF WATER AND SOIL SAMPLES
                        FROM A SEEP NEAR TRENCH 7 (DUGUID, 1975)

                                      Water Samples
                          Sample
                           Date       Count Rate, dpm/ml
                          3-5-73            £2.0
                          3-19-73           £4.0
                          5-11-73            6.7
                                     Soil  Samples
                                   Count Rate,  dpm/q
0-3
3-6
6-9
9-12
12-15
15-18
18-21
£690
£240
£120
£41
£18
£13
<9.7
     Several equilibrium adsorption experiments between a radioactive melt-
glass, a low-grade copper ore and a copper leaching solution were conducted
by Jackson (1976) under studies for the nuclear chemical  mining of copper.
                                                                   144
The radionuclides that could be measured in.the neltglass included    Ce,
                                   3-20

-------
   ELI,  b Zu and    Eu.  Tracers added to the system to obtain distribution
coefficients included 85Sr, 88Y, 103Ru, 124Sb, 134Cs, 141Ce and 152Eu.
     The chemical compositions of the glass and chalcopyrite ore are given
in Table 3-13.  The ore was mostly composed of quartz and sericite with
minor feldspar, chlorite, biotite, calcite, chalcopyrite, pyrite and
hematite.
                   TABLE 3-13. ' CHEMICAL COMPOSITIONS OF
                                MELT GLASS AND PRIMARY
                                COPPER ORE (JACKSON, 19716)
                 Constituent
                     Cu
                     Fe
                     S
                     Si0
                     CaO
                     MgO
                     A1203
                     co2
                     TiO,
                     Ba "
                     Sr
Glass, wtf;
    0.33
    0.05
    0.004
Ore. vit%
-
1.5
-
72.3
4.3
4.3
1.9
0.47
16.1
_
0.7
3.2
2.2
66.9
1.5
2.5
0.9
0.5
15.5
0.4
     The chemical composition of the leaching solution is listed in Table 3-14.
Distilled water was used instead of leaching solution in a single experiment.
     Autoclaves of 300 ml and 3.785 liters capacity were charged with various
combinations of leaching solution or water, ore and glass, and then run for
lengths of time varying from 47 days to 256 days.   Oxygen flow (4 cm3/sec) was
maintained through the system at a temperature of 353 K (90°C).   Except for the
first experiment, glass and ore leaching was accomplished separately.  Glass
                                   3-21

-------
          TABLE 3-14.  COMPOSITION OF LEACHING SOLUTION  INFLUENT
                        (JACKSON,  1976)
     Species     Concentration, mg/1      Species     Concentration, mg/1
       Cu+2            4000                Zn+2            50
       Mg+2            2000                Ni+2            30
       K+               900     .           Na+             20
       Al+3             800                Cc+2            20
       Ca+2             500        '        S04"2       25,000
       Mn*2             200                NOj            200
       Fe*3             100 '"             Cl"             20
                           pH adjusted to 1.7 with H2S04

                                                      124
dissolving rates are listed on a surface area basis.     Sb distribution was
computed as a surface area dependent function (Kds) from the measured Kd
values for both the ore and the glass.  Antimony was one of the most strongly
sorbed alements.  The ore phases adsorbed antimony activity much more strongly
than leaching glass surfaces.   In the experiment using water for leaching,
the pH of the solution soon fell as a result of reactions with the sulfides
                 124
in the ore.   The '   'Sb Kd increased in all  cases with reaction time, suggest-
ing that antimony was being removed by the decomposition products forming in
the system, such as gypsum, anhydrite or jarosite.  In one case, 53.2% of
             124
the original    Sb was found in the solid products.  The jarosite. composition
was nominally KFe^SO-MOHL, but varied from this as a function of run tem-
perature and growth location.  The enrichment of antimony in ion-rich solids
could be expected from the similar occurrence in soils and shales.
Migration Results
     Field Studies—
                    /
     Haney and Linderoth (1959)-and Haney (1967) studied the disposition of
radionuclides beneath  several ground disposal waste faci'lities at Hanford.
             125
For example,    Sb was detected to a depth of 24.4 m in soil sampling wells
drilled beneath the 216-BY covered trenches.   These trenches were in service
from December 1954 to  December 1955 and received 3.4 x 10  liters of U-plant,
high salt, scavenged waste (total  dissolved soTids = 350 g/1, chiefly NaNO.,;
p'H = 9.5) whose  °Sb  content averaged 2 yCi/ml.     Sb was the major
                                   3-22

-------
subsurface soil contaminant from the bottom of the covered trenches to
24.4 m.  Soil contamination averaged 1.5 x 10"  uCi/g to 18.3 m and fell off
rapidly to 1.5 x 10""" yCi/g at 24.4 m.  The 216-BC disposal trenches received
4x10  liters (0.9 column volumes) of the same type of waste on a specific
retention basis, which relies on long-term storage of the waste in the pore
space of the soil column above the water table.  Approximately 1740 Ci of
 ~ Sb were discharged to the BC-3 covered trench of the 8C facility.  The
   Sb soil contamination was 1.8 x 10~^-uCi/g at 6.1 m, 4 ;< 10   yCi/g at
9.1 m and 10"  uCi/g at 12.2 m.  For comparative purposes, the    Cs and
90
  Sr soil contamination maxima beneath the BY trenches fell by 5 orders of
                                               T ^ C
magnitude in the same vertical distance as the  L Sb maxima fell  2 orders of
magnitude.  The antimony, present.in a larger concentration (2 uCi/ml) than
   Cs or   Sr (0.5 to 0.005 uCi/ml each), does not adsorb on the soil with
nearly the Kd value of cesium or strontium,  and travels much further down
the soil column as a consequence.
     The leakage of 435,000 liters of Hanford high salt, high pH (1M NaOH,
5M NaNO-J high level waste from the 241-T-106 underground waste storage tank
in-1973 (Anon., 1973) afforded an opportunity to study the migration of the
various radionuclides in this waste through  the soil column underlying the
tank.  A series of wells were drilled to obtain contaminated soil  samples for
radioanalysis.  The results showed that antimony, cobalt and ruthenium moved
the most rapidly through the soil column.  All  of these radionuclides tend
to form complex species which are neutral or negatively charged,  or hydrolyzed
species that also are neutral to negatively  charged.  In either case, the
mobility of the antimony can be expected to  be relatively high.   The caustic
content of this waste type is high enough to sclubilize a portion of the
soil  column (Shade, 1974).
     Magno et al. (1970) investigated the fate of the radionuclides from the
Nuclear Fuel Services plant in western New York state.   The plant lagoon sys-
tem waters and soils were sampled and analyzed.  They estimated from these
                                   I or
data that approximately 90% of the    Sb in  the plant effluent passed through
the lagoon system and into nearby surface streams.  Ninety-eight percent of
    IOC
the    Sb was described as "dissolved" (in solution), exiting from the last
lagoon.  The antimony is not, therefore, associated with suspended solids,
but is in solution, as Figure 3-4 postulates that it should be.
                                   3-23

-------
     Laboratory Studies--.
     Laboratory studies on antimony migration through rocks or soils are lack-
                           125
ing.  There are studies of    Sb plant availability due to chelation (Hale and
                                                  121
Wallace, 1970; Wallace, 1969) and even the use of    Sb as a groundwater tracer
(Jennings and Sch.roeder, 1968), but little of primary concern with antimony
migration.
                                                               124
     .Saas and Grauby (1973) studied the transfer mechanisms of    Sb found
in reactor cooling water and other wastes to river, irrigated soil and ground-
water.   The river water contained organomineral pollution in the form of
                                                                124
industrial and municipal sewage wastes that interacted with the    Sb to
form chelates, an exchangeable fraction and hydrosoluble components defined
as the totality of the organic and mineral components of a soil that are
                                                            104
soluble in water.   The percentage of the above h-ydrosoluble    Sb fraction
was plotted versus depth in cm for an alluvial, calcareous soil column
          124                                      124
receiving    Sb in river water.   Ten to 30% of the    Sb was in the hydro-
soluble form.  The point of this work was demonstrating the interactions
                                          124
between pollutants in the river water and    Sb that results in increased
antimony migration.   These pollution interactions with radionuclides should
be addressed in nuclear reactor siting studies.
Summary
     The principal aspect of antimony chemistry that governs its adsorption
reactions with soils and rocks over the pH range 4 to 8 is the prevalence of
neutral and complexed species (Figure 3-4).   Low Kd values are found in high
salt, high pH solutions as well  (Serne, 1973), with relatively rapid migration
through soils (Haney and Linderoth,  1959; Haney, 1967; Anonymous, 1973) and
surface waters (Magno et al., 1970).  There are indications that antimony can
be precipitated or coprecipitated in solids (Serne, 1973), even at low pH
values  (Jackson, 1976).  In low redox conditions (02 < 10"   atmospheres),
antimony trihydroxide may precipitate and may be a stable compound that con-
trols antimony concentrations (Figure 3-3).   Many of the antimony species
associated with organopollutants are water soluble (Saas and Grauby, 1973)
and migrate readily with the water.   Thermodynamic data on the antimony com-
plexes  formed with soil organic materials do not exist at present because the
complexes are chemically poorly defined,  and in many cases, unknown.
                                   3-24

-------
 References

 Ahrens, 1. H.  1952.  The Use of  lonization Potentials.  Part  I.   Ionic  Radii
. of the Elements.  Geochim. et Cosmochim. Acta.  2:155.

 Anonymous.  1973.  241-T-106 Tank Leak  Investigation.  ARH-2874.

 Arnold, W. D. and D. J. Grouse.   1971.  Radioactive Contamination  of Copper
 Recovered from Ore Fractured with Nuclear Explosions.  ORNL-4677.

 Baes, C. F. and R. E. Mesmer.  1976.  The Hydrolysis of Cations.   John Wiley
 and Sons, New York.

 Boyle, R. W..  1965.  Geology, Geochemistry, and Origin of the  Lead-Zinc-Silver
 Deposits of the Keno Hill-Galena  Hill Area, Yukon Territory.   Geol. Surv.
 Canada Bulletin III.

 Cotton, F. A. and G. Wilkinson.   1962.  Advanced Inorganic Chemistry.  Inter-
 science Publishers.

 Duguid, 0. 0.  1975.  Status Report on  Radioactivity Movement  from Burial
 Ground in Melton and Bethel Valleys.  ORNL-5017.

 Duguid, J. 0.  1976.  Annual Progress Report of Burial Ground  Studies at
 Oak Ridge National Laboratory:  Period  Ending September 30, 1975.  GRNL-5141.

 Essington, E. H. and W. E. Nork.   1969.  Radionuclide'Contamination Evalua-
 tion-Milrow Event.  NVO-1229-117.

 Garrels, R. M. and C. L. Christ.   1965.  Solutions, Minerals and Equilibria.
 Harper and Row.. New York.  pp. 213-216.

 Giacoletto, C. and C.. Triulzi.  1970.   Radiochemical Studies for Radioactivity  .
 Determinations on Marine Sediment Samples:  Cs-137, Sb-125, and Co-60.
 CISE-N-118.

 Hale, V. 0. and A. Wallace.  1970.  Effect of Chelates on  Uptake of Some
 Heavy Metal Radionuclides from Soil  by  Bush Beans.   Soil  Sci.   109:262-2.53.

 Haney, W. A. and C. E.  Linderoth.  1959.  Exploratory Field Study of a Ground
 Waste Disposal Facility.  HW-60115.

 Haney, W. A.  1967.  Final Report on the Effects of Ben Franklin Dam.  BNWL-41.2.

 Hawkes, H. E.  1954.  Geochemical Prospecting Investigations in the Nyeba
 Lead-Zinc District, Nigeria.   U.S.G.S.  Bulletin 1000-8.

 Izrael, U. A., V.  N. Petrov, A. A. Pressman, F. A.  Rovinsky, E. D. Stukin,
 and A. A. Ter-Saakov.  1969.   Radioactive Contamination of the Environment
 by Underground Nuclear Explosions and Methods of Forecasting It.

 Izrael, Yu.A. and F. Ya. Rovinskii.   1970.  Hydrological  Uses  of Isotopes
 Produced in Underground Nuclear Explosions for Peaceful Purposes.  UCRL-
 Trans-10458.
                                    w ™ £ 3

-------
Jackson, D. D.  1976.  Radiochemical Studies for. the Nuclear Chemical Mining
of Copper.  UCRL-52025.

Jennings, A. R. and M. C. Schroeder.  1968.  Evaluation of Selected  Isotopes
as Ground Water Tracers.  Watar Resour. Res.  4:329-838.

Latimer, W. M.  1952.  The Oxidation States of  the Elements and Their Poten-
tials  in Aqueous Solutions.  -Prentice-Hall, Inc.

Magno, P., T. Reavey, and J. Apidianakis.   1970.  Liquid Waste Effluents  from
a Nuclear Fuel Reprocessing Plant.  6RH-NERHL-7Q-2.   .

Osterberg, C., N. Cutshall, V. Johnson, J.  Cronin, D. Jennings, and
L. Frederick.  1966.  Some Non-Siological Aspects of Columbia River  Radio-
activity.  IN:  Disposal of Radioactive Wastes  into Seas, Oceans  and Surface
Waters.  IAEA-SM-72/T8A, pp. 321-335.

Saas,  A. and A. Grauby.  1973.  Mechanisms  for  the Transfer to Cultivated
Soils  of Radionuclides Discharged  by Nuclear Power Stations into  the System:.
River-Irrigated Soil—Ground Water.  IN: ' Environmental Behavior  of  Radio-
nuclides Released in  the Nuclear Industry.  IAEA-SM-172/57, pp. 255-269.

Shade, J. W.  1974.   Reaction of Hanford Sediments with Synthetic Waste,  A
Reconnaissance Study.  ARH-CD-176.

Sillen, L. G. and A.  E. Martell.   1964.  Stability Constants of. Metal-Ion
Complexes.  The Chemical Society.

Turekian, K.  K. and K. H. Wedepohl.  1961.   Distribution of the Elements  in
Some Major Units of the Eartn's Crust.   Bull.  Geol.  Soc. America.   72:175.

Wagman, D. D., W. H.  Evans, V.  8.  Parker, I. Halon,  S.  M.  Bailey,  and
R. H. Schumm.  1968.  Selected Volumes of Chemical Thermodynamic Properties.
U.S.  Department of Commerce.

Wallace, A.  1969.   Behavior of Certain Synthetic Chelating Agents in Soil
and Biological Systems.   UCLA-34-P-51-26, 92 p.

Ward, F.  N. and H.  W. Lakin.   1954.  Determination of Traces of Antimony in
Soils and Rocks.   Analytical  Chemistry.   26:1168-1173.

Weast, R.  C., Editor.  1976.   Handbook  of Chemistry  and  Physics.   The Chemical
Rubber Co., Cleveland, Ohio,  pp.  B301-B302.
                                    3-26

-------
CERIUM
Natural Soil and Rock Distributions
     The abundance of cerium in rocks is given in Table 3-15.  Cerium generally
increases in concentration from basaltic to granitic rocks.  In a study of
Russian platform soils, Balashov et al. (1954) found that the soils with
highest cerium contents were alkaline, suggesting precipitation as the
hydroxide.  The acid soils were lower in cerium because cerium had been
removed and had migrated.  Soils derived from granitic rocks could be expected
to show the highest cerium contents if the weathering resulted in an alkaline
environment.  Vinogradov (1959) gave the average cerium content of soils as
50 ppm.  Half, or more, of the cerium in rocks is found in accessory minerals
such as apatite, while the rema-inder also substitutes for calcium but only in
the main- stage minerals such as plagioclase.

                   TABLE 3-15.   CERIUM ABUNDANCE IN ROCKS
	Rock Type	•       Ce, ppm
Continental basalts         "          59
Kimberlite                            1T9
UHramafic                              0.1
Basaltic                               48
Granitic, high Ca                      81
Granitic, low Ca                       92
Syenites                              160
Shales                                 59
8616 Russian platform shales           67
6051 Russian platform sandstones       33
11205 Russian platform limestones       6.5
average of above                       57
Sandstones                             92
Dolomite and limestones                11.5
         Reference
Frey et al.,  1968
Burkov and Podporina, 1966
Turekian and Wedepohl, 1961
Turekian and Wedepohl
Turekian and Wedepohl
Turekian and Wedepchl
Turekian and Wedepohl
Turekian and Wedepohl
Ronov et al., 1967
Ronov et al., 1967
Ronov et al., 1967  .
Ronov et al., 1967
Turekian and Wedepohl
1961
1961
1961
1961
1961
1961
Turekian and Wedepohl, 1961
                                   3-27

-------
Brief Chemistry
     Cerium occurs naturally in four stable isotopes shown in Table 3-15.
                  141       144
The radioisotopes    Ce and    Ce are both present in radioactive wastes as
fission products with half-lives of 33 days and 284 days, respectively.  After
100 years, little fission product cerium remains.

                  TABLE 3-16.  STABLE ISOTOPES OF CERIUM
                               (INGHRAM ET AL,, 1947)

                     Isotope     Natural Abundance, '•
                     136Ce               0.15
                  •  138C2               0.25
                     14°Ce              88.5
                     142Ce              11.1
     The usual oxidation state of cerium is tripositive with .ionic radius
of 1.03 A (Ahrens, 1952).
     The solubility product of Ce(III) hydroxide was given by Vickery (1953)
            20
as 1.5 x 10"  .   For comparison, the solubility products of Al(OH)., and
                     1 c             ^"1Q      \                     J
Fe(OH)3 are 3.7 x 10    and 3.8 x 10   , respectively (D'Ans and Lax, 1967).
     In geologic environments, cerium forms weak complexes.  Cerium complexes
of interest to soil environments will be reviewed in a subsequent section.
The most important and the most common cerium minerals include monazite (light
rare earths, Th) P04 and bastnaesite (light rare earths, Th) FCO-j.  Many
other cerium-containing minerals exist but most are relatively rare.
Solid Phase and'Solution Equilibria
     The relative stability of various cerium solid phases is shown in Fig-
ure 3-5.  The thermodynamic data of the compounds used in Figure 3-5 were
obtained from Schumm et al. (1973), for Ce203 and CeO^, Baes and Mesmer
(1976) for Ce(OH)3 and Sillen and Martell (1964) for CePO^.  Among the solids
depicted in Figure 3-5, Ce02 is the only one that changes its position with
a change in oxidation-reduction conditions.  In an oxidizing environment
(p02 < 23.68 atm), the solids in increasing order of stability are ^2^3'
                                   3-28

-------
                    -10
                  u
                  S1
                    -12
                    -14
                    -16
                    -13
                                                AT pCj-0.68

                                        Ce(OHL:Ce02ATp02 -77.22
                           ATp02 "0.68
                                       9
                                       PH
                                             10
                                                         12
       Figure 3-5.  The activity of Ce    in  equilibrium with phosphate
                    levels from Variscite and  Gibbsite  (V and G),  Dical:
                    cium Phosphate Dihydrate (DCPD)  and OctacaTcium
                    Phosphate  (OCP).
Ce(OH)3, CeP04, and CeOg.  Ce2S3  is  too  soluble  to  plot in Figure 3-3 in an
oxidation-reduction environment where  S04    will  be the most dominant sulfur
species.  In oxidation-reduction  conditions  corresponding to pG^ of >E3.68,
CePO, would be the most stable solid phase.
     The activity of various solution  species  of  cerium in equilibrium with
CePO,(solid) and under such concentrations of  F", Cl~,  S0^~, NO^, and PO^"
as commonly found in the environment is  shown  in  Figure 3-6.  The thermody-
namic data for all of the hydrolysis species were obtained from Baes and
                                  2+
Mesmer  (1976).  The data for CeNO^   were obtained from  Si 11 en and Martell
(1964).  All the other data were  obtained from Shumrn et al.  (1973).  Since
Figure  3-6 is based on solution complexes in equilibrium with Ce(III) as
CePO,,  Ce(IV) ions and complexes  would decrease  in  activity with increase in
reducing conditions.  In the environmental pH  range of  interest the Ce(IV)
ions and complexes do not contribute significantly  to the total cerium con-
centration in solution.
                                   3-29

-------
      Figure 3-6.  The activity of various cerium species in equilibrium
                   with CeP04fs,) in an oxidizing soil environment Cp02(g).
                   = 0.68 atmj, pF" = 4.5, pCl" = pS04-  = 2.5, pNOs' =3.0
                   and pH9PO,,~ = 5.0.
Trivalent cerium species in increasing order of importance are CeCl  ,
CeiNO,"1", CeF  , Ce  , and CeSOt.  Thus, CeSO. would be the most dominant
    3                        44
solution species in the environmental pH range of interest (pri 4 to 8).
Experimental Adsorption Results
     Nishita et al. (1956) studied five soils and two clay minerals in systems
with 144Ce.  Pretreatment of bentonite with dilute acid gave.much less cerium
uptake by the clay than the untreated bentonite.  The hydrogen ion was more
                       144
strongly adsorbed than    Ce.  The pH of the leaching solution had a profound
effect on release of cerium by the soils.  At pH values less than 5, 99.5%
              144
recoveries of    Ce were reported.
                                         144
     Rhodes (1957) studied the uptake of    Ce on a Hanford soil as a function
of pH.  The Kd values rose steadily from 3.0 at pH 1.6 to >1980 at pH 6.1.  .
The Kd then decreased to a minimum of about 100 at pH 10 and increased again to
>1980 at pH 12 and above.  The reduced uptake by soil of cerium between pH
7 to 12 did not occur in the presence of macroconcentrations (3M) of sodium
ions.  Thus, the    Ce was present either as a radiocolloid in this region,
which was flocculated by the high concentration of sodium ions, or as an ion
                                   3-30

-------
species that was altered by the sodium ions to increase uptake by the soil.
                                _p
The complete removal of 4.2 x 10  M cerium at pH 5.5 to 6 was reported, which
at this concentration exceeded the capacity of the soil by ten times.  The
precipitation of cerium hydroxide was indicated.
     Bensen (1960) reported that cerium adsorbed on the soil could not be
exchanged above pH 7.4 and was difficult to exchange at pH 5.5 to 7.4.  The
fraction of cerium exchanged by barium at pH 5.5 was 10% and virtually none
at pH 7.4.  All of the cerium could be removed by ion exchange at pH 2.2,
with less removal as a function of pH to 7.4 when none of the cerium was
exchangeable from the soil.  Above pH 7.0, the presence of 0.5M alkali metal
cations and 0.25M alkaline earth metal cations did not affect cerium removal
from solution by soil.  Below pH 3.0, the adsorption of cerium was depressed
similarly by accompanying salts, implying that the cerium is ionic below
pH 3.0 and removed by ion exchange on the soil.
     Bensen (1960) reported that the soil uptake of cerium added as Ce(III)
was identical to that added as Ce(IV).  However, when an oxidizing agent
(Na3i03) was added, uptake of Ce(IV) increased to 99% at pH 1, while only
58% of the Ce(III) was. adsorbed at this pH.  Bensen concluded that the
absence of sufficient holding oxidant resulted in an immediate reduction of
Ce(IV) to ee(III) by contact with the soil.
     Kokotov et al. (1962) investigated    Ce adsorption by soils and found
that cerium adsorption was much less below pH 2.0 due to increased hydrogen
ion competition.
   .  Bochkarev et al. (1964) reported that 85% of 144Ce applied to soil  sur-
faces was accounted for 2 years following the application in the top 5 cm of
medium and coarse clay loams.
     Kampbe-11 (1964) investigated    Ce uptake by colloidal suspensions of
kaolinite, illite and montmorillonite type clays, all of which had a high
             144
affinity for    Ce.  Only under very acid conditions did the cerium remain
in solution.  DTPA also was effective in keeping cerium in solution at
trace levels.
                                   3-31

-------
     Dlouhy (1967) studied the distribution coefficients of cerium on
Casaccia soil and..tuff both in the laboratory and in field experiments.  The
laboratory distribution coefficients ranged from 1050 to 1300 cm /g for the
                                      3
cerium on the soil and 3000 to 3800 cm /g for cerium on the tuff.  The solu-
tion conditions and soil and tuff were not characterized.  Coring of seepage
pits showed the cerium to be essentially contained in the upper 6 cm of the
pit soil.
     Molchanova (1968) studied the adsorption of    Ce by soils and found
batter uptake at acidic pH values above 2.0 than at neutral to alkaline values.
Cerium was more easily removed from soils by iron salts than aluminum or cop-
per salts.   The presence of iron colloids in the soil reduced the sorption of
144
   'Ce more sharply than did iron salts.
     Brown et al.  (1969) studied the formation of    Ce colloids and the
adsorption of'cerium by humic acid and bentonite suspensions.  As ionic
strength and cerium concentration decreased and pH increased, equilibrium
shifted toward radiocolloidal cerium.   At pH 6.5 and ionic strength of 10" ,
about 90% of the cerium was in the radiocolloidal form.  A multiple regression
equation is given that shows the effects of pH, ionic strength and cerium
concentration on radiocolloidal cerium in solution.
Migration Results
     Field Studies—
     Magno et al.  (1970) determined the forms and relative mobilities of
radionucl'ides from the lagoon system of the Nuclear Fuel  Services plant in
                                                       144
Western New York state.  They reported that 0.02 Ci  of    Ce were discharged
                                                                       144
from the last lagoon from May to October 1969.   None of the-discharged    Ce
                                                 144
activity was dissolved in the water.   All of the    Ce activity discharged
                                                            144
was found in the suspended solids.  Approximately 8.5 Ci  of    Ce were dis-
charged from the plant to the lagoon system over the same time period (May-
October 1969).  Thus, although all of the cerium was in particulate form,
about 3% migrated through the lagoon system in the form of suspended solids.
     The migration of 435,000 liters of liquid high-level  waste (^5M NaN03,
1M NaOH) from 241-T-106 underground tank through Hanford soils was followed
by the drilling of several sample wells -after the event (Anon., 1973).   The

                                   3-32

-------
contaminated soil zone was determined on  the basis of  these  analyses  and
the relative migration distances of  the radionuclides  evaluated.   The rare
earths,  including cerium, were more  mobile  than  plutonium and cesium  but
less mobile than antimony, cobalt and ruthenium.  Cerium could  be  expected
to be at least partially complexed in such  a high salt, high pH solution.
     Brookins  (1976) reported that the rare earth elements including  cerium
have been retained at the site of a  1.8 billion  year old fossil nuclear
reactor at Oklo, Gabon.  The rock surrounding the reactor is shale infilling
a fracture system in organo-argillaceous  sandstone.
     Laboratory Studies—
     Nishita et al. (1956) studied the extractability  of     Ce  from clays and
soils with distilled water and 1M ammonium  acetate.  The fraction  remaining
on the soil or clay mineral (bentonite and  kaolinite)  after  the acetate
elution or leaching solution, was passed through  the column was  called  the
                144                                             144
nonexchangeable    Ce fraction.  For both the soils and clays,     Ce was
hardly moved by the water leaching or exchanged  from soils or bentonite.
                                                 144
Kaolinite showed a large portion of  exchangeable    Ce and a relatively small
                          144                 144
-amount-of—nonexchangeable    Ce.  Most of the    Ce on the soils and on the
bentcnite were in the nonexchangeable fraction.  The somewhat more acid
nature of the kaolinite probably accounts for the larger percentage of cerium
associated with it as exchangeable cerium.
     Schulz (1965) included cerium and the  other rare  earths under the cate-
gory of immobile radioactive elements.  These elements were  precipitated as
hydroxides or carbonates or were very strongly bound by soil clays.  Perhaps
                                                                144
this view should be somewhat tempered by the previous  cases  of     Ce migration
as a particulate or in complexed species.
     Eichholz et al. (1967) studied  the fractionation  of several radioactive
elements between several natural waters and the  suspended solids contained
in them.  The results gave an indication of the  ability of cerium  to
migrate as a suspended solid or adsorbed on the  suspended solids already
present in most natural waters.  The suspended solids  are characterized in
Table 3-17.  The most ob.vious relationship  is a direct one between the sus-
pended solids content and cerium adsorption.  However, there also  is a

                                   3-33

-------
              TABLE 3-17.  ADSORPTION OF 144Ce ON SUSPENDED
                           SOLIDS (EICHHOLZ ET AL., 1967)
                                           Suspended     Dissolved
Water Source
Colorado River
Camp McCoy
Bayou Anacoco
Lodgepole Creek
Chattahoochee River
1 Ce Removed, %
72.0
44.6
36.9
96.9
26.1
Solids, mg/1
299
12
24
965
131
Solids, mg/1
350
60
63
200
31
£H_
7.5
6.9
6.2
6.8
7.3
suggestion of an inverse relationship between cerium adsorption and solution
pH.  At the lower pH, more of the cerium should be present as cationic species
and hence better able to adsorb on the suspended solids.  It should also be
pointed out that the Lodgepole Creek solids are largely montmorillonite, a
high exchange capacity clay mineral.
Summary
     Mo information is available on solid compounds of cerium that may be
presant in soils and sediments.  It appears thatCePO* (Figures 3-5 and 3-6),
especially in alkaline conditions, may form in soils and sediments and may
control cerium concentrations.
     In all terrestrial  environments Ce(IV) and its complexes would have
insignificant effects on cerium concentration (Figure 3-6).   Thus, cerium
would be expected to be present as Ce(III) (Figure 3-6 and Ahrens, 1952;
Bensen, 1960).  In the absence of strong complexing 1igands,~and especially
in acidic environments,  cerium would be expected to be present as Ce   (Fig-
ure 3-6; Bensen, 1960) and ion exchangeable.   The increase in pH has been
shown to increase the cerium adsorption (Rhodes, 1957; Bensen, 1960;
Nishita, 1956; Kokotov et al, 1962; Kampbell, 1964).  As expected, the
presence of complexing ligands such as DTPA increases cerium concentration
in solution and reduces  adsorption (Kampbell, 1964).  In slightly acidic
(pH 6.5) to alkaline conditions, cerium is thought to be present as radio-
colloid which has been shown to increase adsorption (Rhodes, 1957; Brown
et al., 1969).  Kampbell (1964) reported that kaolinite, montmorillonite,
and ill He have a strong affinity for cerium.  Cerium adsorbed on suspended

                                   3-34

-------
solids has been shown to migrate in field and laboratory studies  (Magno.
et al., 1970; Eichholz et al., 1967).  However, this type of movement would

be expected to be dependent upon the particle size of the suspended  solids
                             i
and the pore size distribution of soils and rocks.

References

Ahrens, L.  H.  1952.  The Use of lonization Potentials.  Part I.  Ionic Radii
of the Elements.  Geochim. et Cosmochim. Acta.  2:155.

Anonymous.  1973.  241-T-106 Tank Leak Investigation.  ARH-2874.

Baes, C. F., Jr. and R. E. Mesmer.  1976.  The Hydrolysis of Cations.  John
Wiley and Sons, New York.

Balashov, Y. A., A. B. Ronov, A. A. Migdisov, and N. V. Turanskaya.  1964.
The Effects of Climate and Facies Environments on the Fractionation  of the
Rare Earths During Sedimentation.  Geochemistry International.  10:951-969.

Bensen, D. W.  1960.  Review of Soil Chemistry Research at Hanford.  HW-6720!.

Bochkarev, V. M., Z. G. Antropova, and E. I. Belova.  1954.  Migration of
Strontium-90 and Cerium-144 in Soils of Different Texture.  Soviet Soil
Science.  9:936-938.

Brookins, D. G.  1976.  Shale as a Repository for .Radioactive_Was_te:__ The
Evidence from Oklo.  Environmental Geology.  1:255-259.

Brown,  R. E., R. E. Franklin, and R. H. Miller.  1969.  Reactions of Cerium-144
in Solution and Suspensions of Soil Humic Acid and Bentonite.  Soil  Science
Society of America, Proceedings.  33:677-681.

Burkcv, V. V. and Ye. K. Podporina.  1966.  First Data on Rare Earths in
Kimberl.ite. Dole1.. Acad. Sci. USSR.  Earth Science Section.  171:215-219.

D'Ans,  J. and E. Lax.  1967,  Taschenbuch fiir Chemiker and Physiker.  3 Auf1.,
Bd. I.  Springer.

Dlouhy, A.  1967.  Movement of Radionuclides in the Aerated Zone.  _IN_:  Dis-
posal of Radioactive Wastes into the Ground.  IAEA-SM-93/18, pp. 241-249.

Eichholz, G. G., T. F. Craft, and A. N. Galli.  1967.  Trace Element Frac-
tionation by Suspended Matter in Water. Geochim. et Cosmochim. Acta. 31:
737-745.

Frey, F. A., M. A. Haskin, J. A. Poetz, and L. A. Haskin.  1968.  Rare Earth
Abundances in Some Basic Rocks.  J. Geophys. Res.  73:6085-6098.

Inghram, M. G., R. J. Hayden, and D. C. Hess, Jr.  1947.  The Isotopic Consti- '
tution of Lanthanum and Cerium.  Phys. Rev.  72:967-970.


                                   3-35

-------
Kampbell, D. H.  1964.  Cerium, Iron, and- Manganese Sorption by Soil Colloids
and Uptake by Plants.  Thesis. Univ. of Missouri.

Kokotov, Yu. A., R. F. Popova, I. Tsing-Chih, and M. Shih-tsi.  1962.  Scrp-
tion of Long Lived Fission Products by Soils and Argillaceous Minerals.  II.
Sorption of Cerium-144 by Soils.  Radiokhimya.  4:227-228.

Magno, P., T. Reavey, and J. Apidianakis.  1970.  Liquid Waste Effluents
from a Nuclear Fuel Reprocessing Plant..  BRH-NERHL-70-2.

Molchanova, I. V.  1968.  Behavior of Cerium-144. in Various Tvpes of Soil.
Tr. Inst. Ekol. Rast. Zhivotn.  61:4-11.

Nishita, H., B. W. Kowalewsky, A. J. Steen, and K. H. Larson.  1956.  Fixation
and Extractability of Fission Products Contaminating Various Soils and Clays:
I.  Strontium-90, Ruthenium-106, Cesium-137, and Cerium-144.  Soil Science.
81:317-326.

Rhodes, P. W.  1957.   The Effect of pH on the Uptake of Radioactive Isotopes
from Solution by a Soil.  Soil Science Society of America, Proceedings.
21:389-392.

Ronov, A. B., Yu. A.  Balashov, and A. A. Migdisov.  1967.  Geochemistry of
the Rare Earths in the Sedimentary Cycle.  Geochemistry International.  4:1-17.

Schulz, R. K.  1965.   Soil Chemistry of Radionuclides.  Health Physics.  11:
1317-1324.

Schumrn, R. H., D. D.  Wagman, S. M. Bailey, W. H. Evans,"and V. B.  Parker. 1973.
Selected Values of Chemical Thermodynamic Properties.  Tables for the Lanthanide
(Rare Earth) Elements (Elements 62 through 76) in the Standard Order of Arrange-
ment.  NBS Technical  Note 270-7.

Sillen, L. fr. and A.  E.  Martell.  1964.  Stability Constants of Metal-Ion
Complexes.  Special Publication No. 17.  The Chemical Society, London.

Turekian, K. K. and K. H. Wedepohl.  1961.  Distribution of the Elements in
Some Major Units of the Earth's Crust.  Bull. Geol.  Soc. America.  72:175.

Vickery, R. C.  1953.  Chemistry of the Lanthanons.   Butterworths Science
Publications.

Vinogradov, A. P.  1959.  The Geochemistry of Rare and Dispersed Chemical
Elements in Soils.  Consultants Bureau, Inc.
                                   3-36

-------
CESIUM
Natural Soil and Rock Distribution
     The cesium content of several rock types is given in Table 3-18.   Bertrand
and Bertrand (1949) studied the soils of France and Italy and found the largest
concentration of cesium in soils over igneous rocks and in alkaline soils.  .A
range of 0.3 to 25.7 ppm Cs was reported.   Approximately 5 ppm Cs was  reported
by Vinogradov (1959) in soils of the Russian Plains and 1 ppm in Japanese
soils.  An average of 1.1  ppm Cs was found in suspended material and 2.6 ppm
Cs in bottom sediments of seven western United States streams (Sreekumaran
et al., 1968).   Hirst (1962) studied the cesium content of modern marine sands
and clays, and reported them to contain 1.0 to 3.2 ppm Cs and 7.9 to 13 ppm
Cs, respectively.   In general, cesium is enriched in the potassium minerals,
although enrichment may only consist of a  few hundred ppm cesium.   The large
size of the cesium cation  makes it difficult for it to substitute in crystal
positions normally occupied by potassium,  its nearest geochemically abundant,
alkali metal relative.  Hence, cesium often becomes concentrated enough to
form its own mineral phase (pollucite).
                     TABLE 3-18.   CESIUM CONTENT OF ROCKS

         Rock Type	     Cs Content, ppm     	
                                      1.1
                                                         Reference
Basalt (estimated average)

Basalt (estimated average)
Basalt (estimated average)
Basalt (estimated average)
Granodiorite
Granite
Granite
Granites, Russia

Marine shales
Sandstone
Limestone
Modern sediments
                                      1.7
                                      1.0
                                      1.0
                                      3.0
                                      6.8
                                      5.0
                                      5.5

                                      5.0
                                      1.0
                                      1.0
                                      6.0
Turekian and Wedepohl,
1961
Heier and Adams, 1964
Taylor, 1964
Taylor and White, 1966
Heier and Adams, 1964
Heier and Adams, 1964
Taylor, 1964
Beus and Fabrikova,
1961
Horstman, 1957
Horstman, 1957
Horstman, 1957
Horstman, 1957
                                   3-37

-------
 Brief Chemistry
     The only stable isotope of cesium is    Cs.  Fission product cesium
 radioisotopes include four main cesium isotopes, with only    Cs,    Cs and
   Cs (half-lives of 2.05 years, 3 x 10  years and 30.23 years, respectively)
 of significant concentrations 10 years after separation from PWR (Pressurized
 Water Reactor) fuels (Schneider and Platt, 1974).  Pollucite, (Cs, Na)
 (AlSipOg) x HLO containing 22 to 36 wt% cesium, is the only independent cesium
 mineral of any importance.
     In all its natural compounds, cesium occurs as a monovalent cation with
                o
 a radius of 1.67A (Ahrens, 1952).  There is, little, if any, tendency for
 cesium to form complexes in natural environments, and the solubility of most
 cesium compounds in water is very high.
 Solid Phase and Solution Equilibria
     Thermodynamic data was collected for the following solid phases of cesium:
 Cs(OH), Cs20, CsCl, CsC104, Cs2S04, Cs2C03, CsHCOp CsNO,, CsF.   All are highly
 soluble and, therefore, solid phase diagrams for cesium are not presented.
     Baes and Mesmer (1976). report that cesium may be associated with OH" ions
 in solution and'that the extent of this association cannot be estimated accu-
 rately.   Chariot (1957) reported that cesium rarely forms solution complexes.
Therefore cesium would be expected to appear in solutions predominantly as
Cs .   Only 1.5% of the cesium from underground nuclear explosions was found
 (Izrael  et a!., 1970) to be colloidal,' which may have been adsorbed on other
colloids.  The main soil and rock reaction with cesium is expected to be ion
exchange.
Experimental Adsorption Results
     The effect of cesium concentration and pH on cesium adsorption by a
Hanford calcareous soil was studied by McHenry (1954).  The data indicated
 that trace cesium concentrations are essentially completely adsorbed above
 pH 4.0.   When accompanied by 4M NaCl, however, only up to 75% of the trace
cesium was adsorbed and the adsorption was nearly independent of pH over a
wide range.  When the concentration of cesium and competing ion  was at 50%
of the soil ion exchange capacity, competing cation depressed, cesium adsorp-
tion by soil in the order:   H > La £ Sr > Na > NH^.
                                   3-38


-------
     At 1% of the soil capacity and lower cesium concentration, competing
cation effects on cesium adsorption were slight.  Trace concentrations of
cesium were adsorbed to a greater degree and were more difficult to displace
from the soil by competing cations than when the cesium was adsorbed in.
greater concentrations.  The presence of small amounts of mineral phases in
the soil with high cesium selectivities may account for this behavior (Ames,
1959; Jacobs and Tamura, 1960; Tamura and Jacobs, 1950; Schultz et al., 1960;
Sawhney, 1964; Gaudette et al., 1966).  Mica-like minerals such as illite
                                                                            o
tend to "fix" cesium.  When expanded layer silicates were collapsed to a 10 A
mica spacing by potassium saturation and heating, increased cesium selectivity
resulted (Tamura and Jacobs, 1961; Tamura, 1963; Coleman et al., 1963).   Some
                                                               o
researchers have considered the exchange of trace cesium on 10 A. spaced
minerals to be irreversible (Klechkovski and Gulyakin, 1958; Spitsyn et al.,
1963), but the reverse rate may only be much slower than the adsorption rate
(Routson, 1973).
     Several authors have determined cesium distribution coefficients  between
various solutions and rocks.  These are listed in Table 3-19.   Others  have
determined adsorption isotherms on several natural  materials,  and from the
isotherms, derived thermodynamic functions for the specific cation exchange
reactions.  Some of the thermodynamic studies, include Thomas (1967), Tamers
and Thomas (1960), Merriam and Thomas (1956), Gaines and Thomas (1953),
                                          »
Lewis and Thomas (1963) and Eliason (1966) on cesium ion exchange on systems
of cations with clays.  Ames (1959), Frysinger (1962) and Howery and Thomas
(1965) studied ion exchange and cesium thermodynamics on the natural zeolite
clinoptilolite.
     Prout (1958) studied the effects of system variables on Savannah  River
soil (80%. sand, 20% clay) uptake (Kd values) for trace cesium, strontium and
Plutonium.  The pH and adsprbed ion concentration were some of the variables
examined.  Cesium adsorption fell  off rapidly with pH at less  than 6 and
decreased as more than 10% of the soil cation exchange capacity was occupied
by cesium.
     Berak (1963) has contributed much of the cesium adsorption work on dif-
ferent rock types and minerals.  The standard equilibrating solution used is
                                           -5
given in Table 3-20.  Carrier cesium 1 x 10  M as CsCl was used along with the

                                   3-39

-------
TABLE 3-19.  CESIUM DISTRIBUTION COEFFICIENTS FOR SEVERAL ROCKS,
             MINERALS AND SOIL TYPES
 Medium
CsKd, ml/q
Condition
Reference
Al luvium
Central Nevada
Desert Alluvium
Hot Creek Valley,
Nevada
Tuff, Rainier Mesa,
NTS, Nevada
Tuff, Rainier Mesa,
NTS, Nevada
Carbonate, Yucca
Flat, NTS,. Nevada
Granodiorite,
Climax Stock,
NTS, Nevada
Alluvium, INEL,
Idaho
Granite,
Central Nevada
SasaH, NTS,
Nevada
Basalt
Amchitka
Island
Shaley siltstone
New Mexico
Sandstone,
New Mexico
SUPERSATURATED ROCKS
Rhyolites,
Czechoslovakia










121-3165

70-2640


1020

12,100-17,800

13.5

8-9
1030-1810

285-360
450-950
34.3

792-9520

280
39
6.5(seawater)
309

102

(containing quartz)

100,000
33
15 "
67
20
354
4
72
50,000
107
57
500-4000 -jm

500-4000 urn


>400 urn

100-200 mesh

>4000 urn

100-200 mesh
0.5-1.0 mm

(Laboratory soil )
(Field soil) •
>4000 urn

32-80 mesh

500-4000 pm
500-4000 urn
500-4000 urn
>4000 pm

>4000 um



0.1-0.2 mm



c






Nork et al . , 1971

Nork and Fenske, 1970


Nork and Fenske, 1970

Goldberg, et al., 1962

Nork and Fenske, 1970

Beetem et al., 1962


Schmalz, 1972

Nork and Fenske, 1970

Angelo et al., 1962

Essington and Nork,
1969

Nork and Fenske, 1970

Nork and Fenske, 1970



Berak, 1963










                                3-40

-------
                               TABLE  3-19.   (continued)
        Medium
   Central Slovakia
Volcanic Glasses,
  East Slovakia
Central Slovakia
  Iceland
  CsKd.  Ml/q
 40
 31
 32
 11
 9
 50,000
 23
 12,400
 42
 25
 26
 45
 41
 42
 39
 23
 18
:23   ;     -
 13
 31
 10
 34

 13
 34
 16
 33
 72
 12
    Condition
      Reference
0.1-0.2 mm
Berak,  '963
Rhyodacites,
  East Slovakia
Dacites, Slovakia
 45
 49
 85
 96
 41
                  0.1-0.2 mm
                     Berak, 1963
                                         3-41

-------
                     TABLE  3-19.  (continued)
Medium
CsKd, ml/q
Condition
Reference
Quartz Porphyries
Central Bohemia
Bohemia

North Bohemia
SATURATED ROCKS (contain
Trachytes,
North Bohemia
Slovakia
Andesites,
Central Slovakia

Pyroxenic, East
Slovakia




Garnetic, East
Slovakia
Altered, East
Slovakia
Pyroxenic,
Central Slovakia
Propyllitized
Granul. pyrox.,
vitreous
Spillites,
albitized
Diabases,
Bohemia







North Moravia
52 0.1-0.2 mm
32
730
104
32
no quartz or feldspathoids)
0.1-0.2 mm
89
16
54 0.1-0.2 mm
52
77
34
58
' 73
27
39
160
277
56
317
10 0.1-0.2 mm
61 0.1-0.2 mm "
14
37
93
300
640
540
640
19
Berak, 1963


Berak, 1963
Berak, 1963











Berak, 1963
Berak, 1963








                                3-42

-------
                      TABLE 3-19.   ('continued)
Medium
CsKd,  ml/g
Condition
Reference
Melaphyres,
basalt
North Bohemia


Slovakia
Basalts, North
Bohemia,
Vitreous




UNDERSATURATED ROCKS
Picrites, North
Moravia








Teschenites with
olivine, North
Moravia





Peridotite,
North Moravia
Pyroxenites, NW
Bohemia

0.1-0.2 mm
1330
108
37
122
900 0.1-0.2 mm
246
130
910
122
167
(nonfeldspathoidal )
456 0.1-0.2 mm
257
526
89
185
257
526
1330
456
30
318 0.1-0.2 mm
199
56
234
258
70
96 0.1-0.2 mm
2 0.1-0.2 mm
0
Berak, 1963



Berak, 1963





' Berak, 1963








Berak, 1963





Serak, 1963
Berak, 1963

                                3-43

-------
                    TABLE 3-19.   (continued)
Medium
CsKd. ml/g
Condition
Reference
Monchi quite,
with olivine,
North Moravia
Ouachitite,
with apatite
North Maravia

Olivine basalts,
North Bohemia

East Slovakia
Central Slovakia
North Bohemia
Andesitic, Central
Slovakia

UNDERSAT'JRATED ROCKS
Phono lites, North
Bohemia



Tephrites, North
Bohemia
Chabazite
Leuci te
Chabazite
Nephel ine
Nepheline-leucite
Leuci te

Basanites, North
Bohemia, Nepheline-
leucite
Nephel ine

95 0.1-0.2 mm Berak, 1963
28 0.1-0.2 mm Berak, 1963
52
1215 . 0.1-0.2 mm Berak, 1963
1370
1685
75
1023
170
185
(feldspathoidal)
55 0.1-0.2 mm Berak, 1963
34
1150
1600

594 0.1-0.2 mm Berak, 1963
140
589
517
675
1134
1062
397 0.1-0.2 mm Berak, 1963
98
44
                               3-44

-------
                              TABLE 3-19.   (continued)
        Medium
Leucite
Nepheline
Nephelinites, NW
  Bohemia, Metilite
Leucite
Glivine-nepheline
  basalt
Hauyne-nepheline
  basalt
Leucitites, NW Bohemia
Nepheline
Melilitites, olive-
  hauyne basalt
MINERALS
Quartz
Agate
Chalcedony
Opal
01ivine
Humi te
Chondrodite
Thaumasite
Leucophanite
Zircon
Topaz
Kyanite
Sillimanite
Staurolite
 CsKd, ml/q
1550
3240
52

34
329

117
571

700
3746
1328
398
109

145
0
2
11
8-61
5
8
0
16
2
3
4
43
3
1
    Condition
      Reference
0.1-0.2 mm
Berak, 1963
0.1-0.2 mm
Berak, 1963
0.1-0.2 mm
0.1-0.2 mm
Berak, 1963
Berak, 1963
                                         3-45

-------
                                TABLE  3-19.   (continued)
        Medium
Garnet, alamandine
Garnet, grossular
Garnet, andradite
Garnet, uvarovite
Hydrogarnet
Vesuvianite
Sphene
Rinkita
Axinite
Hemimorphite
Beryl
Oioptase
Tourmaline
Wollastonite
Rhodonite
Hypersthene
Aegirine
Augite
Diopside
Enstatite
Jeffersonite
Tremolite
Chrysoti1e
Sepiolite
Attapulgite
Polygorskite
Zoisite
Talc
Pyrophyllite
Biotite
Muscovite
Penninite
 CsKd. ml/q
0
3
0   •
5
54
16
3
9
0
6
7
8
3  .
0
3
18
50
14
6
1
4
8
0
900
89
150
33
0
8
21
15
2
Condition
                                                                        Reference
                                          3-46

-------
                     TABLE 3-19.   (continued)
Medium
CsKd. ml/q
Condition
Reference
Delessite
Sen' cite
mite
Glauconite
Celadonite
Serpentine
Kaolinite
Halloysite
Allophane
Hisingerite
Smectites






Nontronite
Apopnyllite
Melilite
Feldspars
Ortnoclase
Sanadine

Albite
Oligoclase
Andes ine
Laboradorite
Scapolite
Leucite
Analcite
Pollucite
Nepheline
Soda lite
4070
21
400
87
84
17
45
81
276
257
4900
2400
355
425
545
377
345
388 0.1-0.2 m Berak, 1963
8. -
68
0.1-0.2 mm . Berak, 1963
9
58
6
9
12
28
24
26
56
54
5
20
8
                                3-47

-------
                                TABLE 3-19.   (continued)
	Medium
Cancrinite
  Zeolites:
    Mordenite
    Stilbite

    Heulandite
    Faujasite
    Hannotome
    Phillipsite
    Chabazite
    Natrolite
    Scolecite
    Thomsonite
Clinch River, TN
bottom sediments
Silty Clay,  ID

Sandstone,fine
light gray

Shaley siltstone,
carbonaceous,
black
Sandstone,  fine
light gray

Sandstone,  very
fine, silty, dark
gray
CsKd, ml/q
67
25,000
19,900
193
1230
230
416
9900
4900
20
9
7050
2326 - 1 hr
50,152 - 3 days
88,048 - 7 days
3169-1 hr
50,152 - 3 days
82,769 - 7. days
3000
139
166
389
309
472
541
102
141
298
346
456
630
Condition

0.1-0.2 mm










pH fi
pH 9
Quartz, ill ite,
apatite
4000 urn
500 pm
<62 um
4000 um
500 um
<62 um
4000 um
500 ym
<62 um
4000 um
500 um
<62 um
     Reference
Berak, 1963
Carrigan, et al,
1967
Wilding and Rhodes,
1963
Sokol, 1970
Sokol, 1970

Sokol, 1970

Sokol, 1970
                                        3-48

-------
tracer cesium.  A 1% solid suspension was  equilibrated  for 1  day at 20°C.
Results indicated that the secondary silicates  (clays,  zeolites) removed
cesium from solution much better than the  primary  silicates  (feldspars, quartz,
etc.).

            TABLE 3-20.  THE EQUILIBRATING SOLUTION  COMPOSITION
                         PLUS TRACE 137Cs  THAT  WAS UTILIZED  IN
                         ALL OF BERAK'S  (1963)Kd WORK WITH 24
                         HOUR EQUILIBRATIONS  AT 20°C
Ions
Ca2+
Mg2+
Na"1"
SO2"
NOj
Cl"
HCO;
mg/1
80.2
12.2
23.0
96.0
62.0
35.4
61.0
0
0.0001
0.001
0.01
0.02
0.03
0.04
0.05
0.08
0.10
154
147
137
357
483
695
727
634
373
265
     Jacobs (1962) used vermiculite to  illustrate  the  favorable  10  A spacing
caused by potassium added to the cesium in the  influent,  solution.   Cesium
selectivity (Kd) increases through.a maximum while the cation  exchange
capacity of the vermiculite decreases as shown  in  Table  3-21.
            TABLE 3-21.  EFFECT OF ADDITION OF  POTASSIUM TO THE
                         INFLUENT ON THE SORPTION  OF CESIUM BY
                         VERMICULITE FROM 0.5M  NaNOa CONTAINING
                         THE MASS EQUIVALENT OF 2  uc 137Cs/ml
                         (JACOBS, 1962)
                      Potassium        CsKd,      Final CEC,
                   Concentration. M     ml/g      meg/100  q
                                                 75.4
                                                 74.9
                                                 70.2
                                              '   60.0
                                                 38.1
                                                 26.2
                                                 17.9
                                                 15.3
                                                  9.1
                                                  8.6
                                   3-49

-------
     The cation exchange capacities of five standard clays and seven soil
samples were determined using cesium, ammonium and manganese ions by Beetem
et al. (1962).  These authors preferred the use of cesium chloride-   Cs to
accurately determine clay and soil capacities for samples as small as 30 mg.
     Rhodes and Nelson (1957) studied the effects of pH and sodium nitrate
concentrations on cesium distribution coefficients using a glaciofluviatile
silty sand composite recovered from Hanford wells.  The results are given in
Tables 3-22 and 3-23.  In high salt solutions cesium Kd values were much less
affected by pH variations than cesium tracer in water only.

              TABLE 3-22.  EFFECT OF pH ON 137Cs ADSORPTION.  SOIL
                           WAS A COMPOSITIE GLACIOFLUVIATILE SILTY
                           SAND FROM HANFORD, WASHINGTON.  (RHODES
                           AND NELSON, 1957)
                                      CsKd, ml/g
H2° NaCl, 4M
36.5
138.0
200
>200 •
>200
>200
>200
>200
14.0
12.2
17.4
18.0
18.0
12.0
16.0
18.8
                           0.4
                           1.8
                           3.7
                           4.9
                           7.0
                           7.8
                           8.5
                          10.1

       TABLE 3-23.  EFFECT OF NaNOa CONCENTRATION ON 137Cs ADSORPTION.
                    SAME COMPOSITE QLACIOFLUVIATILE SOIL WAS USED AS
                    IN TABLE 3-20 (RHODES AND NELSON, 1957)
NaN03,M
0.25
0.50
1.0
2.2
3.6
4.8
6.0
CsKd, n
245
195
100
50
30
20
18
                                   3-50

-------
     Wahlberg and Fishman (1962) determined cesium Kd values with kaolinite,
illite, montmorillonite and halloysite over a large cesium concentration
range and in competition with sodium, potassium, calcium or magnesium.  The
type and amount of competing ions and the cesium concentration determined
the cesium Kd variation with potassium having the most depressing effect.
     Parsons (1962) determined a    CsKd value of 100 in sand, sandy loam,
silt and humus at a groundwater pH of 5.4 to 6.6.
     Janzer et al. (1962) determined cesium Kd values on a number of 100 to
200 mesh rock samples.  The samples were divided into two parts.  One part
of the sample was equilibrated with the solution without cesium tracer, and
finally with a cesium tracer plus the same solution.  The other sample half
was directly equilibrated with the cesium traced solution.  There were minor
differences between the Kd ranges of the two sample groups.  The five
equilibrating solutions varied from 1,000 to 45,000 mg/1 in dissolved solids.
Equilibration times were up to 30 days and each experiment was run in tripli-
cate.  The rock type descriptions and cesium Kd values are given in Table
3-24 as minimums for the highest dissolved solids solutions to maximums for
lowest dissolved sol.ids solutions.  A membrane technique was used to.keep
solid and cesium traced solution phases separated.

       TABLE 3-24.  SUMMARY OF CESIUM Kd VALUES (JANZER ET AL., 1962)
                               Kd. ml/q
>le Number
45
95
320
331
350
365
380
479
510
526
552
620
653
680
Minimum
613
1,034
9
841
27
36
445
419
51
1,176
536
661
302
396
Maximum
10,300
13,400
48
15,530
170
1,140
6,657
3,600
2,824
8,990
7,150
8,250
3,766
5,280
Rock Description
Sandstone, calcite cement
Siltstone, sandy


Gypsum rocks and siltstone

Dolomite
Gypsum rock
Dolomite


Siltstone, silty sandstone,
dolomite cement
Gypsum rock, siltstone,
                                            sandstone
                                   3-51

-------
     Baetsle et al. (1964) reported several cesium  Kd  values  as  a  function  of
pH and water composition for Mol , Belgium, soil.  The  Kd  values  are  given  in
Table 3-25 and may be directly compared with strontium Kd values on  the  same
soils.  The causes of the low deionized water Kd value relative  to the cesium
Kd values with mains (tap?) water are not known.  Mercer  (1967)  gave cesium Kd
values as a function of sodium and calcium normality for  clinoptilolite, a
natural  zeolite.  The Kd values varied from about 5 x  10  ml/g at  0.1N NaCl to
about 3 x 105 ml/g at 0.001N NaCl .  At 0.001N CaCl2 the clinoptilolite cesium
Kd value was about 3 x 10  ml/g and 3 x 104 ml/g at 0.1N  CaCl.  Certain of
the secondary silicates, such as some of the zeolites, efficiently  remove
cesium from solution even when the concentrations of competing  ions are  rela-
tively large.  Rancon (1967) reported cesium Kd values of 21,000 on illite,
300 on kaolinite and 2000 ml/g on montmorillonite.
    TABLE 3-25.   VARIATION OF TRACE CESIUM Kd VALUES WITH pH  (BAETSLE
                 ET AL., 1964)
              Soil  Type
           Eollan Sand
             Horizon A
             Horizon B
             Horizon C
           Mol White Sand
           Mol Lignitic Sand

     Ames and Hajek (1966) statistically analyzed cesium adsorption data.
Hajek and Ames (1968) also showed the problems associated with determining
cesium distribution coefficients on the A horizon of Burbank  loamy fine  sand
without first equilibrating the soil with the solution minus  the cesium  tracer.
The resulting cesium.Kd values depend on the solution to soil ratio used in
the equilibration, a's seen in Tables 3-26 and 3-27.

     A curve for trace cesium loading on a 1 g Burbank soil is also given.
Even these small columns, when used for Kd determinations, should be  pre-
equilibrated with  the solution minus the radioactive tracer.
                                    3-52
Kd, ,nl/q
Deionized
Water, pH4
25-38
'31-32
226
50
-
Mains Water,
PH7.7 pH3
51 22.4
145 . 68
89 79
30
3.9-7.1 -
Groundwater,
pH4
10.1 '
-
-
1.3-3.2
3.4-4.3

-------
             TABLE 3-26.
Solution
3N NaN03
0.5N Nad
0.5N CaCl
SOLUTION:SOIL RATIO EFFECT ON TRACE
CESIUM EQUILIBRIUM DISTRIBUTION
COEFFICIENTS (HAJEK AND AMES, 1968)

   	CsKd,  ml/q	
   	So1ution:Soi1	
TOO
              50
                                                 TO
                 466
                1091
                5211
          380
         1445
         5362
                       271
                      1354
                      2420
                                                         206
                                                        1173
                                                        1260
        TABLE 3-27.  CESIUM EQUILIBRIUM DISTRIBUTION COEFFICIENTS
                     FOR SAND, SILT AND CLAY FRACTIONS OF BURBANK
                     LOAMY FINE SAND IN 0.5N NaCl (HAJEK AND AMES,
                     1968)
CsKd, ml/g
Clay
Solution: Soil
500 200 100
6557 6008 5364
Silt
Solution: Soil
50 10 5
548 472 408
Sand
Solution:Soil
50 10 5
1657 1014 893
     Three soils representative of the Hanford Site, the Ritzville silt loam,
Burbank loamy sand and Ephrata sandy loam, were studied in detail  for their
ion exchange properties with cesium (Routson, 1973).  The three soils were
thoroughly characterized in terms of their physical  properties (size fractiona-
tion), chemical properties (exchangeable Ca, Mg, Na and K in each soil horizon
and the total cation exchange capacity of each) as well as their mineralogical
properties (semi quantitative mineralogy of each size fraction, plus organic
carbon and calcite content of each fraction).  Cesium equilibrium distribu-
tion coefficients from 0.2N NaCl were determined on the "as received" soil
horizon samples, sodium-based horizon samples and horizon samples with organic
carbon removed, calcite removed and both removed.  The results are given in
Table 3-28.  Soil samples were pre-equilibrated with a 0.2N NaCl  except for
                                   3-53

-------
the "as received" cases.  The Ritzville soil averaged about 12 meq/100 g, the
Ephrata soil about 7 meq/100 g and Burbank about 5 meq/100 g cation exchange
capacities.   The Kd values were directly correlated with the soil cation
exchange capacities.

            TABLE 3-28.   TRACE CESIUM ADSORPTION BY THREE HANFORD
                         SOILS FROM A 0.2M NaCl SOLUTION (ROUTSON,
                         1973)
Ritzville
A12
81
Cca
C
Burbank
A12
AC
AC2
1C
Ephrata
A12
81
IB2
IIC
As Received

4869
4332
3111
3529

2775
2680
2573
1588

3941
4696
3491
1919
Na-based

4869
4930
5455
. 6169

1964
2571
3533
1633

2385
2410
1950
1553
Organic C
Removed

2750
2154
2608
3510

2467
2682
2513
1155

3027
2835
2402
2012
Calcite
Removed

5980
6486
4060
5174

2540
2448
2070
1473

5230
4071
2316
2423
Organic C+
Calcite
Removed

4069
3152
2592
2578

3024
2338
2036
1010

2671
2901
1850
836
     R. J. Serne of PNL, 1973, also determined cesium distribution coefficients
from high salt, high pH solutions.  The solution compositions and sediments
are the same as those described in Table 3-9 and 3-10. The results of the high
salt work are provided in Table 3-29.   Each cesium Kd was determined in tripli-
cate.  The values given in Table 3-29 represent the average of the three Kd
values.
                                   3-54

-------
Solution
I
13.41
9.37
9.06
8.11
10.93
7.88
8.74
II
6.00
3.47
3.42
3.08
3.70
2.04
2.79
III
1.80
1.37
1.27
1.37
1.83
0.93
1.14
IV
0.85
0.61
0.58
0.73
2.92
1.20
1.47
V
12.02
7.14
7.21
7.68
11.30
7.97
9.49
        TABLE 3-29.  CESIUM  Kd VALUES BETWEEN SEVEN GLACIOFLUVIATILE
                     SEDIMENTS AND THE SOLUTIONS OF TABLES 3-9 AND
                     3-10
          Sediments
              1
              2
              3
              4
              5
              6
              7
Migration Results
     Field Studies--
     Mawson (1956) and Evans (1958) reported the results of studies on the
movement of fission products disposed to ground in a sand at Chalk River,
Ontario.  The rate-of movement was followed with a series of wells from which
groundwater samples were collected and analyzed.  Cesium moved the least and
at the lowest velocity of several radionuclides including strontium, cerium
and ruthenium in dilute, neutral wastes.  Cesium was also the slowest radio-
nuclide to move in acid wastes as well because neutralization was soon accom-
plished due to soil buffering.
     The drilling of core wells into abandoned covered trench disposal sites
at Hanford was used to determine cesium distribution in the field.  Raymond
(1965) found that the bulk of the 137Cs was contained in the upper 3 ft
beneath the disposal site.
     Brown (1967) reviewed the hydrology and geology of the Hanford areas and
gave well profiles of    Cs at two disposal sites.  Anomalies in the radionu-
clide distribution were correlated with stratigraphic features such as a
caliche (CaC03) bed and a loess lens.  The average rate of downward moisture
movement decreased from 1.5 m/yr in 1958 to 1959 to 0.5 m/yr from 1963 to. 1966,
as a result of disposal site retirement in 1956.  A study to determine    Cs
                                   3-55

-------
leaching by groundwater of contaminated soils from core wells showed that
                                                    137
50 column volumes were required to leach 11% of the    Cs.  Another 450
column volumes removed an additional 4% of the    Cs.  Trace    Cs was added
to the groundwater.  Distribution coefficients on sediments from below the
groundwater table were 300 ml/g for    Cs.  Eight relatively long-lived radio-
nuclides were detected in the groundwater including    Cs.     Cs was adsorbed
                                                                     Q
in the upper 10 m of 216-S-l and -2 covered trenches.  About 1.5 x 10  liters
of low salt wastes containing about 2000 Ci of    Cs were disposed to 216-S-l
and -2 prior to shutdown in 1956.
     Haney (1967) reported the disposition of 137Cs in the 216-BY covered
trenches that had received 3.4 x 10  liters of U-Plant high salt scavenged
waste that contained 4.1 x 10  Ci of undifferentiated 6 emitting radionuclides
including 3300 Ci of    Cs.  The 8 BY Covered trenches were in service from
December 1954 to December 1955 at Hanford.  Monitoring and sampling wells
drilled into and near the ground disposal facility showed that the cesium peak
occurred at 28.7 uCi/g soil at about 3 meters below the bottom of the trench
                                                              90
and rapidly decreased with depth.  Approximately 12,900 Ci of   Sr also were
                             90
contained in the waste.  The   Sr peak soil loading occurred at about 5 meters
below the bottom of the trench.' Groundwater at the site was 67 m below
ground surface.  Soil samples below 23 m contained less than 2 x 10"  uCi/g
of either cesium or strontium.
                      137
     The migration of    Cs from Oak Ridge seepage pits was studied by
                                                       137
Lomenick et al. (1967).  It was found that most of the    Cs in the pits was
tied up in an illitic clay sludge.   Water movement was found to be 0.15 m/day
while    Cs movement was 0.127 cm/yr.
     Magno et al. (1970) determined radionuclide migration through the efflu-
ent lagooning system of the Nuclear Fuel Services plant in Western New York.
From their analytical data, they estimated that 75% of the 134Cs and 137Cs
discharged from the plant were deposited in the sediments in the lagoon system.
From 40 to 95% of the cesium at the lagoon system exit was associated with
suspended solids.
     In 1973, the 241-T-106 underground liquid high level waste storage tank
at Hanford leaked about 435,000 liters of waste into the surrounding sediments
(Anon., 1973).  A series of wells were drilled to ascertain the locations and
                                   3-56

-------
movements of the various radionuclides in the sediment column.  Monitoring
and core sample analysis showed the relative mobilities of the radionuclides
in the 1M NaOH, 5M NaNO., solution.  Plutonium was the least mobile, but
cesium mobility was only slightly more rapid.  None of the radionuclides
migrated below 27 m.  The water table was at 62 m depth.
     Brookins et al. (1975) and Brookins (1976) have reported that a study
of the 135Ba and T37Ba daughters of 135Cs and 137Cs found at the prehistoric
natural reactor site at Oklo in Gabon indicate the probable retention of fis-
sion product cesium at the reactor site.
     Laboratory Studies—
     Nishita et al. (1960) studied the effects of potassium and cesium on the
           137
release of    Cs from three soils to plants.  Addition of potassium was found
to be inefficient in reducing    Cs uptake by plants.  The addition of non-
                                                    137
radioactive cesium always increased plant uptake of    Cs. even when the
cesium level was injurious to the plant.
     Davis (1961) reported that the soil exchange capacity was essentially
constant above pH 3.5.   Cesium was most effectively desorbed from soils by
neutral salts, and more readily by potassium than by sodium or calcium.
Trace amounts of cesium were held more strongly by soils' than trace amounts
of strontium.
     Nishita et al. (1962) used an equilibrium method in a followup study to
determine the effects of stable cesium and potassium on the movement of
                     137       42
tracer quantities of    Cs and   K.  The concentration of stable cesium and
                                         137
potassium was an important factor.  When    Cs was only a negligible fraction
of the cesium in solution, only a negligible fraction of    Cs was adsorbed.
In equimolar mixtures of cesium and potassium, the adsorption of cesium
relative to potassium decreased as the ionic strength increased.  In low con-
                                                                   137
centrations, cesium was more effective than potassium in releasing    Cs from
soils.  In high concentrations, potassium can be as effective as cesium in
releasing    Cs.
     The adsorption of cesium by soils and its displacement by salt solutions
were investigated by Coleman et al. (1963).  Montmorillonite, illite and kao-
linite loaded with cesium were readily leached by IN KC1 or IN CaCl2.

                                   3-57

-------
Vernriculite and heated (600°C for 12 hr) potassium montmorillonite bound
cesium tightly while contacting solutions of CaCU and AlCU but not KC1 or
NH.C1 solutions.  Prolonged leaching with IN KC1 removed 97.5% of the cesium
adsorbed on heated potassium montmorillonite while IN CaC12 replaced 88%.
The exchange-displacement behavior of cesium suggests that its adsorption in
interlayer spaces allows admission of potassium and ammonium ions but
restricts the entry of calcium.  Apparent specific adsorption of cesium
against exchange with CaCl- occurred at large cesium concentrations.
     Miller and Reitemeier (1963) published some of their earlier work on
strontium leaching adding some additional cesium leaching work.  Leaching
treatments consisted of applications of 30 and 300 in. of deionized water,
0.005N NaCl and 0.005N CaCl9.   There was little downward movement of cesium.
                                       134
Assays showed that 96.6 to 100% of the    Cs was in the surface 1.4 in.  of
the soil columns (Norfolk loamy sand, Hagerstown silt loam, Fort Collins loam,
Miami silt loam, Huntley clay) after 300 in. of leaching by the above
solutions.
                         134
     Cesium K values for    Cs in the saturated subsoil at Mol, Belgium, also
were given by Baetsle et al. (1964).  The K values are listed in Table 3-30
                                                                       89
and may be compared with similar values obtained on the same soils for   Sr
    152
and    Eu.  Mol sand is nearly pure quartz.  The effect of pH on cesium migra-
tion in relation to water can be seen in the K values.  In all cases-, except
for the perturbed profile, K values decreased (cesium mobility increased in
relation to water velocity) as the pH decreased.  K values are markedly dif-
ferent even with single pH unit.
     Cesium distribution coefficients at pH 3 and the rate of cesium movement
in relation to the groundwater movement in Houthulst clays are given in
                                                                         89
Table 3-31.  These values may be compared for the same soil samples with   Sr
and 152Eu.  Note that the K values for cesium decreased in the Houthulst
clays by an order of magnitude as compared to the Mol sands.   It is unfortu-
nate that the minerals constituting the Houthulst clays were not specified.
     Carlile and Hajek (1967)  reported an example of physical  transport
through a soil column.  About 0.1% of the    Cs was found in 2 cm diameter
by 40 cm length N-Area Hanford soil columns.  Until 5 column volumes of
                                   3-58

-------
 TABLE  3-30.   K VALUES OF CESIUM  IN THE SATURATED SUBSOILS
               AT MOL, BELGIUM  (BAETSLE ET AL., 1964)
               K IS DEFINED ON  PAGE  2-30.
                 Soil Types
            Mol  sand
            Perturbed profile
            Mol  sand + eolian sand
            Mol  sand + lignite
            Mol  sand
            Mol  sand
            Mol  sand
            Mol  sand
            Mol  sand
            Mol  sand
pH4
9
125
-
76
38
48
55
38
31
25
pH3
6
431
24
20
26
32
19
14
15
13
TABLE 3-31.   CESIUM DISTRIBUTION COEFFICIENTS AND RELATIVE
              MIGRATION RATES IN HOUTHULST CLAYS AT pH 3
              (BAETSLE ET AL., 1964)


Depth, m Kd. ml/g
0-1
1-2
2-2.5
2.5-3
3-4
4-5
5-6
6-7
7-8
3-9
9-10

0-1
1-2
2-3
*3-4
4-5
5-6
6-7
7-8
8-9
9-10
111.5
77.0
83.0
47.7
-
145.2
133.5
107.0
-
103
163

-
31
29
165
342
570
693
494
462
441
Core Well
I
_£ 	 Depth, m
372
257
278
160
-
485
466
357
-
344
543
Core Well
-
104
98
550
1140
1899
2309
1646
1536
1470
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
18-19
18-19

II
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
18-19
19-20

Kd, ml/a
198
295
386
475
427
742
765
484
470
250


536
527
664
374
247
291
158
220
193
241

1C
560-
933
1286
1583
1423
2472
2548
1613
1566
834


1786
1756
2212
1246
824
970
527
733
644
804
                            3-59

-------
Columbia River water (dissolved solids =119 mg/1) had passed through the
soil, leakage did not fall below detection limits  (10  % leakage).  Effluent
samples were centrifuged, treated with hydrogen peroxide to destroy organic
material, and recentrifuged.  The peroxide treatment did not affect cesium
removal by centrifugation, which removed 50 to 75% of the cesium in either
case.  It was found that the low ionic strength river water dispersed the
soil colloids on which the    Cs was adsorbed.  The breakthrough of cesium
continued until the colloids were flushed from the column.  Preleaching with
river water rid the soil columns of colloids and resulted in no initial
cesium leakage.
     Eichholz et al. (1967) investigated the partitioning of dissolved radio-
nuclides between suspended sediment particles and aqueous solutions.  Their
work does not necessarily have a direct application to particulate migration
through soils, but it does give a feeling for the types and amounts of radio-
nuclides that can be transported on suspended solids.   Several bodies of
water were sampled and utilized by the authors in the adsorption study as
sources of suspended solids.  Some of the properties of the natural water.
samples are listed in Table 3-32.  A fission product mixture was added; the
system allowed to come to equilibrium; and the water was recovered and passed
through a mixed bed ion exchange resin.  The resin and sediment were then
counted with the radionuclides extraction results shown in Table 3-33.  It
                                                                      »
is noteworthy that a large portion (80 to 90%) of the radionuclides was asso-
ciated and traveled with the particulate matter in a highly concentrated form.
There is a direct correlation between cesium adsorption and the suspended
solids concentration, with some effect of dissolved solids in the water.
     Levi and Miekeley (1967) reported that experiments with vermiculite and
the cesium fixation phenomenon showed that a lattice contraction was not
responsible for cesium uptake as previously suggested by others.  Cesium fixed
with respect to isotopic exchange was shown to become mobile by exchange with
many other ions.  Two different "defixation" mechanisms were indicated, one
caused by strongly hydrated ions and the other by small nonhydrated ions.
Cesium ions can be exchanged for potassium, but not potassium ions for cesium.
Strontium isotopic exchange curves were obtained with mixed Cs-Sr vermiculite.
Strontium ions can be exchanged in the system but at a reduced rate in their

                                   3-60

-------
   TABLE 3-32.   PROPERTIES OF  NATURAL WATER  SAMPLES (EICHHOLZ ET AL.,  1967)
                   Suspended    Dissolved          Conductivity
      Source       Solids, ppm   Solids, ppm    pH      umhos
 Colorado River,
 Utah
 .Camp McCoy,
 Wisconsin
 Bayou Anacoco,
 Lousiana
 Lodgepole Creek
 Nebraska
 Chattahoochee
 River, Georgia


 Billy's Lake
 Georgia (swamp)
                    229
                     12
                     24
350
 60
 63
                                  68
7.5
6.9
6.2
          4.2
540
 80
 60
965
131
200
31
6.8
7.3
300
45
           45
     Solids
   Composition

95% quartz,
5% calcite,
feldspar, 11 lite,
kaolinite

30% quartz +
feldspar,
6% kaolinite, 24%
illite

30% quartz,
2% kaolinite,
6% smectite

20% illite,
80% smectite

33% quartz,
44% kaolinite
23% illite

2% quartz,
balance was
amorphous
and/or organic
        TABLE 3-33.   ADSORPTION OF  RADIONUCLIDES  ON SUSPENDED  SOLIDS
                      (EICHHOLZ ET AL., 1967)
                                  Source
                                             137
                                               Cs,
Colorado River
Camp McCoy
Bayou Anacoco
Lodgepole Creek
Chattahoochee
River
Billy's Lake
83.9
3.3
16.2
96.3
26.1
0.3
mobility.   The authors  suggest that  it may be possible to slow  migration of
various  radionuclides  in soils by mixing them with  cations fixed  like cesium

in vermiculite.

     Nishita and Essington (1967) continued work  on the affects of chelating
agents  on  89Sr, 91Y,
                       106Ru, 137Cs  and 144Ce migration in different types of
soils.   Movement by water leaching, irrespective of  soils, was  in  the order
                                      3-61

-------
 137Cs, 91Y,  144Ce < 89Sr « 106Ru.  Except in the Hanford sandy loam  (a
 California soil), practically no movement in    Cs,   Y or    Ce occurred by
 water leaching.  EDDHA was generally the least effective, while the effective-
 ness in promoting leaching of DTPA and EDTA varied with the soil type and
 radionuclide.  Soil pH probably had a large effect on complexing agent
 effectiveness, but the causes were not determined.
     Knoll (1969) reported that two waste organic solutions had detrimental
 effects on retention of radionuclides by the soil.  One was a mixture of 0.4M
 di-(2-ethythexyl) phosphoric acid and 0.2M tributyl phosphate in paraffin
 hydrocarbons (C-JQ to C^), and the other was hydroxyacetic acid.  With the
 0.4M D2EHPA-0.2TBP solution, strontium removal from a 5.0 cm  soil column
 was 70% at 10 column volumes.  Only about 12% of the cesium was removed in the
 same throughput, as was 80% of the adsorbed americium and 30% of the plutonium.
 With a 0.25M hydroxyacetic acid solution, pH 3.7, plutonium, strontium and
 americium removal was essentially complete at 10 column volumes.  As the con-
 centration of hyd'roxyacetic acid decreased, the migration rate of the strontium,
 plutonium and americium also decreased.  None of the cesium on the column was
 eluted at any of the hydroxyacetic acid concentrations.
     Salo and Saxen (1974) investigated the role of humic substances in the
 transport of radionuclides.  Atomic absorption and X-ray fluorescence were
 used to determine the stable elements bound to the humic materials fractionated
 into different molecular size groups.  Surface waters from Bothnian Bay,
 Finland, were used to study the molecular size distributions of humic sub-
 stances, and Kd values were determined by gel chromatography.   The distribu-
 tion of strontium and cesium into different molecular size groups of the
 humic substances were discussed.  Cesium concentration was directly correlated
 with the colored colloids, unlike the strontium, manganese and zinc.  It was
postulated that the Cesium was adsorbed on the colored colloids.  A wavelength
 of 350 mm was used to determine color intensity.
 Summary
     Cesium mainly exists in solutions as Cs .  Therefore, the principal
 reaction mechanism of cesium adsorption on soils and rocks is expected to be
 cation exchange.  In some instances cesium shows a direct Kd-cation exchange
 capacity relationship (Routson, 1973), and in other instances, no relation
                                   3-62

-------
between Kd and exchange capacity (Jacobs, 1962).  At trace cesium concentra-
tions competing cation effects are minor (McHenry, 1954), while at 10% or
more of the exchange capacity occupied by cesium, the Kd value beings to fall
off (Prout, 1958).  Potassium is better than sodium or calcium for desorbing
cesium (Davis, 1961).  There is a tendency for cesium to become "fixed" in
  o
10A mica-like minerals (Tamura and Jacobs, 1961; Jacobs, 1962; Tamura, 1963;
Coleman et al., 1963; Lomenick, 1967), which may be due to a slow desorption
rate rather than fixation (Levi and Miekeley, 1967; Routson, 1973).  The
secondary silicate minerals such as zeolites and clay minerals generally show
much higher cesium Kd values than the primary silicate minerals found in
igneous rocks (Berak, 1963).
     The migration rate of cesium in the groundwater is faster at lower pH
values but slower through clays than through sands due to larger clay Kd
values (Baetsle et al., 1964).  Much of the cesium that migrates in surface
waters does so adsorbed on suspended solids (Eichholz et al.,  1967; Magno
et al., 1970).
     Occasionally cesium can migrate through the soil column adsorbed on sus-
pended solids (Carlile and Hajek, 1967).  . Chelating agents and organic soil
materials have little affect on cesium migration through soils (Mishita and '
Essington, 1967; Knoll, 1969; Salo and Saxen, 1974).

References
Ahrens, L. H.  1952.  The Use of lonization Potentials.  Part I.   Ionic
Radii of the Elements.  Geochim. et Cosmochim.  Acta.   2:155.
Ames, L. L., Jr.  1959.  Zeolite Extraction of Gesium from Aqueous Solutions.
HW-62607.
Ames, L. L., Jr. and B. F. Hajek.  1966.   Statistical Analysis of Cesium
and Strontium Sorption on Soils.  BNWL-CC-539.
Angelo, C. G., N. C. Goldberg, V. J. Janzer, and W. A. Beetem.  1962.  Summary
of Kd's for Fission Products Between Ground Water and Basaltic Rocks.  IN:
Geology of the U18a Site, Buckboard Mesa, NTS Part VI.  U.S.G.S.  Report
Project Danny Boy, WT  1828.  pp. 66-81.
Anonymous.  1973.  241-T-106 Tank Leak Investigation.  ARH-2874.
Baes, C. F., Jr. and R. F. Mesmer.  1976.  The Hydrolysis of Cations.
John Wiley and Sons, New York.
                                   3-63

-------
Baetsle, L. H., P. Dejonghe, W. Maes, E. S. Simpson, J. Souffriau, and
P. Staner.  1964.  Underground Radionuclide Movement.  EURAEC-703.

Beetem, W. A., V. J. Janzer, C. G. Angelo, and M. C. Goldberg.  1962.  Sum-
mary of Distribution Coefficients for Fission Products Between Ground Water
and Granite Rocks, Climax Stock, NTS.  U.S.G.S. Technical Letter NTS-13.

Berak, L.  1963.  The Sorption of Microstrontium and Microcesium on the
Silicate Minerals and Rocks.  UJV-528-63.

Bertrand, G. and D. Bertrand.  1949.  Sur la Presence et la Teneur en Cesium
des Terres Arables. Compt. Rend.  229:533-535.

Beus, A. A. and E. A. Fabrikova.  1961.  Distribution of Cesium in the
Granites of the USSR.  Geochemistry.  1961-970-976.

Brookins, D. G.  1976.  Shale as a Repository for Radioactive Waste:  The
Evidence from Oklo.  Environmental Geology.  1:255-259.

Brookins, D. G., M. J. Lee, B. Mukhopadahyay and S. L. Bolivar.  1975.  Search
for Fission-Produced Rb, Sr, Cs and Ba at Oklo:  in IAEA Symp. 204, p. 401-414.

Brown, D. J. 1967.  Migration Characteristics of Radionuclides Through Sedi-
ments Underlying the Hanford Reservation.  ISO-SA-32.

Carlile, B. L. and B. F. Hajek.  1967.  Soil Radionuclide Adsorption and
Particulate Filtration in an N-Area Soil.  BNWL-CC-995.

Cardigan, P. H., Jr., R. J. Pickering, T. Tarmura, and R. Forbes.   1967.  Radio-
active Materials in Bottom Sediment of Clinch River:  Part A.  Investigations
of Radionuclides in Upper Portion of Sediment.  ORNL-3721, Suppl.  2A.

Chariot, G.  1957.  Qualitative Analysis and Solution Reactions.  Fourth
Edition.  Masson, Paris.

Coleman, N. T., R. J. Lewis, and D. Craig.  1963.  Sorption of Cesium by
Soils and its Displacement by Salt Solutions.   Soil Science Society of
America, Proceedings,  pp. 290-294.

Davis, J. J.  1961.  Cesium and Its Relationships to Potassium in Ecalogy.
_IN_:  Radioecology. V. Schultz and A. W. Klement (eds.).  pp. 539-556.

Eichholz, G. G., T. F. Craft, and A. N. Galli.  1967.  Trace Element Frac-
tionation by Suspended Matter in Water.  Geochim. et Cosmochim. Acta.  31:737-
745.

Eliason, J. R.  1966.  Montmorillonite Exchange Equilibria with Strontium-
Sodium-Cesium.  Am. Mineral.  51:324-335.

Essington, E. H. and W. E. Nork.  1969.  Radionuclide Contamination Evaluation -
Mil row Event.  NVO-1229-117.

Evans, E. J.  1958.  Chemical Investigations of the Movement of Fission
Products in Soil.  CRER-792.        , 64

-------
 Frysinger,  G.  R.   1962.   Cesium-Sodium Ion Exchange on Clinoptilolite Nature.
 194:351.

 Gaines, G.  L.  and H. C.  Thomas.  1953.  Adsorption Studies on Clay Minerals.
 II.  A Formulation of the Thermodynamics of Exchange Adsorption.  J. Chem.
 Phys.  21:714-718.

 Graudette,  H.  E., R. E.  Grim, and C. F. Metzger.  1966.  Illite:  A Model
 Based on the Sorption Behavior of Cesium. Am. Mineral.  51:1649-1656.

 Goldberg, M. C., V. J. Janzer, C. G. Angelo, and W. A. Beetem.  1962.  The
 Effect of Sodium Ion Concentration on Distribution Coefficients for Tuffs
 from NTS.  U.S.G.S. Technical Letter NTS-16.

 Hajek, B. F. and L. L. Ames, Jr.  1968.  Trace Strontium and Cesium Equilib-
 rium Distribution Coefficients:  Batch and Column Determinations.  BNWL-SA-843.

 Haney, W. A.  1967.  Final Report on the Effects of Ben Franklin Dam on
 Hanford.   BNWL-412.  pp. 8-10.

 Heier, K. S. and J. A. S. Adams.  1964.  The Geochemistry of the Alkali Metals.
 Phys. Chem. Earth.  5:235-380.

 Hirst, D. M.  1962.  The Geochemistry of Modern Sediments from the Gulf of
 Paria.  II.  The Location and Distribution of Trace Elements.  Geochim. et
 Cosmochim.  Acta.   26:1147-1188.

 Horstman, E. L.   1957.  The Distribution of Lithium, Rubidium, and Caesium
'in Igneous  and Sedimentary Rocks.  Geochim. et Cdsmochim. Acta.   12:1-28.

 Howery, D.  G.  and H. C.  Thomas.  1965.  Ion Exchange on the Mineral  Clinoptilo-
 lite.  J. Phys.  Chem.  69:531-537.

 Izrael, Yu.A.  and F. Ya. Rovinskii.   1970.  Hydrological  Uses of Isotopes
 Produced in Underground  Nuclear Explosions for Peaceful Purposes.  UCRL-
 Trans-10458.

 Jacobs, D.  G.   1962.  Cesium Exchange Properties of Vermiculite.  Nuclear Sci.
 Eng.   12:285-292.

 Jacobs, D.  G.  and T. Tamura.  1960.   The Mechanism of Ion Fixation Using Radio-
 isotope Techniques.  JJJ:  7th International Cong, of Soil Sci. II:  206-214.

 Janzer, V.  J., M. C. Goldberg, C. G. Angelo, and W.  A. Beetem.  1962.  Summary
 of Distribution  Coefficients for Fission Products Between Ground Water and Rocks
 from Project Gnome.  I,N_:  Hydrologic and Geologic Studies for Project Gnome.
 Part IV,  U.S.G.S.

 Klechkovskii,  V.  M. and  I. V. Gulyakin.  1958.  The Behavior in  Soils and
 Plants of Traces of Strontium, Cesium, Ruthenium, and Zirconium.  Soviet
 Soil  Sci.  3:219-230.

 Knoll, K. C.  1969.  Reactions of Organic Wastes and Soils.  BNWL-860..

 Levi, H.  VI. and N. Miekeley.  1967.   Studies on Ion Diffusion in Vermiculite.
 IN:  Disposal  of Radioactive Wastes  into the Ground.  IAEA-SM-93/12, pp. 161-
 T67.

                                     3-65

-------
Lewis, R. J. and H. C. Thomas.  1963.  Adsorption Studies on Clay Minerals.
VIII.  A Consistency Test of Exchange Sorption in the Systems Sodium - Cesium
Barium Montmorillonite.  J. Chem. Phys.  67:1781-1783.

Lomenick, T. F., D. 6. Jacobs and E. G. Struxness.  1967.  The Behavior of
Strontium-90 and Cesium-137 in Seepage Pits at ORNL.  Health Physics.  13:
897-905.

Magno, P., T. Reavey, and J. Apidianakis.  1970.  Liquid Waste Effluents from
a Nuclear Fuel Reprocessing Plant.  BRH-NERHL-70-2.

Mawson, C. A.  1956.  Waste Disposal into the Ground.  Proc. of the Inter.
Conf. on the Peaceful Uses of Atomic Energy.  9:676-678.

McHenry, J. R.  1954.  Adsorption and Retention of Cesium by Soils of the
Hanford Project.  HW-31011.

Mercer, B. W.  1967.  Ion Exchange Equilibria of Trace Cesium-137 and
Strontium-85 in Multicomponent Systems.  BNWL-SA-1173.

Merriam, C. N., Jr. and H. C. Thomas.  1956.  Absorption Studies on Clay
Minerals.  IV.  Alkali Ions on Attapulgite.  J. Chem. Phys. 24:993.

Miller, J. R. and R. F. Reitemeir.  1963.  The Leaching of Radiostrontium
and Radiocesium Through Soils.  Soil Sci. Soc. Am. Proc.  27:141-144.

Nishita, H. and E.  H. Essington.  1967.  Effect of Chelating Agents on the
Movement of Fission Products in Soils.- Soil Science.  102:168-176.

Nishita, H., E. M.  Romney, G. V. Alexander, and K. H. Larson.  1960.  Influ-
ence of K and Cs on Release of Cs 137 from Three Soils.  Soil Science.
89:167-176.

Nishita, H., P. Taylor, G. V. Alexander, and K. H. Larson.  1962.  Influence
of Stable Cs and K on the Reactions of Cs 137 and K 42 in Soils and Clay
Minerals.  Soil Science.  94:187-197.

Nork, W. E. and P.  R. Fenske.  1970.  Radioactivity in Water - Project
Rulison.  NVO-1229-131.

Nork, W. E., E. J.  Forslow, and E. H. Essington.  1971.  Radioactivity in
Water, Central Nevada Test Area.  NVO-1229-175.

Parsons, P. J.  1962.  Movement of Radioactive Waste Through Soil. 5.  The
Liquid Disposal Area.  CRER-1089.

Prout, W. E.  1958.  Adsorption of Radioactive Wastes by Savannah River
Plant Soil.  Soil Science.  86:13-17.

Rancon, D.  1967.  Mechanism of Radioactive Contamination in Consolidated
Impermeable Rocks or Rocks of Very Low Permeability.  IN:  Disposal of
Radioactive Wastes  into the Ground.  IAEA-SM-93/14, pp. 179-199 (in French).

                                   3-66

-------
Raymond, J. R.  1965.  Cesium and Strontium Distribution Beneath Liquid
Waste Disposal Sites.  BNWL-235.

Rhodes, D. W. and J. L. Nelson.  1957.  Disposal of Radioactive Liquid
Wastes from the Uranium Recovery Plant.  HW-54721.

Routson, R. C.  1973.  A Review of Studies on Soil-Waste Relationships on
the Hanford Reservation from 1944 to 1967.  BNWL-1464.

Salo, A. and R. Saxen.  1974.  On the Role of Humic Substances in the Trans-
port of Radionuclides.  Report SFL-A20, Inst. Rad. Phy., Helsinki, Finland.

Sawhney, B. L.  1964/  Sorption and Fixation of Microquantities of Cesium by
Clay Minerals:  Effect of Saturating Cations.  Soil Science Society America,
Proceedings.  28:183-186.

Schmalz, B. L.  1972.  Radionuclide Distribution in Soil Mantle of the
Lithosphere as a Consequence of Waste Disposal at the National Reactor
Testing Station.  IDO-10049.

Schneider, K. J. and A. M. Platt (eds).  1974.  High-Level Waste Management
Alternatives.  BNWL-1900, Volume I, Appendix 2.D, pp. 2.D.7-2.D.8.

Schulz, R. K., R.  Overstreet, and I. Barshad.  1960.  On the Soil  Chemistry
of Cesium-137.  Soil Science.  89:16-27.

Sokol, D.  1970.  Groundwater Safety Evaluation - Project Gasbuggy.  PNE-1009.

Spitsyn, V., V. C; Balukova, and T. A. Ermanova.  1963.  Studies on Sorption
and Migration of Radioactive Elements in Soil.  IN:  Treatment and Storage
of High-Level Radioactive Wastes.  IAEA, Vienna, pp. 569-577 (in Russian).

Sreekumaran, C., K.  t. Pillai, and T. R. Folsom.  1968.  The Concentrations
of Lithium, Potassium, Rubidium and Cesium in Some Western American Rivers
and Marine Sediments.  Geochim. et Cosmochim. Acta.  32:1229-1234.

Tamers, M. A. and H. C. Thomas.  1960.  Ion-Exchange Properties of Kaolinite
Slurries.  J. Chem.  Phy.   64:29-32.

Tamura, T." 1963.   Cesium Sorption Reactions as Indicator of Clay Mineral
Structures.  IN.:  Proc. Natl. Conf. Clays, Clay Minerals.  10:389-398.

Tamujra, T.  1963.   Selective Ion Exchange Reactions for Cesium and Strontium
by Soil Minerals.   Colog. Intern. Retention Migration Ions Radioactifs Sols,
Saclay, p. 95-104,

Tamura, T. and D.  G. Jacobs.  1960.  Structural Implications in Cesium
Sorption.  Health Physics.  2:391-398.

Tamura, T. and D.  G. Jacobs.  1961.  Improving Cesium Selectivity of Bentonites
by Heat Treatment.  Health Physics.  5:149-154.
                                   3-67

-------
Taylor, S. R.  1964.  Abundance of Chemical Elements in the Continental
Crust:  A New Table.  Geochim. et Cosmochim. Acta.  28:1273-1285.

Taylor, S. R.  1964.  Trace Element Abundances and the Chondritic Earth
Model. Geochim. et Cosmochim. Acta.  28:1989-1998.

Taylor, S. R. and A. J. R. White.  1966.  Trace Element Abundances in
Andesites.  Bull. Volcano!.  29:177-194.

Thomas, H. C.  1967.  The Thermodynamics of Ion Exchange on Colloidal Mate-
rials with Applications to Silicate Minerals.   Thesis.  University of North
Carolina.

Turekian, K. K. and K. H. Wedepohl.  1961.  Distribution of the Elements in
Some Major Units of the Earth's Crust.  Bull.  Geol. Soc. America.  72:175.

Vinogradov, A. P.  1959.  The Geochemistry of Rare and Dispersed Chemical
Elements in Soils.  Consultants Bureau, Inc.

Wahlberg, J. S. and M. J. Fishman.  1962.  Adsorption of Cesium on Clay
Minerals.  U.S.G.S. Bulletin 1140-A.

Wilding, M. W. and D. W. Rhodes.  1963.  Removal of Radioisotopes from   .
Solution by Earth Materials from Eastern Idaho.  IDO-14624.
                                   3-68

-------
COBALT
Natural Soil and Rock Distributions
     The average cobalt concentrations found in igneous and sedimentary rocks
are listed in Table 3-34.  As can be seen, the cobalt content of rocks steadily
decreases from the dark ultramafic rocks to the light-colored granitic rocks.
The ultrabasic rocks usually contain the bulk of the cobalt in olivines and
pyroxenes, but an occasional sulfide body in these rocks provides a workable
ore deposit (Young 1957).  Cobalt normally is enriched in areas of low redox
potential in the environment.
      TABLE 3-34.  COBALT CONCENTRATIONS IN IGNEOUS AND SEDIMENTARY
                   ROCKS IN ppm (YOUNG, 1957)
	Igneous	  	Sedimentary	
                         Granitic
Ultramafic  Basaltic  High Ca  Low Ca  Syenite  Shale  Sandstone  Limestone
    150        48        71        1      19       0.3        0.1

     During weathering processes, cobalt is separated from manganese and
iron because its (III) oxidation state is normally unstable unless cotn-
plexed.  However, certain bacteria are known to liberate chelated Co(III),
as stable amino acid complexes from the soil.  Little is known, as yet, of
these reactions; but the-technique of predicting mineral solubilities for
cobalt based on thermodynamic data may not be entirely applicable if the
bacterial reactions are  ignored (Moore, 1964).  -Vinogradov (1959) reported
the average soils content of cobalt as 8 ppm.  Cobalt contents of up to
1.96 wt% CoO have been reported for calcite (Hacquaert, 1925).
Brief Chemistry
                                     eg
     Only one stable cobalt nuclide,   Co, is known to exist in nature.
Cobalt-60 is the principle cobalt activation product radionuclide.  Cobalt
oxidation states include Co(II) and Co(III).  Cobalt is geochemically similar
to the other ferromagnetic and closely related transition metals, iron and
nickel.
     Sulfides of nickel  and cobalt constitute the chief cobalt ore minerals.
   , «                                                 o
Co   ionic radius in octahedral coordination is 0.74 A, intermediate between

                                   3-69

-------
Ni+2 (0.70 A) and Fe+2 (0.77 A) (Shannon and Prewitt, 1969).  The relative
stabilities of the Co(II) and Co(III) states are greatly affected by complex-
ing in aqueous solutions (Latimer, 1952).  Although Co   cannot exist in
aqueous solution, complexing can stabilize the trivalent state which would
normally decompose water.
Solid Phase and Solution Equilibria
     The sources for thermodynamic data for various solid phases were:  Sillen
and Martell (1964), Co(OH)3, CoC03, CoOOH; Wagman et al.  (1969), CoO, CoHP04;
Robie and Waldbaum (1968), CO^POJ^; Chase et al. (1975), Co,0..  The sta-
bility of cobalt solid phases depends on the pH and the oxidation-reduction
environment.  For example, in acidic (pH 6) and oxidizing (p02 0.68) environ-
ments, the solids in increasing order of stability are:   Co(OH)3, Co(OH)2
pink, CoHP04, Co3(P04)2, CoC03, Co304, and CoOOH as shown in Figure 3-7.
With the change in oxidation-reduction environment, all  the minerals except
Co(OH)3, Co30», and CoOOH would stay in their relative positions.  The latter
minerals would decrease in stability.  With the change in oxidation-reduction
conditions from the extremely oxidizing condition to moderately well oxidized
or reducing conditions, (p02 > 20), CoC03 would be the most stable mineral.
Under extremely reducing conditions such as pO^SO, CoS would be the most
stable mineral.  Cobalt sulfide (CoS) is the chief cobalt ore mineral.  In
general, the stability of all the cobalt minerals increases with an increase
in pH.
                        2+       3+
     Cobalt exists as- Co   and Co   and forms solution complexes with common
                         2
soil anions (OH , Cl , S0|~, N03).  The relative concentration of these solu-
tion species in equilibrium with CoCO^ and in an oxidizing environment (09 =
   n CQ                              *                                   £
10     ) is depicted in Figure 3-8.  If a solid phase other than CoC03 or a
constant concentration of Co is chosen for the diagram,  the lines will shift
up and down, while the relative concentration of various  solution species
would stay the same.  The thermodynamic data for CoS04 were obtained from
Sillen and Martell (1964) the hydrolysis species data were obtained from
Baes and Mesmer (1976).  Data for all the other species were obtained from
Wagman et al. (1968, 1969).
                                   3-70

-------
               3456
Figure 3-7.  The relative stability of cobalt solids in an oxidizing
             soil environment CpOz(g) = 0.68 atm], pCa2+ = 2.5 and
             phosphate levels in equilibrium with Variscite and
             Gibbsite (V and G), Dicalcium Phosphate Oihydrate (DCPD)
             and Octacalcium Phosphate (OCP).
       -24
                                    r
 Figure 3-8.   The activity of various  cobalt ions in equilibrium with
              CoC03(s)  in an oxidizing soil  environment Cp02(g)  =0.68
              atm],  pC02(q) = 1.52 atm, pCl" = pS042 = 2.5,  pN03 = 3.0
and
                        - = 5.0.
                                3-71

-------
                                                            2+
     It can be seen from the figure that cobaltinous ion (Co  ) and its solu-
tion complexes are the most stable and dominant solution species in the oxidiz-
ing environment.  The most dominant solution species up to a pH value of
                       2+
approximately 9.5 is Co  .
                                                                 2+
     Beyond pH 9.5, Co(OH)° becomes dominant.  The activity of Co   ion in
solution decreases 100-fold with an increase of a pH unit, whereas the acti-
vity of Co(OH)°2 does not change with the pH.  These predictions are similar
to Baes and Mesmer (1976).   If CoCOg was the solid phase that controlled the
solubility, the total cobalt concentration in solution at pH 7 would be
                -7                      -1?
approximately 10" M and approximately 10  M at pH 9.5.
Experimental Adsorption Results
     Spencer and Gieseking (1954) studied cobalt adsorption on and release
from Wyoming bentonite and Swygert clay.  Cobalt was adsorbed more strongly
than calcium.   They suggested that cobalt was adsorbed on the clays as
monovalent hydroxy-cobaltous ions and as divalent cobaltous ions.
     Haney (1957) described a test disposal of TBP (tributyl phosphate)
scavenged wastes produced by uranium recovery from the Bismuth Phosphate
Process.  The Bismuth Phosphate Process recovered only weapons grade plutonium
from the irradiated, solubilized fuel elements.   Rhodes and Nelson (1957)
referred to the same waste as Uranium Recovery Plant Waste and gave a cobalt
content of from 4 x 10"5 to 4 x 10   yc/cm  of   Co.  Glaciofluviatile soil
column studies were undertaken and little   Co was removed by adsorption on
the soil.  The   Co was reported to form a complex ion, probably a cobalti-
cyanide complex, that did not undergo ion exchange reactions with the soil.
The presence of nonexchangeable   Co became a limiting factor in the disposal
of some wastes.  Sorathesn et al. (1960) reported that   Co behaved like a
colloid at pH 6 to 9, and gave a   Co Kd value of 23,624 on ill He.
     Friend (1963) reported that suspended solids in a pond environment took
up   Co rapidly, with suspended illite having the most selectivity for cobalt
removal of the clay minerals.  Wilding and Rhodes (1963) studied the effects
of 100 ppm citrate and EDTA on cobalt equilibrium distribution coefficients
with the zeolite clinoptilolite and Wei 1-81 sediments.  At pH 6.5, citrate
enhanced cobalt adsorption on clinoptilolite, the Kd value being 48 with no

                                   3-72

-------
treatment and 102 with 100 ppm citrate added.  With EDTA addition at pH 6.5,
the .cobalt Kd value fell from 100 to 1 ml/g.  Well-81 sediments adsorbed
about the same amounts with and without citrate.  At pH 7.7, the cobalt Kd
was 30 ml/g with no citrate and 26 ml/g with citrate.  At 7.7, the cobalt Kd
was 1000 ml/g with no EDTA and 1.2 ml/g with EDTA.  Tiller et al. (1963)
found no correlation between cobalt adsorption and soil  mineralogy.   Hawkins
(1964) described a process of cobalt removal from waste solutions by scaveng-
ing on manganese dioxide precipitate and removal of the remaining cobalt by
ion exchange on lignite, soil and the natural zeolite clinoptilolite.  Dis-
tribution coefficients for cobalt were 56, 800 and 24 ml/g for Idaho soil,
lignite and clinoptilolite, respectively.
     Gonzalez and Gomez (1961) found that most of the cobalt in Andalusian
soils was found in oxide and hydroxide gels associated with clays.   There was
a positive correlation between cobalt content of the soil and Feo^3  content.
     Basu and Mukherjee (1965) obtained cobalt isotherms on montmorillonite
clay.  Cobalt exchange behavior was very similar to that of potassium,  ammo-
nium and nickel cations.
     Humic acid was isolated from the A horizon of the Emory silt loam and
equilibrated-wfth Co   and   Co chloride solutions by Dunigan and Francis
(1972).  Subsequent salt solution washings of the humic acid removed 91 to
99% of the 60Co.
     James and Healy (1972) showed the adsorption isotherm for 1.2 x 10"  M
  +2
Co   on silica at 25°C, as a function of pH.  Adsorption is 10% at pH 6 and
greater than 90% at pH 8.  Computed hydrolysis data at this cobalt concen-
tration are also shown.
                                                           +2
     Tewari et al. (1972) investigated the adsorption of Co   by Fe^O^, A^O-j
and Mn02 as a function of concentration, solution pH and temperature in con-
nection with the transport of cobalt adsorbed on corrosion product oxides.
                +2
Hydrolysis of Co   was suggested as a possible mechanism for the marked
dependence of cobalt adsorption on pH and temperature.
                                   3-73

-------
Migration Results
     Field Studies—
     Haney (1S67) reported the disposition of radioactivity beneath several
Hanford ground disposal sites, among them the 216-BY covered trenches.  The
eight BY trenches were used from December 1954 to December 1955 and received
3.4 x 10  liters of radioactive U-plant, high-salt scavenged waste (O.IM^
POA3", total  dissolved solids = 350 g/liter, mostly NaNO,, pH = 9.5, 60Co =
      -3          -53                      *
4 x 10"  to 4 x 10~  yCi/cm ).  The waste contained enough phosphorus (some
of it as degraded tributyl phosphate) to complex most of the cobalt in a neu-
tral to anionic species that essentially passed through the soil column
without reaction.  The groundwater beneath the trenches was contaminated with
1 x 10   uCi/cm  of   Co.   The   Co on the soil column (68 m to the water
table) was uniformly distributed at concentrations of from 1 x 10"  uCi/g of
              -4
soil to 1 x 10   uCi/g of soil.
     Magno et al. (1970) reported that   Co constituted only <0.1% of the
beta activity in surface drainage from the Nuclear Fuel Services plant Lagoon
system in western New York State.  Even that small amount of   Co constituted
40% of the   Co entering the system of lagoons because of the small amount of
  Co disposed to the lagoon.  The   Co in the effluent was relatively small,
but was all in solution and thus able to travel with the surface water.   The
implication is that the "soluble"   Co was migrating as a complexed species.
     Duguid (1976) studied the seepage waters from burial grounds at Oak Ridge
used for the disposal of intermediate-level liquid wastes from 1962 to 1965.
Trench 7 seepage contained   Co combined in an organic complex that was  not
readily adsorbed by the surrounding shale.  Groundwater, as well as surface
     , contained these ree
     Laboratory Studies—
     The chemical mechanism of   Co transport in groundwater from intermediate-
level  liquid waste Trench 7, Oak Ridge, Tennessee, was studied by Means  et al.
(1976).  A small volume groundwater seep 50 m east of Trench 7 contained   Co
in soil concentrations of 10  to 10  dpm/g and water concentrations of
10  dpm/ml.  The   Co was transported as a complex with molecular weight less
than 700 according to Sephedex gel G-10 studies.  The G-10 is designed to

                                   3-74 .
water, contained these readily-transported   Co complexes.

-------
retard the passage of organic solutes with molecular weights less than 700.
The implication is that the   Co is complexed by low molecular weight natural
organics.  Migration of   Co would be negligible in the absence of organic
complexation because the soil adsorption capacity for inorganic forms of   Co
was reported to be extremely high.  Manganese oxides in the Conasauga shale
were the principal adsorbents of ionic and weakly complexed b Co.
     Saas and Grauby (1973) studied the migration of cobalt and its transfer
mechanisms from nuclear reactor to river water, irrigation water and finally
groundwater.  Three fractions of   Co were differentiated including a chelated
fraction, an exchangeable fraction and a hydrosoluble fraction.  The hydro-
soluble fraction was defined as the totality of the organic and mineral com-
ponents of the soil that are water soluble.  The different fractions in a
brown, calcareous soil were given.  The hydrosoluble fraction constituted
about 70%, the exchangeable fraction about 20% and about 10% was chelated
from river water containing   Co.  The hydrosoluble form on an alluvial
calcareous soil column varied from 10% at the point of input to 60% at 17 cm
depth.  The water soluble material was only slightly biodegradeable, even-
tually migrating to the groundwater through application in irrigation water.
The pollutants in the' river water,, including industrial  wastes and mun-icipal
sewage, are responsible for the rapidly migrating hydrosoluble fraction.   The
authors suggest that the quality of surface drainage water will have to be
taken into account during nuclear reaction siting.
Summary
     Among the various soil components, FegO^ (Gonzalez and Gomez, 1964)  and
illite (Friend, 1963; Sorathesn et a!., 1960) have been shown to have a strong
affinity for cobalt.  The adsorption of cobalt by soil minerals have been
attributed to mainly an ion exchange phenomenon (Hawkins, 1964; Basu and
Mukherjee, 1965).  With the increase in pH (especially 6 to 9) the adsorption
of cobalt increases (James and Healy, 1972; Tewari, Campbell and Lee, 1972).
This increase in adsorption with pH has generally been explained as due to
the formation of cobalt colloids (Sorathesn et al., 1960) and hydrolysis  of
  ?+
Co   (Tewari et al., 1972).  The presence of various complexing ions such as
EDTA, cyanide, and fulvic acids have been shown to reduce the adsorption  by
soil minerals (Wilding and Rhodes, 1963; Rhodes and Nelson, 1957; Dunigan and

                                   3-75

-------
Francis, 1972).  These laboratory results are consistent with the theoretical
predictions (Figure 3-8) which indicate that in acidic pH values cobalt would
                         2+
predominantly exist as Co   and with the increase in pH it hydrolyzes to
CoOH* and Co(OH)|.

     Discrete cobalt minerals have not been identified in soils.  High concen-

trations of cobalt are required to precipitate cobalt minerals presented in

Figure 3-4 under acidic pH values.  It is not likely that any of the minerals
would exist in the acidic soil environment.  However, in highly alkaline con-

ditions, CoCOg would maintain fairly low concentrations of Co in solution and

thus could exist in the soil environment.

     Cobalt is apparently relatively easily complexed by natural organics
(Duguid, 1976; Means et al., 1976) and by synthetically-produced organics

(Haney, 1967; Saas and Grauby, 1973)', and as such is able to migrate with
relative ease through the soil column with which it would normally react

by adsorption.

References

Baes, C. F., Jr. and R. E. Mesmer.  1976.  The Hydrolysis of Cations.  John
Wiley and Sons, New York.

Basu, A. N. and S. K. Mukherjee.  1965.  Interaction Between Montmorillonite
Clay and Trace Element Cations.  II.  Exchange Behavior of Cobalt, Nickel and
Chromium Ions in Clays.  J. Indian Soc. Soil Sci.  13:251-256.

Chase, M. W., J. L. Curnutt, H. Prophet, R. A.  McDonald and A. N. Syverud.
1975.  JANAF Thermochemical Tables, 1975 Supplement.  Journ. Phys. and Chem.
Ref. Data 1974.  (1)4.

Duguid, J.. 0.  1976.  Annual Progress Report of Burial Ground Studies at Oak
Ridge National Laboratory:  Period Ending September 30, 1975.  ORNL-5141.

Dunigan, E. P. and C. W. Francis.  1972.  Adsorption and Desorption of
Cobalt 60, Strontium 85, and Cesium 137 on Soil  Humic Acid.  Soil Science.
114:494-496.

Friend, A. G.  1963.  The Aqueous Behavior of 85Sr, 137Cs, 65Zn, and 60Co as
Determined by Laboratory Type Studies.  .IN.:  Transport of Radionuclides in
Fresh Water Systems.  TID-7664.

Gonzalez, G. F. and A. M. Gomez.  1961.  Geochemistry of Cobalt in Soils of
Western Andalusia.  IV.  Cobalt Clays and Correlation Between Cobalt and
Iron, Clay and Manganese Contents.  An. Edafol.  Agroboil.  23:563-572.


                                   3-76

-------
Hacquaert, A. L.  1925.  Pink Calcite from Tantara (Katanga).  Natuurw. Tijds.
Antwerp, vol. 7, p. 102.

Haney, W. A.  1957.  Disposal of High Cobalt-60 Scavenged Wastes.  HW-48862.

Haney, W. A.  1967.  Final Report on the Effects of Ben Franklin Dam on
Hanford.  BNWL-412, pp. 8-13.

Hawkins, D. B.  1964.  Removal of Cobalt and Chromium by Precipitation and
Ion Exchange on Soil, Lignite and Clinoptilolite from Citrate-Containing
Radioactive Liquid Waste.  IDO-12036.

James, R. 0. and T. W. Healy.  1972.  Adsorption of Hydrolyzable Metal Ions
at the Oxide-Water Interface.  J. Colloid Interfac. Sci.  40:65-81.

Latimer, W. M. 1952.  The Oxidation States of the Elements and Their Potentials
in Aqueous Solutions.  Prentice Hall, Inc.

Magno, P., T. Reavey, and J.  Apidianakis.  1970.  Liquid Waste Effluents'from
a Nuclear Fuel Reprocessing Plant.  BRH-NERHL-70-2.

Means, J. L., D. A. Crerar, and J. 0. Duguid.  1976.  Chemical Mechanisms of
60co Transport in Ground Water from Intermediate - Level Liquid Waste Trench 7:
Progress Report for Period Ending June 30, 1975.  ORNL/TM-5348.

Moore, L. R.  1964.  The Microbiology, Mineralogy and Genesis of a Tonstein.
Prac. Yorkshire Gebl. Soc.  34:235-292.

Rhodes, D. W-. and J. L. Nelson.   1957.  Disposal'of Radioactive L-iquid Wastes
from the Uranium Recovery Plant.   HW-54721.

Robie, R. A., and D. R. Waldbaum.  1968.  Thermodynamic Properties of Minerals
and Related Substances at 298.15°K (25°C) and One Atmosphere (1.013 bars)
Pressure and at Higher Temperatures.  Geol.  Surv. Bull. 1259.

Saas, A. and A. Grauby.  1973.  Mechanisms for the Transfer to Cultivated Soils
of Radionuclides Discharged by Nuclear Power Stations into the System:  River-
Irrigated Soil-Ground Water.   IN;  Environmental Behavior of Radionuclides
Released in the Nuclear Industry.  IAEA-SM-172/57, pp. 255-269:

Shannon, R. D. and C. T. Prewitt.  1969.  Effective Ionic Radii in Oxides
and Fluorides.  Acta. Cryst., B.   25:925.

Sillen, L. G. and A. E. Martell.   1964.   Stability Constants of Metal-Ion
Complexes.  Special Publication No.  17.   The Chemical Society, London.

Sorathesn, A. G. Bruscia, T.  Tamura, and E.  G.  Struxness.   1960.   Mineral and
Sediment Affinity for Radionuclides.  CF-60-6-93.

Spencer, W. F., and J. E. Gieseking.  1954.   Cobalt Adsorption and Release
in Cation Exchange Systems.  Soil Sci.  78:267-276.

Tewari, P. H., A. B. Campbell, and W. Lee.  1972.  Adsorption of Cobalt (+2)
by Oxides from Aqueous Solution.   Can. J. Chem.   50:1642-1648.
                                   3-77

-------
Tiller, K. G., J. F. Hodgson, and M. Peech.  1963.  Specific Sorption of
Cobalt by Soil Clays.  Soil Science.  95:392-399.

Vinogradov, A. P.  1959.  The Geochemistry of Rare and Dispersed Chemical
Elements in Soils.  Consultant Bureau.

Wagman, 0. D., W. H. Evans, V. B. Parker, I. Halow, S. M. Bailey, and
R. H. Schumm.  1968.  Selected Values of Chemical Thermodynamic Properties.
Tables for the First Thirty-Four Elements in the Standard Order of Arrange-
ment.  MBS Technical Note 270-3.

Wagman, D. D., U. H. Evans, V. B. Parker, I. Halow, S. M. Bailey and
R. H. Schumm.  1969.  Selected Values of Thermodynamic Properties.  Tables
for Elements 35 Through 53 in the Standard Order of Arrangement.  MBS Tech-
nical Note 270-4.

Wilding, M. W. and 0. W. Rhodes.  1963.  Removal of Radioisotopes from Solution
by Earth Materials from Eastern Idaho.  IDO-14624.

Young, R.  S.  1957.  The Geochemistry of Cobalt.  Geochim. et Cosmochim.
Acta.  13:28-41.
CURIUM

Natural Soil and Rock Distributions

     The element curium has not been reported to occur naturally in soil or

rocks.

Brief Chemistry
                                                238      250
     Thirteen isotopes of curium are known from    Cm to    Cm, most with

relatively short half-lives.  There are no naturally-occurring curium isotopes.
                                   3-78

-------
                                                                  239
The curium isotopes can be produced in reactors by n reactions on    Pu or by
decay of californium isotopes  (Keller, 1971).  The LWR reprocessing waste iso-
topes are given in Table 3-35.  Curium-250 also has a long half-life, but
yields are normally very low.  Curium can exist in Cm(III) or Cm(IV) oxidation
states, but stabilization of Cm(IV) requires high concentrations of fluoride
                                               _2
ions to form the anionic complexes CmFg and CmF ^ (Keller, 1971).  Only tri-
valent curium is normally stable in aqueous solutions.  The ionic radius of
  +3                                   °
Cm   in coordination number 6  is 0.986 A.  Trivalent curium forms many com-
plexes with inorganic ligands  found in wastes and in groundwaters.

                   TABLE 3-35.  CURIUM RADIONUCLIDE DATA
                                (WEAST, 1976)
                                              Decay Mode
                                                  a
                                                  a
                                                  a
                                                  a
                                                  a
                                                  a

Solid Phase and Solution Equilibria
     Curium forms  hydride,  hydroxide,  halides,  oxides,  and organometallic  com-
pounds (Keller, 1971).   Except for CmF^,  a search  for thermodynamic  data  for
these compounds was unsuccessful.   Therefore,  predictions  regarding  the  sta-
bility of formation of various compounds  of curium in soils cannot be  made
at this time.
     Curium is found in two oxidation  states [Cm(III) and  Cm(IV)] in aqueous
solution (Keller,  1971).  Curium (IV)  is  not stable  in  solutions  because  of
self-rad.iation and rapid changes to Cm(III).   Only trivalent curium  is stable.
in aqueous solutions.   Keller (1971)  report's that  the behavior of Cm  in
solution is similar to lanthanide  elements.   Most  of the lanthanides form
                                                                      -19
trivalent hydroxides in natural waters with solubility  products of <10
(Vickery,  1953).   Thomas and Jacobs (1969) found that the  activity of  curium
in 0.1M NaCl  solutions which contained approximately 10    M/l  of curium

                                   3-79
Isotope
242Cm
243Cm
.244Cm
245Cm
246Cm
247Cm '
Half-Life
163 days
32 years
17.6 years
9,300 years
5,500 years
470,000 years -

-------
decreased at pH values above 3.  They attributed  this  decrease  in  activity in
pure solutions to either precipitation or adsorption on walls.   If this  is
assumed  to be due to  precipitation,  the  solubility  product  of curium hydroxide
would be at least as  low as the  lanthanides,  and  probably much  lower.
     Since thermodynamic data for the curium  compounds is not available,  it
was decided to plot the solution species in equilibrium with  Cm(OH)3 with
an assumed log K° of  Z for the dissolution of Cm(OH),  [Cm(OH),  + 3H+ £ Cm3+
+ 3H20] in order to show the relative activity of various solution species
of curium with the changes in pH.  Keller (1971)  reviewed the curium data
and stated that little work has been performed on the  complex chemistry of
curium and since a wide variation occurs in stability  constants  determined
by various authors, these values should be regarded as relative  rather than
absolute.  The thermodynamic data used to develop Figure 3-9  were  obtained
from the following sources:  Trotman-Dickinson (1973), CmCl2"1", CmF2+, CmSO/;
Jones and Choppin (1969), CmN032+; Shalinets and Stepanov (1972),  CmOH2+,
Cm(OH)2+ and Moskin (1969), CmH2P042+.
               z-io
               Z-12
               z-w
               Z-16
               Z-18
               Z-20
               1-22
                                    6      7
                                       pH
                                             10
    Figure  3-9.
The activity of various curium ion species in equilibrium
with Cm(OH)3(s)  with pF" = 4.5, pCl  = pSO*2' = 2.5,
pNO^ =3.0 and pH0P07 = 5.0.
                                   3-80

-------
      It can be  seen  from Figure 3-9 that curium forms various  solution  com-
plexes with OH",  F", Cl", NO^, SO*", and P0^~.  The activity of all  the com-
plexes decreases  with the increase in pH.  The solution complexes  of P0?~,
Cl",  NOg, do not  contribute significantly to the total curium  concentration
in solutions.   The most dominant solution species  in pH ranges of  <3.3,  3.3-
5.7,  5.7 and up are  Cm + , Cm(OH)  , and Cm(OH)t, respectively.
Experimental Adsorption Results
                                       242
     The adsorption  characteristics of    Cm on several clays was measured in
distilled water and  0.1M NaCl by Thomas and Jacobs (1969).  Adsorption  was not
a function of ion exchange capacity because greater adsorption occurred in
0.1M  NaCl solutions  (Jacobs et al., 1966) than in distilled water  solutions.
When adsorption studies were made as a function of pH, some curium was removed
from solution even without the addition of clay at pH values above 3.  This
was due to precipitation or to adsorption on glass walls.   Addition of clay
caused a considerable increase in adsorption possibly because increased surface
                                                              242
area  was available for deposition.  Column loading studies of    Cm  tagged
0.1M  NaCl showed  normal ion exchange chromatographic breakthrough at  pH  1.
Above pH 1, the effluent deviated from normal chromatographic curves  and
                                                                242
remained constant during the entire run.  The effluent/influent    Cm concen-
tration ratio at  pH  3 was about 0.008; at pH 7 the ratio was 0.028; and at
                                     242
pH 10 the ratio was  0.150.  Although    Cm precipitated from stock solution
at pH values above 3, the column leakage of curium increased with, increasing
pH.   The leakage  is  probably caused by hydroxy and oxo-complexes forming
radiocolloids.  Movement of curium through soil is probably restricted  by
filtration and  surface adsorption of radiocolloids rather than by ion exchange.
                                                                    244
      Sheppard et al. (1976) investigated the distribution values of     Cm on
several Washington and South Carolina soils.  The 50 day distribution results
for trace    Cm in a distilled water solution are given in Table 3-36.  The
distribution for  curium between soil and solution increased with time.  The
distribution values  were computed in the same manner as a Kd value.   However,
Kd values are defined as obtained at soil-solution equilibrium which  these
values did not  attain.  The slow increase in distribution values were inter-
preted by the authors as due to the presence of colloids.  It was also  shown
that  iron and manganese oxides were effective scavengers for curium.

                                   3-81

-------
             TABLE 3-36.  CURIUM 50-DAY DISTRIBUTION COMPUTED
                          FROM SHEPPARD ET AL.,  (1976)

                              Distribution,    Monthly Change in
           Soil Identity          ml/g           Distribution

          Muscatine               1330                  +8%
            Silt Loam

          Burbank                  106                 +27%
            Loamy Sand

          Ritzville                704                 +40%
            Silt Loam

          Fuquay Sand             1850                 +26%
            0-5 cm Depth

          Fuquay Sand             1850                 +28%
            5-15 cm Depth

          Fuquay Sand             1240                 +18%
            15-50 cm Depth


Migration Results

     Field Studies—

     Duguid (1976) reported that Bondietti had shown that the alpha contamina-

tion in-water from seep S-4 along the south'side of burial ground 5, Oak Ridge,

Tenr
238r
Tennessee, was due to 3.2 x 10"7 yCi/ml of 244Cm and 3.2 x 10~8 yCi/ml of
   Pu.  The burial trenches in burial ground 5 were overflowing because pre-
cipitation infiltrated the trenches, reached the less permeable trench bottom
                                                           244       238
and flowed out the lower end of the trench in a seep.  The    Cm and    Pu
were presumably picked up by the water in its passage through the waste-filled
burial trench.  Bondietti and Reynolds (1977) reported that seepage water,
                                                             244
presumably from burial ground 5, contained 700 ± 70 dpm/1 of    Cm in the

filtrate from a 0.45 ym filter.  •

     Laboratory Studies—

     Nishita et al. (1976) determined the extractability of plutonium and
curium from a contaminated soil as a function of pH and soil components.  The
organic matter and pH influences hydrolysis, precipitation, coprecipitation
and adsorption of the radionuclides.  In the work, 2 g of soil were suspended
in 25 ml of extracting water, in duplicate, and the pH adjusted with NaOH or
HN03.  The Kd values computed from Nishita et al.  (1976) are given in Table 3-37

                                   3-82

-------
for the untreated Aiken clay loam, a kaolinitic soil, at several pH values.
The Aiken clay loam also was treated sequentially with progressive treatments
to remove salts, organic matter, carbonates, manganese oxides, ion oxides,
free silica and alumina and amorphous aluminosilicates.   The organic matter
was shown to have the greatest ability to retain curium, but this was a func-
tion of pH as well.  Below pH 4.5, curium was less strongly adsorbed whether
organic matter was present or not.

            TABLE 3-37.  Kd VALUES FOR 242Cm IN UNTREATED AIKEN
                         CLAY LOAM (NISHITA ET AL., 1976)

                    242,
                          Kd, ml/g      pH      fc^Cm Kd.  ml/g
1.21
2.12
2.56
4.69
85.7 ± 12
2,457 ± 6
6,803 ± 46
100,000 ± 30,000
8.54
9.43
10.31
11.25
4608 ± 1635
1776 ± 142
690 ± 38
358 ± 2.7
            5.55    100,000 ± 10,000   12.22      190 ± 3.3
            7.08     71,429 ± 10,204   13.25      272 ± 23

Summary
     No information is available on possible curium compounds in soils  and
sediments.  It appears that curium precipitation as Cm(OH)., in alkaline solu-
tions may control curium concentrations in those solutions.  Theoretical
calculations (Figure 3-9) indicate that Cm   hydrolyzes even in acidic  solu-
tions (pH 3.3 and above).  Very little reliable information is available  on
interactions of curium with soils and sediments, and none on curium interac-
tions with rocks.  What little data are available indicate that curium  adsorp-
tion is not a function of ion exchange capacity of the soil (Thomas and
Jacobs, 1969; Jacobs et a!., 1966), and that precipitation and/or formation
of radiocolloids may predominately influence curium adsorption reactions
(Thomas and Jacobs, 1969; Sheppard et al., 1976).
                                   3-83

-------
References

Bondietti, E. A. and S. A. Reynolds.  1977.  Field and Laboratory Observations
on Plutonium Oxidation States.  BNWL-2117, pp. 505-538.

Duguid, J. 0.  1976.  Annual Progress Report of Burial Ground Studies at Oak
Ridge National Laboratory:  Period Ending September 30, 1975.  ORNL-5H1.

Jacobs, D. G., Y. E. Kim, and 0. M. Sealand.  1966.  Application of Mineral
Exchange to Reactor Technology.  ORNL-4007, pp. 27-33.

Jones, A. D.  and G. R. Choppin.  1969.  Complexes of Actinide Ions in Aqueous
Solutions.  Actinide Reviews.  1:311-336.

Keller, C.  1971.  The Chemistry of the Transuranium Elements.  Vol. 3.
Kernchemie in Einzeldarstellungen.  Verlag Chemie GmbH.

Moskin, A. I.  1969.  Complex Formation of the Actinides with Anions of Acids
in Aqueous Solutions.  Soviet Radiochemistry.   11:447.

Nishita, H.,  M. Hmailton and A. J. Steen.  1976.   Extractability of 238Pu and
   Cm from a  Contaminated Soil as a Function of pH and Certain Soil Components.
HNOs - NaOH System.  Presented at Annual Meeting of Soil Science Society of
America, Houston, Texas.

Shalinets, A. 8. and A. V. Stepanov.  1972.  Investigation of Complex Forma-
tion of the Trivalent Actinide and Lanthanide Elements by the Method of Electro-
migration. XVII.-  Hydrolysis.  Radiokhimiya.  14:280-282.
                                                              •
Sheppard, J.  C., J. A. Kittrick, and T. L. Hart.   1976.  Determination of
Distribution  Ratios and Diffusion Coefficients of Neptunium, Americium and
Curium in Soil-Aquatic Environments.  RLO-2221-T-12-2.

Thomas, W. A. and D. G. Jacobs.  1969.  Curium Behavior in Plants and Soil.
Soil Science.  108:305-307.

Trotman-Dickinson, A. F. (Executive Ed.).  1973.   Comprehensive Inorganic
Chemistry.  Pergamon Press, New York.

Vickery, R. C.  1953.  Chemistry of the Lanthanons.  Butterworths Sci. Pub!.,
London.

Weast, R. C., Editor.  1976.  Handbook of Chemistry and Physics.  The Chemi-
cal Rubber Co., Cleveland, Ohio, pp. B349-B350.
                                   3-84

-------
EUROPIUM
Natural Soil and  Rock Distributions
     The europium content of several common rock types are given  in  Tables 3-38
and 3-39.
             TABLE 3-38.   EUROPIUM CONTENT OF IGNEOUS  ROCKS  IN  ppm
                          (HASKIN AND SCHMITT, 1967)
                   Low K   Hawaiian                                Nepheline
        Peridotlte Tholeiite Tholeiite Andesite Granodiorite  Leucogranite Syenite
           0.2        1.9      1.3       1.0       1.2        0.17      10
          TABLE 3-39. EUROPIUM CONTENT OF SEDIMENTARY ROCKS IN ppm
                        (HASKIN ET AL., 1966)
                    Shale     Quartzite    Limestone    Subgraywacke
                     2.0        0.09        0.14          1.3

     Haskin and Schmitt  (1967)  gave the value of 2.9 ppm Eu in  a composite
of 40 North American  shales, with a variation in sedimentary rocks of  about
five times that value.
     In a study of rare  earths  in Russian platform soils, Balashov et  al.
(1964) reported high  europium contents in alkaline soils, the europium precipi-
tating as hydroxide.  Acidic soils prob.ably allow complexing and removal of
the europium and  other rare  earths.
Brief Chemistry
     Europium occurs  in  two  stable isotopes out of a total of 17 isotopes,
   Eu (47.8%) and     Eu  (52.2%).   The radionuclides of interest are  given
in Table 3-40.  The  oxidation states of europium are trivalent  with
            o                                o
radius 0.95 A and  divalent with radius 1.09 A (Ahrens, 1952).   The
                                                                      +2
divalent state  is  typical  of europium occurrence, substituting  for Ca
and K  in main  stage feldspars  (plagioclases, microcline, orthoclase).  The
potential of the  Eu(III)/Eu(II) couple is -0.43 volts (Latimer, 1952).

            TABLE 3-40.   EUROPIUM RADIONUCLIDE DATA (WEAST, 1976)
                                               Decay Mode
                                               6*, EC, 8"
Isotope
152Eu
154Eu
155Eu
156Eu
Half-Life
13.6 years
3.6 years
1 .81 years
15 days
                                      3-85

-------
 Europous  ions  are stable in  water solutions  at low hydrogen ion concentrations
.(basic pH),  and can  be formed from Eu   by zinc reduction.   However,  the
 europium  in  soil  solutions  is more apt to be in the trivalent than divalent
 state.
 Solid Phase  and Solution Equilibria
      Europium  forms  oxides,  hydroxides,  and  salts  with chlorides and  sulfates
 (EuO, Eu203, Eu304,  Eu(OH)3,  EuCl3'6H20, and Eu2(S04)3'SH20).  Thermodynamic
 data  for  these solid phases  were  used to plot the  stability of the minerals
 (Figure 3-10).   It can be seen that,  except  for the hydroxides [Eu(OH)3],  all
 the other minerals considered are too soluble in environmental pH ranges of
 interest  (pH 4-10).   In Figure 3-10,  Eu(OH)3> Eu203, Eu2(S04)38H20 are the
 only  solids  shown.   The thermodynamic data for Eu203 and Eu(OH)3 were obtained
 from  Shumm et  al.  (1973), and for Eu2(S04)3'8H20 were obtained from Latimer
 (1952).   The other solids were too soluble and fall  beyond  the graph  bounda-
 ries.   Of all  the solids considered in Figure 3-10,  Eu(OH)3 is the most stable
 in alkaline  environments. The results of Serne and Rai  (1976) indicate that
 kinetics  of  Eu(OH),  precipitation is  very fast and that Eu(OH), is the most
                                                     3+
 likely europium solid to maintain the activity of  Eu   in solution in slightly
 acidic to alkaline conditions.  However, where Eu(OH)3 does not precipitate
 or the europium concentration is  below the Eu(OH)3 solubility line, cation
 exchange  reactions are the probable europium adsorption  mechanism.
                              2+       3+              3+
      Europium  can exist as Eu  and Eu  .  However,  Eu   is the only  state
 which is  stable in water (Baes and Mesmer, 1976).   The activity of various
 solution  species  of  europium in equilibrium  with Eu(OH)3 and at assumed con-
 centrations  of various anions commonly found in soils is given in Figure 3-11.
                                                      2+        2+
 The thermodynamic data of all  the species except EuOH   and EuF   were obtained
                                              2+        2+
 from  Schumm  et al. (1973).   The data  for EuOH   and EuF  was obtained from
 Baes  and  Mesmer (1976) and Walker and Choppin (1967), respectively.
      In a general  way the solution complexes of europium in decreasing order
                                                 ?+      +          "4"      ?+
 of importance  can be arranged as  follows:  Eu,P907 , EuSO,, Eu(SO.):,, EuF   ,
    2+          2+
 EuCl   , and  EuNOt .   Under the conditions assumed  in Figure 3-11, the pre-
 dominant  solution species in  pH ranges of <4.5, 4.5-7.75, 7.5-8.85, and >8.35
 will  be EuSoJ,  Eu^Oy"1", EuSOj, and EuOH  ,  respectively.  Thus, significant

                                    3-86

-------
              -4
              -6
              -8
             -10
             -12
                                  6      7
                                    PH
                                            10
     Figure 3-10.  The  relative stability of various europium
                   solids  at  pSQ2-  =2.5.
                 -2
              ff  -6
                 -S
                -ID
                -12
                                                EuOH
                                                   ,2+
                                                 10     11
                                    pH
Figure 3-11.
The activity of various  europium species in equilibrium
with Eu(OH)3(s).  Other  soil  solution conditions included
pS042- = pCl" = 2.5, pN03~  =  3.0,  pF" = 4.5 and phosphate
levels were from Variscite  and  Gibbsite (V and G),
Dicalcium Phosphate Dihydrate (DCPD)  and Octacalcium
Phosphate (OCP).
                                3-87

-------
quantities of uncomplexed europium (Eu  ) can only be expected in soils with
very low levels of sulfate and phosphate.  The solution complexes of europium
with fluoride, chloride, and nitrate do not contribute significantly to total
europium concentration in solution and hence can be ignored.
Experimental Adsorption Results
     Bensen (1960) reported that the uptake of rare earths  (cerium, promethium
and europium) was virtually complete and unaffected by accompanying 0.5M
alkali metal and 0.25M alkaline earth metal chlorides at a  pH above 7.  Below
pH 3, however, the adsorption of all  rare earths was depressed similarly by
the accompanying salts.  These results were interpreted to mean that the rare
earths, including europium, are ionic at pH below 3 and that they are taken
up principally by ion exchange on the soil.  Above pH 7, they are precipitated.
The rare earths, including europium,  appear to exist as tripositive ions.
     Baetsle and Dejonghe (1962) gave a europium Kd range in Mol  soil (mostly
quartz sand) at pH 7.7 as 228 to 705, and reported that at pH >3, europium
hydrolyzed.  Therefore, above pH 3, the authors did not consider that adsorp-
tion was an ion exchange phenomenon.   However, a plot of pH versus log Eu
activity (Figure 3-11) does not support the above conclusion.
     Baetsle et al.  (1964) gave Kd values for trace europium in three differ-
ent waters.  The compositions of these waters were not given.  Kd results are
shown in Table 3-41.   The europium Kd values are much larger than the cesium
and strontium Kd values for the same soils.

            TABLE 3-41.  VARIATION OF TRACE EUROPIUM Kd VALUES '
                         ON MOL SOILS (BAETSLE ET AL., 1964)
                                           Kd, ml/g
              Soil Type
          Eolian Sand
           Horizon A
           Horizon B
           Horizon C
          Mol White Sand
          Mol Lignitic Sand
 Deionized
Water, pH 3
Tap Water,
  pH 7.7
214-244
61-75
69-89
_—
705
430
384
801
Groundwater,
    pH 3
                           130-380
                           165-315
                            27-1830

-------
     Serne and Rai (1976) studied the adsorption-precipitation behavior of
europium in soils and pure solutions.  The pH of the various europium solu-
tions (0.15M CaCl2 plus trace to 50 mg Eu/1) was increased in 0.05 pH incre- .
ments until precipitation was observed by the Tyndall Beam method.  At the
point of Eu(OH)3 precipitation, the europium activity and pH was determined.
The results are plotted in Figure 3-12 along with the theoretical Eu(OH)3
solubility line.  The results indicate that:
 1)  the kinetics of precipitation of Eu(OH)3 is rapid,
 2)
 3)
the laboratory results agree well with the theoretical Eu(OH)-, solubility,
  3+                          -                               J
Eu   does not appear to hydrolyze appreciably because total europium
                                                                    3+
     activity in equilibrium with Eu(OH)3 is similar to the calculated Eu
     activity, and
 4)  for meaningful Kd values, the theoretical  solubility of Eu(OH)3 should
     not be exceeded.









^
LU
s













-2


-3


-4


-5

-6




-7


-8



\
\
e\
I \
• \
•\

\
\

\
\
\
\
\
\
\
\
\
V-
\Eu(OH)3
\
\
• \

3456 789
pH
       Figure 3-12.
                The influence of pH on the activity of europium
                in solution (Serne and Rai, 1976)
                                   3-89

-------
     Europium adsorption  experiments where Eu(OH)3 did  not  precipitate were
conducted, as shown  in  Table 3-42.   The results show that the europium Kd
increases with an  increase  in pH. and decreases with increasing europium con-
centration, below  the point of Eu(OH)3 precipitation.   These reactions sug-
gest that ion exchange  is an active mechanism beTow the point of Eu(OH)3
precipitation.
           TABLE 3-42.  EFFECT OF pH AND EUROPIUM CONCENTRATION
                        ON  EUROPIUM Kd BY BURBANK SAND*
                        (SERNE AND RAI, 1976)

Eu, ppm**
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
Final
JEiL
4.82
4.82
4.90
5.00
5.05
5.05
5.08
5.10
5.10
5.10
5.10
5.10
5.15
5.15
5.15
5.15
5.20
5.20
5.20
5.25
5.30
5.30
5.50
4.75
4.75
4.80
5.00
5.00
5.00
5.08
5.08
5.10
5.25
Kd,
mg/1
46
48
51
75
70
69
75
84
78
78
82
75
93
90
81
93
94
98
98
107
106
97
153
32
38
38
63
53
53
56
62
67
71

Eu, ppm**
0.05
0.05
0.05
0.05
0.05
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
Final
nW
5.25
5.30
5.30
5.30
5.35
4.75
4.80
4.80
5.00
5.05
5.05
5.10
5.10
5.10
5.15
5.15
5.20
5.20
5,20
5.25
5.25
5.30
5.35
4.80
4.80
4.80
4.95
5.00
5.05
5.05
5.10
5.10
5.15
Kd,
mg/1
73
83
82
84
98
28
26
27
35
35
41
42
41
42
• 43
45
53
48
45
48
57
54
60
12.4
11.3
12.4
16.9
18.4
15.7
14.8
15.9
18.7
16.4

Eu, ppm**
5.0
5.0
5.0
5.0
50.0
50.0
50.0
50.0
50.0
50.0
50.0
50.0
50.0
50.0 .
50.0
50.0
50.0
50.0
50.0














Final
pH
5.20
5.20
5.25
5.32
4.70
4.70
4.73
4.88
4.90
4.90
4.90
4.90
4.90
5.00
5.00
5.00
5.15
5.15
5.20














Kd,
rnq/1
17.7
15.9
17.0
20.0
6.9
6.9
6.5
5.8
5.8
5.9
6.3
6.2
6.2
7.2
7.0
7.1
8.2
8.2
7.9














         * One  gram samples  free of CaC03 and soluble  salts were used
        ** Various concentrations of Eu  were added to  0.15M CaCl2 solu-
           tions spiked with Eul52.  Samples were equilibrated first
           for  48 hr and then for 24 hr  after adjusting the pH.
                                      3-90

-------
Migration Results
     Field Studies--
     Brookins (1976) reported that europium was retained in the shale that
enclosed the fossil nuclear reactor zones at the Oklo mine in Gabon along
with the other rare earths produced there by the nuclear reactions 1.8 bil-
lion years ago.
     Laboratory Studies—
     Baetsle et al. (1964) determined the K values associated with europium
movement in relation to groundwater movement is given in Table 3-43.   These
data also may be compared to strontium and cesium K values for the same Mol
soils.  The europium K values are very pH-sensitive, as the data of Fig-
ure 3-12 indicate.

            TABLE 3-43.  K VALUES OF EUROPIUM IN THE SATURATED
                         SUBSOILS AT MOL, BELGIUM (BAETSLE
                         ET AL., 1964)  K IS DEFINED ON PAGE 2-30.
               	Soil  Type	       pH 4        pH .3
               "MoFsand                    100         58
               Perturbed Profile            1963         511
               Mol  Sand & Eolian Sand         —         280
               Mol  Sand & Lignite             —         997
                 Mol  Sand                   2472         225
                 Mol  Sand                    640         139
                 Mol. Sand     _	       1282         78
                 Mol  Sand                  10600         74
                 Mol  Sand                   2145         53
                 Mol  Sand                    697         64
               Mol  sand is nearly pure quartz.
Summary
     No information on possible europium compounds in soils and sediments is
available.  Seme and Rai (1976) have shown that the kinetics of precipitation

                                   3-91

-------
of  Eu(OH)- from pure solutions is rapid.  The concentration of europium  in
equilibrium with Eu(OH)g decreases 1000-fold with an  increase of one  pH unit.
The activity of europium in equilibrium with Eu(OH).,  is low under highly
alkaline conditions and Eu(QH)., may control europium  concentration  in alkaline
soils and sediments.
     Bensen (1960) and Baetsle and Dejonghe (1962) reported that europium
appears to be taken up principally by ion exchange below pH 3.  Baetsle and
Dejonghe (1962) suggested that ion exchange was not the adsorption  mechanism
above pH 3 because of europium hydrolysis.  However,  data presented in Fig-
ure 3-11 and the results of Serne and Rai (1976) are  at variance with the
above conclusion.  Serne and Rai (1976) have shown that for meaningful
determination of Kd values in soils, one must be below the solubility line of
Eu(OH)g.  Their results, thus obtained, showed that between pH 4.7  and 5.5
the Kd decreases with an increase in pH and Kd decreases as europium concen-
tration increases, which suggests an ion exchange as  an adsorption  mechanism.
References
Ahrens, L. H.  1952.  The Use of lonization Potentials.  Part I.  Ionic Radii
of the Elements.  Geochim.  et Cosmochim. Acta.  2:155.
Baes, C. F., Jr. and R. E. Mesmer.  1976.  The Hydrolysis of Cations.  John
Wiley and Sons, New York.
Baetsle, L. and P. Dejonghe.  1962.  Investigations on the Movement of
Radioactive Substances in the Ground.  Part III:  Practical Aspects of the
Program and Physiocochemical Considerations.  IN:  Ground Disposal of Radio-
active Wastes.  TID-7628, pp. 198-210.
Baetsle, L. H., P. Dejonghe, W. Maes, E. S. Simpson, J. Souffriau, and
P. Staner.  1964.  Underground Radionculide Movement.  EURAEC-703.
Balashov, U. A., et al.  1964.  The Effects of Climate and Facies Environment
on the Fractionation of the Rare Earths During Sedimentation.  Geochemistry
International.  10:951-969.
Bensen, D. W.  1960.  Review of Soil Chemistry Research at Hanford.   HW-67201.
Brookins, D. G.  1976.  Shale as a Repository for Radioactive Waste:  The
Evidence from Oklo.  Environmental Geology.  1:255-259.
Haskin, L. A. and R. A. Schmitt.  1967.  Rare-Earth Distributions.  IN.:
Researches in Geochemistry.  Vol. 2.  P. H. Abelson (ed.).   John Wiley &
Sons, Inc.  pp. 234-259.

                                   3-92

-------
Haskin, L. A., et al.  1966.  Meteoric, Solar, and Terrestrial Rare Earth
Distributions.  IN:  Physics and Chemistry of the Earth.  L.  H. Ahrens et
al.  (eds.)  Vol. 7.  Oxford, Pergamon Press.

Latimer, W. M.  1952.  The Oxidation States of the Elements and Their Poten-
tials in Aqueous Solutions.  Prentice-Hall, Inc.

Seme, R. J. and Dhanpat Rai.  1976.  Adsorption-Precipitation Behavior of
Eu in Soils and Standard  Clays.  Agronomy Abstracts,  p.  132.

Schumm, R. H., D. D. Wagman, S. M. Bailey, W. H. Evans, and V. B.  Parker.
1973.  Selected Values of Chemical Thermodynamic Properties.   Tables for the
Lanthanide (Rare Earth) Elements (Elements 62 through 76)  in  the Standard
Order of Arrangement.  NBS Technical Note 270-7.  Schwille, F., W.  Lippok and
D. Weisflog.  1967.  Model Experiments on Fluid Flow in the Transition Zone
from Unsaturated to Saturated Soil.  IAEA-SM-93/11, pp. 151-159.

Walker, J. B., and G. R. Choppin.  1967.  Thermodynamic Parameters  of Fluoride
Complexes of the Lanthanides.  Advances in Chemistry Series No. 71.  The
American Chemical Society, Washington D.C.  pp. 127-140.

Weast, R. C., Editor.  1976.  Handbook of Chemistry and Physics.  The Chemi-
cal Rubber Co., Cleveland, Ohio, p. B313.
                                   3-93

-------
IODINE
Natural Soil  and  Rock Distributions
     The  iodine content of volcanic rocks  is given  in Table 3-44.  Yoshida
et al.  (1971)  tried leaching the iodine from volcanic rocks, and usually
removed less  than 20% of the total iodine  in the  rock.   The average I value
for igneous rocks probably falls within the range of 75 to 150 ppb.  Nearly
the same  average  value also would be applicable to  metamorphic rocks.

            TABLE 3-44.   IODINE CONCENTRATION  IN  VOLCANIC ROCKS
                  Rock Type
              Basalt  (BCR-1)
               Columbia River, WA
              Basalts, Japan, 15
              Andesites, Japan, 10
              Andesite (AVG-1)
               Guano Valley, OR
              Oacites, Japan, 3
              Rhyolite, Japan, 2
              Obsidians, Japan, 3
              Obsidian, Oregon
              Obsidian, Arizona
              Obsidian, Utah
              Obsidian, Mexico
I. PPb
160
      Reference
Becker and Manuel, 1972
29-140
22-260
270
43-220
26-320
20-65
730
730
540
1900
Yoshida et
Yoshida et
Becker and
Yoshida et
Yoshida et
Yoshida et
Seeker and
Becker-and-
Becker and
Becker and
al, 1971
al., 1971
Manuel, 1972
al., 1971
al., 1971
al., 1971
Manuel, 1972
Manuel-, -1972
Manuel, 1972
Manuel, 1972
     The abundance  of iodine in sedimentary rocks  is  shown in Table 3-45.
Note that the units have changed from ppb in igneous  rocks to ppm in sedi-
mentary rocks.   Pre-1960 iodine analytical values  may have been as much as
35 times too high  (Brehler and Fuge, 1974).
     While the  iodine content of marine sediments  depends  on the carbon to a
large extent, marine sediments are much richer in  iodine than sedimentary
rocks.  Walters  and Winchester (1971) examined iodine binding in dry marine
sediments and found that most of the iodine was surface-adsorbed or covalent
bonded to carbon.   About 23% could be extracted with  organic solvents.
     There is a  marked increase in the soil iodine content compared to the
rocks they are  derived from.  For example, the average iodine content of
soils was noted  as  5 ppm by Vinogradov (1959), who also suggested that much
of the soil iodine  is atmospherically derived.
                                    3-94

-------
                 TABLE  3-45.   IODINE IN  SEDIMENTARY ROCKS
              Sedimentary Rock
I, ppm
Sandstones, 11,
Bashkiria, USSR .
Sandstone, white,
Klondyke, MO
Sandstone, red,
Potsdam, NY
Argillaceous Sandstone,
Portageville, NY
Argillaceous Shale,
Rochester, NY
Calcareous Shale,
Lima, NY
Limestones, 6,
Paleozoic, OK
Sediments,
0.5-1.5
0.068
0.1.4
37.6
13.0
38.0
4.2
405
Reference
             SW Barents Sea
                                             Itkina and Lygalova, 1964

                                             Becker et al., 1972

                                             Becker et al., 1972

                                             Becker et al., 1972

                                             Becker et al., 1972

                                             Becker et al., 1972

                                             Collins et al., 1971

                                             Price et al.,  1970
     Many of the  iodine  studies  have noted the correlation of iodine content
and humus in the  soil  (Sinitskaya,  1969;  Irinevich et al., 1970; Pennington
and Lishman, 1971).  De  et al.  (1971)  contacted soils and soil clays with
iodide solutions  at 20°,  30°  and 40°C  containing 0.127 g I"/l, 0.635 g l"/l
and 1.269 g l"/l.  Only  the clay minerals showed iodide uptake, with illite
adsorbing more iodide  than kaolinite or montmorillonite.
Brief Chemistry
     There are 24 known  isotopes of iodine with 18 of these isotopes having
half-lives of less than  1 day.   The only  stable isotope is 127I.  Only 129I,
with a half-life  of 1.7  x 10  years is of long-term interest in high-level
waste disposal.   The fission  product   I with a half-life of 8.07 days, and
   I with a half-life of 60 days,  are  often  a  short-term disposal hazard
because of their anionic character and resulting lack of soil adsorption.
     Although iodine is known  in the  (-1),  (I),  (III), (V)  and -(VII) oxidation
states, its usual occurrence is as the (-1)  (iodide)  state  in fresh waters with a
               o
radius of 2.20 A (Goldschmidt  et al . ,  1926)  and  as iodate-iodine in marine or simv
lar aqueous environments (Sugawara and Terada, 1958).  I" tends to be a dispersed
element in many environments as a  result of  its  large size  in comparison with
fluorine, chlorine, bromine and hydroxyl  ions.   Oxidation of iodide ions to
                                    3-95

-------
produce the iodate ion (103) is easily accomplished in basic solution by
the reaction:  3I2 + 60H" * 51" + I0~ + 3H20 (Cotton and Wilkinson, 1962).
Iodine can form complexes with metal ions, but these are generally the least
stable of all the halide complexes, with a few exceptions.  Iodine also is a
volatile element, subliming at atmospheric pressures without melting.
     Examples of iodine minerals include marshite (Cul), iodargyrite (Agl) and
coccinite (Hglg), bellingerite  [Cu(I03)2L salesite [CuI03(OH)] and lautarite
[Ca(I03)2] (Fleischer, 1966).   Most such iodine minerals are confined to
unusual  environments such as the Chilean nitrate deposits.
Solid Phase and Soil Solution Equilibria
     Most of the compounds of iodine are very soluble.   Some of the insoluble
or sparingly soluble compounds include the iodides of Pb and Pd, the hypo-
iodites of Ag and Hg, and Ba periodates (Pourbaix, 1966).  In normal soils
the concentration of most of these elements (Pb, Pd, As, Hg and Ba) that
iodine forms compounds with is very low.  Therefore, they would not be expected
to be present in soils.
     Iodine can exist in (-1), (I), (III), (V)  and (VII) oxidation states.  Out
of all of these oxidation states, (-!•) (I", iodide) is  the most important and
its domain of predominance extends all along the pH scale, almost completely
covering a large part of the stability domain of water  (Pourbaix, 1966).  Thus,
it is not surprising that iodine in aqueous solutions free from oxidizing
agents generally exists as iodide.
Experimental Adsorption Results
     Raja and Babcock (1961) examined the behavior of    I with two California
soils, kaolinite and bentonite.  Results of pretreatment by autoclaving,
                                                        131
peroxide oxidation, alcohol digestion and extraction of    I with various salt
solutions indicated that the bulk of the iodide retained by soils was due to
reaction with organic matter.
     Goldberg et al. (1962) determined iodide adsorption on Rainier tuff
ground to 100 to 200 mesh from simulated groundwater.   The value obtained
using 131I was 1.10 ral/g.
                                   3-96

-------
                                                 1 2]
     Kepak  (1965,  1966)  studied  the sorption of    I on hydrated aluminum
and ferric  oxide.  One wt%  silver oxide was added  to the ferric oxide adsor-
bent.  The  adsorbent of  ferric oxide was 0.06 to 0.1 mm in grain size and
the influent solution contained  0.1M NaNO, at pH 6.9.  Slow column break-
                131                                 131
through of  the     I began immediately even from a     I-distilled water solu-
tion.  Sixty-six free column  volumes (void fraction = 0.66) were loaded to
1% column breakthrough and  758 free column volumes to 90%    I breakthrough.
     The most extensive  study of iodine retention  by soils was made by
Wildung et  al.  (1975).   Iodine was applied to the  soils as iodide (I") and
methyl iodide (CHjI), a  colorless transparent liquid when fresh, with one
part soluble in 50 parts of water.   The solution normally remained nonionic
during the  distribution  studies.   Twenty-two soil  types were collected for
use in the  study in Oregon, Washington and Minnesota with a range of proper-
ties as listed  in-Table  3-46.  The    I" and CH3   I were used in trace quan-
tities in 0.01M CaCl2 solution.

  TABLE 3-46.  RANGE OF  SURFACE  SOIL PROPERTIES USED IN THE METHYL IODIDE
               AND IODIDE ADSORPTIONS  (WILDUNG ET  AL.,  1975)
                                            Contents,_wt;L
        Cation Exchange   Soil Paste          Organic
      Capacity, meg/100 q     pH    Carbonate  Carbon    Sand     Silt     Clay
         5.5 - 90.0     3.6 - 8.9   0 - 6.5   0.23-28.8 14.1-73.1  17.6-58.0  38-46.6

     The equilibrium distribution coefficients  were treated statistically
(Table 3-47).  Methyl iodide  retention  by soils was largely a function of
organic carbon and clay content,  while  iodide adsorption was a partial function
of silt content.  The mechanisms  and causal  relationship between the silt and
iodide were under further  study  but were  not covered in the report.  The
mechanisms underlying the  cation  exchange capacity-methyl iodide relationship
are difficult to envision.  Regression  equations were listed to obtain
iodide and methyl iodide Kd values over the  soil parameter ranges studied.
Migration Results
     Field Studies-
     Brown (1967), in the  study  of groundwater  beneath disposal sites at
                                   125
Hanford, mentioned the presence  of    I in the  groundwater.  The groundwater
                                                  125
samples were taken from monitoring wells  and the    I concentration were all
less than 2 x 10"5 uCi/mK
                                   3-97

-------
     TABLE 3-47.  Kd VALUES  AND CORRELATIONS RELATING SOIL PROPERTIES
                  IODIDE AND METHYL IODIDE ADSORPTION (WILDUNG ET AL.,
                  1975)
                                 Correlation Coefficients of Independent and
                                 	Dependent (Kd) Variables	
               Independent
              Soil Variables               Iodide    Methyl Iodide
          Silt wtS                         0.47*       0.21
          Clay wtS                         0.22       0.43*
          Organic Carbon wtS                 0.06       0.85**
          Cation Exchange Capacity            0.21       0.79**
          pH                            -0.05       -0.41*
            Statistically significant at the 5% confidence level
            Statistically significant at the \% confidence level
     Laboratory  Studies-
     Jacobs  (1965)  studied the desorption  of radioiodine from clays.   The
clays and Clinch River (Oak Ridge) floodplain sediment were contaminated with
iodine vapor  from the oxidation of     I-tagged KI.   Desorption  studies indi-
cated that the adsorption iodine was  in  the form of a mixture of  HI,  I,, and
adsorbed iodine.   The rate of desorption decreased with .decreased surface
concentration, and after 24 hr, the iodine remaining was proportional  to the
reciprocal of the absolute air temperature.   Desorption occurs  most rapidly
in high humidity air.  Most of the iodine  was readily removed with tap water.
After ten 25  ml  increments of Oak Ridge  tap water through a 10  g, iodine
contaminated  Clinch River floodplain  sediment, less than 5% of  the iodine
remained on  the  sediment.  However, removal  from 10% to 5% required 60% of
the water leach  volume, indicating that  part of the iodine is relatively
tightly held.  Some preliminary work  with  methyl iodide suggested that this
chemical form of iodine would move readily through the ground without being
adsorbed.

Summary
     Because  the anion exchange capacities of most soils are minimal  over a
normal pH range  of 6 to 8, the adsorption  of iodide (I"), iodate  (IQ^) and
organic-iodine molecules is normally  also  minimal (Wildung etal., 1975).
However, with soils of low pH (4 to 6),  the iodide Kd value can rise to

                                    3-98

-------
50 ml/g.  Wildung et al. (1975) reported no statistical correlation between

organic content and iodide Kd, as they did for methyl iodide.  The range of
methyl iodide Kd values was from 0.1 to 3.1, much lower than the iodide ion
Kd range, even though the correlation between methyl iodide Kd values and
soil organic content was statistically significant.  Others (Sinitskaya, 1969;
Irinevich et al., 1970; Pennington and Lishman, 1971) have shown a positive

correlation between iodine content and organic material in the soil where the

iodine is apparently covalently bonded to carbon (Walters and Winchester, 1971)
It should be kept in mind that iodine, especially when present as an organic-
iodine molecule, can change to a vapor phase and migrate much more rapidly
than in an aqueous phase.

References

Becker, V. J., J. H. Bennett, and 0. K. Manuel.  1972.  Iodine and Uranium in
Sedimentary Rocks.  Chem. Geol.  9:133.

Becker, V. J. and 0. K. Manuel.  1972.  Chlorine, Bromine, Iodine, and
Uranium in Tektites, Obsidians and Impact Glasses.   J. Geophys. Res.  77:5353.

Brehler, B. and R. Fuge.  1974.  Iodine.  IN:  Handbook of Geochemistry,
Vol. II/3, Springer-Verlag, New York.  p. 53-1.

Brown, D. J.  1967.  Migration Characteristics of Radionuclides Through Sedi-
ments Underlying the Hanford Reservation.  ISO-SA-32.

Collins, A. G., J.. H. Bennett, and 0. K. Manuel.  1971.  Iodine and Algae in
Sedimentary Rocks Associated with Iodine-Rich Brines.  Bull.  Geol. Soc.  Am.
82:2607.

Cotton, F. A. and G. Wilkinson.  1962.  Advanced Inorganic Chemistry.   Inter-
science Publishers.

De, S. K., S. S. Rao, C. M. Tripathi, and C. Rai.  1971.   Retention of Iodide
by Soil Clays.  Indian Journal of Agricultural Chemistry.   4:43-49.

Fleischer, M.  1966.  Index of New Mineral Names, Discredited Minerals and
Changes of Mineralogical Nomenclature in Vols. 1-50 of the American Mineralo-
gist.  Am. Mineral.  51:1248.

Goldberg, M. C., V. J. Janzer, C. G. Angelo, and W. A. Beetem.   1962.   The
Effect of Sodium Ion Concentration on Distribution Coefficients for Tuffs
from NTS.  U.S.G.S. Technical Letter NTS-16.

Goldschmidt, V. M., T. Barth, G. Lunde, and U. W. Zachariasen.   1926.   Geo-
chemische Verteilungsgesetze der Elemente.  VII.  Skrifter Norske Videnskaps-
Akad.  Oslo, I. Mat-Naturv. Kl., No. 2.

                                   3-99

-------
Irinevich, A. D., I. Z. Rabinovich, and V. A. FiTKov.  1970.  Iodine in
Moldavian Soils.  Pochvovedenie.  58-.

Itkina, E. S. and V. W. Lygalova.  1964.  Geochemistry of  Iodine and Bromine
in the Carboniferous Horizon and in the Domanik and Bavly  Formations of Some
Petroliferous Areas of Bashikiria.  IH:  The Geochemistry  of Oil and Oil
Deposits.  L. A. Gulyayeva (ed.).  Israel Program for Scientific Translations.

Jacobs, D. G.  1965.  Desorption of Radioiodine from Clays.  ORNL-P-1531.

Kepak, F.  1965.  Sorption of Small Amounts of Radioiodine as  Iodide Anions
on Hydrated Ferric Oxide Containing Silver.  Collection Csechoslov.  Chem.
Commun.  31:1493-1500.

Kepak, F.  1966.  Sorption of the Radioisotopes 35S, 131I, and  06Ru on
Hydrated Oxides in Laboratory Columns.  Collection Czechoslov. Chem. Commun.
31:3500-3511.

Pennington, W. and J. P. Lishman.  1971.  Iodine in Lake Sediments  in Northern
England and Scotland.  Biol. Rev.  46:279-313.

Pourbaix, M.  1966.  Atlas of Electrochemical Equilibria in Aqueous Solutions.
Pergamon Press, Oxford, England.

Price, N. B., S. E. Calvert, and P. G. W. Jones.  1970.  The Distribution of
Iodine and Bromine in the Sediments of the Southwestern Barents Sea.  J.
Marine Res.  28:22.

Raja, M. E. and K; L. Babcock.  1961.  On the Soil Chemistry of Radioiodide
Soil Science.  91:1-5.

Sinitskaya, G. I.  1969.  Iodine Content of the Zeya-Bureya Plain   Soils.
Uch. Zap. Dal'nevost.  Gos. Univ.  27:72.


Sugawara, K. and K. Terada.  1958.  Oxidized Iodine in Sea Water.  Nature.
182:250.

Vivogradov, A. P.  1959.  The Geochemistry of Rare and Dispersed Chemical
Elements in Soils.  Consultants Bureau,Inc.

Walters, L. J. and J. W. Winchester.   1971.  Neutron Activation Analysis of
Sediments for Halogens Using Szilard-Chalmers Reactions.  Anal. Chem. 43:1020.

Wildung, R. E., R. C. Routson, R. J.  Serne, and T.  R.  Garland.    1975.
Pertechnetate, Iodide and Methyl Iodide Retention by Surface Soils.  BNWL-
1950.  Pt. 2.  pp. 37-40.

Yoshida, M., K. Takahashi, N.  Yonehara,  T. Ozawa, and I. Iwasaki.   1971.
The Fluorine, Chlorine, Bromine and Iodine Contents of Volcanic Rocks in
Japan.   Bull. Chem. Soc. Japan.  44:1844.


                                   3-100

-------
NEPTUNIUM
Natural Soil and Rock Distributions
                              0-57    0-57   ooo       10
     Very small quantities of "xNp r0/Npr°0U = 10"'^) were reported by
Keller (1971) but for practical  purposes neptunium does not occur naturally
in soils or rocks.
Brief Chemistry
                                                   228      241
     There are 16 known isotopes of neptunium from    Np to    Np (Keller,
1971).  Only 237Np> a neutron reactor product of 238U (n,2n)  and  235U (n.y)2,
with a half -life of 2.14 x 106 years and 239Np (2.35 days)  are of interest.
                         236
The >5000 year half-life    Np is a cyclotron product only  while  the 22-hr  half-
     23fi
life    Np yields    Pu as a daughter.   Neptunium exists  in  aqueous  solutions  in
five oxidation states, Np(III), Np(IV), Np(V),  Np(VI)  and Np(VII).   In  the
absence of complexing agents, the first four oxidation states  exist  as  hydrated
ions Np*3-aq, Np+4-aq, NpOt • aq and Npot2 • aq (Keller,  1971).   Np(VII)  is  a
                                                                         -3
strong oxidizing agent that is stable in strong alkaline  solutions as
The most stable oxidation state in solution is  the pentavalent one  where  nep-
tunium exists as a single charged neptunyl  ion, NpO? »aq,  with symmetrical
                       ±                           ^
linear bonding -(0-Np-O) .  It hydrolyzes only at a pH of greater than  7,  dis-
proportionates only at high acid concentrations and forms  no  polynuclear  complexes
(Keller, 1971).  The NpO« ion is a poor complexing agent with inorganic ligands.
Solid Phase and Solution Equilibria
     The thermodynami c data are lacking for neptunium compounds other  than
oxide and hydroxides (Burney and Harbour, 1974).   Figure 3-13 relates  the
activity of NpOp to pH under an oxidizing environment (p02 0.68) in equilibrium
with neptunium oxide and hydroxides.   Since Figure 3-13 was made using Np(V)
(Np02) as the y axis, the solubility  of Np(IV)  minerals would decrease and
Np(VI) minerals would increase with an increase in reducing conditions.   In
an oxidizing environment CpOgfq) s °-68 atm^ tne soll'ds i° an increasing
order of stability are Np(OH)4, Np02, NpOgOH, Np02(OH)2.  Under the oxidation-
reduction conditions represented by p02(g)  > 3.8 atm, Np02 would be the most
stable compound.  Since Np02 maintains very high concentrations of  neptunium
in solution, especially in acidic and very oxidizing environments,  it  is
unlikely that Np02 would be found as  a discrete solid in such terrestrial
environments.  It may exist in very reducing conditions.
                                   3-101

-------
               S
                 -2
                                     I      I
                     NpOjOH; Np02 AT pO. • 2.08 atm; Np02(OH)2 AT p02 • 3.84 atm
                                                             10
                                       PH
  Figure 3-13.  The relative stability of various  neptunium  solids  in  an
                oxidizing soil environment  [pO
2(g)
0.68 atm]
     The activity of various solution complexes  of  neptunium in equilibrium
with Np02(s) in an oxidizing environment  (p02 =  0.68)  and  assumed weather-
ing environment is given in Figure 3-14.  Thermodynamic data for NpO«  HPOT
were obtained from Sillen and Martell (1964).  The  remaining data were from
Burney and Harbour (1974).  The solution complexes  of  Np(IV),  Np(V), and
Np(VI) in order of increasing importance are Np(IV), Np(VI),  and Np(V).   The
activity of Np(IV) and Np(VI) complexes is so small that these complexes  can
be safely ignored since they would not contribute significantly to the total
activity of Np in solution (Burney and Harbour,  1974).  The  lines of Np(V)
and Np(VI) complexes will shift downward with the increase in  reducing envi-
ronment but still be parallel to the present lines.  Under the conditions
assumed in Figure 3-14, NpO  is the most dominant solution species in  a pH
range of 0 to approximately 9.  Beyond pH 9, NpOgHPO" and NpOpHCO^ would
control the total concentration of neptunium in solution.
          is the usual form of charged neptunium species up to pH  9.  Above
pH 9, an uncharged bicarbonate complex is formed.  According to  Keller  (1971)
neptunium (V) does not hydrolyze below a pH of 7.  Hence, a singly charged
                                   3-102

-------
    Figure 3-14.  The activity of various neptunium species in equilibrium
                  with NpO?(s) in an oxidizing soil environment [p02(g) =
                  0.68 atm], pCT = pS042' = 2.5, pH03" = 3.0, pF~ = 4.5
and
                                5.0.
neptunyl ion is the usual form.  Hence, NpO« would be expected to enter into
                            •f                                   +2
ion exchange reactions.  NpQ2 does not compete favorably with Ca   and other
common divalent ions, and the distribution coefficients are usually rela-
tively low on most soils (Routson et al., 1976; Sheppard et al., 1976).
Experimental Adsorption Results
     Bensen (1961) examined the adsorption on minerals of radionuclides in
                                                                      237
reactor effluent cooling water.  One of the radionuclides studied was    Np,
which was adsorbed on 25 different common sulfide, silicate and carbonate
minerals, 0.05 to 0.25 mm, in equilibrium tests with trace amounts in Columbia
River water at 80°C.  According to Bensen, neptunium was adsorbed poorly or
not at all by the minerals tested.
     Robertson (1974) determined the speciation of neptunium in the cooling water
effluent from the Hanford N reactor.  Through the use of filter membranes and
                                   3-103

-------
cation and anion exchange resins, the following distribution was observed:
26% participate, 70% cationic, <3% anionic and <1%. non-ionic.
     Routson et al. (1975, 1976) determined neptunium Kd values for a
Washington sand and a South Carolina sandy clay.  The properties of these
soils are given in Table 3-48.  The neptunium Kd values are given in Table 3-49.
Pre-equilibrations of the soils with nonradioactive solutions prior to contact
                                    237
with the traced solution containing    Np were carried out.  Calcium nitrate
and sodium nitrate salts were used as calcium and sodium ion sources.  Cation
exchange of NpO  cannot be the principal adsorption mechanism because Na  con-
centration essentially does not affect the neptunium Kd values from no sodium
competition to 3.0M Na  competition.  Calcium has some affect on the neptunium
Kd, but much less than there would be if neptunium removal and adsorption were
due to ion exchange.

                   TABLE 3-48.  PROPERTIES OF SOIL SAMPLES
                                (ROUTSON ET AL., 1976)

                                                             CEC,
        Soil          CaCQq.  mq/q   Silt,  wt%  Clay. wt%   meg/ 100 q    pH
   Washington Soil         0.8         10.1        0.5        4.9    • 7.0
   (Surbank sandy loam)
   South Carolina         <0.2          3.6        37.2        25     51
   Subsoil

             TABLE 3-49.  NEPTUNIUM Kd (ml/g)  AS A FUNCTION
                          OF SOIL AND SOLUTION (ROUTSON
                          ET AL., 1976)
                                    Ca               Na
Soil
Washington
South Carolina
0.002 M
2.37
0.25
0.2 M
0.36
0.16
0.015 M
3.9
0.7
3.00 M
3.2
0.4
     Sheppard et al. (1976) determined neptunium distributions on several
                                                                237
Washington and North Carolina soils.   The distilled water-trace    Np solu-
tions and soils were equilibrated over long periods of time to obtain a rate
of change per month in the distribution values.   The 50-day values are given
in Table 3-50.  A positive value for the monthly change in the distribution
                                   3-104

-------
means an increase in the size of the distribution value.  The distribution
was determined in the same way that a Kd value is determined, but does not
have the connotation that equilibrium has been attained as the Kd does.
       TABLE 3-50.  NEPTUNIUM 50-DAY DISTRIBUTION VALUES COMPUTED
                    FROM SHEPPARD ET AL. (1976)
    Soil Identity
    Muscatine
      silt loam
    Burbank
      loamy sand
    Ritzville
      silt loam
    Fuquay sand
      0-5 'cm depth
    Fuquay sand
      5-15 cm depth
    Fuquay sand
      15-50 cm depth
Np Distribution, ml/g
        127

         15.4

         20.2

         33.7

         37.2

         32.4"
Monthly Change
in Distribution
     +10%

     +48%

     + 28%

     + 25%

     +25%

     +19%
     The neptunium Kd value of about 3.9 ml/g was given by Routson et al.
(1975, 1976) for the Burbank sandy loam without sodium in the solution.
Sheppard et al. (1976), apparently for the same conditions, gives a neptunium
50 day distribution value of 15.4 ml/g for Burbank and states that this value
is increasing at the rate of +48%/month.  At 100 days, the distribution was
approximately 50 ml/g on the Burbank loamy sand.  The differences in neptunium
distribution between soil and solution probably resulted from the experimental
differences because the Burbank sandy loam of Sheppard was obtained from
Routson.  However, Sheppard shows the Burbank of his sample to contain 21.2%
silt and 2.8% clay with a pH of 8.1 and a cation exchange capacity of
5.94 meq/100 g compared to Routson's Burbank soil sample (Table 3-48).
Sheppard1s Burbank must have contained calcite (CaC03) to attain a pH of 8.1.
                                   3-105

-------
There is no evidence in Sheppard et al. (1975) that the soil samples were
pre-equilibrated in any way with the nonradioactive solution before the .
neptunium tracer was added, or that a blank solution without the soil was
run with the soil equilibrations for reference and use in counting as the
original solution condition without the soil present.  In addition, Sheppard
                  237
reported that his    Np(V) in the aqueous phase equilibrating with Ritzville
soil was partially filterable on Whatman number 50 filter paper.  Routson,
on the other hand, reported that upon filtering previously centrifuged
samples of solution containing neptunium and in contact with Burbank or
South Carolina subsoil  through 0.01 and 0.45 urn filters, no evidence of col-
       237
loidal    Np was found.  This is further evidence that Sheppard1s high nep-
tunium distribution values require further elucidation in the light of
Routson's results.
Migration Results
     Field Studies-
     No reports of field studies of neptunium migration through soil or rock
were found in the literature.
     Laboratory Studies-
     No reports of laboratory studies of neptunium migration through soil  or
rock were found in the literature.
Summary
     The existing thermodynamic data (Figure 3-14) show that neptunium should
exist in an oxidizing soil environment as Np(V) in the form of NpO^.  However,
existing adsorption results,  essentially only a study by Routson et al.  (1975,
1976) and Sheppard et al.  (1976), inferred that there is little evidence of
NpO- ion exchange as a neptunium adsorption mechanism.  Sheppard et al.  (1976)
reported neptunium colloids present in the soil-solution environment, while
Routson et al.  (1975, 1976) specifically showed that they were absent from
his solutions that had contacted soils.  It is fairly certain that neptunium
Kd values are generally low.   Little more can be said based on the adsorption
and migration studies now available in the literature.  Additional  research
is required on neptunium-soil and rock reactions.
                                   3-106

-------
References

Bensen, D. W.  1961.  Mineral Adsorption of Radionuclides in Reactor Effluent.
HW-69225.

Burney, G. A. and R. M. Harbour.  1974.  Radiochemistry of Neptunium.  NAS-
NRC Nuclear Sci .  Ser. NAS-NS-3060.

Keller, C.  1971.  The Chemistry of the Transuranium Elements.  Vol. 3.
Kernchemie in Einzeldarstellungen.  Verlag Chemie GmbH.

Robertson, D. E.   1974.  Physicochemical Characterization of N-Reactor
Effluent Radionuclides in Soil and Water Systems.  BNWL-1950, Pt. 2,
pp. 82-85.
                                                                   99
Routson, R. C., G. Jansen, and A. V. Robinson.  1975.  Sorption of   Tc,
237^ and 241 Am on Two Subsoils from Differing Weathering Intensity Areas.
BNWL-1889.
                                                       qq    0-37
Routson, R. C., G. Jansen, and A. V. Robinson.  1976.  * Tc,    Np, and   'Am
Sorption on Two Subsoils from Differing Weathering Intensity Areas.  BNWl-
2000.  Pt. 2, pp. 50-52.

Sillen, L. G. and A. E. Martell.  1964.  Stability Constants of Metal-Ion
Complexes.  Special Publication No. 17.  The Chemical Society, London.
                                                  %
Sheppard, J. C., J. A. ICittrick, T. L. Hart.  1976.  Determination of Dis-
tribution Ratios and Diffusion Coefficients of Neptunium, Americium and
Curium in Soil -Aquatic Environments.  RLO-221-T-12-2.
                                   3-107

-------
PLUTONIUM
Natural Soil and Rock Distributions
     Except for very small quantities of plutonium in the natural environments
given below, plutonium does not occur in any significant amount in soils or
                                                      239
rocks.   Cherdyntsev et al. (1968), for example, lists    Pu values in natural
                                      -17   ?^Q              -T5   239
rocks and minerals varying from 7 x 10 " g "*Pu/g to 1 x 10    g    Pu/g.
Brief Chemistry
     Fifteen isotopes of plutonium are known, including 10 isomers with dif-
ferent decay modes (Keller, 1971).  Most of the isotopes are formed by multiple
neutron capture, and therefore, are not ordinarily encountered in natural
environments.  An exception is the natural  reactor at Oklo, Gabon (IAEA, 1975)
and very small  amounts (Pu/U = 10"  ) found in the uranium mineral pitchblende
(Cleveland, 1970).
     The plutonium isotopes of concern to waste disposal, either because they
are contained in the fission product wastes or are a part of an impotant decay
chain for waste constituents, are listed in Table 3-51.  In the short term,
almost any plutonium radionuclide would be of concern in waste-radionuclide
                                                    242  •
reactions although, on a weight basis, lon-ger-lived    Pu is not nearly as
             239
hazardous as    Pu.  Long irradiation times and high neutron fluxes are
                                           242
required to produce significant amounts of    Pu, so that a prevalence of
the higher mass Pu isotopes usually will  not occur in waste solutions from
commercial fuel treatment facilities.
                 TABLE 3-51

                  Isotope
                   236,
         PLUTONIUM  RADIONUCLIDE  DATA
         (KELLER, 1971)
                   238
                   239
                   240
                   241
                   242
                   243
'Pu
 Pu
 Pu
 Pu
 Pu
 Pu
 Pu
   HaIf-Life
   2.85 years
   86.4 years
 24,400 years
  6,600 years
   14.1 years
387,000 years
   4.98 hours
Decay Mode
   a, SF
   a, SF
   a, SF
   a, SF
   8", a
   a
                                   3-108

-------
     Plutonium exists in five oxidation states in aqueous solutions that
include Pu(III), Pu(IV), Pu(V), Pu(VI) and Pu(VII).  These states can occur
                       +3         +4          +          +2             -3
as the hydrated ions Pu   • aq, Pu   • aq, PuCL • aq, PuCL  • aq and PuOg  •
aq (Keller, 1971).  The most stable state of plutonium in aqueous solution
is Pu(IV).  However, disproportionation, in which Pu(III), Pu(IV), Pu(V) and
Pu(VI) states can coexist in unequal quantities, tends to further complicate
plutonium chemistry in its reactions with soils and sediments.  According to
Cleveland (1970), plutonium in near-neutral solutions occurs in the order:
Pu   > PuOl"  > Pu   > PuOt.  Hydrolysis reactions lead to the formation of
positively charged to neutral colloidal polymers in a stepwise manner.  For
example, PuH + OH" * PuOH*3 + OH" * Pu(OH)+2> etc., leads finally to
Pu(OH)4 which loses water to product thermodynamically stable Pu02 (Keller,
1971).  The kinetics of the hydrolysis reactions leading to PuO« are not
known.  It is known that depolymerization of colloidal Pu(IV) is a very slow
process, having a "depolymerization half-life" of 320 hr at 25°C in 5M HN03
(Keller, 1971).  Complexing agents such as fluoride or sulfate ions can accel-
erate the process.  Plutonium can form complexes with most of the ions com-
monly encountered in soil solutions.  Thus disproportionation, complex forma-
tion and hydrolysis reactions, all relatively pronounced with aqueous plutonium
solutions, combine to add to plutonium chemistry complexity.
Solid Phase and Solution Equilibria
                                           4+
     Figure 3-15 relates the activity of Pu   to pH in an oxidizing environ-
ment CpOo/n) = 0-68 atm] in equilibrium with various Pu solid phases.  The
parameters indicated in parentheses after the mineral formulas refer to the
additional conditions of equilibrium.  For example, (V&G) indicated after
the Pu(HPO.)2 mineral formula denotes that the mineral is considered to be
in equilibrium with variscite and gibbsite.  Any solid in Figure 3-15 that
lies below another solid, at a given pH, is the more stable.  Thus, for any
                                                                    4+
two solid phases at a specific pH, the solid that maintains lower Pu   activity
is more stable.  Under the assumptions outlined in Figure 3-16 and at pH 6,
the plutonium compounds in order of increasing stability are:  Pu(OH)3;
BPu203; PuF4; PuF3; Pu(HP04)2; PuOgOH; Pu02C03; Pu(OH)4; Pu02(OH)2; and
Amont the solids reported in Figure 3-15, Pu02 is the most stable mineral
under oxidizing conditions at all pH values.

                                   3-109

-------
                                        Pu(HP04)2
-------
     The stabilities of plutonium solid compounds in a reducing environment
        = 80 atm] are represented in Figure 3-16.  In general, the reducing
environment substantially changes the stabilities of the solids.  Comparing
Figures 3-15 and 3-16, it can be seen that the Pu(IV) solids remain unchanged,
while Pu(III) solids increase and Pu(V and VI) solids decrease in their sta-
bilities with the decrease in the oxygen pressure.  Plutonium solids in the
reducing environment at pH 6 can be arranged in order of increasing stability
as follows:  Pu02(OH)2; PuOgOH; PuF4; Pu(HP04)2; Pu(OH)3; 6Pu203; Pu02(OH)2;
PuF^; and Pu02.  PuF-j would be the most stable solid compound below pH 4.
Above this pH value, Pu02 would be the most stable solid (Figure 3-1.6).  Based
upon the Eh-pH diagrams for plutonium oxides, and hydroxides, Polzer (1971)
predicted Pu02(s) to be the most stable solid phase under conditions generally
found in the environment.  However, the relative stabilities of Pu compounds
other than hydroxides and oxides, and predictions regarding weathering
sequences, were not considered by Polzer (1971).
     Knowledge of the most stable compound present in the soil is necessary
in order to predict the behavior of fate of the element in the soil.  An exten-
sive review on actinides indicated that numerous workers have studied the
plutonium concentration and distribution with depth in sofls.  However, with  •
the exception of Price and Ames (1975), none have made any attempt to identify
the solid phases of plutonium present in soils.  Price and Ames (1975) reported
the presence of Pu02 under both alkaline and acidic oxidizing environments.
Their observations were made approximately 20 years after the acidic radio-
active waste was disposed to the alkaline soils.  Tamura (1974) inferred the
presence of Pu02 in aridisols or entisols from the Nevada Test Site.  Thus,
the Price and Ames, and Tamura results confirm theoretical predictions made
regarding the most stable plutonium compound expected to be present in soils.
However, if the soil solution activity of plutonium is lower than the solu-
bility product of Pu02 and any other possible compound, the final solution
activities will depend mainly on cation exchange processes and the copre-
cipitation of plutonium with other soil minerals.
     Figure 3-18 demonstrates the activity of various plutonium species
expected in eauilibrium with PuO«(s) and in an oxidizing environment (p09 16
                                c                               3+    4+    +
atm).  Plutonium in solution exists in four oxidation states (Pu  , Pu   Pu02,

                                   3-111

-------
 Pu02+) and forms complexes with OH', d", F", SO*', H2PO~ and Co*' ions.   In
 general, the activity of all  positively charged ions and ion-complexes and a
 few neutral  ion-pairs decreases with the increase in pH, while the activity
 of all negatively charged ion-complexes increases with the increase in pH.
 In an oxidizing environment,  Pu + and Pu   and their ion-pairs have an insig-
 nificant effect on the total  activity of plutonium in solution.   At any given
 pH,  the total  activity of plutonium in solution can be obtained  by adding
 the  appropriate activities of various plutonium species.   Since  the lines for
 some plutonium species are based on concentration constants,  the position of
 the  lines may  change  when converted to thermodynamic equilibrium constants.
 Figures 3-17 and 3-18 were constructed by assuming a weathering  environment
 (pC02 = 3.53;  pd" =  p$0*~ =  2.5;  pF~ = 3.5)  and Pu02(s)  as the  compound  that
 may  be present in  the soil  at equilibrium.  When diagrams  are  constructed
 for  an actual  weathering  environment  of a  given  soil,  all  the  organic  and
 inorganic ligands  present in  the  soil  solution  should  be  considered. Fig-
 ures  3-17 and  3-18 were developed  for selected  inorganic  ligands  only.
                        -is
Figure 3-17.
The activity of various plutonium species in soil solution in
equilibrium with Pu02(s) at pH 8, pCOj = 3.52 atm, pCT =
     - =2.5 and pHjPO^f =5.0 (Rai and Serne, 1977)

                     3-112

-------
Figure 3-18.
The activity of various plutonium species in soil solution in
with Pu02(s) in a mildly oxidizing soil environment [pOg = 16
atm], pC02 = 3.52 atm, pCl = 2.5, pF~ =3.5 and pHgPO^ =5.0
(Rai and Serne, 1977)
     Figure 3-17 relates the activities of (a^, moles/liter) of various solu-
tion species at pH 8 to the oxidation potential as represented by the various
oxygen pressures.  In general, with a decrease in oxygen pressure (increase
in reducing conditions), the kind of activity of the solution species
changes:  1) Pu   species increase in activity, 2) Pu * species are unaffected
          i         OL
and 3) Pu02 and PuO^  species decrease in activity.  However, in doing so,
the total activity of plutonium in solution at any given pH would not change
significantly because the activity of species for which thermodynamic
equilibrium constants are known is very large as compared to the others.
When the concentration constants are converted to thermodynamic constants,
the new lines would shift downward or upward but remain parallel to the
present lines.  The activities of various species would still vary in the same
direction with the changes in pH as reported in Figure 3-18.
                                   3-113

-------
     A few of the reactions are not plotted in the figures because either
the activities of the species fall beyond the graph boundaries or the equilib-
rium constant data are not reliable.  Cleveland (1970) suggested that the
                                          2+
values for the formation constants of PuCO^  and Pu(OH)4° were suspect; the
values for Pu02(OH)3, Pu02(0)2°, and Pu02OH+ should be viewed with skepticism;
and the value for PuO,OH° should be considered only approximate.  Under the
                     £                               '          o
assumptions of Figure 3-18, the lines for Pu(OH).0 and Pu02(OH)2 would be
parallel to the X-axis with 10~12'51 and lO"19-1^ moles/1 activity, respec-
tively.  Many researchers (Andelman and Rozzell, 1970; Fukai and Murray, 1974;
Grebenshchikova and Davydov, 1961, 1965; Sheidina and Kovarskaya, 1970) have
reported that plutonium in solution also exists as colloidal species over a
wide pH range.  Colloidal species and neutral  ion-complexes [Pu(OH)°»,
Pu02(OH)°2]-are not reported in Figures 3-17 and 3-18.  If present in large
quantities, these species would significantly influence the nature of the
species and the total concentration of plutonium in solutions.
     For the conditions specified in Figure 3-18, the dominant plutonium
species in low and high pH ranges are Pu02 and Pu02C03OH~, respectively.
The colloid complex of soils is mainly negatively charged.  Therefore, posi-
tively charged species are usually absorbed by soils in cation exchange pro-
cesses.  Since positively charged Pu02 species predominate only in solutions
of pH values <6, cation exchange adsorption of plutonium by soils would be
limited to acidic soils (pH <6).  The negatively charged Pu02C03OH~ complex
that predominates above pH 8 [even when the activity of Pu02CO.,OH~ is compared
to Pu(OH)°4] should not be readily adsorbed by the soils.  Thus, the activity
of plutonium in soil  solutions of pH >8 would  be relatively high because of:
1) little soil adsorption of negatively charged ions and 2) increases in the
activities of negatively charged ions with the increase in pH.  Similarly,
under extremely low pH conditions the activity of plutonium in soil  solutions
in equilibrium with PuOgU) may also be high.   Rhodes (1957) found low adsorp-
tion of plutonium by soils in the pH range of  approximately <2 and 8-13.  Thus,
his results confirm part of the predictions based on Figure 3-18; continuous
decrease in adsorption is predicted with the increase in pH above 8, whereas
Rhodes' data indicated first a decrease in adsorption to pH 12 and then an
unexplained continuous increase in adsorption  above pH 12.
                                   3-114

-------
      It  should be mentioned that  in addition  to cation, exchange,  the  total
retention of plutonium wastes percolating through soils would depend  on many
other factors, such as flow rate, precipitation of solid phases,  physical
entrapment of particulate and colloidal plutonium, soil properties, and waste-
water composition.  Therefore, the above adsorption predictions based only on
cation exchange should be viewed  as an approximation.  As the data on actual
adsorption isotherms and mechanisms and various soil-actinide interactions
accumulate, more accurate predictions will be possible.
Experimental Adsorption Results
     Thorburn (1950) reported a sorption experiment in which plutonium in
water was percolated through a soil column.  The soil was then wet sieved and
the activity of each particle size fraction counted.  In general  the decrease
in particle size increased the amount of Pu adsorbed (Table 3-52).  In other ad-
sorption experiments, Thorburn placed small aliquots of actual plutonium wasta
on top of soil columns and eluted with distilled water or saturated magnesium-
calcium  carbonate.  There were no significant differences in the  elution curves.
It is interesting to note that the elution curves had higher plutonium activity
in the first volume out of the column and always showed a small leakage which
Bright signify physical transport  of fine particulates, neutral polymers or
colloids.  Thorburn took the effluent from the distilled water leached column
and percolated it through a second soil column at various flow rates.  The
data show that Pu concentration in effluents through the second column varied
with flow rate.  The faster the flow the more plutonium in the effluent.  Fil-
tration  of the influent through a fine filter paper did not change the results.
Explanations of these phenomena are obscure but would seem to indicate plutonium
transport as polymer.  Final adsorption studies of plutonium in water versus
pH by soil columns showed 87% adsorption at pH 3.8, 77% at pH 7.2 and 24% at
pH 10.   Note that this falling of plutonium adsorption with increasing pH is
supported by the thermodynamic data of Figure 3-18.
     Evans (1956) reported percentage of plutonium removed from a 0.001 N HNO^
solution by various grain sizes of quartz.  The plutonium Kd values may be com-
puted from these data, and are listed in Table 3-53.  Data also were given on
plutonium adsorption on different minerals of coarse clay size (2 to 0.2 um).
                                    3-115'

-------
                  TABLE  3-52.   PLUTONIUM ADSORPTION  VERSUS
                                PARTICLE SIZE  (THORBURN,  1950)

                        Size,  nun              Pu.  d/min/g
                    >1.19                           39.3
                     1.19 - 0.25
                     0.25 - 0.15                 1,171
                     0.15 - 0.10                   929
                    <0.105                       8,578
                    Uncentrifugible
                      Silt and Clay             62,750
           TABLE 3-53.  PLUTONIUM Kd VALUES FOR QUARTZ OF VARIOUS
                        PARTICLE SIZES FROM A 0.001 N HN03-
                        PLUTONIUM SOLUTION (EVANS, 1956)

                  Quartz Particle Sizes             Kd,  ml/g
                  Coarse sand                          5.6
                  Fine sand                           10.0
                  Coarse silt (50-20 vat)               11.3
                  Medium silt (20-5  ym)               33.5
                  Fine silt (5-2 ym)                  48.8
                  Coarse clay (2-0.2 ym)               80.9
Plutonium Kd values were computed from these data and are listed in Table 3-54.
Evans suggested that the lower Kd for plutonium by montmorillonite is probably
due to experimental difficulties and believed that Rhodes' (1955) data may
represent preferred plutonium Kd results.  Rhodes' data, as reported by Evans,
are given in Table 3-55, assuming a 1 ml to 1 g solution to soil ratio.  This
ratio was not given, but the results are proportional in any case.
     Rhodes (1952, 1957a, 1957b) performed similar plutonium Kd determinations
on a Hanford sand soil (92% sand) with 2% calcium carbonate content.  Using
a soil to solution ratio of 1:20 and neutral pH, Rhodes found rapid adsorption.
Analysis versus time yielded an equation for the amount adsorbed versus contact

                                    3-116

-------
          TABLE 3-54.  PLUTONIUM Kd VALUES FOR DIFFERENT MINERALS
                       OF COARSE CLAY SIZE (2 to 0.2 urn) FROM
                       0.001 N HN03-PLUTONIUM SOLUTION  (EVANS,  1956)

                       Mineral                   Kd, ml/g
                    Feldspar                       170.4
                    Quartz                           82.0
                    Glauconite                        «
                    Montmorillonite                 157.0
                    Kaolinite                     1091.0
                   TABLE 3-55.  RHODES' PLUTONIUM Kd DATA AS
                                REPORTED BY EVANS (1956)
                          Material                  Kd.  ml/g
                    Montmorillonite                  4990
                    Fine  micaceous sand               49
                    Kaolinite                        42.5
                    Soil                              36
                    Coarse sand                      .0.87
time, %PuADS = 81 + 5.2 log time/m,-n\-  For example, after 1 min, 20 min and
60 min, 81%, 88%, and 90%, respectively, of the solution plutonium concentra-
tion was adsorbed.  Increases in the soil to solution ratio were also observed
to increase the percent adsorbed.  Rhodes also obtained adsorption versus pH
data and found rapid decreases in adsorption below pH 2 but almost total adsorp-
tion (Kd > 2000) between pH 3 and 8.5.  Decreases at pH 10 to 12 were believed
to indicate changes in plutonium speciation.  In other experiments, coarse sand
was observed to adsorb only 47% of the plutonium present compared to 97 to
99.8% for soils and clay minerals.  Rhodes felt all these data indicated ion
exchange adsorption mechanisms.  Yet plutonium adsorption from distilled water,
a synthetic high salt waste and an actual high salt waste all showed greater
than 98% removal.  Plutonium solution loading experiments onto soil columns
at various pH's (1,4, 10) showed good removal at the first two pH's until the
                                    3-117

-------
high acid case had destroyed the CaC03 buffering capacity.  As the effluent
pH dropped to pH 1, increasing plutonium breakthrough was observed.  Plutonium
breakthrough at pH 10 was highly variable and rapid, possibly caused by fine
Plutonium particulate dispersion or polymer formation.  Plutonium adsorption
from distilled, water, 4M NaN03 and 2M(NH4)2HP04 solutions was above 96%, but
a 4M ammonium acetate solution showed only 59% adsorption.  Acetate complexing
of plutonium was probably responsible for the diminished adsorption rather
than any effects from competing ions.
     Further studies showed increasing the acetate concentration from 0.01M
to 8M dropped plutonium adsorption from 74% to 21%.  Acetate solutions were
also observed to have the capability of extracting plutonium adsorbed on soils.
Bensen (1960) presented data showing that oxalate salts also inhibit plutonium
adsorption by soluble Pu-oxalate complex formation.  Bensen also re-interpreted
Rhodes'  data and concluded precipitation reactions, not ion exchange, probably
control  plutonium adsorption on soils.
     Other investigations on plutonium were performed at Savannah River by
Prout (1953, 1959).  Plutonium adsorption by a soil (80% sand-20% clay) with
the clay mineral kaolinite dominant was determined versus pH for distilled
water traced with plutonium (MO  M).  The soil/solution ratio was 1:10 and
the mixtures shaken for 2 hr before phase separation by centrifugation.  Adsorp-
tion of plutonium depended on pH and the valence state of the plutonium added
to solution.  Adsorption was more than 90% complete from solutions of Pu(III)
and Pu(IV) between pH 2.5 and 12 and from solutions of Pu(VI) at pH's greater
than 6.   Strong adsorption at mildly acidic to mildly basic pH's is probably
due to a combination of cation exchange and precipitation of hydrolysis products.
Decreased adsorption above pH 9 possibly represents the formation of soluble
negatively charged Pu polymers which do not readily exchange with the soil.
     Knoll (1965, 1969) and Hajek and Knoll (1966) reported studies on high
salt wastes with organic contaminant? typical"of fuel  reprocessing.  An acid
(0.3M) high salt (5.4M NO^) waste spiked with  actual americium and plutonium
wastes were percolated through soil columns.   After 0.3, 1.5 to 4.0, and 20
column volumes of effluent, the plutonium breakthrough was 20 to 30%, 50%,
and 70 to 98%, respectively, for several columns.  Batch adsorption tests on
the high salt acid wastes from storage tanks showed low adsorption Kd

                                    3-118

-------
 values of from 2.4 to 2.9 ml/g with resultant soil pH's of 2 to 3.  If the
 waste was neutralized and the supernatant was used in batch adsorption tests,
 the Kd increased.   Addition of suspected organic contaminants to a concentra-
 tion of 20% by volume decreased the Kd for the neutralized waste to 1.4.   Water
 leachates of the sludge produced on neutralization contained some plutonium.
 Upon soil contact  the Kd for this leachate rose to 1540 ml/g.  If the high salt
 acid waste was neutralized in the presence of molar quantities of citric  acid,
 no precipitation was observed.  If this citrate-neutralized waste was perco-
 lated through soil  columns, most of the plutonium was removed.  A constant low
 leakage percentage of plutonium was observed and may have been caused by  a
 solubility or particle transport mechanism.
      In other experiments (Knoll, 1969) tap water spiked with plutonium was
 percolated through soil columns to allow plutonium adsorption.  The soil
 columns were then  leached with various organics:  TBP-CCl^, DBBP-CC1,, Fab Oil,
 D2EHPA in hydrocarbons C,Q-C-,^, and hydroxyacetic acid.   These organics are
 all commonly used  in various process streams during fuel reprocessing. TBP in
 carbon tetrachloride after 130 column volumes had caused only 5% of the soil
 sorbed plutonium to leach; DBBP removed 40% of the soil  sorbed plutonium  in
-80 column-volumes;  Fab Oil removed less than 4% in .30 column volumes;  D2EHPA
 removed about 30%  of the plutonium; and hydroxyacetic acid removed no less
 than 40% of the plutonium depending on the acid concentration.  When the  plu-
 tonium was added to the organics and then percolated through the soils, very
 little soil  adsorption was observed.
      Many investigations have been performed on plutonium adsorption by sedi-
 ments in both fresh and marine waters.  Schneider and Block (1968) found
                                •       j.0                 iO      J.
 plutonium to adsorb more than I  and Sr   but less than  Zn   or Cs  on Rhine
 River sediments.  Equilibrium was reached within 24 hr.
      The location  of plutonium at ground disposal sites  on the Hanford Reserva-
 tion has been closely monitored for over 20 years.  Liquid wastes disposed to .
 the unsaturated sandy soil in several instances included trace amounts of
 plutonium.  Typically, plutonium has not been found to percolate through  the
 soils to any extent.  Typical data (Brown, 1967) show 99.9+% of long-lived
 radionuclides such as plutonium to be adsorbed within the first 10 m of soil.
                                     3-119

-------
Crawley (1969) reports on plutonium penetration at two facilities."  Plutonium
in a sand to which slightly acidic high salt wastes were released had pene-
trated at least 18 ft.  Plutonium in a sand to which neutralized low salt
water had been released penetrated less than 2 ft.
     Desai and Ganguly (1970) determined the ability of humic and fulvic acid
in ammoniacal solutions to solubilize plutonium.  In this basic solution with-
out humic or fulvic extract, only 13% of added plutonium remained in solution;
with humic extract 54% of the added plutonium remained in solution; and
further testing showed that the soluble plutonium was noncationic.   The fulvic
acid extract solubilized 58% of the plutonium, and again the complex was
noncationic.
     Rozzell and Andelman (1971) studied the adsorption-desorption  of aqueous .
             -8      -7
plutonium (10   to 10" M) on quartz and other silica surfaces.  Adsorption
continued typically for 12 to 15 days before apparent equilibrium was reached.
At pH 7, adsorption increased with increasing ionic strength, but decreased
when HCOg was added.   The amount of adsorption varied at pH 5 and 7, but dif-
ferently at high and low ionic strengths, as well as with age of the solution.
Plutonium desorption indicated that there were two basically different adsorbed
species.  There was a great net desorption at pH 5 versus pH 7 to 9.
     Tamura (1972) discussed the adsorption of plutonium on several materials
including the soil constituents listed in Table 3-56.  The suspension pH repre-
sents the final system pH.

              TABLE 3-56.  REMOVAL OF PLUTONIUM FROM pH 7 WATER
                           BY SEVERAL SOIL MINERALS  (TAMURA, 1972)

                 Constituents        Kd,  ml/g     Suspension pH
              Attapulgite              4,370          9.60
              Montmorinonite            630          9.20
              Alumina, activated         755          8.35
              Kaolinite                  352          6.25
              Illite                      129          5.90
              Quartz                        0          6.35
                                     3-120

-------
     Fried et al. (1973, 1974, 1974) found that plutonium adsorbed on lime-
stones, shales, and sandstones would very slowly migrate through the pores.
The velocity of migration relative to the leaching water was 3 x 10   for
                    -5
limestone and 6 x 10   for basalt.  If the leaching solution was saturated
with C02 an increased elution from the limestone and sandstone was observed.
Studies of plutonium movement along a fissure crack in basalt shows plutonium
                                                                      +4
movement in two phases:  the slower moving phase adsorbs similar to Pu  , but
a faster moving phase with only about 1/10 the tendency to adsorb was also
observed.  Fried et al. hypothesized a polymeric state for this phase.
     Fried et al. (1974, 1974) immersed disks of limestone and basalt in
Pu(N03)4 solution at 4 x 10  M Pu.  After equilibration the activity in the
solutions was re-measured.  Losses in plutonium were equated to adsorption on
the rocks.  Basalts adsorbed more plutonium than limestone on a surface area
basis.   The effects of salt solutions of Na , Ca  , Sr  ,  La  , and Zr   were
also studied.  In all  cases the salts displaced some of the plutonium from
the rocks.  The higher the concentration and the larger the valence charge,
the easier was the displacement.   This was similar to the  trends expected for
ion exchange mechanisms.  A leaching study showed eight column volumes of
0.5M HC1 removed 30% of the plutonium adsorbed on_shale.  Eight additional
column "volumes of 4M HC1 removed an additional 20%.
     Two review articles which contain discussions on adsorption reactions
of plutonium and soils are available, Francis (1973) and K. Price (1971, 1973).
     Ames (1974), Price and Ames (1975) and Ames (1976) characterized the
actinide-bearing sediment underneath a liquid waste disposal facility which
received high salt, acidic wastes.  The distribution in plastic impregnated
core samples showed a surface soil plutonium concentration of near 0.5 mCi
239                          "                           239
   Pu/g.  This concentration decreases to about 0.6 uCi    Pu/g within the
first 2 m of underlying sediment and to less than 60 pCi/g at the maximum
depth sampled (9 m).  Examination of the contaminated soils showed at least
two types of plutonium present:  1) discrete Pu02 particles (>70 wt% Pu02) and
2) plutonium (<0.5 wt% Pu02) associated with silicate hydrolysis.  The Pu02
particulate form "filtered out" within the upper 1 m of sediment.  The
"hydrolysis" type of plutonium penetrated deeper within the sediment and
deposited in association with silicate hydrolysis products as the soil minerals

                                    3-121

-------
reacted with-the acid waste.  The highest concentrations of nonparticulate
Plutonium about 1 m below the bottom of the disposal facility receiving acidic
wastes were associated with smectites present in the soil before waste solu-
tion disposal.  The smectites higher up in the column tended to be partially
to totally altered because of the acidic nature of the waste solutions.  Plu-
tonium was uniformly distributed in the sectioned smectite particles, suggest-
ing adsorption of charged plutonium species by the clays.
     Emery and co-workers (1974, 1974, 1975) have studied the ecological
behavior of one Hanford liquid waste pond.  Ninety-five percent of the incoming
low-level waste percolates through the desert sands.  The sediments are the
principal repository, with the top 10 cm containing an average 390 pCi Pu/g.
The overlying water contains 0.01 pCi Pu/1.  Assuming that these values con-
stitute equilibrium conditions, an estimate of the plutonium Kd can be made.
The resultant Kd value is 3.7 x 10  ml/g.   Plutonium in the interstitial
waters appears to be mainly cationic or nonionic forms.  Plutonium in the pond
water appears to be mainly fine particulates possibly signifying that the waste
stream plutonium content is predominantly Pu02 particles.
     Miner et al. (1973, 1974), Polzer and Miner (1974, 1977) and Glover et al.
(1977) described-experiments in progress to measure the adsorption of both
"soluble" plutonium and particulate plutonium (Pu09) in numerous soils at
                                  6        8
three plutonium levels between 10   and 10  M.  The plutonium adsorption was
rapid and quite high:   59% of the time, adsorption was greater than 99% (Kd >
430) and only 5% of the time was it below 90% adsorption (Kd < 39).  Correla-
tions were found between soil characteristics associated with soil ion exchange
or acidity.   The adsorption also depends on the initial plutonium concentrations
for which several possible causes are discussed including precipitation with
increasing plutonium concentration and the formation of carbonate complexes.
     Experiments at Savannah River (Savannah River Laboratory, 1975)  with
degraded TBP-DBBP-kerosene wastes containing plutonium showed rapid migration
through soils (Kd ^ 0.4 ml/g).  When small portions of the solvent waste were
allowed to dry on soils and leached by groundwater, however, very little plu-
tonium migration was observed (Kd ^ 6000 ml/g) in the effluent.
                                    3-122

-------
     Duursma et al. (1971, 1973, and 1974) have determined the adsorption of
Plutonium in the marine environment.  The adsorption of plutonium is inter-
mediate; less than Cs, Rb, Zn, Fe, Zr-Nb, Ru and Pm and greater than Ca and
Sr.  Plutonium has been observed to penetrate sediments at least to 8 to 14 cm,
and in the early 1970's, inventories in the sediment appear to show losses.
Biological processes or upward migration in interstitial waters as a reduced
species are two possible explanations, although Mo and Lowman (1975) showed
data which find reducing conditions tend to lower plutonium mobility.  Changes
in the americium to plutonium ratio from waters to sediments in the ocean led
Duursma to conclude americium is preferentially adsorbing or not being remobilized
as fast as plutonium.  In seawater Pu02(C03)3~  may predominate and cause the
                                     237
lower plutonium adsorption.  Using a    Pu tracer in valence states +3, +4, and
+6 in both oxic and anoxic seawaters at pH 7.8 to 8.0, Duursma found the'Kd
for marine sediments to be about 10  ml/g.  Valence states and solution Eh
did not cause marked differences.
     Fukai and Murray (1974) performed plutonium adsorption-desorption experi-
ments for both freshwater and saline conditions.  Plutonium (III) adsorption
experiments from river water to river sediments (100 ml traced solution/g
sediments) were performed.  The initial stage of adsorption was rapid; 50% .
within 10 hr.  For pH 3 to 11 over 90% of the Pu(III) was adsorbed by the
river sediment.  Desorption experiments were performed wherein the sediments
used in the adsorption experiments were suspended in seawater.  Over the pH
range 4 to 10 about 10% of the adsorbed plutonium was released without apparent
pH differences.  Similar adsorption-desorption experiments were performed with
river sediments contacted with various dilutions of freshwater sewage effluent
at pH 8.1.  The Pu(III) adsorption increased from 85% in river water to over
95% at 100% sewage effluent.  These sediments were next contacted with sea-
water and desorption monitored.  At equilibrium, river sediments that had not
been contacted with sewage effluent released only 5% of the plutonium, whereas
sediments contacted with 5 'to 100% by volume sewage effluent released 12%.
     Van Dal en et al. (1975) determined the plutonium distribution coefficient,
Kd, for several Dutch subsoils from 90% saturated NaCl solutions between pH 7
                                                                   4
and 8.  Clay samples of mainly illite and kaolinite had a Kd of MO  ml/g and
                                    3-123

-------
for a river sand a Kd of 200 ml/g.  Gypsum-bearing and clay-bearing sandstones
were intermediate.  The pH dependence on adsorption of plutor.ium was minor
between pH 5 and 8.
     Mo and Lowman (1975) placed piutoniurn-contaminated calcareous sediment
in aerated seawater and anoxic seawater and stirred the system until solution
activities were at equilibrium.  The resultant plutonium Kd values were 1.64
x 104 and 3.35 x 105 ml/g.
     Tamura (1975, 1975) evaluated the plutonium content of various size frac-
tions of contaminated Nevada Test Site and Oak Ridge soils and Mound Laboratory
sediments.  These results should be considered exploratory in nature because
the results are based on only a few observations; the source terms differed;
and the particle size fraction separations were technique dependent.  The NTS
soils showed different particle sizes with plutonium associations depending on
location from ground zero.  In general, the coarse silt (20 to 53 um) and medium
silt (5 to 20 ym) fractions contained about 80 to 85% of the total plutonium.
The Oak Ridge plutonium was more evenly distributed in the silt and clay sizes,
with clay slightly higher.  The Mound Laboratory sample showed 70% of the
plutonium in the clay fraction.
     Bondietti et-al., (197-5.)-- found, purified soil humates" to absorb'plutonium
strongly, 99.9+% while reference clays (montmorillonite and kaolinite) adsorbed
about 98 and 96% of plutonium added to a water solution at 163 iag Pu/g of solid.
If the organic matter and iron oxide coating were removed from clay material,
plutonium adsorption was 95% versus 99.9% while intact.  Leaching solutions,
citrate, DTPA, and fulvate removed small amounts of adsorbed plutonium as
shown in Table 3-57.  Thus, it appears that natural soil  organic matter can
very strongly adsorb plutonium.  The ability of soil  organic matter to alter
the oxidation state of plutonium in soil water solutions  was also verified.
Pu(VI), and by inference Pu(V), were unstable in the presence of fulvic acid,
polygalacturonic acid and alginic acid.  Positive proof of reduction to Pu(IV)
was shown at environmental pH's 6.5 to 8.  Plutonium (IV) was the most stable
valence upon interaction with these organics.  Further reduction of Pu(IV)  to
Pu(III) occurred in the presence of humic or fulvic acids, but was not observed
                                    3-124

-------
above pH 3.1 under aerobic conditions.  Bondietti also demonstrated that plu-
tonium is at least partially associated with humic materials in ORNL soil
contaminated 30 years ago.

            TABLE 3-57.  PERCENT PLUTONIUM LEACHED BY EXTRACTANTS
                         (BONDIETTI ET AL., 1975)
                           	Percent Leached
           Extractant
Humate
0.3
30
5
Montmorillonite
8
15
0.9
Kaolinite
11
20
3
         10"3M Citrate
         10"3M DTPA
         10"2M Fulvate
     Bondietti (1974) determined the plutonium Kd for clays separated from a
soil (Miami silt loam) and montmorillonite using 5 x 10" M calcium solutions
at pH 6.5 for two plutonium valence states.  The results are shown in Table
3-58.  The high adsorption of Pu(VI) on the soil clay, in contrast to or as
compared with montmorillonite, indicated that plutonium reduction by clay com-
ponents must have occurred, since adsorption values similar to the Pu(IV) were
obtained.  Pu(VI) appeared to adsorb less than Pu(IV).

         TABLE 3-58.  PLUTONIUM Kd AS A FUNCTION OF OXIDATION STATE
                      (BONDIETTI, 1974)
             Material          Initial Oxidation        Kd, ml/g
        Soil Clay Fraction           Pu(IV)          1.0 - 1.7 x 105
                                     Pu(VI)          7.5 x 104
        Montmormonite              Pu(IV)          2.1 x 104
                                     Pu(VI)          2.5 x 102

     'Relyea and Brown (1975) studied the adsorption-desorption relations of
plutonium solutions added to water at pH 2.  The ratio of plutonium adsorbed
by the soil to that in the equilibrium solution was greater than 99.1.  Dif-
fusion of plutonium in four soils was found to be much slower than for exchange-
able soil cations.

                                    3-125

-------
                                      237
     Volesky and Friedman (1976) used    Pu to study the adsorption of plu-
tonium on Niagara limestone surfaces (k), and the effect of sodium chloride
                                                237
concentrations on k values.   A k is defined as (   Pu activity/ml of solution)/
 237              2                                         2
(   Pu activity/cm  of limestone), with the dimensions ml/cm .   The k value
is essentially constant until NaCl is 1M in concentration.  The k then
increases rapidly to 3M NaCl and again becomes constant.  As the k value
           237
increases,    Pu is being desorbed from the limestone and re-entering the
solution.  The desorption may result from the formation of plutonium chloride
complexes.
     Rai and Seme (1977) used thermodynamic data to develop solid phase-soil
solution equilibria diagrams which can be used to estimate plutonium behavior
in terrestrial environments.  Pu02 was the most stable phase in the pH and
oxygen partial pressure ranges found in soil environments.  Plutonium experi-
mental  results found in the literature were used to verify empirical  predic-
tions,  based on the diagrams, of plutonium adsorption on soils.
Migration Results
     Field Studies—
     Magno et al. (1970), in investigations of radionuclide migration from the .
Nuclear Fuel Services plant in New York State, estimated that more than 90%
       0-30
of the "°Pu,    Pu and    Pu contained in the plant effluent remained in the
sediments of the lagoon system.
     A series of wells were drilled into the contamination produced by the
1973 leakage of 435,000 liters of high level waste from a Hanford storage tank
(Anon., 1973).  Core sample analysis showed the relative mobilities of the
radionuclides contained in the high pH, high salt solution.  Plutonium had
migrated the shortest distance from the tank rupture of all the radionuclides
in the contaminated sediments.
     Emery et al. (1974a, 19746, 1975) have discussed the plutonium inventory
             2
of a 56,657 m  reprocessing waste pond at Hanford.  The pond has received
                                                    239 240
wastewater since 1944, which included about 1 Ci of    '   Pu.   It has
retained greater than 99% of this plutonium in the upper 20 cm of its sediments
even though no water leaves the pond except by evaporation or soil percolation
and the mean residence time of the water is only 40 hr (Emery et al., 1974).
The plutonium is probably present in a parti cul ate form.
                                    3-126

-------
     The plutonium resulting from operation of the natural reactor at Oklo
1.8 billion years ago was reported to have remained in the vicinity of the
original reactor site (Brookins, 1976).
     Bondietti and Reynolds (1975) investigated the species of plutonium found
in Oak Ridge seepage water from a sol'id waste burial ground.  Pu(VI) was iden-
tified as the main Pu species in the seepage solutions from the burial ground.
The implication is that Pu(VI) is more mobile than Pu(IV).
     Water infiltration and plutonium migration at Maxey Flats burial grounds
were studied by Meyer (1975).  A hydrostatic head of 4 to 6 m was found in
some trenches, causing lateral and downward migration of leachates.  Surficial
plutonium migration also had occurred from trench leakage.  Above background
plutonium concentrations have been detected in onsite and offsite surface
soils and stream sediments, in soil cores to 90 cm and in wells 15 to 27 m
deep.  Other radionuclides accompanying the plutonium included tritium,   Co,
89'9°Sr and 134-137Cs.
     Radiological measurements at the Maxey Flats burial  site, Kentucky, also
have indicated some physical radionuclides transport on particulate matter
(Meyer, 1975; Montgomery et al., 1977).  The authors mentioned that 54Mn, 60Co,
90    137    238       239
  Sr,    Cs,    Pu and    Pu were detected in stream sediments, although the
water levels were quite low.  In E-Series test well measurements, all plutonium
within the detection limit was associated with sediment.   The authors suggest
that this raises a question as to the mechanism of plutonium migration from
the disposal trenches.
     Laboratory Studies—
     Lindenbaum and Westfall (1965) prepared several dilute solutions of
colloidal plutonium with citrate to plutonium ratios of 1800 (3.4 x 10" M
citrate) to 1.  The colloid was about 80% ultrafilterable at pH 4.0 and a
citrate to plutonium ratio of 1.  The cellophane used in the ultrafiltration
had a pore size that corresponded to a plutonium hydroxide molecular weight
of about 200,000.  At pH 10, the citrate to plutonium ratio 1 colloid fell  to
65% ultrafilterable and 5% at pH 11.  The citrate to plutonium ratio 1800
colloid behaved in a similar fashion falling from 92% ultrafilterable at pH 4
                                    3-127

-------
to 72% at pH 11.  With no citrate present, the plutonium was about 13% ultra-
filterable at pH 4 and fell to near 0% at pH 11.  Qualitatively, Zr(IV), Th(IV)
and the rare earths may behave in a similar manner, peptizing at higher pH,
as suggested by the works of Rhodes (1957).  The dispersed polymers are less
soil-reactive and should migrate farther through the soil.  Hajek (1965) deter-
mined that the migration rate in soil, relative to groundwater, was 1/100 and
1/1000 times the rate of groundwater flow for strontium and cesium, respectively.
The diffusion velocities of plutonium and americium were 1/10,000 times the
transporting solution velocity (Hajek, 1966).  The apparent self-diffusion
                                      -12   2
coefficient for plutonium was 4.8 x 10    cm /sec, so that the average distance
diffused in 329,000 years was 10 cm (Hajek, 1966).  The 329,000 years was
                    239
13.5 half-lives for    Pu.  Relyea (1977) determined the diffusion coefficient
for plutonium in the same soil using a quick freeze method developed by Brown
et al. (1964).   He obtained an'apparent plutonium diffusion coefficient of
      -10   2
5 x 10    cm /sec at 791 hr on the same Burbank loamy sand.
     Hajek (1966) reported that plutonium mobility from contaminated soils was
low in the Hanford 216-Z-9 covered trench.  About 0.1% of the plutonium could
be leached with groundwater, but about 7.5% of the resident americium was
removed.  No further plutonium was leached after a throughput of 13 column
volumes of groundwater.  At least 3.5% of the plutonium on the soil was removed
by 20 column volumes of IN NaNOg solution.  The leach rate was constant after
20 column volumes.  One column volume was equivalent to 61 cm of leaching
solution.
     The role of microorganisms in the movement of soil plutonium has been
under study by Au (1974).  He reported that the plutonium discrimination factor
was about 4 between agar growth medium at pH 2.5 and Aspergillus spores.   The
pH 2.5 probably aids in maintaining plutonium mobility.  However, the usual
discrimination factor between soil and plants, for example, is 10  to 10
(Francis, 1973).  Acid-producing molds that could increase plutonium solubili-
zation were also identified in the soil.
     Fried et al. (1974), in their plutonium distribution studies in limestones
and basalts, indicated that the adsorption coefficients were dependent on the
types and amounts of other ions in solution.   At least two chemical forms of
                                    3-128

-------
Plutonium were migrating in neutral solutions.  One of these, possibly a
polymerized plutonium oxide, migrated ten times faster through rock fissures
than the other form.
     Jakubick (1975) modeled the migration of plutonium down the soil profile
and found that the migration rate of PuCL (0.8 cm/yr) was about 100 times
faster than plutonium applied to the soil as Pu(N03)4-
     Nishita et al. (1976) determined the extractability of plutonium from a
contaminated soil as a function of pH.  The pH influences hydrolysis and pre-
cipitation of the plutonium.  In the study, 2 g of contaminated soil was agi-
tated in 25 ml of solution, in duplicate, and the pH adjusted with NaOH or
HN03.  The Kd values are given in Table 3-59 for the Aiken clay loam, a
kaolinitic soil, at several pH values.  It is not clear from the report whether
these values are initial or final system pH measurements.  The pH values, at
the high and low ends of the range especially, usually drift considerably
during equilibrations with soils so a final pH measurement is most meaningful.
The Kd values are given in Table 3-59 because it should be of little importance
as to which direction equilibrium is approached.

             TABLE 3-59.  Kd VALUES OF 238Pu IN UNTREATED" AIKEN
                          CLAY LOAM (NISHITA ET AL., 1976)
_pji_
1.21
2.12
2.56
4.69
5.55
7.08
"°Pu Kd, mg/Jl
430 ± 44
862 ± 33
796 ± 59
2591 ± 591
2347 ± 220
3086 ± 19
pH
8.54
9.43
10.31
11.25
12.22
13.25
"°Pu Kd, mq/2,
963 ± 130
512 ± 14
3.02 ± 1.6
213 ± 5.8
138 ± 4.8
207 ± 14
     The migration of plutonium and americium in the lithosphere was investi-
gated by Fried et al. (1976).  Most of the plutonium and americium were strongly
held on tuff, basalt and limestone.  Migration rates of 10   m/m of water flow
were measured in one case, and 30 um/m of water flow for plutonium migration
in another instance into limestone.  A small amount, less than 35%, of the
plutonium appeared to migrate at a rate about 10 times as fast as the rest of

                                    3-129

-------
the plutonium.  This rapidly moving plutonium may be adsorbed on fine mineral
particles.  Also, the chemical environment can drastically increase the migra-
tion rate of plutonium and americium.  Even dilute solutions (10  M) of highly
charged ions cause substantial desorption and migration of plutonium and
americium adsorbed on rock surfaces.
     Relyea and Brown (1975) measured the diffusion coefficients of plutonium
in soil and in water by a capillary tube diffusion cell suggested by Brown et
al., (1964).  The soil diffusion coefficients were measured by contacting cells
containing contaminated and uncontaminated soils for a known time interval,
freezing instantly with liquid air, and slicing and counting of sections of
the untagged soil perpendicular to the diffusion axis.  The apparent diffusion
                                                               -8   2
coefficients were low compared to normal soil cations, 1.4 x 10   cm /sec in a
                              -11   2
sandy soil to less than 5 x 10    cm /sec in a silt loam.  Associated Kd values
were 300 to 500 ml/g for the sandy soil and 10,000 ml/g for the silt loam.
EDTA and DTPA reduced most Kd values by a factor of up to 100.   The aqueous
                                           -7   2
diffusion coefficients varied from 3.1 x 10   cm /sec in solution extracted
                              -5   2
from the silt loam to 2.7 x 10   cm /sec in a solution extracted from the sandy
soil.
     Fried et al. (1977) reported the upper 'limits of the relative migration
ratios (radionuclide distance moved/water distance moved) of plutonium and
americium through rocks to be 100 um/m and 500 um/m of water flow, respectively.
Using a migration ratio of plutonium of 125 um/m of water flow, and a water
flow of 0.6 km/yr, the plutonium would migrate no more than 12.5 cm/yr.  The
241
   Am moves faster but has a shorter half-life.   The authors concluded that
deep burial in geological formations would provide effective isolation of
actinides from the environment.  Several rather large assumptions are contained
in this conclusion.   One of these assumptions is that the small cores used to
obtain the migration data perfectly model a geologic waste repository.
     Further work on the rapidly migrating form of plutonium was reported by
Friedman et al. (1977).   In one case, a HOEHP column was used to remove all
polymeric plutonium species before the experiment, and the rapidly moving plu-
tonium was eliminated.  In another treatment, the plutonium solution was evapo-
rated repeatedly with HN03 with the same experimental results.   Thus, the.
identification of the rapidly moving plutonium form as a polymer was verified.

                                    3-130

-------
A model was devised to predict the flow of wastes containing plutonium through
rock fissures.  The model will be field tested by core drilling a former dis-
posal site at Los Alamos and comparing predicted and actual movement of the
actinides through the tuff.
     Fried et al. (1977) also studied the retention of Pu(VI) on Los Alamos
tuff with a stock solution of plutonium containing 268 ug Pu(VI)/ml.  The
                          +2
adsorption spectrum of PuO   at 830 nm was used to verify the purity of
the stock solution.  A 10 ml tuff cylinder, coated with wax on the exterior,
was -used at a flow rate of 2 ml/hr.  Prior to use, the tuff column was vacuum
evacuated under water to ensure saturation.  Measured amounts of Pu(VI) were
placed on top of the column and allowed to percolate downward.  Water was
slowly passed down the column to simulate the passage of groundwater.  The
effluent water was collected in several fractions and analyzed.  The value
obtained for Pu(VI) migration was 26,500 um/m of water flow through the tuff.
The comparable Pu(IV) migration was 100 um/m of water flow.  After eight free
column volumes (2.4 ml/free column volume) passed through the tuff, a switch
to 4N HNO^ rapidly stripped most of the remaining plutonium.  A small amount
of residual plutonium, evenly distributed in the tuff column, may or may not
have been Pu(VI).
     Wildung et al. (1973, 1977) have considered the influence of soil micro-
bial processes on the long-term solubility, form and plant availability of the
transuranic elements.  Diffusible plutonium in soils, usually less than 0.1%
of the total plutonium, appeared to be particulates of hydrated oxide but
there was evidence of microorganisms affecting the solubility, form and plant
availability of transuranics.  Plutonium toxicity was a function of plutonium
solubility in soil but microorganisms were generally resistant, with toxicity
due to radiation effects.  Highly resistant bacteria, fungi and actinomycetes
were isolated from the soil and shown to transport plutonium into the cell,
altering its form to a soluble, negatively-charged complex.  The plutonium
forms were not well defined, but were dependent on organism type, carbon source
and plutonium exposure time during growth.
                                    3-131

-------
Summary
     In most oxidizing environments, ionic plutonium is expected to be present
in solution predominantly as Pu(V) or Pu(VI) (Figures 3-17 and 3-18).   With
high plutonium concentrations, PuCL could be expected as a stable solid phase
(Figures 3-15 and 3-16).  The plutonium adsorption data indicate that more
than one mechanism is operating during reactions with rocks and soils.  There
is a direct correlation of plutonium adsorption results with soil cation
exchange capacities (Thorburn, 1950; Evans, 1956; Rhodes, 1952, 1957a, 1957b;
Van Oalen et al., 1975; Glover et al., 1977) indicating that part of the
adsorption is due to ion exchange.  However, little effect on adsorption was
obvious from competing cation concentrations that were very high (Rhodes,
1957a; Mo and Lowman, 1975).  Perhaps the principal correlation of plutonium
adsorption is with pH (Rhodes, 1952, 1957a, 1957b; Knoll, 1965; Hajek and
Knoll, 1966; Knoll, 1969).  A relatively rapid soil column breakthrough of
plutonium was obtained with a low pH (<1) solution (Knoll, 1965).  Strong
adsorption occurs over the pH range of from 4 to 8 (Prout, 1958, 1959), but
above pH 8 the formation of negative complexes or polymers with low charge
densities can cause lowered adsorption (Thorburn, 1950; Rhodes, 1957a; Polzer
and Miner, 1977).  In addition, plutonium is easily complexed with- humic acids
(Desai and Ganguly, 1970; Bondietti et al., 1975) and with oxalate (Bensen,
1960), acetate (Rhodes, 1957a) and a large number of phosphorus-bearing organics
used in the nuclear fuels reprocessing industry (Knoll, 1969).   Complexed plu-
tonium also can result from soil  fungal  or microbial  activities (Wildung
et al., 1973, 1977; Au, 1974) all of which result in more mobile plutonium.
Even the precipitated or adsorbed plutonium can be readily moved through the
aqueous environments in the form of particulates or suspended sediments
(Magno et al., 1970; Emery et al., 1974a, 1974b, 1975; Meyer, 1975;
Montgomery et al., 1977).  If the Eh environment allows it, Pu(IV) can be
oxidized to Pu(V) or (VI), with much lower plutonium adsorption values result-
ing (Bondietti, 1974).  The plutonium polymers usually are adsorbed readily,
but can migrate more rapidly than soluble plutonium in some instances  where
the surface charge density is low (Friedman et al., 1977).  Much experimental
work remains to be done to separate and understand the several  reactions
mechanisms occurring in plutonium adsorption on rocks or soils.  A controlled
Eh environment for the adsorption work is important as an experimental condi-
tion when an element disproportionates as readily as plutonium.
                                    3-132

-------
References

Ames, L. L., Jr.  1974.  Characterization of-Actinide Bearing Soils:   Top
Sixty Centimeters of 216-Z-9 Enclosed Trench.   BNWL-1812.

Ames, L. L., Jr.  1976.  Actinide Occurrences  in Sediments Following  Ground
Disposal of Acid Wastes at 216-Z-9 Covered Trench.   BNWL-2117, pp.  39-61.

Andelman, J. B. and T. C. Rozzell.  1970.  Plutonium in the Water Environment.
I. Characteristics of Aqueous Plutonium.   IN:  Radionuclides in the  Environment.
Adv. in Chem. Ser. No. 93. Amer. Chem. Soc., Washington, DC, pp.  118-137.

Anonymous.  1973.  241-T-106 Tank Leak Investigation.  ARH-2874.

Au, F. H. F.  1974.  The Role of Soil Microorganisms in the Movement  of Plu-
tonium.  IN:  Dunaway, P. B. and M. G. White (eds).   The Dynamics of  Plutonium
in DesertTnvironments.  NVO-142, pp. 135-141.

Bensen, D. W.  1960.  Review of Soil Chemistry Research at Hanford.   HW-67201.

Bondietti, E. A.  1974.  Adsorption of Pu(IV)  and Th(IV) by Soil  Colloids.
Agronomy Abstracts.

Bondietti, E. A., S. A. Reynolds, and M.  H. Shanks.   1975.  Interactions of Plu-
tonium with Complexing Substances in Soils and Natural  Waters.  IAEA-SM-199/51.

Bondietti, E. A. and S. A. Reynolds.  1977. Field and Laboratory Observations
on Plutonium Oxidation States.  BNWL-2117, pp.  505-533.

Brookins, D; G.  1976.  Shale as a Repository  for Radioactive Waste:   The
Evidence from Oklo. Environmental Geology.  1:255-259.

Brown, D. A., B. E. Fulton and R. E. Phillips.   1964.  Ion Diffusion:  I, A
Quick-Freeze Method for the Measurement of Ion Diffusion in Soil  and  Clay
Systems.  Soil Sci. Soc. Amer. Proc.  28:628-632.

Brown, D. J.  1967.  Migration Characteristics of Radionuclides Through Sedi-
ments Underlying the Hanford Reservation.  IN:   Disposal of Radioactive Wastes
Into the Ground.  IAEA-SM-93/16, pp. 215-228.

Cherdyntsev, V. V., V. L. Zverev, V. M. Kuptsov, and G. T. Kislitsina.   1968.
Plutonium-239 in Nature.  Geochemistry International.  5:355-361.

Cleveland, J. M.  1970.  The Chemistry of Plutonium.  Gordon and  Breach Science
Publishers.

Crawley, D. T.  1969.  Plutonium-Americium Soil Penetration at 234-5  Building
Crib Sites.  ARH-1278.

Desai, M. V. M. and A. K. Ganguly.  1970.  Interaction of Trace Elements with
Organic Constituents in the Marine Environment.  BARC-488.

Duursma, E. K. and D. Eisma.  1973.  Theoretical, Experimental and  Field Studies
Concerning Reactions of Radioisotopes with Sediments and Suspended  Particles of
the Sea.  Part C:  Applications to Field Studies.  Netherlands Journal  of Sea
Research.  6:265-324.
                                     3-133

-------
Duursma, E. K. and M. G. Gross.  1971.  Marine Sediments and Radioactivity.
IN:  Radioactivity in the Marine Environment.  National Academy of Sciences.
pp. 147-160.

Duursma, E. K. and P. Parsi.  1974.  Distribution Coefficients of Plutonium
Between Sediment and Seawater.  IN:  Activities of the International Laboratory
of Marine Radioactivity.  IAEA-163.  pp. 94-96.

Emery, R.  M., D.  C.  Klopfer and W.  C. Weimer.  1974a.  Ecological  Behavior
of Plutonium and Americium in a Fresh Water Ecosystem.  Phase I.  Limnological
Characterization of Isotopic Distribution.   BNWL-1867.

Emery, R.  M. and T.  R. Garland.  1974b.  The Ecological Behavior of Plutonium
and Americium in a Fresh Water Ecosystem:  Phase II.  Implications of Differ-
ences in Transuranic Isotopic Ratios.  BNWL-1879.

Emery, R.  M., D. C.  Klopfer, T. R.  Garland and W. C. Weimer.  1975.  Ecological
Behavior of Plutonium and Americium in a Freshwater Pond.  BNWL-SA-5346.

Evans, E.  J.  1956.   Plutonium Retention in Chalk River Soil.  CRHP-660.

Francis, C. W.  1973.  Plutonium Mobility in Soil and Uptake in Plants:  A
Review.  Journal Environmental Quality.  2:67-70.

Fried, S.  M., A. M.  Friedman, and L. H. Fuchs.  1973.  Retention of Actinide
Elements on Soils and Rocks in the Natural  Environment.  ANL-7996.  pp. 6-7.

Fried, S.  M., A. M.  Friedman, and L. A. Quarterman.  1974.  Annual Report on
Project AN0115A, Fiscal Year 1974.   ANL-8115.

Fried, S., A. M. Friedman, and R. Weeber.  1974.  Studies on the Behavior of
Plutonium and Americium in the Lithosphere.  ANL-8096.

Fried, S., A. M. Friedman, J. J. Mines, R.  W. Atcher, L. A. Quarterman and
A. Volesky.  1976.  The Migration of Plutonium and Americium in the Lithosphere.
IN: Actinides in the Environment, Chapter 3.  ACS Symposium Series 35,
A. Friedman, ed.

Fried, S., A. M. Friedman, J. Mines, G. Schmitz, and M. Wheeler.  1977.  Dis-
tribution of Plutonium and Americium at an Old Los Alamos Waste Disposal Site.
IN: Dynamics of Transuranics in Terrestrial and Aquatic Environments.

Friedman, A. M., S. Fried, J. Mines, and L. Quarterman.  1977.  Migration of
Pu and Am.  BNWL-2117, pp. 73-102.

Fukai, R. and C. N. Murray.  1974.   Adsorption and Desorption of Plutonium
and Americium in Freshwater - Sediment and Seawater - Sediment Systems.  IN:
Activities of the International Laboratory of Marine Radioactivity.  IAEA-163,
pp. 96-103.

Glover, P. A., F. J. Miner, and W.  L. Polzer.  1977.  Plutonium and Americium
Behavior in the Soil/Water Environment.  I. Sorption of Plutonium and Americium
by Soils.  BNWL-2117.
                                     3-134

-------
Grebenshchikova, V. I. and Y. P.  Davydov.   1961.   A Study of the State of Pu(IV)
in Dilute Solutions of Nitric Acid.   Radiokhimiya.  3:155-164.

Grebenshchikova, V.I. and Y. P.  Davydov. '  1965.   State of Pu(IV) in the Region
of pH 1.0-12.0 at a Plutonium Concentration of 2.0 x 10~5M.   Radiokhimiya.
7:191-195.

Hajek, B. F.  1966.  Plutonium and Americium Mobility in Soils.   BNWL-CC-925.

Hajek, B. F. and K. C. Knoll.  1966.   Disposal Characteristics  of Plutonium
and Americium in a High Salt Acid Waste.  BNWL-CC-649.

IAEA.  1975.  International Atomic Energy Agency.   The  Oklo  Phenomenon.   Proc.
Sym., Libreville, 23-27.   June 1975.

Jakubick, A. T.  1975.  Migration of Plutonium in  Natural  Soil.   IAEA-SM-199/3.

Keller, C.  1971.  The Chemistry of the Transuranium Elements.   Vol.  3.
Kernchemie in Einzeldarstellungen.  Verlag  Chemie  GmbH.

Knoll, K. C.  1965.  Reaction of High Salt  Aqueous Plus Organic  Waste with
Soil.  BNWL-CC-313.

Knoll, K. C.  1969.  Reactions of Organic Wastes  in Soils.   BNWL-860.

Lindenbaum, A. and W.  Westfal\.  1965.  Colloidal  Properties of  Plutonium in
Dilute Aqueous Solution.   InteV.  J.  App. Radiation and  Isotopes.   16:545-553.

Magno, P., T. Reavey,  and J. Apidiunaki-s.   1970.   Liquid Waste  Effluents from
a Nuclear Fuel Reprocessing Plant.  BRH/NERHL-70-2.

Meyer, G. L.  1975.  Preliminary Data on the Occurrence of Transuranium  Nuclides
in the Environment at  the Radioactive Waste Burial Site.  Maxey  Flats, KY.
IAEA-SM-199/105.

Miner, F. J., P. A. Glover, and H. W. Miller.  1973.  Plutonium  Behavior in the
Soil/Water Environment.  RFP-2004-A,  pp. 1-3.

Miner, F. J., P. A. Glover, and W. L. Polzer.  Plutonium Behavior in the Soil/
Water Environment. I.  Sorption of Plutonium by Soils.  Agronomy  Abstracts,  p.  35.

Mo, T. and F. G. Lowman.   1975.  Laboratory Experiments on the Transfer  Dynamics
of Plutonium from Marine Sediments to Seawater and to Marine Organisms.   CONF-
750503-5.

Montgomery, D. M., H,  E.  Kolde, and R. L. Blanchard.  1977.   Radiological Mea-
surements at the Maxey Rats Radioactive Waste Burial Site 1974-1975.  EPA-
520/5-76/020.

Nishita, H., M. Hamilton, and A.  J.  Steen.   1976.   Extractability of 238Pu  and
242cm from a Contaminated Soil as a Function of pH and  Certain Soil  Components.
Presented at Meeting of Soil Science Society of America.  November 28-
December 3.  Houston,  Texas.
                                     3-135

-------
Polzer, W. L.  1971.  Solubility of Plutonium in Soil/Water Environments.
CONF-710401, pp. 411-430.

Polzer, W. L. and F. J. Miner.  1974.  Plutonium Behavior in the Soil/Water
Environment.  II.  Mechanisms of Plutonium Sorption by Soils.  Agronomy
Abstracts,  p. 37.

Polzer, W. L. and F. J. Miner.  1977.  Plutonium and Americium Behavior in the
Soil/Water Environment.  II.  The Effect of Selected Chemical and Physical
Characteristics of Aqueous Plutonium and Americium on Their Sorption by Soils.
BNWL-2117, pp. 255-286.

Price, K.  R.  1971.  A Critical Review of Biological Accumulation, Discrimina-
tion and Uptake of Radionculides Important to Waste Management Practices.
BNWL-B-148.

Price, K.  R.  1973.  A Review of Transuranic Elements in Soils, Plants, and
Animals.  Journal Environment Quality.  2:62-66.

Price, S.  M. and L. L. Ames.  1975.   Characterization of Actinide-Bearing
Sediments  Underlying Liquid Waste Disposal Facilities at Hanford. IAEA-SM-
199/87.

Prout, W.  E.  1958.  Adsorption of Radioactive Wastes by Savannah River Plant
Soil.  Soil Science.  86:13-17.

Prout, W.  E.  1959.  Adsorption of Fission Products by Savannah River Plant
Soil.  DP-394.

Rai, Dhanpat and R. J. Seme.  1977.  Plutonium Activities in Soil Solutions
and the Stability and Formation of Selected Plutonium Metals.  Journal
Environmental Quality.  6:89-95.

Relyea, J. F.  1977. .The Behavior of Plutonium in Soil.  Ph.D. Thesis,
University of Arkansas, Fayetteville, AR.

Relyea, J. F. and D. A. Brown.  1975.  The Diffusion of Pu-238 in Aqueous and
Soil Systems.  Agronomy Abstracts,  p. 124.

Rhodes, D. W.  1952.  Preliminary Studies of Plutonium Adsorption in Hanford
Soil.  HW-24548.

Rhodes, D. W.  1957a.  The Adsorption of Plutonium by Soil.   Soil Science.
84:465-471.

Rhodes, D. W.  1957b.  The Effect of pH on the Uptake of Radioactive Isotopes
from Solution by a Soil.  Soil Science Society of America, Proceedings.
21:389-392.

Rozzell, T. C. and J. B. Andelman.  1971.  Plutonium in the Water Environment.
II. Sorption of Aqueous Plutonium on Silica Surfaces.  Advances in Chemistry
Series.  TQ6:280-98.
                                     3-136

-------
 Savannah  River Laboratory Quarterly Report Waste Management.   April-June 1975.
 DPST-75-125-2.

 Schneider,  H.  and  W.  Block.   1968.   On  the Question  of the  Capacity of the
 Rhine  for Radioactive Nuclides.   Sorption  of Radionuclides  by  Sediments  of
 the  Rhine.   Gas Gasserfach.   109:1410-15  (in German).

 Sheidina, L.  D.  and  E.  N.  Kovarskaya.   1970.   Colloidal  State  of  Pu(IV)  in
 Aqueous Solutions. Sov.  Radio.  Chem.  12:229-233.

-Tamura, T.   1974.  Distribution  and Characterization of Plutonium in Soils
 from Nevada'Test Site.   p.  24-42.    IN.:   P.  B.  Dunaway and  M.  G.  White (ed.)
 The  Dynamics  of Plutonium in  Desert Environments.  NVO-142.

 Tamura, T.   1972.  Sorption Phenomena Significant in Radioactive  - Waste
 Disposal.  IN:   Underground Waste Management and Environmental  Implications.
 T. D.  CookTid.).  Am.  Assoc.  Petrol. Geol.  pp.  318-330.

 Tamura, T.   1975.  Characterization of  Plutonium in  Surface Soils from Area
 13 of  the Nevada Test Site.   IN:  The Radioecology of Plutonium and Other
 Transuranics  in Desert Environments.  M.  G.  White and P.  B. Dunaway (eds.).
 NVO-153,  pp.  27-41.

 Tamura, T.   1975.  Physical and  Chemical  Characteristics  of Plutonium in
 Existing  Contaminated Soils and  Sediments.   IAEA-SM-199/52.

 Thorburn, R.  C.  1950.   Absorption  on Hanford Soil and Related Soil Properties.
 HW-15655. '                      -   .                            .

 Van  Dalen,  A.,  F.  DeWitte,  and J. Wiskstra.   1975.   Distribution  Coefficients
 for  Some  Radionuclides Between Saline Water and Clays, Sandstones and Other
 Samples from the Dutch Subsoil.   Reactor  Centrum Nederland.  pp.  75-109.

 Volesky,  A. F.  and A. M.  Friedman.   1976.   Niagara Limestone Absorption  of
 Plutonium from Salt  Solutions.   ANL.

 Wildung,  R. E.  and T. R.  Garland.   1973.   Influence  of Soil Microbial  Activity
 on the Uptake and  Distribution of Plutonium in  the Shoots and  Roots of Barley.
 BNWL-1850,  Pt.  2,  pp. 22-25.

 Wildung,  R. E.  and T. R.  Garland.   1977.   The Relationship  of  Microbial  Pro-
 cesses to the Fate and Behavior  of  Transuranic  Elements in  Soils  and Plants.
 IN:  The  Transuranium Elements in the Environment.   W. C. Hanson  (ed.).
                                     3-137

-------
PROMETHIUM
Natural Soil and Rock Distributions
     Promethium has not been found to occur naturally in soils and rocks.
Brief Chemistry
     There are 16 known radioactive isotopes and isomers of promethium with
mass numbers from 141 to 154 and half-lives ranging from 34 sec to 17.7 years
(Lavrukhina and Pozdnyakov, 1970).  Promethium-147 (half-life 2.5 years) is
                                                           235
the only promethium isotope obtained in large amounts from    U fissioning by
thermal neutrons, and is present in fuel reprocessing fission products.
     Promethium is a lanthanide and like most lanthanides exists in a trivalent
state.  Only the (III) oxidation state of promethium is stable in aqueous soil
solutions.  Starik and Lambert (1958) studied the state of micro-amounts of
promethium in aqueous solutions at different pH values by ultrafiltration and
centrifugation.  The results showed that promethium was present as Pm   at
pH < 3.  With an increase in pH, hydrolysis products gradually accumulate as
multiply-charged aggregates.m  After a maximum is reached at pH of about 7, the
charged aggregates begin to flocculate.  At higher pH values, the aggregates
become neutral or negatively charged.  Electrophoteric measurements showed that
from pH 5.3 to 6.7, the bulk of the polynuclear particles was positively
charged.  At pH 9, no mobility is seen, indicating neutrally charged particles.
At pH 5 to 8, no promethium can be centrifuged, indicating a colloid in solu-
tion.  At pH 9, uncharged forms such as [Pm(OH)x(N03)3_x]  predominate (Starik
et al., 1959).
Solid Phase and Soil Solution Equilibria
     Promethium forms crystalline compounds with F", Cl", Br", I, OH", P04 ",
AsO 3~, B033~, Cr04, Cr03> Mo042~, W042", V043~, Sc033" (Lavrukhina and
Pozdnyakov, 1970; Wheelwright, 197.3).  However, the thermodynamic data are
available for Pm(OH)3 (solid) (Sillen and Martell, 1964) and Pm203 (Pourbaix,
1966).  It is evident that Pm(OH)3 is more stable than Pm203 (Figure 3-19).
The Pm   activity in equilibrium with Pm(OH)3 decreases a thousand-fold with
a one unit increase in pH.
                                    3-138

-------
     Promethium is a lanthanide and like most lanthanides exists in a trivalent
               No data were found on solution species of promethium except for
state (Pm3+).
       (PnT  + NOZ * PmNO,  log K° at I = 1  is 2.48) (Lavrukhina and
    J                                                                3+
Pozdnyakov, 1970).  The solution species is  probably predominantly Pm  , much
like europium.  If it assumed that promethium does not form any other complexes,
then the concentration of promethium in solution will be governed mainly by
Pm  .  Under these conditions, ion exchange  can be an important promethium
removal mechanism in acidic environments and at trace promethium concentrations.
     Figure 3-19.  The relative stability of various promethium solids.

Experimental Adsorption Results
     Bensen (1960) reported the complete uptake of promethium and other triva-
lent rare earths at pH > 6.  Promethium behavior in soil systems was said to
be identical to cerium behavior.  The uptake of promethium is nearly complete
and unaffected by 0.5M alkali metal and 0.25M alkaline earth metal cations at
above pH 7.  Below pH 3, the adsorption of all rare earths including promethium
is depressed similarly by accompanying salts.  The results indicate that
promethium and other rare earths are ionic species below pH 3, and begin to
precipitate as charged polymers above pH 3.  Bensen (1960) also reported that
the species and amount of accompanying anion affect promethium adsorption by
soils.  'Citrate, acetate and carbonate ions were found to inhibit rare earth
                                    3-139

-------
adsorption on soils.  This was attributed to formation of complex ion species.
Chloride and nitrate ions had no appreciable effect on adsorption, and sulfate
had very little effect.
Migration Results
     Field Studies—
     Magno et al. (1970) investigated migrating radionuclides from the Nuclear
Fuel Services plant in western New York State.  They estimated from analytical
                                  147
data that greater than 90% of the    Pm discharged from the plant was deposited
by sedimentation in the lagoon system.  These results imply that most of the
promethium was a precipitate, probably Pm(OH)3, or adsorbed on other solids
in the water that settled out in the lagoons.
     Laboratory Studies—
     Ames (1960) reported the removal of promethium during a replacement
reaction of calcite with phosphate in solution.  The promethium was finally
                                               2+                     3+
immobilized in apatite, Ca^POJgOH, where 2 Ca   are replaced by a Pm   and
Na  or seme other univalent cation.
     Schulz (1965) classified promethium with other rare earths as immobile
in soils due to either very strong adsorption by clay particles or present
as insoluble hydroxide.
Summary
     Low concentrations of promethium are expected in equilibrium with
in slightly acidic to slightly alkaline conditions (Figure 3-9) so that
PmfOHjj precipitation may control promethium concentrations in soil  and sedi-
ment solutions.  Promethium, like europium, would be expected to be ion
exchangeable below the concentration required for Pm(OH)- precipitation.   Only
limited data are available on promethium adsorption on soils and sediments,
but these data tend to support the above predictions.
     The removal of promethium is nearly complete on soils and rocks above
pH 6 (Bensen, 1960).  The effect of competing cations on promethium adsorption
above pH 7 is minimal.  Below pH 3, the adsorption of all rare earths including
promethium are similarly depressed by competing salts.  Charged polymers  can
form under the proper conditions during Pm(OH)3 precipitation (Starik et  al.,
                                    3-140

-------
1959; Bensen, 1960).  Citrate, acetate and carbonate ions inhibited rare

earth adsorption on soils., probably due to the formation of complex ion species.
Over 90% of the promethium discharged to a lagoon system remained in the lagoon

(Magno et al., 1970) as strongly adsorbed on clay particles or present as an

insoluble hydroxide (Schulz, 1965).  Promethium can be removed from solution
during replacement reactions (Ames, 1960).

References

Ames, L. L., Jr.  1960.  Some Cation Substitutions During the Formation of
Phosphorite from Calcite,  Economic Geology.  55:354-362.

Benson, D. W.  1960.  Review of Soil Chemistry Research at Hanford.   HW-67201.

Lavrukhina, A. K. and A. A. Pozdnyakov.  1970.  Analytical Chemistry of
Technetium, Promethium, Astatine and Francium.  Translated by R. Kondor.
Ann Arbor - Humphrey Science Publishers.

Magno, P., T. Reavey, and J. Apidianakis.  1970.  Liquid Waste Effluents from
a Nuclear Fuel Reprocessing Plant.  BRH-NERHL-70-2.

Pourbaix, M.  1966.  Atlas of Electrochemical Equilibria in Aqueous  Solutions.
Pergamon Press, Oxford, England.

Schultz, R. K.  1965.  Soil Chemistry of Radionuclides.  Health Physics.
11:1317-1324.      .                                             '

Sillen, L. G. and A. E. Martell.  1964.  Stability Constants of Metal-Ion Com-
plexes.  Special Publication No. 17.  The Chemical Society, London.

Starik, I. E., N. I. Ampelogova, F. L. Ginzburg, M. S.  Lambert, I.  A.  Skulskii,
and V. N. Shchebetkovskii.  1959.  On the Molecular State of Ultra-Small
Quantities of Radioelements in Solutions.  Radiokhimiya.  1:370-378  (in
Russian).

Starik, I. E. and M. S. Lambert.  1958.  State of Microquantities of Promethium
in Aqueous Solutions.  ZhNKh.  3:136.

Wheelright, E. J.  1973.  Promethium Technology.  Am.  Nucl. Soc. Publication.
                                    3-141

-------
RADIUM
Natural Soil and Rock Distributions
     The radium content of various igneous and sedimentary rocks is given in
Table 3-60.  Vinogradov (1959) reported that the radium content of soils
ranged from 0.5 x 10"  to 1.1 x 10"  ppm.  The small amount of radium present
in soils has become separated from parent thorium and uranium during rock
weathering processes, and most closely follows barium.  The radium content of
igneous rocks increases by 100 times from ultra-basic to granitic rocks.  The
radium content of most sedimentary rocks is about the same as that for
granites.
          TABLE 3-60.  AVERAGE RADIUM CONTENT OF VARIOUS ROCK TYPES
             Rock Type
       Ultrabasic Igneous
       Basic Igneous
       Intermediate, Igneous
       Granitic Igneous
       Sandstones
       Shales
       Limestones
  Ra, ppm
            Reference
0.009 x 10
  0.6 x 10
0.917 x 10"
1.395 x 10
 0.71 x 10
 1.08 x 10
 0.42 x 10
-6
-6
-6
,-6
-6
,-6
Davis, 1947
Evans et al., 1942
Senftle and Keevil, 1947
Senftle and Keevil, 1947
Bell et al.,  1940
Bell et al.,  1940
Evans and Goodman, 1941
Brief Chemistry
                                          213      230           218
     There are 16 isotopes of radium from    Ra to    Ra with no    Ra or
229
   Ra.  All of the isotopes of radium are unstable, and all of the naturally-
                                                             228    '24
occurring radium isotopes occur in the thorium decay series (   Ra, "  Ra),
    23fl                       226             235
the    U-radium decay series (   Ra), and the    U-actinium decay series
 223
(   Ra) (Vdovenko and Dubasov, 1973).  Radium radionuclide data are given in
                                           228
Table 3-61.  All are alpha emitters except    Ra which is a beta emitter.
With the present uranium fuel cycle and the much longer half-life, only the
226
   Ra isotope is of long-term concern in waste disposal.  If future energy
                                        228
production includes thorium fuels, then    Ra also must be included in waste
disposal management plans.
                                    3-142

-------
                   TABLE 3-61.   RADIUM RADIONUCLIDE DATA
                                (VDOVENKO AND DUBASOV,  1973)
Isotope
223Ra
224Ra
225Ra
226Ra
228Ra
Half-Life Decay Mode
11.43 days
3.64 days
14.8 days
1622 years
5.77 ± 0.02 years
ct
a
6"
a
B"
     Radium is a homolog of the alkaline-earth elements, with a (II) oxidation
state.  The radius of the +2 radium ion is 1.52 A compared to 1.43 A for Ba+2.
The compounds formed by radium and their solubilities are similar to barium.
For example, the solubility product of radium sulfate is 4.25 x 10    at 25°C
(Vdovenko and Dubasov, 1973) and the solubility product of barium sulfate is
1.08 x TO  ° at 25°C (Weast, 1976).  Metallic radium dissolves in water with
the evolution of hydrogen and the formation of readily soluble Ra(OH)9.  Of
                                    +2
the alkaline-earth metal cations, Ra   shows the least tendency for complex
formation, although 1:1 complexes with citric, tartaric, succinic and several
other acids were detected at pH 7.2 to 7.4 by Schubert et al. (1950).  It may
                  +2
be assumed that Ra   is not hydrolyzed in aqueous solutions, in an analogy
       +2
with Ba  , although there is no specific literature on the subject.
Solid Phase and Solution Equilibria
     The thermodynamic data for radium compounds are available only for radium
nitrate, chloride, iodate and sulfate (Parker et al., 1971).  However, all
of the compounds except sulfate are very soluble.  Therefore, solid phase
diagrams are not presented for radium compounds.  The solubility product for ,
RaS04 is 10~10'37 compared to 10"9'96 for BaS04-
     No thermodynamic data were located for radium hydrolysis or complex ion
species.  It is expected the radium will behave in the soil  solution much like
                               2+
strontium does.  The species Ra   is expected to be the most important over
the normal soil pH range from 4 to 8.
Experimental Adsorption Results
     Stead (1964) gave a radium Kd value of 6700 ml/g for NTS tuff.   Arnold
and Grouse (1965) ran batch adsorption tests on some exchange materials that
included the natural zeolites, clinoptilolite and chabazite, as represented
by a pelletized molecular sieve (AW-500) and barite (barytes), a natural  BaSOd.
                                    3-143

-------
The results of the adsorption tests were recomputed as distribution coefficients
or Kd values.  The solution was a lime-neutralized waste that contained 4100
pCi 226Ra/l, 500 mg/1 Ca+2, 80 mg/1 Mg+2, 1000 mg/1 Na+, 2500 mg/1 SO^2 and
900 mg/1 Cl" at pH 7.7.   The radium Kd values are given in Table 3-62.

       TABLE 3-62.  RADIUM Kd VALUES FROM LIME-NEUTRALIZED WASTE;
                    1.25 g EXCHANGER/1 OF WASTE (ARNOLD AND GROUSE,
                  _J965)
                                     Exchanger Loading,
          Exchanger      Mesh Size   _ pCi/g _  Kd,ml/q
        Clinoptilolite     20-50             2650           646
        Chabazite          20-50             2900           707
        Barite             20-50             2000           490

     R.  J.  Seme of PNL, 1974, used soils from Utah and simulated river water
to determine radium distribution coefficients.   The soils were pre-equi libra ted
by four washings with the simulated river water composition  shown in Table
               ?2fi                                         226
3-63 minus "the    RaCl.  The fifth solution contained the    Ra as well as
the other constituents, and was used for the radium Kd determinations.  The
Kd values were determined in triplicate to allow measurement of precision.
The Utah soils contained 2 to 5% calcite, with 'quartz and feldspar constituting
the bulk of these sandy, arid soils.  Minor constituents included hydromica
and a small amount of a smectite clay.  The radium Kd results are listed in
Table 3-64.  The Kd correlated with the cation exchange capacities of the soils.

                TABLE 3-63.  SIMULATED RIVER WATER COMPOSITION
                             (SERNE, 1974)
Constituent
Ca
Mg
Na
K
. HC°3
so4
Cl
u
Ra
Added as-
CaS04-2H20, CaCl2
Mgso4
NaCl
KC1
NaHC03
CaS04»2H20, MgS04
CaCl2, NaCl, KC1
uo3
RaC19 in HC1 solution
mg/1
82
26
75
3.4
171
246
57
1
7 ug/1
                                     3-144

-------
       TABLE 3-64.  RADIUM DISTRIBUTION COEFFICIENTS WITH THE SOLUTION
                    OF TABLE 3-63 (SERNE, 1974)
Son
i


ii


in


IV


Final pH
7.9
7.9
8.0
7.6
7.7
7.6
7.8
7.9
7.8
7.8
7.6
7.8
Kd,

354


289


467


214

ml/g

± 15


t 7


t 15


± 15

Migration Results
     Field Studies--
     Granger et al. (1961) and Granger (1963) showed that radium had migrated
out of Ambrosia Lake, New Mexico, uranium ores and had been partly reconcentrated
in barite (BaSOJ and cryptomelane (KMn +Mn % 9C01C'H90) found in and near
                             226  2+         /.to 10  (.*,
some of the ore bodies.  The    Ra   occurred in the Ba   position in barite
	..   2 L	                                     '              ? 2 fi
and the Mn   position in cryptomelane.  The high concentrations of.   Ra-were
not associated with parent uranium, which is good evidence for the recent
migration of radium.  The strongly to weakly radioactive cryptomelane partially
replaced mudstone that occurred near the ore bodies and was relatively low
in uranium content and enriched in lead.   This suggests that the lead is
                                          226
radiogenic and has also migrated with the    Ra.  The mechanism involved in
                       ??fi                          226  2+
reconcentration of the    Ra is the substitution of    Ra   for the chemically
               2+                 2+
very similar Ba   in barite and Mn   in cryptomelane.  Analyses of the outer
5 to 10 cm of mudstone layers near ore disclosed anomalously high radioactivity
coupled with an abnormally high lead content.  Within mudstone layers, however,
the radioactivity was essentially in balance with the uranium content,' and
the lead content was low.
     Hansen and Huntington (1969) determined radium and thorium distributions
in a series of morainal soils in Bench Valley, California.  Thorium accumulated
immediately beneath horizons, containing a high amount of organic material.
The thorium apparently migrated as organic complexes.  Radium was distributed
with the uranium in the high organic layers.

                                     3-145

-------
     The ground and surface water sampling and analyses for radium given by
Wruble et al. (1964) and Kaufmann et al. (1975, 1976) for waters of the
Colorado Basin and Grants area, New Mexico, respectively, also illustrate the
tendency of the radium daughter to become separated from uranium due initially
to uranium mining and milling operations, and to continue migrating due to
inherent geochemical differences between uranium and radium.
     Laboratory Studies —
                                                              226
     Several reports have been concerned with the leaching of    Ra from
uranium recovery process tailings and wastes (Whitman and Porter, 1958; Anonymous,
1960; Feldman, 1961), and other authors have shown that radium can be leached
from stream sediments, minerals and uranium mill wastes (Starik and Polevaya,
1958; Starik and Lazerev, 1960).  The factors that influence Teachability of
226
   Ra from uranium mill waste solids and river sediments were investigated by
Shearer (1962) and Shearer and Lee (1964).   Less than 1.5 wt% of the river
sediments and mill waste solids were greater than plus 20 mesh size and less
than 30 wt% were minus 140 mesh in particle size.  The amounts of radium
leached with distilled water versus time showed that essentially no radium
was leached after 15 min.  Diffusion of radium from the interior of the parti-
cles was relatively insignificant.  By varying the liquid to solid ratio in
distilled water-solid Teaching equilibria,  it was shown that the ratio affects
the amount of radium leached.  The largest ratio effect was shown with leach-
ing of the acid leach process tailings, less with alkaline leach process tail-
ings and the least from river sediments.  It was demonstrated that sulfate was
present in the waste solids and that the sulfate was easily solubilized.  Trace
amounts of barium present led to precipitation of BaSO, and the coprecipitation
of RaSO,.  If radium was added prior to the solids-distilled water equilibria,
it too was removed from solution by coprecipitation with BaSO..   One-hundredth
molar solutions were used in leaching equilibria (100 ml/g river sediment)  to
determine effects on radium leaching.  MgClo. KC1 ,  NaCl , HC1 and water solu-
tions all leached less than 1 uug of radium while CaCl2 leached 1.2 uug Ra,
SrCl9 6.3 uug Ra and Bad, 30 uuRa.   Apparently the radium on river sediments
                    +2
is in the form of Ra   and is exchangeable.
     Havlik et al .  (1968a) investigated the leaching of    Ra from uranium mill
solids and uranium ores.  The first report concerned the effects of pH on
                                    3-146

-------
 leaching  rates.   The  authors  found,  Vike  Shearer and Lee  (1964),  that equilib-
 rium  leaching was  rapidly  accomplished  in  15 to 30 min.   Homogenized uranium
 ores  (540 pCi Ra/g and  85  pCi  Ra/g)  and milling tailings  (14  pCi  Ra/g) were
 shaken  for varying lengths of time as a 3g solid/30 ml solution ratio.  The
 pH  was  modified  from  1  to  14  with hydrochloric acid, boric acid and sodium
                                2?fi
 hydroxide.   At pH  1,  22% of the    Ra was  liberated.  At  pH 9, the amount
                                                                226
 leached had decreased to 2.8%.  At pH 13,  the amount of leached    Ra increased
 to  5%.
      The  second  report  by  Havlik et  al. (1968b) studied the leaching of radium
 from  the  same solids  as affected by  leaching solution composition in addition
 to  acidity.  Unlike the leaching results of Shearer and Lee (1964), Havlik
 et  al.  found that  the largest concentrations of radium were leached by IN KC1
 and IN  NaCl  solutions.  Bad7. SrCU and CaCU also were  used, but with much
 less  radium liberated.  The IN KC1 leached 100% of the radium in mill tailings
 and NaCl,  95%.   Uranium ore leaching results were lower,  showing 22% and 31%
 leached by KG! and 14%  and 17% leached  by NaCl.  In all cases, the one normal
 salt  solutions removed more radium than ten normal salt solutions.
 Summary
                                                      '~   • i      • —	
      Radium is present  as  Ra   over  the normal soil pH range  (4 to 8) and
 shows little tendency to form complex species (Schubert et al., 1950).  Radium
 would be  expected  to  substitute for  other divalent cations during replacement
-or  precipitation reactions (Granger  et  al., 1961; Granger, 1963).  A direct
 correlation of cation exchange capacity with adsorption (Arnold and Grouse,
 1965; Serne, 1974)  and  leaching studies with different types of competing
 cations (Starik  and Polevaya,  1958;  Starik and Lazerev, 1960; Shearer, 1962;
 Shearer and Lee, 1964;  Havlik et al.,.1968b) suggests that an important reac-
 tion  mechanism for radium  adsorption is cation exchange.  Radium could be
 expected  to migrate in much the same manner as strontium.

 References

 Anonymous.   1960.   January, 1960.  Summary Report on I.   The Control of Radium
 and Thorium in the Uranium Milling Industry, II.  Radium- 226 Analysis Princi-
 ples, Interference and  Practice, III.   Current Winchester Laboratory Projects.
 WIN-112.
                                     3-147

-------
Arnold, W. D. and D. J. Grouse*  1965.  Radium Removal from Uranium Mill
Effluents with Inorganic Ion Exchangers.  I&EC Process Design and Development.
4:335-337.

Bell, K. G., C. Goodman and W. L. Whitehead.  1940.  Radioactivity of Sedimen-
tary Rocks and Associated Petroleum.  Bull. Am. Assoc. Petroleum Geol. 24:1529.

Butler, J. N.  1964.  Ionic Equilibrium.  Addison-Wesley, Reading, MA.
pp. 180-183.

Davis, G. L.  1947.  Radium Content of Ultramafic Igneous Rocks.  I. Laboratory
Investigation.  Am. J. Sci.  245:677.

Evans, R. D. and C. Goodman.  1941.  Radioactivity of Rocks.  Bull. Geol. Soc.
Am.  52:459.

Evans, R. D., C.  Goodman, N. B. Keevil.  1942.  Radioactivity:  The Earth's
Heat and Geological Age Measurements.  IN:  Handbook of Physical Constants.
Geol. Soc. Am. Special Papers 36, Sec. 18, p. 267.

Feldman, M. H.  1961.  Summary Report, 1959-1961, Winchester Laboratory,
Winchester, Massachusetts and Grand Junction, Colorado.  WIN-125.

Granger, H. C.  1963.  Radium Migration and Its Effect "on the Apparent Age
of Uranium Deposits at Ambrosia Lake, New Mexico.  U.S. Geol. Survey Prof.
Paper 475-B, pp.  B-60-B-63.

Granger, H. C., E. S. Santos, B.  G. Dean and P. B. Moore.  1961.  Sandstone--
Type Uranium Deposits at Ambrosia Lake, New Mexico - an Interim Report.- Econ.
Geol., vol. 56, no. 7, pp.  1179-1209.

Hansen, R. 0. and G. L. Huntington.   1969.  Thorium Movements in Morainal
Soils of the High Sierra, California.  Soil Science.  108:257-265.

Havlik, B., J. Grafova, and B. Nycova.  1968a.  Radium-226 Liberation from
Uranium Ore Processing Mill Waste Solids and Uranium Rocks into Surface
Streams - I. Health Physics.  14:417-422.

Havlik, B., B. Nycova, and J. Grafova.  1968b.  Radium-226 Liberation from
Uranium Ore Processing Mill Waste Solids and Uranium Rocks into Surface
Streams - II.  Health Physics.  14:423-430.

Kaufmann, R. F., G. G. Eadie and C. R. Russell.  1975.  Summary of Ground Water
Quality Impacts of Uranium Mining and Milling in the Grants Mineral Belt,
New Mexico.  U.S.E.P.A., Off. Rad.  Prog.-Las Vegas Facility.  Tech. Note
ORP/LV-75-4.71 p.

Kaufmann, R. F., G. G. Eadie and C. R. Russell.  1976.  Effects of Uranium
Mining and Milling on Ground Water in the Grants Mineral Belt, New Mexico.
Ground Water.  14(5):296-308.
                                    3-148

-------
Parker, V. B., D. D. Wagman, and W. H. Evans.  1971.  Selected Values of
Chemical Thermodynamic Properties.  U.S. Department of Commerce.  NBS Technical
Note 270-6.

Schubert, A. J., E. R. Russell, and L. S. Myers.  1950.  Dissociation Constants
of Radium Inorganic Acid Complexes Measured by Ion Exchange.  J. Biol. Chem.
185:387-398.

Senftle, F. E. and N. B. Keevil.  1947.  Thorium-Uranium Ratios in the Theory
of Genesis of Lead Ores.  Trans. Am. Geophys. Union. 28:732.

Shearer, S. D., Jr.  1962.  The teachability of Radium-226 from Uranium Mill
Solids and River Sediments.  Thesis.  Univ. Wisconsin.

Shearer, S. D., Jr. and G. F. Lee.  1964.  teachability of Radium-226 from
Uranium Mill Solids and River Sediments.  Health Physics.  10:217-227.

Starik, I. E. and K. F. Lazarev.  1960.  Effect of Crushing of Minerals on
the Extraction of Radioactive Elements.  Radiokhimiya.  11:749-752.

Starik, I. E. and N. I. Polevaya.  1953.  The Leachability of ThX and Rd Th
from Minerals.  AEC-tr-4208, pp. 108.

Stead, F. W.  1964.  Distribution in Groundwater of Radionuclides from Under-
ground Nuclear Explosions.  IN:  Proc. Third Plowshare Symp. Engineering with
Nuclear Explosives.  April 21^23, 1964.  TID-7695, p. 127-138.

Vdovenko, V. M. and Yu. V.- Dubasov.  1973... Analytical Chemistry of Radium.
Israel Program for Scientific Translations.

Vinogradoy, A. P.  1959.  The Geochemistry of Rare and Dispersed Chemical
Elements in Soils.  Consultants Bureau.

Weast, R. C.  1976.  Handbook of Chemistry and Physics.  The Chemical
Rubber Co.

Whitman, A. and E. S. Porter.  1958.  Chemical Stream Pollution from Uranium
Mills.  WIN-99.

Wruble, D. T., S. D. Shearer, D. E. Rushing and C. E. Sponagle.  1964.  Radio-
activity in Waters and Sediments of the Colorado River Basin, 1950-1963.  Rad.
Health Data, 5(ll):557-567.
                                    3-149

-------
RUTHENIUM
Natural  Soil  and Rock  Distributions
     Few data on natural  distribution  of ruthenium are available.   Data avail-
able on  ruthenium in rock-forming minerals are listed  in Table 3-65.   Platinum
group metals, including  ruthenium, are known to be concentrated  in  chromite;
but these are older analytical values  and may be much  higher than modern
neutron  activation ruthenium analyses.   There are no ruthenium data on sand-
stones and limestones.   However, the ruthenium contents are probably very low.
The meager data on ruthenium content of shales are given in Table 3-66.  To
put the  ppb values into  perspective, 100 ppb amounts to 1  x 10   g  of ruthenium
per gram of rock or mineral.  The behavior and distribution of natural  ruthe-
nium during weathering processes is largely unknown.
            TABLE 3-65.   ABUNDANCES OF  RUTHENIUM IN  IGNEOUS ROCKS
                          AND ROCK-FORMING MINERALS IN  ppm
                Mineral
            01ivine
            Chromite, Norway
            Chromite, Texas
            Chromite, Penn.
            Columbite, Norway
            Ilmenite
            Rutile
            Tantalite
            Ounite, Russia

            Dunite, Russia

            Apodunite, Russia

            Mica  shonkinite
  Ru. ppb
     10
    500
    500	
    500
     10
      4
     10
      2
    2-30
(5 average)
    3-6
(4.5 average)
   0.9-4
(2.2 average)
      0.5
         Reference
Noddack and Noddack, 1931
Qoldschmidt and Peters, 1932
Goldschmi'dt and Peters, 1932
Goldschmidt and Peters, 1932
Noddack and Moddack, 1931
Noddack and Noddack, 193;
Noddack and Noddack, 1931
Noddack and Noddack, 1931
Razln et al., 1965

 Razin et al., 1965

 Razln et al., 1965

 Razin et al., 1965
                                      3-150

-------
                 TABLE 3-66.  RUTHENIUM CONTENTS OF SHALES

           	Shale	   Ru, ppb         Reference
           Kupferschiefer,           3      Noddack, 1936
           Mansfeld,  Germany
           Black  Shale  (3)          60      Tischendorf,  1959
                                    80
                                    40
           Bleached Shale  (3)       40      Tischendorf,  1959
                                     5
                                    20
Brief Chemistry
     The 16 isotopes of ruthenium include the radioactive fission  products
93Ru, 94Ru, 95Ru, 97Ru, 103Ru, 105Ru, 106Ru,  107Ru and 108Ru.   The natural,
stable ruthenium isotopes and their abundance are shown in Table 3-67.
            TABLE 3-67.   ATOMIC PERCENTAGES OF STABLE RUTHENIUM
                         ISOTOPES (HEATH,  1976)
                96      98
     Atom %    5.51     1.87
     Ruthenium-103 with a 39.6-day half-life and 106Ru with a  367-day half-
life are the only radioisotopes of ruthenium tha't persist long enough to  be
of concern in waste disposal.
     Ruthenium exhibits several oxidation states varying from  Ru(-II) to
Ru(VIII), with Ru(III) and Ru(IV) the most common oxidation states  in aqueous
solutions.  Ru(II) is less common.  For ruthenium, there is little  evidence
for simple aquo ions.  Nearly all aqueous solutions,  whatever  the anion,  may
be considered to contain complex ions (Cotton and Wilkinson, 1962).   Ru(II),
for example, is known to form a unique series of nitrosyl  complexes,  nitric
oxide complexes, ammonia and sulfur complexes, carbonyl  complexes,  etc.
     In an alkaline medium, Ru(IV) forms the insoluble hydrated oxide Ru02 •
XH20 which does not tend to be reduced to the trivalent state.   Ruthenium (III)
hydroxide is readily oxidized in air to Ru(IV) (Ginzburg et al., 1975).
Ruthenium exists in nitric acid solutions either as the Ru(III)  nitrosonitrates

                                    3-151

-------
[RuNO(N03)n(H20)5_3]3"n or [RuNO(N03)5_nHn(OH)m •  (HgO^P', or as the poly-
meric Ru(IV) aquohydroxo cations [Ru(OH)x(H20)g-x] (N03)4_x (Ginzburg et al.,
1975).  Thus, from acidic nitrate solutions, which are the usual form of fuel
reprocessing wastes, ruthenium solutions may be obtained that contain cationic,
anionic and neutral complexes.  Hence, the chemical  forms of the ruthenium in
a given media are a product of the chemical history of that media.
Solid Phase and Solution Equilibria
     Ruthenium is generally present in association with platinum group metals.
The abundance of ruthenium in various rocks and soil  forming minerals is sum-
marized in Tables 3-60 and 3-61.  Ruthenium also forms discrete solid compounds
such as Ru02, RuO^, Ru(OH)3, Ru(OH)., RuCl3, and RuS2>  The relative stability
of the discrete solids in oxidizing conditions (p02/q\ = 0.68 atm) is given
in Figure 3-20.  The thermodynamic data for Ru02 and RuO, were obtained from
Wagman et al. (1969).  The data for Ru(OH)3 and Ru(OH)4 were selected from
Sillen and Martell (1964).  The solids in increasing order of stability for
the conditions outlined in Figure 3-20 are:  Ru(OH)3> RuO^, Ru(OH)4, and Ru02
(amorphous, hydrated).  RuCU and RuS2 do not appear in Figure 3-20 because
they are too soluble and fall beyond the boundaries  of the figure.  From
Figure 3-20 it can be ascertained that Ru02 (amorphous, hydrated) would be
most stable over the pH range and up to reducing environments equivalent to
p02 of 50.  In extremely reducing conditions (p02  =  80) where HLS (aq) would
be the most stable sulfur species, Ru$2 would maintain lower activities than
Ru02 and hence would be the most stable.
     The activity of various solution complexes of ruthenium in equilibrium
with Ru02 (amorphous, hydrated) in oxidizing (p02  0.68) and other conditions
spelled out in the graph are given in Figure 3-21.  The thermodynamic data
used to construct Figure 3-21 were selected from several  sources including
Sillen and Martell (1964) for RuOH3*, Ru4* and Ru3"1",  Baes and Mesmer (1976)
for Ru04(OH)~, and Wagman et al. (1969) for the remaining species.  Baes and
Mesmer (1976) indicate that the thermodynamic data on ruthenium complexes'are
either not available or is not very accurate.  However, it was decided to
construct Figure 3-21 using the existing data to estimate the nature of the
species that may be present in solution.
                                   3-152

-------
               -8
               -10
               -12
               -14
               -16
               -18
               -20
  - Ru0;
       AND

   p02 • SO. 8
                                                        uO,
  3
                                                               10
                                       pH
Figure 3-20.
The relative  stability of various ruthenium solids in an oxidiz-
ing soil environment [p02(g) = 0.68 atm],  pCl" = pS042~ = 2.5
and pH$ =3.0.
               -8
              ,-12
              -16
              -20
              -24
                                                        Ru04°
                                    6      7
                                        pH
                                                 10
Figure 3-21.   The activity of various  ruthenium species in equilibrium with
                    (amorphous hydrate)  in  an oxidizing soil environment
                                    3-153

-------
     Figure 3-21 indicates that in an oxidizing environment, significant
                          4+
activity of uncomplexed Ru   will be present only in very acidic conditions
(pH <2).  Some of the ruthenium species which may be present in solution are
Ru04~, Ru042~, Ru04°, Ru(OH)22+, Ru04(OH)-, RuCl5(CH)2~, Ru4+, and Ru3+.
Among these species, Ru   would be predominant in pH <2, Ru(OH)?   in pH 2 to
                                      o                        Cm
5.2, Ru04  in pH 5.2 to 10.5, and Ru04   in pH >10.5.  With an increase in
reducing conditions, the activity of Ru   would increase, the activity of
    o-2-
RuO. , RuO. , RuO*   and Ru(LOH  would decrease and the rest of the species
would remain unchanged in activity.   The nature of the predominant solution
species is highly dependent upon the oxidation-reduction conditions, as is
the most stable solid compound.  The predominant ruthenium species are cationic
below pH 5 and anionic above pH 5.  Hence, ion exchange on soils and rocks,
mainly cation exchange, would be of importance only below pH 5 under these
conditions.
Experimental Adsorption Results
     Rhodes (1957) reported the variation in    Ru distribution coefficients
with solution pH on a typical Hanford subsoil.  The soil had a cation exchange
capacity of 5 meq/100 g, 2 wt% calcium carbonate, with about 2% clays of
mainly montmorillonite and lesser amounts of chlorite.  The Kd values for
106Ru are presented in Table 3-68.  It should be noted that the 106Ru was
obtained from Oak Ridge National Laboratory, probably as RuCU in HC1 solution.
    TABLE 3-68.  RUTHENIUM DISTRIBUTION COEFFICIENTS ON A HANFORD SOIL AS
                 A FUNCTION OF pH (RHODES, 1957)

                 pH       Kd,  ml/g         pH         Kd,  ml/g
                                                        40
                                                        39
                                                        44
                                                        92
                                                        27
                                                       101
                                                       752
                                    3-154
0
1.3
2.6
4.7
6.6'
7.4
8.5
0
0
26
101
752
460
274
9.2
10.1
10.4
11.0
11.3
13.0
14.0

-------
     The soil adsorption of Ru   in a chloride medium cannot be similar to the
adsorption of the nitro- and nitrato complexes of nitrosylruthenium normally
present in acidic nitrate waste solutions.  Rhodes and Nelson (1957) make this
same point when they discuss the soil adsorption of ruthenium present in
uranium recovery plant waste.  Sampling of groundwater beneath a discontinued
disposal facility showed that the concentration of    Ru remained constant for
several months.  This indicated little adsorption after the initial break-
through of    Ru.  Klechkovsky (1956) also investigated    Ru adsorption on
soils but did not specify the type of ruthenium or soils used.
     Wilding and Rhodes (1963) reported that the addition of 100 ppm citrate
to a synthetic waste-cationic ruthenium solution reduced the Kd value for
ruthenium on Idaho well sediments to 27 ml/g from 472 ml/g with no citrate.
For the same solutions, the reduction on the natural zeolite clinoptilolite
was to a Kd value of 4 ml/g from a Kd value of 48 ml/g with no citrate present.
Addition of 100 ppm EDTA to the synthetic waste (MTR cooling water) lowered
the ruthenium Kd value from about 472 ml/g with no EDTA to 100 ml/g with EDTA.
                                                  • o       4.3
Kepak (1966) reported the adsorption Kd for Ru(NO)   and Ru   on hydrated
                        3             3
ferric oxide as 8.4 x 10  and 4.8 x 10 , respectively.
     Jenne and Wahlberg (1968) examined the sediments present in White Oak
Creek, Tennessee, identifying their mineralogy and determining a Kd for several
elements.  The Kd was determined by analyzing the radionuclide content of
White Oak Creek water and sediment, and assuming that an equilibrium between
the two was attained.  The water contained 230 pCi    Ru/1, while no measur-
able    Ru was found in the sediment.  It was concluded that the    Ru in
White Oak Creek water was not being adsorbed by the contacting sediment.
     The form of ruthenium in the Clinch River, downstream of White Oak Creek,
was suggested by Pickering et al. (1966) to be nitrosyl ruthenium hydroxide.
They reported 16 Ci of    Ru in Clinch River bed sediment as of 1962.
Lomenick and Gardiner (1965) reported 1,038 ± 88 Ci  of    Ru concentrated in
a small area of White Oak Lake coinciding with two inlet streams that drain
the waste seepage pit area.  White Oak Lake is above White Oak Creek in the
drainage system around Oak Ridge, Tennessee.
                                   3-155

-------
     Izrael and Rovinskii (1970) studied the groundwater leaching of radio-
nuclides from Russian bomb blast rubble and debris.  They used dialysis,
electrodialysis extraction and ion exchange to examine the chemical state of
several radionuclides.  Ruthenium was found to be 19.4% colloidal, 5.1%
cationic and 75.5% anionic.  The complex makeup of    Ru showed nearly the
same distribution in groundwater solutions of different origin, and had an
insignificant adsorptive capacity.
     Arnold and Grouse (1971) described problems associated with radioactive
contamination of copper recovered from ore fractured with nuclear explosives.
The fractured ore was leached with dilute sulfuric acid and the copper was
recovered by replacement of scrap iron as copper cement.  The    Ru consti-
tuted 42% of the total activity in the leach solution and only the    Ru was
removed along with the copper during the iron replacement reaction.
     Amy (1971) studied the adsorption and desorption rates of three ruthenium
compounds on acidic and calcareous soils (see Brown, 1976, for a summary of
Amy's work).  The three compounds were cationic Ru NON02 (N03)2 (H20)2,
anionic Ru NO(N02)2 OH(H20)2 and RuCl3 which produced a preponderance of
anionic forms.
     R. J. Serne of PNL, 1973, determined    Ru equilibrium distribution
coefficients with the simulated tank waste compositions given in Table 3-9,
and seven well sediments described in Table 3-10 under antimony.  The    Ru
was added to the simulated tank solutions as RuCU in HC1 solution and Kd
                                                         1 nc
values obtained as shown in Table 3-69.  The form of the    Ru after addition
to the.simulated tank solutions is unknown.  The traced solution was allowed
to remain at 70°C for 48 hours before contact with the sediments.
  TABLE 3-69.  EQUILIBRIUM DISTRIBUTION COEFFICIENTS BETWEEN SEVEN GLACIO-
               FLUVIATILE SEDIMENTS AND THE SOLUTIONS IN TABLE 3-9 CONTAINING
               106Ru.  EACH VALUE IS AN AVERAGE OF THREE EQUILIBRATIONS.
                                      Solution
merits
1
2
3
4
5
6
7
	 I_
2.14
2.01
2.18
2.60
1.04
1.03
1.19
II
0.13
0.09
0.16
1.79
0.30
0.24
0.41
III
0.27
0.23
0.18
0.28
0.53
0.40
0
IV
0.60
0.53
0.30
0.03
0.58
0.54
0.15
	 v_
0.34
0.24
0.29
0.53
0.68
0.44
0.68
                                    •3-156

-------
Migration Results
     Field Studies—
     Spitsyn et al. (1958) used an alkaline solution (4 to 8 g NaOH/Jl,
200 g NaN03A) and an acidic solution [6 to 8 g HNOg/A, 200 g Al(N03)3/&]
'in field studies to determine radionuclide migration rates through soils.
Practically no ruthenium was adsorbed from acidic solutions, and no anionic
ruthenium was adsorbed by soils.  Various radionuclides including ruthenium
were injected into the ground in a mixture and migration followed by wells
along the migration route.  Ruthenium traveled about half as far as strontium.
     Haney, Brown and Reisenauer (1962) and Brown and Haney (1964) estimated
from the direction and rate of the groundwater flow that 7 to 8 years were
required for    Ru contamination in the groundwater to move from Hanford Purex
Plant site to the Columbia River, a distance of nearly 17 miles.  The half-
life and travel time of    Ru result in less than 1% of the    Ru reaching
the Columbia River.
     Lomenick (1963) discussed ruthenium flow into the drained White Oak Lake
bed at Oak Ridge from the intermediate waste pits by surface runoff and ground-
water movement.  Movement of groundwater in the upper 2 ft of lake bed soil
was from T to 5 ft/day, while at 2 to 5 ft depth, movement was 0.05 to 0.25 ft/
day.  The lake bed then (1963) contained about 1200 Ci of ruthenium, with
about 70% of the ruthenium in the top 2 ft, or rapid water movement zone,
presumably originating at the intermediate level waste pits.
     Raymond (1964, 1965) investigated the vertical migration and lateral
spread of radionuclides beneath three Hanford waste disposal sites by using
scintillation logging of nearby monitoring wells over a period of time.
                                      g
     A disposal site received 7.6 x 10  liters of condensate waste containing
about 4500 gross beta curies, mostly    Ru, to 1965.  One week after disposal
started, contamination was present to 105 ft.  Ruthenium had migrated to
130 ft by June 1958, and the entire soil column was contaminated with ruthe-
nium by June 1959.  Ruthenium traveled at nearly the same velocity as the
soil water.
                                    3-157

-------
     Brown (1967) gave some profiles of    Ru at disposal sites.  The    Ru
moved at essentially the same velocity as the groundwater, with little or no
soil adsorption.  Radiochemical analyses of the groundwater yielded a ruthe-
nium pattern of contamination comparable to the size and shape of the tritium
and technetium contamination patterns.  Limits defined by the 1.0 pCi    Ru/
  3
cm  isoconcentration contour in 1966 are given.  The disposal site ruthenium
concentration was 1000 pCi    Ru/cm  of waste.
     Brendakov et al.  (1969) added    Ru to a mountain meadow soil and followed
the vertical migration due to precipitation.  The average displacement distance
from the surface was 3 mm a year after amendment, 7 mm 2 years after amendment
and 14 mm 3 years after amendment.  The average displacement was defined as:
    ?* * *
X* = ^P1X1, where pi = fractional % of the cadionuclide in a layer (3 mm thick-
ness) and xi = the depth of the layer in mm.  The displacement distance of
   Ru was about the same as the    Cs and    Sb also used in the study.  These
results are not typical  for    Ru migration that ordinarily has been reported
to much exceed the migration velocity of    Cs.
     In 1973, the 241-T-106 underground liquid high level waste storage tank
at Hanford leaked about 435,000 liters of waste into the surrounding sediments
(Anon., 1973).  A series of wells were drilled to ascertain the locations and
movements of the various radionuclides in the sediment column.  Monitoring and
core sample analysis showed the relative mobilities of the radionuclides in
the high pH, high salt solution.  Ruthenium was the most mobile of the radio-
nuclides,  having migrated to 27 m (35 m above the water table).  The one yCi
   Ru/1 soil isopleth was used to delineate the contaminated zone of about
25,000 m  of sediments.   One of the reactions of the high pH waste with the
sediment column involved the chemical attack and partial  dissolution of soil
minerals (Shade, 1974).
     Magno et al. (1970), investigating the radionuclides from the Nuclear
Fuel Services plant in Western New York that were migrating through the
effluent lagooning system, estimated that 70% of the    Ru passed through the
lagoon system and into nearby surface streams.
     Brookins (1976) reported that ruthenium was retained in the shale that
surrounded the natural reactor site at Oklo in Gabon.   Approximately 1.8 bil-
lion years had elapsed since the reactor was active.
                                   3-158

-------
     Laboratory Studies—
     Nishita et al.  (1956) leached ruthenium from soils and clay minerals with
water and a salt solution.  The    Ru was the most readily leached radionuclide
from the soils with water.  The    Ru also was the most easily water-leached
from bentonite and kaolinite.  The ruthenium tended to be the least exchange-
able in both the soils and the bentonite and kaolinite.  A persistent fraction
of the bentonite and kaolinite adsorbed ruthenium was nonexchangeable.
     Essington et al. (1965) used fallout material on soils to determine the
leaching effects of water, chelating agents and HC1 on    Ru.  Chelating
agents (DTPA, CDTA, EDDHA), increased the soluble radionuclides by a small
amo
103
amount as compared to water.  The    Ru tended to remain in the column.  Some
   Ru was reported in leachates in CaDPTA-leached soils.
     According to Schulz (1965), ruthenium tends to react like manganese,
cobalt, zinc, iron and chromium in reactions in a soil environment.  They are
most soluble in acidic conditions and tend to precipitate as hydroxides as
the pH rises.  Ruthenium tends to form complexes that are uncharged or nega-
tively, charged and would therefore tend to be quite mobile in soils.
     Dlouhy (1967) gave ruthenium elution curves from Casaccia soil that were
given as the volume of water put through the column versus the fraction eluted
at flow velocities of 136 cm/hr and 270 cm/hr.  The characteristics of Casaccia
soil were not given.  The velocity of the ruthenium down the column was
obtained in relation to water velocity and a theoretical profile on the soil
column derived for field application and estimates.  The Kd values for the
                                   3                              3
ruthenium ranged from 1.8 to 2.4 cm /g on the soil and 10 to 15 cm /g on the
tuff.  The velocity of the ruthenium divided by the velocity of the water was
experimentally determined as 0.11 at the 136 cm/hr water velocity and 0.10 at
the 270 cm/hr water velocity.
     Eichholz et al. (1967) studied the distribution of ruthenium between
suspended sediment particles and aqueous solutions.  Natural water samples
were characterized and used in the adsorption studies as sources of suspended
solids, as shown in Table 3-70.  A fission product mixture was added to the
natural water samples, the system equilibrated and the water and sediments
separated.  The adsorption-of    Ru is shown in Table 3-71.  A portion of the

                                   3-159

-------
  6Ru was  associated with  the suspended  sediments, though  to a lesser  extent

than the other radionuclides with the exception of     I.
      Source
  Colorado River,
  Utah
Camp McCoy,
Wisconsin
Bayou Anacoco,
Lousiana
  Lodgepole Creek
  Nebraska

  Chattahoochee
  River, Georgia


  Billy's Lake
  Georgia (swamp).
                TABLE 3-70.
                 Suspended
                Solids, ppm

                   229
                      12
                      24
                            PROPERTIES  OF  NATURAL WATER  SAMPLES
                            (EICHHOLZ ET AL., 1967)
 Dissolved         Conductivity
Solids, ppm    pH       pmhos
   350
    60
    63
                                 68
7.5
6.9
6.2
             4.2
540
 80
                                                     60
965
131
200
31
6.8
7.3
300
45
           45
    Solids
   Composition

95% quartz,
5% calcite,
feldspar, illite,
kaolinite

30% quartz +
feldspar,
6% kaolinite, 24%
illite

30% quartz,
2% kaoUnite,
6% smectite

20% illite,
80% smectite

33% quartz,
44% kaolinite
23% illite

2% quartz,
balance was
amorphous
and/or organic
    TABLE  3-71.   ADSORPTION OF 106Ru ON  SUSPENDED SOLIDS  IN NATURAL WATERS
                  (EICHHOLZ ET AL., 1967)
                 Source
           Colorado River

           Camp McCoy

           Bayou Anacoco

           Chattahoochee River

           Billy's Lake
                                    -EL
                                    7.5

                                    6.9

                                    6.2

                                    7.3

                                    4.2
                                                 106
                     Ru Adsorption, %

                           25.0

                            5.7

                           -0

                           10.4

                            0
      It  is  obvious that  ruthenium adsorption was a function of pH of  the  water.

The higher  percentages of adsorbed ruthenium are directly correlated  with

water pH.
                                      3-160

-------
     Nishita and Essington (1967) continued work on    Ru migration in dif-
ferent types of soils, comparing the relative movement of   Zn,   Sr,   Y,
   Ru,    Cs,    Ce and    Pm.  Irrespective of the soil type, water leaching
resulted in the greatest movement for    Ru.
     Bovard et al. (1968) reported that the free functional groups (carboxylic,
phenolic) were the cause of organic matter complexing and fixation of radio-
nuclides.  For humic acids, the mean order of complexing was Ce>Fe>Mn>Co>Ru>^
Sr>Cs.  In the case of fulvic acids, no universal order existed.  Two types of
soils were used in the studies, a calcareous alluvial soil and a humoferrugi-
nous podzol.  Leaching of fission products, particularly ruthenium, by rain-
water occurred essentially in the form of fulvates.
     Amy (1971) studied the migration rates of different ruthenium forms in
calcareous and acidic soil in columns 30 cm in height by 4.5 cm in diameter.
The nitrbdinatrato complexes of ruthenium, RuNONO,(NO^)., (H.,0),, were much
                                                                      137
more mobile than cesium in calcareous soils.  About 98 to 100% of the    Cs
remained in the top 4.5 cm of the soil columns, while 38 to 80% of the nitro-
dinatrato complexes were found in the top 4.5 .cm with distribution down the
column to 1 to 2% of the ruthenium in the last column section.  From 2 to 33%
of this-ruthenium form passes through the column.  For comparison, ruthenium
as RuCl3 adsorbs only 10 to 50% in the top 4.5 cm of the soil  column,  and the.
dinitro complex, RuNOtNOgJgOhKHgOjg, only 11 to 36%.  The fraction from RuCl3
that migrates down the entire column length varied from 3 to 68%, depending
on the soil type.  In summary, in calcareous soils maximum ruthenium mobility
was attained when the ruthenium was in the form of an anionic complex in per-
meable soils with a sandy texture and low in organic material.  Three forms
were identified in the column work referred to as A, B, and C.  Form A was
cationic with a very rapid adsorption rate.  Form B was adsorbed very slowly by
the soil and was essentially adsorbed to the same degree throughout the column.
It may be an anionic form that is evolving toward a. cationic form.  Form C
was present only at the bottom of the soil column if flow rates are very slow.
This also was predominately an anionic form also changing toward a more easily
adsorbed form.  The RuNONO^NO-j^HgO^ contained Forms A and B in about equal
proportions.  The RuCl3 contained 15% of Form A, 75% of Form B and 15% of
Form C.  The RuCNOg^OHMH?^ contained only 5* °f P0'™ A and 85% and

                                   3-161

-------
Forms B and C, respectively.   Less than 2.5% of any of the above forms  passed
through the Barthelasse soil  which was rich in organic matter.   The acidic
soils always adsorbed less ruthenium in any form than the calcareous soils.
However, like the calcareous  soils, the adsorption of ruthenium decreased with
the increasing proportion of  anionic forms.  The calcareous soils,  in order  to
adsorb 50% of the ruthenium,  required from 1 to 4 min for the nitrodinatrato
complexes, 100 to 300 min for RuCl3 and 200 min to 1 day for the dinitro com-
plexes.  The acidic soils, on the other hand, required 7 to 15  min  to adsorb
50% of the ruthenium from the nitrodinatrato complexes, 500 min to  1 day from
RuCl3 and 1 to 3 days from dinitro complexes.  There were indications that  the
organic matter in acidic soils favored the mobility of all forms of ruthenium.
     Champlin and Eichholz (1976) have shown that finely particulate soil
materials, such as kaolinite, can adsorb a variety of trace radionuclides
including    Ru and adhere to the soil matrix surface by Van der Waal's forces
(Champlin, 1969).  Further, these fine particulates were mobilized  by laundry
detergents and softening agents, so that they were transported  quickly  through
the. soil column without reaction with contacting soils.
Summary
                     3+
     Adsorption of Ru   peaks at pH 6.6 and 14, with a definite correlation
between pH and ruthenium Kd for all ruthenium species (Rhodes,  1957; Eichholz
et al., 1967; Amy, 1971).  Adsorption of all ruthenium species  is low at low
pH, and little if any adsorption of anionic species occurs at any pH (Spitsyn
et al., 1958; Haney et al., 1962; Brown and Haney, 1964; Raymond, 1964, 1965;
Brown, 1967; Magno et al., 1970; Schulz, 1965).  Some instances of  migration
on suspended solids have been noted (Champlin and Eichholz, 1976; Eichholz
et al., 1967).  Hydrated ferric oxide is a good adsorbent for ruthenium (Kepak,
1966).  Low ruthenium Kd values have been reported in high salt solutions
(Anonymous, 1973; Serne, 1973), probably as a result of nitrate complexintj:
Migration of most ruthenium complexes through the soil column is relatively.
rapid  (Pickering et al., 1966; Izrael and Rovinskii, 1970; Amy, 1971).   Most
of the thermodynamic data for the ruthenium complexes do not exist.  Thus an
estimation of their relative  importance in waste disposal  operations or spills
is difficult, requiring experimental studies for evaluation in^each case.
                                   3-162

-------
References

Amy, J. P.  1971.  Contribution to the Study of the Migration of Ruthenium-106
in Soils.  RFP-Trans-140.

Anonymous.  1973.  241-T-106 Tank Leak Investigation.  ARH-2874.

Arnold, W. D. and D. J. Grouse.  1971.  Radioactive Contamination of Copper
Recovered from Ore Fractured with Nuclear Explosions.

Avtokratova, T. D.  1963.  Analytical Chemistry of Ruthenium.  Israel Program
for Scientific Translations.

Baes, C. F., Jr. and R. F. Mesmer.  1976.  The Hydrolysis of Cations.
John Wiley and Sons, New York.

Bovard, P., A. Grauby and A. Saas.  1968.  Chelating Effect of Organic Matter
and Its Influence on the Migration of Fission Products.  IN:  Isotopes and
Radiation in Soil Organic Matter Studies.  CONF-680275, pp. 471-495.

Brendakov, V. F., A. V. Dibtseva, V. I. Svishcheva, and V.  N. Churkin.  1969.
Vertical Distribution and Evaluation of the Mobility of Fission Products in
Certain Soil  Types of the Soviet Union.  AEC-tr-7030, pp. 152-170.

Brookins, D.  G.  1976.  Shale as a Repository for Radioactive Waste:  The
Evidence from Oklo. Environmental Geology.  1:255-259.

Brown, D. J.   1967.  Migration Characteristics of Radionuclides Through Sedi-
ments Underlying the Hanford Reservation.  ISO-SA-32.

Brown, D. J.  and W. A. Haney.  1964.  Chemical Effluents Technology Waste
Disposal Investigations July-December, 1973 - The Movement  of Contaminated
Ground Water from the 200 Areas to the Columbia River.   HW-80909.

Brown, K. W.   1976.  Ruthenium:  Its Behavior in Plant  and  Soil  Systems.
EPA-600/3-76-019.

Champlin, J.  B. F.  1969.  Transport of Radioisotopes by Fine Particulate
Matter in Aquifers.  PB-232179.

Champlin, J.  B. F. and G. G. Eichholz.  1976.  Fixation and Remobilization of
Trace Contaminants in Simulated Subsurface Aquifers.   Health Physics.
30:215^219.

Cotton, F. A. and G. Wilkinson.  1962.  Advanced Inorganic  Chemistry.
Interscience Publishers.

Dlouhy, Z.  1967.  Movement of Radionuclides in the Aerated Zone.   IN:
Disposal of Radioactive Wastes into the Ground.  IAEA-SM-93/18, pp. 241-249.

Eichholz, G.  G., T. F. Craft, and A. N. Galli.  1967.  Trace Element Fractiona-
tion by Suspended Matter in Water.  Geochim. et Cosmochim.  Acta. 31:737-745.
                                    3-163

-------
Essington, E. H., H. Nishita and A.  J.  Steen.   1965.   Release and Movement of
Radionuclides in Soils Contaminated  with Fallout Material  from an Underground
Thermonuclear Detonation.  Health Physics.   11:689-698.

Ginzburg, S. I., N. A. Ezerskaya, I. V. Prokof'eva, N. V.  Fedorenko,
V. I. Shlenskaya, and N. K. Bel'skii.  1975.   Analytical  Chemistry of Platinum
Metals.  Translated by N. Kaner. Israel Program for Scientific Translations.

Goldschmidt, V. M. and C. Peters.  1932.  Zur  Geochemie der Edelmetalle.
Nachr. Adad. Wiss. Gottingen, Math-Physik.  K1.IV:377.

Haney, W. A., D. J. Brown, and A. E. Reisenauer.  1962.   Fission Product
Tritium in Separations Wastes and in the Ground-Water.  HW-74536.

Heath, R. L.  1976.  Table of Isotopes.  IN:   Handbook of Chemistry and
Physics.  R. C. Weast (ed.), Chemical Rubber Co.

Izrael, Yu. A. and F. Ya. Rovinskii.  1970.  Hydrological  Uses of Isotopes
Produced in Underground Nuclear Explosions  for Peaceful  Purposes.  UCRL-
Trans-10458.

Jenne, E. A. and J. S. Wahlberg.  1968.  Role  of Certain Stream-Sediment
Components in Radioion Sorption.  U.S.G.S.  Professional  Paper 433-F.

Kepak, F.  1966.  Sorption of the Radioisotopes 35S,  131I  and 106Ru on
Hydrated Oxides in Laboratory Columns.   Collection Czechoslov. Chem.  Commun.
31:3500-3511.

Klechkovsky, V. M.  1956.  On the Behavior  of  Radioactive Fission Products in
Soil.  Their Absorption by Plants and Their Accumulation in Crops. AEC-TR-2867.

Lomenick, T. F.  1963.  Movement of  Ruthenium  in the  Bed of White Oak Lake.
Health Physics.  9:835-845.

Lomenick, T. F. and D. A. Gardiner.   1965.  The Occurrence and Retention  of
Radionuclides in the Sediments of White Oak Lake.  Health. Physics. 11:567-577.

Magno, P., T. Reavey, and-J. Apidianakis.  1970.  Liquid Waste Effluents  from
a Nuclear Fuel Reprocessing Plant.  BRH-NERHL-70-2.

Nishita, H. and E. H. Essington.  1967.  Effect of Chelating Agents on the
Movement of Fission Products in Soils.   Soil Science.   102:168-176.

Nishita, H., B. W. Kowalewsky, A. J. Steen, and K. H.  Larson.  1956.   Fixation
and Extractability of Fission Products Contaminating  Various Soils and Clays.
Soil Science.  81:317-326.

Noddack, I.  1936.  Uber die Allgegenwart der  Chemischen Elementer. Angew.
Chem.  49:835.

Noddack, I. and W. Noddack.  1931.  Die Geochemie des  Rheniums.  Z. Physik.
Chem.  54:207.
                                    3-164

-------
Pickering, R. J., P. H. Carrigan, T.  Tamura, H.  H.  Abee, J.  W.  Beverage, and
R. W. Andrew.  1966.  Radioactivity in Bottom Sediments of the  Clinch -
Tennessee Rivers.  IAEA-SM-72/4.

Raymond, J. R.  1964.  Investigation  of the Disposition and  Migration of Gross
Gamma Emitters Beneath Liquid Waste Disposal Sites.   HW-81746,  pp.  4.32-4.36.

Raymond, J. R.  1965.  Cesium and Strontium Distribution Beneath Liquid Waste
Disposal Sites.  BNWL-235.

Razin, L., V. Khvostov, and V. Novikov.  1965.  Platinum Metals in  the Essen-
tial and Accessory Minerals in Ultramatic Rocks.   Geochemistry  International.
2:118.

Rhodes, D. W.  1957.  The Effect of pH on the Uptake of Radioactive Isotopes
from Solution by a Soil.  Soil Science Society of America, Proceedings.
21:389-392.

Rhodes, D. W. and J. L. Nelson.  1957.  Disposal  of Radioactive Liquid Wastes
from the Uranium Recovery Plant.  HW-54721.

Schulz, R. K.  1965.  Soil  Chemistry  of Radionuclides.   Health  Physics.
11:1317-1324.

Shade, J. W.  1974.  Reaction of Hanford Sediments with Synthetic Waste.
ARH-CD-176.

Si 11 en, L. G. and A. E. Martell.  1964.  Stability Constants of Metal-Ion
Complexes.  Special Publication No. 17.  The Chemical Society,  London.

Spitsyn, V. L, V. D. Balukova, A. F. Naumova, V. V. Gromov, F. M.  Spiridonov,
E. M. Vetrov, and G. I. Gravfov.  1958.  A Study of the Migration of Radio-
elements in Soils.  IN:  Proc. Second Annual U.N. Conf. Peaceful Uses of
Atomic Energy.  18:43T-448.

Tischendorf, G.  1959.  Zur Genesis einiger Selenidvor Kcmmen Insbensondere
von Tilkerode im Harz. Freiberger Forschungsh.  C69:l.

Wagman, D. D,, W. H. Evans, V. B. Parker, I. Halow,  S.  M. Bailey and
R. H. Schumm.  1969.  Selected Values of Chemical Thermodynamic Properties.
fables for Elements 35 Through 53 in  the Standard Order of Arrangement.
Nat. Bur. Stds. Technical Note 270-4.

Wilding, M. W. and D. W. Rhodes.  1963.  Removal  of Radioisotopes from Solu-
tion by Earth Materials from Eastern  Idaho.  IDO-14624.
                                    3-165

-------
STRONTIUM
Natural Soil and Rock Distribution
     Average strontium contents of igneous and sedimentary rocks are given in
Tables 3-72 and 3-73.  Note that the strontium contents are often high when
compared to the trace concentrations of strontium that may come in contact
with them in waste solutions.  Vinogradov (1959) reports the average strontium
                                                     90
content of soils as 300 ppm.  The 300 ppm present as   Sr in a waste solution
                 90
would be 42.3 Ci   Sr/1, for example.  Limestone and other calcium-rich sedi-
ments and sedimentary rocks tend to be the highest in strontium content due
to limited substitution of strontium for calcium in calcite.  Strontium car-
bonate crystallizes in an aragonite type structure when present in sufficiently
high concentrations.

           TABLE 3-72.  AVERAGE STRONTIUM CONTENT OF IGNEOUS ROCKS
                        (TUREKIAN AND WEDEPOHL, 1961) IN ppm
                                          Granitic
         Ultramafic     Basaltic     High Ca     Low Ca     Syenite
             1.0           465         440         100-       200
  TABLE 3-73.  AVERAGE STRONTIUM CONTENT IN SEDIMENTARY ROCKS-AND SEDIMENTS
               (TUREKIAN AND WEDEPOHL, 1961) IN ppm

                                               	Pelagic
                   Sandstone     Limestone     Carbonate     Clay
                       20           610        .  2000        180
Brief Chemistry
                                                                84
     The four naturally-occurring isotopes of strontium include   Sr (0.55%),
QC           Q7               00                          Q7
bbSr(9.75%), b/Sr (6.96%) and abSr (82.74%).  The isotope b/Sr can be locally
                                                  87
variable in abundance because it is a daughter of   Rb.  The other isotopes
of strontium between   Sr and   Sr are radioisotopes.  Only   Sr fission
                                            90
product (half-life 28.1 years), decaying to   Y, is of sufficiently high yield
and long half-life to be of concern in waste disposal operations.   The
                                    3-166

-------
             on
radionuclide   Sr also is obtained in high yield but the half-life is too
                                                                      85
short (52 days) to create a persistent disposal  problem.  The isotope   Sr
(half-life 64 days) is not a fission product, and is usually not encountered
except in laboratory waste solutions.
     Strontium (II) is the only oxidation state  to be encountered in soil-
                                                                       +?
contacted solutions.  According to Ahrens (1952) the ionic radius of Sr w is
1.12 A, very close to that of Ca+2 at 0.99 A and Ba"1"2 at 1.34 A.  Strontium
tends to follow calcium in weathering and other  geochemical  processes with
some exceptions.  For example, when present in sufficient concentrations,
strontium does form a carbonate that is not isostructural with CaC03.  The
other main strontium ore mineral, celestite (SrSO,), is isostructural with
its calcium counterpart, anhydrite (CaS04).  However, limestone or calcite
(CaCQ3) can allow the substitution of several hundred parts  per million
strontium before there is any tendency for strontianite (SrC03) to form.   As
the limestone recrystallizes, there is a tendency for the strontium content
to decrease.
     There is little, if any, tendency for strontium to form complexes with
inorganic ligands.  Izrael and Rovinskii (1970), for example, studied by
electrodialysis the chemical state of strontium  leached by groundwater from
rubble produced in a nuclear explosion.  They found that 100% of the strontium
      +2
was Sr   , with no colloidal or anionic strontium present in  the leachate.
Solid Phase and Solution Equilibria
     Strontium is an alkaline-earth and forms several salts.  Thermodynamic
data for various strontium compounds [SrO, Sr(OH)2, SrF2, SrClg, SrCl2-2H20,
•SrCl2-3H20, SrCl2-H20, SrHP04, Sr3(P04)2, SrS04, Sr(N03)2,  SrC03> SrSi03,
SrSi04, SrZr03] were considered in order to determine their stability.  Except
for SrSi03, SrHP04, Sr3(P04)2, SrC03, and SrS04, the strontium solid phases
are too soluble to show in Figure 3-22.  The thermodynamic data for all of
these compounds except Sr3(P04)2 were obtained from Parker et al. (1971).
The data for Sr3(P04)2 were selected from Sillen and Martell (1964).  In an
acidic environment, most of the strontium solids will be highly soluble, and
                     2+                       -4
if the activity of Sr   in solution exceeds 10   moles/liter, SrS04 can
                                    3-167

-------
precipitate and would be a stable phase.   However,  in  alkaline conditions,
SrC03 would be the stable phase and the compound  that  may maintain  strontium
activity in soi.l solutions.
 CVi
                             r-Sr3(P04)2(V&G)
                     Sr3(P04)2(OCP)
                                                                       2+
 Figure 3-22.  The relative stability of various strontium solids at pCa   =
              pS042' = 2.5, pH4Si04 = 3.1, and pC02(g) = 1.52 atm in equilib-
              rium with Variscite and Gibbsite (V and G), Dicalcium Phosphate
              Dihydrate (DCPD) and Octacalcium Phosphate (OCP).  Redox
              potential has little effect on strontium solubilization.

      The  activity of various solution species in equilibrium with SrC03 and
                                    o                        M
 at anionic  activities of pCl" = pS04 " = 2.5, pN03  » 3.0, pF  = 4.5,
 pH2P04" = 5 and  pC02/  > = 1.52 atm are plotted in Figure 3-23.  The thermo-
 dynamic data for SrN03^ and SrP04~ were obtained from Sillen and Martell
 (1964).   All  the other data were obtained from Parker et al. (1971).  Although
 strontium forms  complexes with the various anions listed above, the solution
 complexes do not contribute significantly to the total strontium activity in
 solution.  Sr2+  would  be the dominant solution species over the environmental
                                    3-168

-------
pH range of interest and would be expected to be adsorbed by an ion exchange
reaction.  Izrael et al. (1970) have indicated that strontium was likely to be
100% cationic and that the principal reaction with soils and rocks is ion
exchange.
 co
 01
 o
The activities of various strontium species in equilibrium with
SrC03(s) in the soil at pCl = pS042' = 2.5, pN03~ = 3.0, pF" =
4.5, pC02(g) « 1.52 atm and pHgPCty" = 5.0.
Figure 3-23.
Experimental Adsorption Results
                                       QQ
     Amphlett and McDonald (1956) used   Sr to study cesium and strontium
removal by a soil made up of montmorillonite, illite and kaolinite.  The
                                               + Sr"1"2 H> 2 Cs"1"
equilibrium constant for the reaction 2
determined as 9.3 x 10  .
                                                  Srsoil was
     Rhodes (1957) showed the effect of system pH on the adsorption of stron-
tium from a distilled water solution by a fluviatile Hanford soil.  A soil
paste pH was 8.6, and the calcite content was 2 wt%.  The clay fraction of
                                    3-169

-------
less than 2 ym was 2 wt% and consisted of primarily montmorillonite, with some
mica and chlorite.  The system was pH adjusted with NaOH or HC1.   The strontium
Kd rose from 5 ml/g at pH 6 to 30 ml/g at 8, and 120 ml/g at pH 10.   Above
pH 10, the sodium added in the.NaOH used for pH adjustment began  to  compete
for exchange sites with the strontium, and strontium adsorption began to level
off.  In 4M NaNO.,, strontium adsorption was much less affected by pH.  At pH 8,
for example, strontium Kd was about 5 ml/g and rose to about 10 ml/g at pH 10.
     Rhodes and Nelson (1957) determined the effects of strontium concentra-
tion on the strontium distribution between soil and synthetic uranium recovery
plant scavenged waste.  The soil used here was the same as that used in the
above study (Rhodes, 1957).  On a log-log plot of strontium concentration in
the waste versus strontium concentration on the soil, a straight  line relation-
ship resulted indicating ion exchange as a reaction mechanism.
     Prout (1958, 1959) showed the effects of pH and strontium concentration
on strontium adsorption by Savannah River soil.  This soil was 80% sand and
20% clay, chiefly kaolinite.  The results were very similar to those of Rhodes
(1957) in that a maximum strontium adsorption was reached at about pH 10,
although this maximum was much higher (Kd = 700 to 800 ml/g).  The effects of
competing cations including several sodium nitrate, concentrations were also
given.  The 30% NaN03 solution is about the same as Rhodes (1957) 4M NaN03
solution, and the results also are comparable in strontium distribution
coefficients.  The amount of competing sodium is apparently so high  that the
soil differences are of lesser importance.
     Spitsyn and Gromov (1958) showed that the adsorption of strontium by
montmorillonite was an ion exchange process.  Various competing cations were
added to the strontium-clay system and the strontium distribution coefficient
                                                                          _2
redetermined.  These authors also added small amounts of anions such as CO,
        -2
and C,0A   that were said to form radiocolloids with strontium.   They believed
                                                                         -5
the enhanced removal of strontium reported by McHenry (1955), when 3 x 10   to
      -4    -3
3 x 10  M PO*  was added to wastes contacting calcareous soils, could be
attributed to the formation of precipitates as a result of the natural calcium
on soil exchange sites reacting with the added phosphate ions. This may be
true at some phosphate concentrations, but later study also demonstrated a
replacement reaction (Ames, 1959) where calcite reacted with the  phosphate to
form an apatite.
                                   3-170

-------
     McHenry (1955, 1958) extensively studied strontium adsorption by a Hanford
soil.  The Hanford soil had an exchange capacity of 5.90 ± 0.16 meq/100 g, a
clay content of 1.15%, a calcite content of 1.50% and a paste pH of 8.4.  The
soil was calcium-based before use.  A comparison of the Hanford composite soil
with others is made in Table 3-74 as strontium Kd values from distilled water.
The effect of initial  strontium concentration on strontium Kd values is also
given.  The influence of am'ons on strontium Kd values is shown in Table 3-^75.
It is probable that replacement of the carbonate in the calcite by oxalate and
                                          *3       9        *3
phospha'te ions was underway at the 3 x 10" M £7^4   an(* ^4  concentrations,
leading to enhanced removal of strontium during the reaction (Ames et al., 1958).

              TABLE 3-74.  STRONTIUM Kd VALUES FOR WESTERN SOILS
                           (McHENRY, 1958)
Soil
Ringold, E. Wash.
Warden, E. Wash.
Bowdoin, E. Mont.
Hall, C. Nebraska
Hanford Composite
Kd, ml/q
400
270
135
600
70
CEC,
meq/100 g
34.0
6.0
13.6
26.3
5.8
Mechanical Analysis, %
>50 um
18.6
78.1
48.5
5.9
84.6
50-2 um
39.0
18.9
34.6
60.6
• 11.9
<2 yun
42.4
3.0
16.9
33.5
3.5'
Paste
7.2
7.3
8.3
6.5
8.3
   TABLE 3-75.  THE INFLUENCE OF ANIONS ON THE STRONTIUM Kd IN THE HANFORD
                COMPOSITE SOIL.  ANIONS, AS SODIUM SALTS, WERE PRESENT IN
                CONCENTRATIONS 10 TIMES AS GREAT AS THE STRONTIUM CATIONS.
                (McHENRY, 1958)
                                       Kd, ml/q
S/2,
3 x
3 x
3 x
3 x
10
10
10
10
M
-4
-5
-6
-7
NOJ
18.
49.
65.
65.
Cl"
0
8
6
6
25
55
70
72
.7
.9
.3
.9
C2042
574
98.
65.
67.

9
6
9
sa
24
61
72
78
-2
.3
.4
.9
.6
PQ43
847
115
70.3
75.7
     Klechkovsky et al. (1958) determined strontium distribution coefficients
on a number of Russian soils, and the effects on strontium Kd of competing
Ca   and K .  The com)
their concentrations.
  +2
Ca   and K .   The competing cations lowered the'strontium Kd in proportion to
                                    3-171

-------
     Sorathesn et al.  (1960) obtained some strontium distribution coefficients
on Clinch River sediments that were in the ten thousands, much higher than for
standard clays equilibrated with the same solution.  The difference was thought
to be due to the organic material in the river sediment.  Benson (1960)
reviewed the strontium work with soils at Hanford up to 1960.
     Baker and Beetem (1961) studied the distribution coefficients for carrier-
free strontium on an Alaskan wind-blown silt with various concentrations of
competing cations which were to be expected at the site.  One or two of the
Kd values exceed the experimental limits by large amounts, but the remaining
values conform to a straight line plot of strontium Kd versus the inverse sum
of the competing calcium and magnesium concentrations as required by ion
exchange theory (Kaufman, 1963).
     Cohen and Gailledreau (1961) investigated strontium adsorption on Saclay
soil consisting of 45% montmorillonite, 45% kaolinfte and 10% illite.  The
cation exchange capacity was 210 meg/100 g.  Graphs were presented of trace
strontium-macrocalcium, trace strontium-macrosodium and trace strontium-
macropotassium systems.  Some work was performed on three-component ion
exchange systems.  It was shown that the Kd values plotted versus the stron-
tium concentration for the above Saclay ~soil yielded-a straight line on log-
log scales as required by ion exchange theory.
     Goldberg et al. (1962) measured the strontium distribution between
ground Rainier Tuff and a synthetic groundwater solution at 1-day and 7-days
equilibration times.  There was marked time dependency on strontium adsorp-'
tion by the 100 to 200 mesh tuff.  Comparable solutions yielded a 7-day
strontium Kd of 3000 ml/g and a 1-day strontium Kd of 500 ml/g.
     Angelo et al. (1962) measured strontium Kd values on 32 to 80 mesh
basalt and obtained values ranging from 16 ml/g to 135 ml/g, depending on
the amount of salt in the contacting solution.  The lesser values correlated
with the higher salt contents.
     Janzer et al. (1962) examined strontium adsorption on Project Gnome
rock samples.  The strontium Kd ranges obtained are listed in Table 3-76.
The rock types are described in Table 3-24.  One sample part was equilibrated
with non-traced solution, with the tracer added later.  The five equilibrating

                                   3-172

-------
solutions varied from 1,000 to 45,000 mg/1 in dissolved solids and each experi-
ment was run in triplicate.  The strontium Kd values are given as minimums for
the highest dissolved solids to maximums for the lowest dissolved solids.  A
membrane technique was used to separate solids and solution.

                 TABLE 3-76.  SUMMARY OF STRONTIUM Kd VALUES
                              (JANZER ET AL., 1962)
                                  	Kd, ml/g
                  Sample
                     45
                     95
                    320
                    331
                    350
                    365
                    380
                    479
                    510
                    5-26
                    552
                    620
                    653
                    680

     Baetsle et al.,-(1962, 1963) gave the range in strontium Kd values with
soils at Mol, Belgium, that consisted chiefly of quartz sand as 1.7 to 38 ml/g
at pH 7.7 with tap water.  Dynamic or column Kd values were compared to static
determinations.  Problems were encountered in obtaining adequate Kd values
derived from column breakthrough data because physical flow phenomena were
superimposed on the effects due to radionuclide adsorption reactions.  Batch
adsorption data were easier to obtain and free of problems associated with
column physical flow phenomena.
     Beetem et al. (1962) determined strontium Kd values on granodiorite from
the Climax stock with synthetic groundwater.  The 100 to 200 mes'h samples

                                    3-173
Minimum
8
8
10
3
8
9
6
8
5
8
7
6
0.6
0.7
Maximum
20
23
29
22
16
16
12
17
14
17
15
14
9
11

-------
were equilibrated for 72 hr and yielded strontium Kd values of 4 to 9 ml/g.
The 0.5 to  1.0 mm granodiorite was equilibrated for 720 hr and yielded Kd
values of from 11 to 23 ml/g, probably due to the longer contact time.
     Kokotov et al. (1961, 1962) gave Kd values for strontium on six Russian
soil types  as a function of pH.  The replacement series Sr > Ca > Mg > K s NH.
> Na was given for strontium competing cation efficiency.
     Stead  (1963) referred to strontium Kd values of 5 to 14 ml/g on a dolo-
mite from the Rustler formation when the groundwater had 4250 mg/1 of total
dissolved solids.  The average Kd was 10 ml/g.
     Berak  (1963) used the same solution and conditions shown in Table 3-20
under cesium to determine strontium Kd values of a number of rocks and min-
erals.  The diameter of the solid particles was 0.1 to 0.2 mm.  The concen-
tration of  carrier strontium was 1 x 10  M as SrCl^-  The strontium distribu-
tion coefficients are given in Table 3-77.  Strontium discrimination factors
were determined on several rocks and minerals by using a solution containing
10"5M Sr (NO,), + 90Sr and 2.5 x 10"3M Ca"1"2.  After equilibration, 90Sr was
            J                  +2
determined  by radioassay and Ca   by complexometric titration.  The fraction
   90                                             +2
of   Sr on  the solid divided by the fraction of Ca   retained on the solid
constitutes the strontium discrimination factor at the strontium and calcium
concentrations and ratios used.  The results are listed in Table 3-78.
     TABLE 3-77.  STRONTIUM DISTRIBUTION COEFFICIENTS FOR SEVERAL ROCK,
                  MINERAL AND SOIL TYPES
             Rock
Alluvium,
  Central Nevada
Silty Clay, Idaho

Silty Clay, Idaho
Clinch River, Tenn.
  bottom sediments
Sr
Kd
, mg/1 Condition
48-2454 500-4000


545
1690
1438
659
3537
4024
240
220
:
-
*

1
3
7
1
3
7
Quartz ,
illite,
apatite
Quartz ,
illite,
apatite
hr
days
days
hr
days
days
pH 6
pH 9
ym




    Reference
Nork et al.,
1971
Wilding and
Rhodes, 1963

Wilding and
Rhodes, 1963

Carrigan et al.,
1967
                                    3-174

-------
                           TABLE 3-77.   (continued)
              Reck
  Sr Kd, mg/1
  Condition
     Reference
Sandstone, fine,
light gray

Shaley siltstone,
carbonaceous, black

Sandstone, fine,
light gray

Sandstone, very fine,
silty, dark gray

1.26
1.13
1,88
8.32
9.41
9.56.
1.37
1.36
2.08
7.76
8.37
9.79
4000 ym
500 ym
<62 ym
4000 ym
500 ym
<62 ym
4000 ym
500 ym
<62 um
4000 ym
500 ym
<62 ym
Sokol ,


Sokol ,


Sokol ,


Sokol ,


1970


1970


1970


1970


 Shaley limestone,
   New Mexico

 Sandstone,.
   New Mexico

 Alluvium, INEL,  Idaho
Tuff, Rainier
  Mesa, Nevada
Tuff, Rainier
  Mesa, Nevada
Tuff, Rainier
  Mesa, Nevada
Tuff, NTS, Nevada
Limestone,
  Yucca Flat, Nevada

Granodiorite,
  Climax stock
Granite, Central
  Nevada

Basalt, Buckboard
  Mesa, Nevada

Basalt, Amchitka


Basalt, Amchitka
      8.32
      1.37
 >4000 ym
 >4000 ym
  7.2-10.5 (Laboratory)
      40   (Field)
     260


  2070-3480


  1700=4300


    4000

    0.19
     4-9
    11-23

    1.7
    16-135
     220
 >400 ym
 100-200 mesh
>4000 ym
100-200 mesh
0.5-1 mm

>4000 ym
32-80 mesh
500-4000 ym
    1220        500-4000 ym
1.1  (seawater)   500-4000 ym
 Nork and
 Fenske, 1970

 Nork and
 Fenske, 1970

 Schmalz, 1972


 Nork and
 Fenske,  1970

 Goldberg et  al.,
 1962

 Kaufman,
 1963

 Stead,  1964

 Nork and
 Fenske,  1970

Beetern
et al.,  1962
Nork and
Fenske,  1970

Angelo et al.,
1962

Essington and
Nork, 1969

Essington and
Nork, 1969
                                     3-175

-------
Rock
TABLE 3-77.  (continued)



    Sr Kd. mg/1     Condition
Reference
SUPERSATURATED ROCKS (contain quartz)
Rhyolites,
Czechoslovakia

















Volcanic Glasses, East Slovakia


.
Central Slovakia
Iceland

Rhyodacites, East Slovakia
Dacites, Slovakia



Quartz Porphyries,
Central Bohemia
Bohemia


North Bohemia
SATURATED ROCKS (contain no quartz
Trachytes, North Bohemia
Slovakia
Andesites, Central Slovakia


56 0.1-0.2 mm
15
7
6
33
30
25
41
12
4
12
9
9
21
12
31
37
0
7 0.1-0.2 mm
32
1
11
13
3
6
11
16 0.1-0.2 mm
43
21
27

18 0.1-0.2 mm
8
47
6
8
or feldspathoids)
39 0.1-0.2 mm
12
11 0.1-0.2 mm
13

Berak, 1963

















Berak, 1963


>•




Berak, 1963




Berak, 1963





Berak, 1963

Berak, 1963

                       3-176

-------
                           TABLE  3-77.   (continued)
             Rock
  Pyroxenic, East Slovakia
  Garnetic, East Slovakia
  Altered,
  Pyroxenic,
  Propylletized
  Granul. pyrox., vitreous
Spillites, albitized
  Proterozoic basalt,  -
    Central Bohemia
Diabases, Bohemia
  North Moravia
Melaphyres, basalt
  North Bohemia
  Slovakia
Basalts, North Bohemia
  Vitreous
                               Sr Kd, mg/1

                                   17
                                   13
                                   37
                                   16
                                   30
                                   32
                                   21
                                   48
                                   20
                                   50
 Condition
Reference
                                    8
                                    0
                                    1
                                   10
                                   25
                                   32
                                   30
                                   39
                                   45
                                    1


                                  212
                                   35
                                    5
                                   10
                                  187
                                    6
                                   22
                                   80
                                   33
                                   32

UNDERSATURATED ROCKS (Nonfeldspathoidal)
Picrites, North Moravia
                                   64
                                   64
                                   43
                                   85
                                   85
                                   64
                                   51
                                  108
                                    9
                                              0.1-0.2 mm    Berak, 1963
0.1-0.2 mm    Berak, 1963
0.1-0.2 mm    Berak, 1963
0.1-0.2 mm    Berak, 1963
0.1-0.2 mm    Berak, 1963
                                     3-T77

-------
Rock
TABLE 3-77.  (continued)




     Sr Kd, mg/1      Condition
Reference
Teschenites with olivine,
North Moravia


.


Picrites, North Moravia

Peridotite, North Moravia
Pyroxenite, NW Bohemia

Monchi quite with olivine,
North Moravia
Ouachitite, with apatite
North Moravia
Olivine basalts, N. Bohemia
East Slovakia
Central Slovakia
North Bohemia
UNDERSATURATED ROCKS (with
Phono! ites, North Bohemia



Tephrites, North Bohemia,
chabazite
leucite
chabazite
nepheline
nepheline-leucite
..leucite
leucite
Basanites, North Bohemia,
nepheline-leucite




Nephelinites, Mel i lite,
NW Bohemia
leucite
olivine-nepheline basalt


74
30
22
38
25
46
92
13
35
0
0
35

7
52
321
322
324
334
feldspathoids)
15
• • 7
140
75

86
26
44
143
83
65
80

32
0
22
38
59
__.0 	
5
22
28
60
0.1-0.2 mm Berak, 1963






0.1-0.2 mm Berak, 1963




0.1-0.2 mm Berak, 1963








0.1-0.2 mm Berak, 1963
. . ... 	 	 . .



0.1-0.2 mm Berak, 1963







0.1-0.2 mm Berak, 1963





0.1-0.2 mm Berak, 1963



                       3-178

-------
                            TABLE 3-77.   (continued)
             Rock
Leucitites, NW Bohemia
Sr Kd, mg/1
  Condition
   Reference
  nepheline
Melitites, olivine-hauyne basalt

MINERALS
  Quartz
  Agate
  Chalcedony
  Opal
  Olivine
  Chondrodite
  Thaumasite
  Leucophanite
  Zircon
  Topaz
  Kyanite
  Sillimanite
  Staurolite
  Garnet, almandine
  Garnet, grossular
  Garnet, andiadite
  Garnet, uvarovite
  Hydrogarnet
  Vesuvianite
  Sphene
  Rinklte
  Axinite
  Hemimorphite
  Beryl
  Dioptase
  Tourmaline
  Woliastonite
  Rhodonite
  Hypersthene
  Aegirine
  Augite
  Diopside-
'  Enstatite
  Jeffersonite
  Tremolite
  Chrysotile
  Sepiolite
  Attapulgite
  Palygorskite
  Zoisite
    56
    82
    33
    16
    29
     5
     9
     0
     0
     3
    10
     0
    18
    35
     8
     0
     9
     8
     0
     0
     0
     2
    24
     0
     0
    10
     0
    14
     0
    15
     0
     8
     5
     4
    14
     2
     3
     3
    16
     6
     2
    30
     5
     8
     0
 0.1-0.2 mm     Berak,  1963
0.1-0.2 mm
0.1-0.2 mm
Berak, 1963


Berak, 1963
0.1-0.2 mm    Berak, 1963
                                    3-179

-------
                         TABLE 3-77.   (continued)


	Rock	   Sr  Kd,  mg/1      Condition         Reference

 Talc                              6~~
 Pyrophyllite                      0
 Biotite                           5
 Muscovite  -                       5
 Chlorite Penninite                2
 Chlorite, Dellesite               o
 Sericite                          2
 Illite                          100
 Glauconite                       25
 Celadon He                       35
 Serpentine                        0
 Kaolinite                        15
 Ha Hoy site                       58
 Allophane                        71
 Hisingerite                      97
 Smectites                       104
                                 150
                                  96
                                 138
                                 150
                                 217
                                 163
   Nontronite                    187
 Apophyllite                      10
rMelvlite -.--.	                5                  '          •
 Feldspars, Orthoclase             10
   Sanidine                       23,11
   Albite    ,                     10
   Oligoclase                     11
   Andesine                       10
   Laboradorite                    8
 Scapolite                        11
 Leucite                          11
 Anal cite                         25
 Pollucite                         7
 Nepheline                         0
 Sodalite                          3
 Cancrinite                       14
 Zeolites:
   Mordenite                      27         0.1-0.2 mm    Berak, 1963'
   Stil bite                      166
                                  25
   Heulandite                     47
   Faujasite                      34
   Harmotone                      12
   Chabazite                '     210
   Natrolite                •       8
   Scolecite                       4


                                   3-180

-------
TABLE  3-78.  MICROSTRONTIUM-MACROCALCIUM DISCRIMINATION ON  SOME 0.1 to  0.2 mm
              ROCKS AND MINERALS (BERAK,  1963)


                                                      Strontium
                                                   Discrimination
              	Mineral	            Coefficient

              Halloysite           '                        1
              Montmorillonite                              1
              Sepiolite                                    2.3
              Chlorite                                     1
              Oellesite               •                     9
              Socialite                                   >9
              Anal cite                                     2
              Cancrinite                                  12
              Leucite                                     3
              Orthoclase                                   9
              Orthoclase                                  11
              Microcline                                   8
              Albite                                      9
              Labradorite                                 9
              Mordenite with quartz                        1.6
              Heulandite                        .          48
              Heulandite                                  55
              Heulandite                                  19
              Mordenite                                    1.3
              Faujasite                                    4
              Harmotome                                   >9
              Brewsterite                               >24
              Phillipsite                                 8
              Chabazite                                   14             .
             . Natrolite                                    9
              Scolecite                                   >9
              Thomsonite                                 >34
              Thomsonite                                  10.2

              SILICATE ROCKS

              Basalt                                    "17
              Nepheline tephirite                          2
              Nephelinite                         .       10
              Nepheline-leucite tephirite                  4
              01ivine hauyne nephelinite                   5
              Leucite tephrite                             5
              Leucite tephrite                             4
              Leucitite with phillipsite                   8
              Tephrite with chabazite                      5
              Basalt mineralogy -  augite, 40%
                                  plagioclases, 50%
                                  magnetite, 8%
                                  apatite, 1%

              Nephelinite - Ti-augite, 68%
                           magnetite, 12%
                           nepheline, 10%
                           plagioclase, 7%
                           leucite, 3%
                                     3-181

-------
     The data of Table 3-77 show that the secondary silicates, such as the
 clay minerals and certain zeolites, give much higher Kd values for strontium
 than the primary igneous minerals found in rocks, such as quartz and feldspars
 Some of the secondary silicates can also discriminate between geochemically
 similar calcium and  strontium, as seen in Table 3-78.
     Tamura and Struxness (1963) outlined several strontium removal mechanisms
 that could apply to  soil solutions.  Two ion exchange reaction types were
 postulated.  One type involves simple ion exchange on a relatively pH-stable
 substrate such as zeolite.  The other type was exemplified by alumina that
 is amphoteric as a substrate for ion exchange.  The differences were further
 elucidated by Tamura (1964).  Strontium Kd values as high as 30,000 were mea-
 sured  in alkaline systems on alumina.  The strong influence of soil sesqui-
 oxides in strontium  removal was emphasized.
     Baetsle et al.  (1964) have given data (Table 3-79) on Kd values for
 strontium with Belgian soils.  The strontium exchange work on Mol, Belgium,
 soils was with deionized water, mains water (tap water?) and groundwater.
 None of the water compositions were given.  The pH was given probably as mea-
 sured  in the eluting solution prior to the equilibrations.  The Kd values
                                     89
 were obtained by adsorbing a band of   Sr on the top of a soil column and
 eluting with the solution indicated.  Samples of the effluent were counted
 and the Kd determined from the C/CQ = 0.5 as indicated above.  The results
 were part of a study to compare static and dynamic Kd values.
         TABLE 3-79.  VARIATION OF TRACE STRONTIUM Kd VALUES WITH pH
                      (BAETSLE ET AL., 1964)
                     	       Kd. ml/g	     	
                      Deionized   Mains Water           Groundwater
    Soil Type        Water, pH4  pH 7.7  pH 3     pH 4      pH  5      pH 7
Eolian Sand
  Horizon A             16-22     38     °7     2.0-2.8   3.6-5.8     5-7
  Horizon B            1.5- 1.6   10     13
  Horizon C            0.5- 0.9    3.5   12
Mol White Sand         0.95        1.7    4.5   2.1-2.5   5.4-5.6   4.7-4.9
Mol Lignitic Sand       --      '    --     —   2.0-3.1    5.0-5.5   6.8-7.4
                                   3-182

-------
     The mains water Kd values are higher due to the higher pH.  However, the
higher mains water Kd values at pH 3 were difficult to rationalize.  They
should be lower than both the deionized water Kd values and the mains'water Kd
values at pH 7.7.
     A Burns, Mississippi, montmorillonite, with the composition shown in
Table 3-80, was used in macroion systems of potassium, calcium and magnesium
to obtain Kd values and calculated Kd values in mixed two and three ion
systems (Baetsle et al., 1964).  The selectivity coefficients given were
KSr ' °-096
""mont. + Sr> and     " °'98 
-------
     TABLE 3-80.  COMPOSITION OF BURNS,  MISSISSIPPI, MONTMORILLONITE.
                   CAPACITY WAS 1.04 meq/g  (BAETSLE ET AL.,  1964).
Constituent

  Sl°2
 A1203
 Fe2°3
  FeO
  Ti02
  CaO
  MgO
                              H20"
                              H20+
                             Total
                                            Wt 1,
51.29
15.96
 2.46
 0.14
 0.18
 2.21
 3.95
 0.05
 0.13
18.00
 5.37
99.74
TABLE 3-81.   COMPARISON OF  EXPERIMENTAL AND  COMPUTED STRONTIUM Kd VALUES
              FOR BURNS MONTMORILLONITE (MERCER, 1967)
Concentration, meq/ml
<-K . CMg ^ca
0.0855 .
0.0404
0.0143
0.00702 .
0.00240
«
—
—
• --
--
0.0885
0.0411
0.0155
0.00612
0.00582
—
—
—
—
..
0.01382
0.00962
0.00604
0.00412
0.00282
0.0180
0.0146
0.0106
0.0068
0.0043
--
«
—
—
—
0.0852
010388
0.0100
0.0046
0.0015
_.
—
«
--
~
0.0769
0.0348
0.0096
0.0032
0.0009
0.0124
0.00850
0.00520
0.00360
0.00248
0.0128
0.0114
0.0074
0.0056
0.0036
Experimental
Kd, ml/g
24.8
59.5
135.5
207.0
369.7
• 12.17
20.6
56.9
114.3
193.5
18.0
50.0
116.0
193.0
336.0
12.3
• 25.8
58.6
100.6
244.1
Baetsle
Kd, ml/g
21
41
102
177
368
12.1
21.3
53.2
106.8
221.8
22.0
48.0
m.o
203.0
300.0
14.5
24.5
55.0
107.0
209.0
Mercer
Kd, ml/g
10
33
102'
191
352
11
21
53
109
212
10
33
102
206
285
11
22
62
104
206
                                    3-184

-------
     Hawkins and Short (1965) completed an experimental design to allow mul-
tiple regression analysis of the effects on strontium and cesium Kd values of
                                                      +2          '     •+•
the following ions and their concentration ranges:  Ca  , 5-500 ppm; Na ,
6-650 ppm; K+, 1-100 ppm; NHt, 1-100 ppm; Mg"1"2, 1-125 ppm; Cs+, 0.05-5 ppm;
  -3
PO, , 0.1-10 ppm; and pH 4-9.  The equation for strontium adsorption may be
applicable to other soils besides INEL, Idaho soils, if they do not contain
abundant sesquioxides.
     Wahlberg and Dewar (1965) studied ion exchange adsorption of trace con-
centration strontium (10   N   Sr) as affected by two competing cations.  The
Kd values calculated with the mass-action equation agreed most closely with
high exchange montmorillonite than with lower capacity kaolinite.
     Dlouhy (1967) investigated strontium and other cations adsorbed by
Cascaccia soil and tuff as a function of pH, time and solution composition.
Kd values were used in chromatographic equations to show the radionuclide
distribution beneath disposal sites.  Strontium Kd values ranged from 6 to
15 ml/g in the soil to 45 to 75 ml/g in the tuff.
     Hajek and Ames (1968) compared batch and small column Kd values for
strontium and cesium on a Burbank loamy fine sand containing up to 2 wt%
calcite.  The soil to solution ratio was varied between 100 to 5 for the
batch Kd determinations.  Strontium Kd values tended to decline as the solu-
tion to soil ratio declined.  The small columns did not yield a real Kd value
for strontium unless it was preleached with at least 450 column volumes of
3N NaNOg.  It is probable that calcium from the calcite was solubilizing and
competing with the strontium during the batch tests.  Pre-equilibration of
the soil with the solution to be used minus the strontium radioisotope is a
must before Kd values are measured.  Gardner and Skulberg (1964) had similar
Kd changes with solid-solution ratio changes.
                                                            90
     Jenne and Wahlberg (1968) found that the adsorption of   Sr was controlled
by in situ precipitation of calcium carbonate in a small stilling basin sample
in White Oak Creek, Tennessee.  The clay minerals controlled only the adsorp-
tion of cesium while manganese and iron oxides controlled   Co removal.
     Cerrai et al. (1969) reported that the strontium Kd on a marine sediment
from a core at 50 m off La Spezia, Italy, was 6.3 ml/g.

                                    3-185

-------
     R. J. Serne (PNL), 1973, determined strontium distribution coefficients
from high salt, high pH, synthetic tank solutions with Hanford core well sedi-
ments.   The chemical composition of the synthetic tank solutions are given in
Table 3-9 and the sediment characteristics in Table 3-10 under antimony.  The
strontium Kd values in Table 3-82 are the average of three values.  The rela-
tively low strontium Kd values obtained were principally due to the high con-
centrations of competing cations present.

                  TABLE 3-82.  STRONTIUM Kd VALUES (ml/g)
                                         Solution
Sediment
1
2
3
4
5
6
7
I
3.91
2.90
2.93
12.40
1.87
2.50
1.86
II
1.74
1.17
1.22
5.44
0.81
0.95
0.87
III
0.023
0.047
0.072
0.074
0.13
0.10
0.08
IV
0.0
0.0
0.0
0.0
0.17
0.14
0.23
V
18.40
8.40
12.87
173.84
12.70
39.20
20.16
     Routson (1973) determined strontium adsorption characteristics of three
Hanford project soils.  The soils were well-character!zed as shown in
Table 3-83.  The clay fractions (0.2 to 2 ym and <0.2 urn) were analyzed by
X-ray diffraction and found to be very similar for the three soils.  The
Ritzville silt loam, Burbank loamy sand and Ephrata sandy loam showed mont-
morillonite as the dominant mineral in the 0.2 to 2 ym fraction, with lesser
amounts of vermiculite and mica.  Montmorillonite was the only clay mineral
detected in any of the <0.2 urn fractions.  There was no evidence of clay
mineral weathering after deposition of the parent materials.  The only indica-
tion of soil weathering is the accumulation of calcium carbonate in the upper
C horizons of all three soils.  The strontium Kd values are given in Table 3-84.
The original soil samples had chiefly calcium in the exchange positions.  Note
that the sodium based soils showed higher strontium Kd values, indicating that
calcium displaced strontium when present in the system.  It can also be seen
in Table 3-84 that calcite removal from the soils was the most effective way
to raise the strontium Kd values.

                                    3-186

-------
TABLE 3-83.
CHARACTERISTICS OF THE SOILS USED IN STRONTIUM Kd
DETERMINATIONS (ROUTSON, 1973)
Particle Size
% of
Sail Fraction
Ritzville
All
0-2 in

A12 2-18 in.
81
Cca
C
Burbank
All
18-27
27-41
41-72

0-3 in
in.
in.
in.

.
A]2 3-16 in.
AC
AC2
1C
Ephrata
All
A12
Bl
IB2
IIC
16-22
22-30
30 in.

0-3 in
in.
in.'

.
3-12 in.
12-15 in.
15-19 in.
19-24 in.
TABLE 3-84.
Sand
23.
47.
54.
39.
55.

76.
91.
88.
82.
79.

44.
59.
57.
62.
92.
3
0
7
8
0

8
2
7
5
8

2
0
4
9
1
TRACE
FROM 0
Silt Clay
64.1 12.6
46.0 7.0
42.5 2.5
57.7 3.3
42.0 3.0

16.8 6.4
7.6 1.2
10.6 0.7
16.4 1.1
16.2 4.0

47.1 8.7
36.9 4.1
39.3 4.3
31.4 5.7
5.4 2.5
CaC03
% of Total %

-
• o
1
8
4

-
0
1
5
3

-
0
0
0
3

-
.8
.3
.9
.2

-
.7
.6
.5
.5

-
.57
.70 .
.87
.28
Fe2°3 Organic Carbon
of Total % of Total

1
1
0
l'
1

0
0
0
0
0

1
1
1
0
0

.50
.44
.89
.09
.11

.80
.84
.62
.52
.39

.22
.25
.20
.98
.46

—
0.34
0.27
0.13
0.17

--
0.16
0.14
0.13
0.20

—
0.18
0.13
0.29
0.26












STRONTIUM ADSORPTION CHARACTERISTICS
.2N NaCl (ROUTSON, 1973)
Equilibrium Distribution Coefficients (ml/a)
No
Pretreatment
Ritzville
A12
Cca
C
Burbank
A12
AC
AC,
1C
Ephrata
A12
81
IB2
IIC








21
27
27
24

12
15
21
17

12
12
17
23
















Na-Based '

109
131
84
58

57
41
53
48

73
68
32
55
Organic Carbon
Removed










37
46
40
43

20
28
35
38

21
22
33
55









Calcite
Removed
189 -
257
71
82

106
91
67
81

115
126
166
145
Organic C
and Calcite
Removed

96
101
61
66

45
61
65
52

44
57
96
70









                           3-187

-------
     Jackson (1976) used tracers to study adsorption of seven radionuclides,
including strontium, on copper ore and decomposition products during copper
leaching, as a function of ore particle size, solution composition, pH and
liquid/solid ratio.  The reactions were carried out in autoclaves for up to
8 months in a controlled oxygen pressure and 363 K.  The solids continued to
concentrate strontium from the liquid for a long period of time as the ore
and gangue surfaces continued to change and grow other mineral phases.  A sub-
stantial amount of activity was adsorbed by the ore and gangue decomposition
products (gypsum, jarosite, anhydrite).  As an example, in one long-term
experiment, the strontium Kd was 0.25 ml/g at time zero and 499.6 ml/g at
zero plus 173.6 days.
Migration Results
     Field Studies —
     Mawson (1956) and Evans (1958) reported the results of studies on fission
product radionuclide movements after disposal in sand at Chalk River, Ontario.
In dilute, neutral pH wastes ruthenium moved much more rapidly through the
sand than did strontium.  When acid wastes were disposed into the same sand,
strontium and ruthenium both moved rapidly and at about the same speed through
the sand.
     Spitsyn et al.. (1958) used an alkaline solution (4 to 8 g NaOH/1,
200 g NaN03/7) and an acidic solution (6 to 8 g HN03/1, 200 g A1(N03)3/1)
in field studies of radionuclide migration.  A mixture of radionuclides were
injected into the sandy soil and the movement traced by wells along the migra-
tion route.  Strontium traveled the farthest, with ruthenium traveling about
half as far as strontium.  The authors concluded that the groundwater into
which the strontium is disposed should have a low calcium and general dissolved
solids content.
     Parsons (1961) described the movement of fission products from a surface
disposal site in Chalk River, Ontario, through the near-surface groundwater,
resulting from a 1954 experimental disposal of 5700 liters of waste containing
2100 kg of ammonium nitrate.  The high concentration of ammonium ions caused
all of the strontium to move off relatively rapidly with the groundwater.
Horizontal movement of the waste was rapid enough to reduce vertical  dispersion,
                                   3-188

-------
and the strontium remained close to the top of the water table and traveling

at 0.027 times the velocity of the groundwater.

     Parsons (1963) described a 1954 disposal of 3800 liters of acid waste
                                                 90
containing complexing agents and over 1000 Ci of   Sr.  The waste was poured

into a pit lined with limestone.  Sampling of the sandy soil and near-surface

groundwater allowed the movement of fission products to be followed.  The

ruthenium migrated rapidly, traveling at nearly the same velocity as the
                  90
groundwater.  The   Sr became separated from the ruthenium, developing into
                                                                        90
a continuous tongue 198 m long, 3 to 4 m thick and containing 800 Ci of   Sr.
                                                        90
The ruthenium entered a shallow lake in 1957, while the   Sr remained more

deeply buried and traveling much more slowly.
                      90
     The migration of   Sr as seepage from Oak Ridge burial pits was investi-

gated by Lomenick et al. (1967).  They reported that most of the strontium on

the side walls and bottom of the pit in Conasauga shale was tied up in the

precipitated calcite (calcium carbonate).  Water moved 0.15 m/day while the
90
  Sr moved 0.24 m/yr.

     Mortensen and Marcdsiu (1963) incubated a silty clay loam with 40 uCi
90
  Sr for 60 days and obtained hot water extractions or hydrolyzed with 6N HC1.
                                                90
The'supernatant liquid was electrodialyzed with   Sr migrating to the anode
                                          90
compartment suggesting negatively charged   Sr-organic matter complexes.  Gel
                       90
filtration showed that   Sr was complexed or held on exchange sites by high

molecular weight polymers.  The radioactivity correlated with the presence of

polyuronides.  Phenols, amino acids and keto acids were separated from the

6N HC1 hydrolysate by paper electrophoresis, with no clear correlation between

radioactivity and these components.

     Brown (1967) reviewed the hydrology and geology of two contaminated waste

disposal trenches at Hanford. .Several wells were drilled in and around the
                       90
disposal sites and the   Sr and moisture determined in the sediment column
                                             QQ
beneath the disposal sites.  The bulk of the   Sr activity was contained in a

15 meter vertical section immediately below the bottom of the disposal sites.
                                          -1     90
The average strontium concentration was 10   uCi   Sr/g sediment in the

15 meter high strontium concentration zone and fell off rapidly below this
          -4     90                                    90
zone to 10   uCi   Sr/g sediment.  Spatially anomalous   Sr distributions
                                    3-189

-------
were correlated with changes in lithology.  Contaminated sediment columns con-
                 go

taining 0.15 yCi   Sr/g were used to determine the effects of groundwater

            90
leaching on   Sr movement.  Fifty column volumes of groundwater removed 4%

       90
of the   Sr from the sediment column and 500 column volumes removed 31% of

    90
the   Sr adsorbed on the sediments.



     Magno et al. (1970) investigated the migration of strontium through the


effluent lagooning system of the Nuclear Fuel Services plant in western New


York State.  They estimated on the basis of analyses that approximately 90%

       90
of the   Sr passed through the lagoon system and into nearby surface streams.

                         90
Further, only 12% of the   Sr discharged from the lagoon system was associated

                                        90
with suspended solids.  Thus 78% of the   Sr discharged from the lagoon system

                                        90
was in solution.  The low adsorption of   Sr was apparently due to a combina-


tion of pH and competing cations lowering ion exchange loading of solids.


For example, when the plant effluent pH was 11 during one sampling period,

           90
75% of the   Sr was removed in the lagoon system.  At another sampling period,

                                            90
plant effluent was pH 7 and only 10% of the   Sr was removed by-the lagoon


system.



     Merritt (1967, 1976) reported on the leaching and migration of radio-


activity from nepheline syenite-waste glasses buried below the water table at

                                                        90
Chalk River, Ontario.  Sampling in 1971 showed that the   Sr front in sandy


soil had reached 33 m.  Since the leach rate from the glass blocks has been


constant for several years, the plume size has reached a steady state.



     Himes and Shufeldt (1970) investigated the effects of soil organics on

90
  Sr migration through soils.  Thirty 0.002-acre soil microplots were deline-

                   QO
ated and 21.64 yCi 3USr added to the surface of each.  Facilities were in.

                                                                90
place to collect runoff water and leachates at 40 cm depth, and   Sr was
                                                                 on

found in the leachate.  The decreasing order of effectiveness of   Sr removal

                                        90
by organic compounds on the quantity of   Sr adsorbed by most soils was phytic


acid > glucuronic acid > citric acid > pectic acid > pyrocatechol > glycine >

                                                              90
dextrose, as shown in equilibrium experiments.  Adsorption of   Sr by soils


high in organic matter treated with HF-HC1 solution to remove clay minerals

                                                                          90
showed that methylation with diazomethane decreased the total quantity of   Sr


adsorbed.  Saponification,. acetylation and remethylation also altered the
                                    3-190

-------
          90                              90
amount of   Sr retained on the soil.  The   Sr contained in plant residues
leached through 30 cm soil columns at a slow but measurable rate with applica-
                                                      90
tions of distilled water.  Drying of soils containing   Sr at 105°C before
                                                   90
extraction increased the fraction of water soluble   Sr that could be extracted,
     Work on the long-term migration of fission products from the natural
reactor site at Oklo (Brookins, 1976) indicated that strontium had partially
migrated from the reactor vicinity in the past 1.8 billion years.
     Laboratory Studies ~
     Nishita et al. (1956) studied the general extraction of radionuclides by
water leaching, by exchange with IN ammonium acetate and remaining in the soil
                                                                     90
or clay mineral.  In both the soils and clays (bentonite, kaolinite)   Sr was
prominent in the water soluble and cation exchangeable fractions.  It showed
the least tendency to remain in the soil or clay.
     Orcutt et al. (1956) examined hydraulic and ion exchange phenomena as
related to the underground movement of radiostrontium.  Ion exchange reduced
the linear velocity of the radionuclide front in relation to the liquid front.
If the solid media are selective for a radionuclide, the radionuclide concen-
tration front becomes sharp.  The front becomes wider with distance traveled
if the radionuclide selectivity is unfavorable.  The latter condition also
will occur if the waste velocity is too fast to allow instantaneous equilibrium
between exchanger and the radionuclide at the concentration front.
     Another article, by Orcutt et al. (1957) includes experimental work on ion
exchange in columns and its theoretical treatment.  Several theories are pre-
sented that quantitatively describe hydraulic dispersion, ion exchange equili-
bria and ion exchange kinetics as they apply to the movement of radionuclides
through natural porous media.  In the opinion of the authors, the theoretical
models, used in conjunction with field tests, served to reduce the uncertain-
ties of ground disposal operations.
     Spitsyn et al. (1960) emphasized again the ease of migration of strontium
in the soil.  A well was drilled and the migration of fission products was
                                                                       +2
followed in the soil at a field site.  At the site, up to 50 ml/1 of Ca   in
the groundwater caused the initial migration rate of strontium to be about
6 cm/day, practically the same speed as the grcundwater.  Drilling 15 months

                                   3-191

-------
later showed that strontium contamination had spread to 6 m distance from the
injection point, and that the radiostrontium migration rate was 1.1 to 1.3 cm/
day compared to 6 cm/day water velocity.
     A mathematical model of strontium movement in soils based on the labora-
tory work of Miller and Reitemeier (1957) was proposed by Thornthwaite et al.
(1960).  Miller and Reitemeier showed strontium distribution curves for
Norfolk fine sandy loam, Miami silt loam, Huntley clay loam and Fort Collins
silt loam after the application in cycles of up to three solutions (deionized
water, 0.005N CaCU, 0.005N NaCl) at rates of 30 and 300 in.  The strontium
in the sandy soil moved farther than in the other three soils.
     To describe these distributions, Thornthwaite et al. (1960) set up a
model that relates leaching efficiency (cycles/in.) to soil cation exchange
capacity.  The model was used on a New York soil to compute the distributions
   90
of   Sr from fallout and compare the results to measured values.  The compari-
son showed good agreement.
     A study of the influence of climatic and hydrologic factors on the move-
ment of strontium through the upper layers of soil was completed by Mather
and Nakamura (1961).  Strontium fallout moved through the upper part of the
soil profile at a rate determined by the volume of leaching solution and the
chemical and cation exchange properties of the soil and soil solution.  A
mathematical model was devised that allows computation of strontium movement
in the.soil.  To use the model, values are needed for the initial  strontium
concentration, leaching solution volume and leaching efficiencies (cycles/in.).
The leaching efficiency varied from high values in most regions to low values
in dry regions for soils of the same texture.  Leaching efficiency is related
to moisture index.  Tables and monograms were prepared to permit determination
of leaching efficiency from readily available climatic data.
     Jacobs (1963) reviewed an earlier paper (Jacobs, 1960) and a paper by
Bovard and Grauby (1960) on retention of radiostrontium by soils.   Bovard'and
                                                                      90   90
Grauby made laboratory and field studies of the downward migration of   Sr-  Y
in undisturbed soils.   The recovered solution was analyzed and the exposed
block of soil autoradiographed after the experiment.  The experimental results
represented a qualitative description of the nature of the soil and system
                                    3-192

-------
hydrodynamics.  The use of a water tracer was suggested by Jacobs to obtain
groundwater velocities.  If heterogeneities are not severe, and flow conditions
are fairly constant, the groundwater velocities and a time transformation term
for radionuclide adsorption would provide a quantitative site evaluation.
     Miller and Reitemeier (1963) studied leaching of strontium with applica-
tions of 76.2 cm and 762 cm of distilled water, 0.005N NaCl and 0.005N CaCK
solutions to columns containing soils contaminated with strontium-90.  The
CaClp solution produced the greatest movement and the deionized water the
least.  The maximum distance the strontium moved with 30 in. of water was
1.3 in., and with 300 in. of water the distance moved was 4.3 in.  Movement
was greatest in the Norfolk loamy sand.
     Some associated K values for the Mol soils are listed in Table 3-85
                                                     i
(Baetsle et al., 1964), where K is a measure of the velocity of strontium
through the soil relative to groundwater velocity.  Mol sand was almost pure
silica.

         TABLE 3.85.  K VALUES OF STRONTIUM IN THE SATURATED SUBSOILS
                      AT MOL, BELGIUM (BAETSLE ET AL., 1964).  K IS
                      DEFINED ON PAGE 2-30.
                     Soil  Type	      pH  7     pH  3
                 Mol  Sand                     5      1
               Perturbed Profile            48      1
               Mol  Sand &  Eolian Sand      525
               Mol  Sand &  Lignite          570
                 Mol  Sand                    6      1
                 Mol  Sand                    71
                 Mol  Sand                    3      1
                 Mol  Sand                    4      1
                 Mol  Sand                    3      1
                 Mol  Sand                    6      1
     The strontium Kd values and associated relative strontium velocities
through the Houthulst clays at pH 3 are given in  Table 3-86.   Houthulst  is

                                    3-193

-------
located in western Belgium about 40 km from the North Sea.  The strontium Kd
values are somewhat higher for the clay but still much less than cesium.
Cesium K averages 1000 while the strontium K averages 100.

   TABLE 3-86.  STRONTIUM DISTRIBUTION COEFFICIENTS AND RELATIVE MIGRATION
                RATES IN HOUTHULST CLAYS AT pH 3 (BAETSLE ET AL., 1964)

Depth, m
0-1
1-2
2-2.5
2.5-3
3-4
4-5
5-6
6-7
7-8
8-9
9-10

0-1
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9'
9-10

Core Well
I
Kd, ml/q K Depth, m
7.2
10.6
14.1
9.4
—
11.8
13.9
15.8
19.4
17.3
20.2

__
3
5
18
18
23
29
31
16
29
24
36
48
32
—
40
47
54
66
59
68
Core Well
._
11
18
61
61
78
98
104
54
98
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
18-19
19-20

II
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
18-19
19-20

Kd, ml/g
21.8
27.5
30.7
33.5
33.5
31.8
31.8
34.0
32.6
31.5


30
30
28
29
31
31
29
31
31
31

K
74
93
102
112
112
107
107
114
no
106


101
101
95
98
104
104
98
104
104
104
     Polyakov et al.  (1969) determined the diffusion coefficients of strontium
                                                      -7           -7   2
in three soils.  The coefficients varied from 0.6 x 10   to 20 x 10   cm /sec.
These authors pointed out that diffusion coefficients vary with the soil type

                                    3-194

-------
and probably change appreciably within the same soil.  Prokhorov (1962) had
shown earlier that the diffusion coefficient in soil varied with the soil/water
content.  In the majority of the cases, however, except for the very long-lived
radionuclides and very slow flow velocities, diffusion does not contribute
greatly to radionucl.ide migration.
     According to Schulz (1965), the radionuclides showing a degree of mobility
include those of strontium.  The amounts and kinds of competing cations have a
marked effect on soil profile distribution as shown by the effects of compara-
                                                          gn
ble concentrations of Nad and CaCl2 on redistribution of   Sr down the soil
profile.  Cad* moves the strontium more readily than Nad, and the more con-
centrated the CaCl2» the more rapid and further the strontium movement.
     Nishita and Essington (1967) continued work on leaching strontium from
soil columns.  All soils showed that movement by water leaching resulted in
more rapid movement than strontium only by ruthenium.
     Carlile and Hajek (1967) reported an example of physical  transport of
                              85
strontium.  About 0.1% of the   Sr was found in the first effluent from N-Area
soil columns (2 cm diameter by 40 cm in length).  Up to 1% random leakage of
strontium was reported until  5 column volumes-of Columbia River water had
passed through-the soil'.  Then leakage fell to below detection limits (10  %).
Columbia River water has an annual mean pH of about 7.8, a mean CaC03 hardness
of 69 mg/1 and a mean total dissolved solids of 119 mg/1.  Effluent samples
were centrifuged at high speeds, treated with HLO^ to destroy organic material
and recentrifuged.  Fifty to 75% of the activity in the leaked effluent was
removed.  Hydrogen peroxide treatment did not affect removal of strontium by
centrifugation, indicating that an organic material was not involved.  Leach-
ing a soil system with low ionic strength river water containing trace radio-
nuclides had the result of dispersing the soil colloids on which much of the
85
  Sr were adsorbed.  Column breakthrough was immediate until the dispersed
colloids were flushed from the column.  Preceding the contaminated river water
by a preleach of noncontaminated river water flushed the dispersed colloids
and prepared the soil for use.  Adding other ions to the system could have
accomplished the same thing,  but would have competed with adsorption of
strontium.
                                   3-195

-------
                                                                  90
     Brendakov et al. (1969) followed the average displacement of   Sr from a
                    90
Chestnut soil.  The   Sr was amended to the soil in the field and sampled at
yearly intervals.  Two years after amendment, the average displacement dis-
tance from the surface was 11 mm and 5 years after amendment, 16 mm.
     The average displacement was defined as:

                             .  x =
where
     pi = fractional % of the radionuclide in a layer (3 mm thickness)
     xi = the depth of the layer.
There was little difference in X" for any of the radionuclides used in the
experiment.
     Chelation and movement of strontium in a calcareous soil were evaluated
in columns of Mohave sandy loam soil (Fuller and L'Annunziata, 1969).  The
                89
displacement of   Sr from the soil columns was 19, 38, 44, 47, and 81% in the
order HEEDTA < DCyTA < EDTA < EGTA < DTPA.  Only 5% of the 89Sr was removed
from the soil with no complexing agents in the leaching solution.  The radio-
strontium existed in the leachate as a negative complex.  Any natural organic
chelating agents in the soil did not influence radiostrontium movement.
     Four soils varying greatly in organic matter content, cation exchange
capacity and texture were studied by Juo and Barber (1970) for their stron-
tium retention characteristics.  The adsorption of strontium by soils increased
with increasing pH within the pH range 4 to 8.  Saturating cations had some
effect on strontium adsorption, with influence in the order Na>K>Mg>Ca>Ba>H.
As the pH increased, an increasing fraction of the adsorbed strontium became
nonexchangeable with ammonium acetate.  The nonexchangeable portion of the
strontium is probably tied up by soil organics and is permanently fixed under
favorable pH conditions.  In a higher pH environment, a larger portion of the
strontium in solution was present as soluble strontium chelates or complexes.
     Varga and Jacobs (1970) examined how the strontium from dissolved (Ca,
85
  Sr) CO, moved through a mineral exchange system as represented by a ver-
                           85
miculite column.  The (Ca,   Sr) CO., was precipitated on the top of the column
                                                  flC    QC
by mixing 50 ml of 0.4M NaOH, 50 ml of 0.7M NaCl, *°Ca, °°Sr, 1  ml 1M NaHC03

                                    3-196

-------
and 1  ml 1M CaCNO^Jot and allowing the mixture to stand 1  hr. '  An HC1  solution
was then used to leach the strontium-contaminated calcite  precipitated on top
of the column.  The behavior of the released calcium (traced with   Ca) and
strontium were very similar.  Less than 1% of the   Ca and   Sr on top of the
column appeared in the effluent prior to acid addition, and no  significant
activity was found in the effluent until breakthrough of the acid front
occurred.  The release of calcium and strontium from the system was directly
related to the amount of acid added in the leaching process.
Summary
     Most of the strontium chemical compounds for which data are available are
very soluble (Figure 3-22).  Only in highly alkaline soils could SrCCL control
strontium activity in solutions.  Strontium in solution is expected to be
                           2+
predominantly present as Sr   (Figure 3-23) and to be exchangeable on  soils.
Laboratory studies show that ion exchange is the principal mechanism of
strontium adsorption by soils and rocks (McHenry, 1955, 1958; Rhodes and
Nelson, 1957; Klechovsky et a!., 1958; Prout, 1958, 1959;  Baker and Beetem,
1961; Cohen and Gailledreau, 1961).  Strontium Kd values are usually directly
correlated with cation exchange capacity of the soil or rock, but not
Invariably (McHenry, 1958).  A direct correlation between  solution pH'and
strontium Kd has also been reported (Rhodes, 1957; Prout,  1958, 1959;
Baetsle et al., 1964; Juo and Barber, 1970), which suggests hydrogen ion com-
                2+
petition with Sr   for exchange sites.  Strontium Kd values fall rapidly from
100 to 200 ml/g or more in low ionic strength salt solutions to less than
5 ml/g in high ionic strength salt solutions (Rhodes, 1957; Seme, 1973).
Strontium and calcium most readily replace trace strontium (Kokotov et al.,
1961, 1962; Schulz, 1965) and cause strontium radioisotopes to  migrate rapidly
when present in groundwater with the strontium (Spitsyn et al., 1960).  Low
pH also results in a relatively high strontium migration velocity in ground-
water (Parsons, 1963; Baetsle et al., 1964) with the strontium  and ground-
water velocity identical at pH 3.  Calcium competition is  the probable cause
of the 90% soluble strontium migration from the lagoons of the  Nuclear Fuel
Services plant (Magno et al., 1970), although strontium also migrates  on soil
particulates (Carlile and Hajek, 1967).  The many strontium Kd  values  deter-
mined by Berak (1963) demonstrated that secondary minerals such as clays and

                                    3-197

-------
zeolites are much better strontium adsorbers and are more selective for stron-

tium from solutions containing strontium and calcium than are the primary
minerals such as quartz, feldspars and pyroxenes.  Another mechanism for

strontium removal from solution is its coprecipitation with calcite (Jenne

and Wahlberg, 1968; Varga and Jacobs, 1970) or adsorption during phosphate or

oxalate replacement of carbonate in calcite (McHenry, 1958; Ames, 1958).
Strontium organic complexes are known that result in a measure of strontium

fixation (Mortensen and Marcusiu, 1963) or enhanced migration rates (Himes

and Shufeldt, 1970).

References

Ahrens, L. H.  1952.  The Use of lonization Potentials.  Part I.  Ionic Radii
of the Elements. Geochim. et Cosmochim. Acta.   2:155.

Ames, L. L., Jr.  1959.  The Genesis of Carbonate Apatites, Economic Geology.
54:829-841.

Ames, L. L., Jr., J. R. McHenry, and J. F. Honstead.  1958.  The Removal of
Strontium from Wastes by a Calcite - Phosphate Mechanism.  IN:  Proc.  2nd UN
Int. Conf. PUAE. 18:76.                                    "~

Amphlett, C. 8. and L. A. McDonald.  1956.  Equilibrium Studies on Natural
Ion-Exchange Minerals-. I:Caesium and Strontium.  J. Inorg.  Nucl. Chem.
2:403-414.

Angelo, C. G., N. C. Goldberg, V. J. Janzer, and W. A. Beetem.  1962.   Summary
of Kd's for Fission Products Between Ground Water and Basaltic Rocks.   IN:
Geology of the U18a Site, Buckboard Mesa, NTS.  Part VI.  U.S.G.S. Report
Project Danny Boy, WT 1828, pp. 66-81.

Baetsle, L. and P. Dejonghe.  1962.  Investigations on the Movement of  Radio-
active Substances in the Ground.  Part III:  Practical Aspects of the Program
and Physicochemical Considerations.  IN:  Proc. Conf. Ground Disposal of Radio-
active Wastes, 2nd, Chalk River, CanadT, September 1961.   J. M. Morgan,
D. K. Jamison, and J. D. Stevenson (eds.)  TID-7628, pp.  198-210.

Baetsle, L., P. Dejonghe, W. Maes, and P. Staner.  1963.   Present Status of
the Study Program on the Movement of Radioelements in the Soil at Mol.
EURAEC-416.

Baetsle, L. H., P. Dejonghe, W. Maes, E. S. Simpson, J. Souffriau, and
P. Staner.  1964.  Underground Radionuclide Movement.  EURAEC-703.

Baker, J. H. and W. A. Beetem.  1961.  Distribution Coefficients for Adsorp-
tion of Carrier-Free Cesium, Strontium, and Iodine by Sample from the Vicinity
of Cape Thompson, NW Alaska, U.S.G.S. Tech Letter Chariot-1.
                                    3-198

-------
Beetem, W. A., V. J. Janzer, C. G.  Angelo, and M.  C.  Goldberg.   1962.   Summary
of Distribution Coefficients for Fission Products  Between Ground Water and
Granite Rocks, Climax Stock, NTS.  U.S.G.S. Technical  Letter NTS-13.

Benson, D. W.  1960.  Review of Soil  Chemistry Research at Hanford.   HW-67201.

Berak, L.  1963.  The Sorption of Microstrontium and  Microcesium on  the
Silicate Minerals and Rocks.  UJV-528-63.

Bovard, P. and A. Grauby.  1960.  A Study of the Aptitude of Soils Under
Natural Conditions to Retain Radiostrontium.  AEC-tr-5453.

Brendakov, V. F., A. V. Dibtseva, V.  I. Svishcheva, and V. N.  Churkin.  1969.
Vertical Distribution and Evaluation  of the Mobility  of Fission Products in
Certain Soil Types of the Soviet Union.  AEC-tr-7030,  pp. 152-170.

Brookins, D. G.  1976.  Shale as a  Repository for  Radioactive  Waste:   The
Evidence from Oklo. Environmental Geology.  1:255-259.

Brown, D. J.  1967.  Migration Characteristics of  Radionuclides Through
Sediments Underlying the Hanford Reservation.  ISO-SA-32.

Carlile, B. L. and B. F. Hajek.  1967.   Soil Radionuclide Adsorption  and
Participate Filtration in an N-Area Soil.  BNWL-CC-995.

Carrigan, P. H., Jr., R. J. Pickering,  T. Tamura,  and  R.  Forbes.   1967.
Radioactive Materials in Bottom Sediment of Clinch River:  Part A, Investiga-
tions of Radionuclides in Upper Portion of Sediment.   ORNL-3721,  Suppl.  2A.
                                                •
Cerrai, E., M. G. Mezzadri and C. Triulzi.  1969.   Sorption Experiments of
Strontium, Cesium, Promethium, and  Europium on Marine  Sediment Samples.
Energia Nucleare.  16:378-385.

Cohen, P. and C. Gailledreau.  1961.   Preliminary  Investigations  of  the Adsorp-
tion of Radiostrontium in Saclay Soil.   Three-Component System.  TID-7621,
pp. 90-99.

Dlouhy, Z.  1967.  Movement of Radionuclides in the Aerated Zone.  IN:  Dis-
posal of Radioactive Wastes into the  Ground.  IAEA-SM-93/18, pp.  24T^"249.

Essington, E. H. and W. E. Nork.  1969.  Radionuclide  Contamination  Evalua-
tion - Mi Trow Event.  NVO-1229-117.

Evans, E. J.  1958.  Chemical Investigations of the Movement of Fission
Products in Soil.  CRER-792.

Fuller, W. H. and M. F. L'Annunziata.   1969.  Movement of Algal-  and  Fungal-
Bound Radiostrontium as Chelate Complexes in a Calcareous Soil.  Soil  Science.
107:223-30.

Gardner, K. and 0. Skulberg.  1964.  Sorption Phenomena of Radionuclides to
Clay Particles in River Water.  Int.  J. Air Wat. Poll.   8:229-241.
                                   3-199

-------
Goldberg, M. C., V. J. Janzer, C. G. Angelo, and W. A. Beetem.  1962.  The
Effect of Sodium Ion Concentration on Distribution Coefficients for Tuffs
from NTS.  U.S.G.S. Technical Letter NTS-16.

Hajek, B. F. and L. L. Ames, Jr.  1968.   Trace Strontium and Cesium Equilibrium
Distribution Coefficients:  Batch and Column Determinations.  BNWL-SA-843.

Hawkins, D. B. and H.  L. Short.  1965.  Equations for the Sorption of Cesium
and Strontium on Soil  and Clinoptilolite.  IDO-12046.

Himes, F. L. and R. Shufeldt.  1970.  Influence of Some Organic Compounds on
the Movement of Sr90 in Soils.  COO-414-11.

Izrael, Yu.A. and F.Ya. Rovinskii.  1970.  Hydrological Uses of Isotopes Pro-
duced in Underground Nuclear Explosions  for Peaceful  Purposes.  UCRL-Trans-10458.

Jackson, D. D.  1976.   Radiochemical Studies for the Nuclear Chemical Mining
of Copper.  UCRL-52025.

Jacobs, D. G.  1960.  Strontium Movement in the Soil.  Nuclear Safety.  2:63-65.

Jacobs, D. G.  1963.  Movement of Radionuclides Through the Ground.  Nuclear
Safety.  5:109-115.

Janzer, V. J., M. C. Goldberg, C. G. Angelo and W. A. Beetem.   1962.   Summary
of Distribution Coefficients for Fission Products Between Ground Water and
Rocks from Project Gnome.  IN:  Hydrologic and Geologic Studies for Project
Gnome.  Part IV.  U.S.G.S.          .

Jenne, E. A, and J. S. Wahlberg.  1968.   Role of Certain Stream - Sediment
Components in Radioion Sorption.  U.S.G.S. Professional Paper 433-F.

Juo, A. S. R. and S. A. Barber.  1970.  The Retention of Strontium by Soils
as Influenced by pH, Organic Matter and  Saturated Cations.  Soil Science.
109:143-148.

Kaufman, W. J..  1963.   An Appraisal of the Distribution Coefficient for
Estimating Underground Movement of Radioisotopes.  HNS-1229-21.

Klechkovsky, V. M., L. N. Sokdova, and G. N. Tselishcheva.  1958.   Sorption of
Microquantities of Strontium and Cesium  in Soil.  IN:  Proc. UN Inter. Conf.
Peaceful Uses of Atomic Energy, 2nd, Geneva, SeptemFer 1-13, 1958.  18:486-493.

Kokotov, Yu.A. and. R.  F. Popova.  1962.   Sorption of Long-Lived Fission
Products by Soils and Argillaceous Minerals III:  Selectivity of Soils and
Clays Toward 90$r Under Various Conditions.  Sov. Radiochem.  4:292-297.

Kokotov, Yu.A., R.  F.  Popova, and A. P.  Urbany.  1961.   Sorption of Long-Lived
Fission Products on Soils and Clay Minerals.  Radiokhimya.  3:199-206.

Lomenick, T. F., D. G. Jacobs and E. G.  Struxness.  1967.  The Behavior of
Strontium-90 and Cesium-137 in Seepage Pits at ORNL.   Health Physics.
13:897-905.
                                    3-200

-------
Magno, P., T. Reavey, and J. Apidianakis.  1970.  Liquid Waste Effluents from
a Nuclear Fuel Reprocessing Plant.  BRH-NERHL-70-2.

Mather, J. R. and J. K. Nakamura.  1961.  The Climatic and Hydrologic Factors
Affecting the Redistribution of Sr90.   AFOSR-1623.

Mawson, C. A.  1956.  Waste Disposal into the Ground.  Proc. of the Inter.
Conf. on the Peaceful Uses of Atomic Energy.   9:676-678.

McHenry, J. R.  1955.  Adsorption and Retention of Strontium by Soils of the
Hanford Project.  HW-34499.

McHenry, J. R.  1958.  Ion Exchange Properties of Strontium in a Calcareous
Soil.  Soil Science Society America, Proceedings.  22:514-518.

Mercer, B. W.  1966.  Adsorption of Trace Ions from Intermediate Level  Radio-
active Wastes by Ion Exchange.  BNWL-180.

Mercer, B. W.  1967.  Ion Exchange Equilibria of Trace Cesium 137 and
Strontium-85 in Multicomponent Systems.  BNWL-SA-1173.

Merritt, W. F.  1967.  Permanent Disposal by Burial of Highly Radioactive
Wastes Incorporated Into Glass.  IN:  Disposal of Radioactive Wastes Into the
Ground.  IAEA, Vienna, pp. 403-408.

Merritt, W. F.  1976.  The Leaching of Radioactivity from Highly Radioactive
Glass Blocks Buried Below the Water Table:  Fifteen Years of Results.
AECL-5317.

Miller, J. R. and R. F. Reitemeier.  1957.  Rate of Leaching of Radiostrontium
Through Soils by Simulated Rain and Irrigation Water.  U.S. Dept. Agr.   ARS
Research Rept. No. 300.

Miller, J. R. and R. F. Reitemeier.  1963.  The Leaching of Radiostrontium and
Radiocesium Through Soils.  Soil Sci.  Soc. Am. Proc.  27:141-144.

Mortensen, J. L. and E. C. Marcusiu.  1963.  Fission Product-Soil Organic
Matter Complexes:  I. Strontium-90 (Yttrium-90) in Water Extracts and HC1
Hydrolysates of Soil.  Soil Sci. Soc.  Amer. Proc.  27:653-656.

Nishita, H., B. W. Kowalewsky, A. J. Steen, and K. H. Larson.  1956.  Fixation
and Extractability of Fission Products Contaminating Various Soils and  Clays.
Soil Science.  81:317-326.-

Nishita, H. and E. H. Essington.  1967.  Effect of Chelating Agents on  the
Movement of Fission Products in Soils.   Soil  Science.  103:169-176.

Nork, W. E. and P. R. Fenske.  1970.  Radioactivity in Water - Project
Rulison.  NVO-1229-131.

Nork, W. E., E. J. Forslow, and E. H.  Essifigton.  1971.  Radioactivity  in
Water, Central Nevada Test Area.  NVO-1229-175.
                                     3-201

-------
 Orcutt,  R.  G.,  I.  Naor, G.  Klein and W. J.  Kaufman.  1956.  Hydraulic and
 Ion-Exchange  Phenomena  in the Underground Movement of Radiostrontium.
 TID-7517 (Pt. 1),  pp. 191-211.

 Orcutt,  R.  G.,  M.  N. E. Rifai,  G.  Klein, and U. J. Kaufman.  1957.  Under-
 ground Movement of Radioactive  Wastes  Presented at Nuclear Engineering and
 Science  Congress;  Philadelphia,. Pennsylvania, March  11,  1957.

 Parker,  V.  B.,  D.  D. Wagman, and W. H.  Evans.  1971.  Selected Values of Chemi-
 cal Thermodynamic  Properties.   Tables  for the Alkaline Earth Elements (Elements
 92  through  97)  in  the Standard  Order of Arrangement.  NBS Technical Note 270-6.

 Parsons, P. J.   1961.   Movement of Radioactive Waste Through Soil.  3.
 Investigating the  Migration of  Fission Products from High-Ionic  Liquids
 Deposited in  Soil.  CRER-1018.

 Parsons, P. J.   1963.   Migration from  a Disposal of  Radioactive  Liquid in
 Sands.   Health  Physics.  9:333-342.

 Polyakov, Yu. A.,  V. F. Gol'tsov,  and  V. G. Grakovskiy.  1969.   The Diffusion
 of  Strontium-90 in Soils.   AEC-tr-7030, pp. 14-25.

 Prokhorov,  V. M.   1962.  Diffusion of  Strontium-90 in Soil and Sand.
 Radiokhimiya.   4:205-211.

 Prout, W. E.  1958.  Adsorption of Radioactive Wastes by Savannah River Plant
 Soil.  Soil Science.  86:13-17.

•Prout, W. E.  1959.  Adsorption of Fission  Products  by Savannah  River Plant
 Soil.  DP-394.

 Rhodes,  D.  W.   1957.  The Effect of pH on the Uptake of  Radioactive Isotopes
 from Solution by a  Soil.  Soil  Science Society of America, Proceedings-.
 21:389-392.

 Rhodes,  D.  W. and  J. L. Nelson.  1957.  Disposal of  Radioactive  Liquid Wastes
 from the Uranium Recovery Plant.   HW-54721.

 Routson, R. C.   1973.  A Review of Studies  on Soil-Waste Relationships on the
 Hanford  Reservation from 1944 to 1967.  BNWL-1464.

 Schmalz, B. L.   1972.   Radionuclide Distribution in  Soil Mantle  of the
 Lithosphere as  a Consequence of Waste  Disposal at the National Reactor Test-
 ing Station.  IDO-10049.                    %         '   .

 Schulz,  R.  K.   1965.  Soil  Chemistry of Radionuclides.   Health Physics.
 11:1317-1324.

 Si 11 en,  L.  G. and  A. E. Martell.   1964.  Stability Constants of  Metal-Ion
 Complexes.  Special Publication No. 17.  The Chemical Society, London.

 Sokol, D.  1970.   Groundwater Safety Evaluation - Project Gasbuggy.  PNE-1009.



                                    3-202

-------
Sorathesn, A., G. Bruscia, T. Tamura, and E. G. Struxness.  1960.  Mineral
and Sediment Affinity for Radionuclides.  CF-60-6-93.

Spitsyn, V. I., V. D. Balukova, A. F. Naumova, V. V. Gromov, F. M. Spiridonov,
E. M. Vetrov, and G. I. Gravfov.  1958.  A Study of the Migration of Radio-
elements in Soils.  IN:  Proc. Second Annual U.N. Conf. Peaceful Uses of
Atomic Energy.  18:439"-448.

Spitsyn, V. I., V. D. Balukova, V. V. Gromov, S. I. Zakharov, B. P. Zhagin
and F. M. Spiridonov.  1960.  Sorption.Regularities in Behavior of Fission-
Product Elements During Filtration of Their Solutions Through Ground.  IN;
Disposal of Radioactive Wastes.  11:429-435.

Spitsyn, V. I. and V. V.  Gromov.  1958.  A Study of the Systematic Adsorption
of Radioactive Strontium by Montmorillonite and Its Fixation by Roasting.
Soviet Atomic Energy.  5:1341-1347.

Stead, F. W.  1963.  Tritium Distribution in Ground Water Around Large Under-
ground Fusion Explosions.   Science.   142:1163-1165.

Stead, F. W.  1964.  Distribution in Groundwater of Radionuclides from Under-
ground Nuclear Explosions.  IN:  Proc.  3rd Plowshare Symp. TID-7695.

Tamura, T.  1964.  Reactions of Cesium-137 and Strontium-90 with Soil  Minerals
and Sesquioxides.  ORNL-P-438.

Tamura, T. and_ E. G. Struxness.  1963.   Reactions Affecting Strontium Removal
from Radioactive Wastes":  Health Physics.  9:697-704.

Thornthwaite, C. W., J. R. Mather, J. K. Nakamura.   1960.   Movement of Radio-
strontium in Soils.  Science.  131:1015-1019.
                 .\

Turekian, K. K. and K.  H.  Wedepohl.   1961.  Distribution of the Elements  in
Some Major Units of the Earth's Crust.   Bull.  Geol. Soc. Am.  72:175.

Varga, J. A. and D. G.  Jacobs.  1970.  Dissolution of Calcium-Strontium
Carbonate and Its Subsequent Movement in the Ground.  ORNL-TM-3130.

Vinogradoy, A. P.  1959.   The Geochemistry of Rare and Dispersed Chemical
Elements in Soils.   Consultants Bureau, Inc.

Wahlberg, J. S., J. H.  Baker, R. W.  Vernon, and R.  S. Dewar.  1965.   Exchange
Adsorption of Strontium on Clay Minerals.  U.S.G.S. Bulletin 1140-C.

Wahlberg, J. S. and R.  S.  Dewar.  1965.  Comparison of Distribution  Coefficients
for Strontium Exchange from Solutions Containing One and Two Competing Cations.
U.S.G.S.  Bulletin 1140-D.

Wilding, M. W. and D. W.  Rhodes.  1963.  Removal of Radioisotopes from Solution
by Earth Materials from Eastern Idaho.  • IDO-14624.
                                   3-203

-------
 TECHNETIUM
 Natural  Soil  and  Rock  Distributions
      According  to Boyd and  Larson  (1956),  technetium occurs  in most abundance
 in  molybdenite  (MoS2), a  common molybdenum ore, as  shown  in  Table 3-87.  The
 technetium content of  all materials  except pitchblende  is of secondary origin,
 formed by the action of cosmic radiation on molybdenum, ruthenium or niobium.
     99                                 238
 The   Tc in pitchblende is  a  product of    I)  spontaneous  fissioning.  Not
                                               99
 everyone accepts  the existence of  the  natural   Tc  as having been proven (see
 Handbook of Chemistry  and Physics, 57th Edition,  p. B-50).

            TABLE  3-87.  TECHNETIUM CONTENT OF  SOME  NATURAL MATERIALS
                        (BOYD AND  LARSON,  1956)
 	Material	  	Tc, g/kg	  Determination Method
 Molybdenite,  Climax, CO       <4  x  10~1                  Spectroscopic
 Molybdenite,  NV              <1.5  x  10 7            ,,     Spectroscopic
 Molydenite, Miami,  AZ         <1.8  x  10" >  8.3  x 10"  ,    Activation
                              1.3  x  NT 1
 Molybdenite,  NV              1.3  x  10_g                  Activation
 Osmium-iridium  concentrate    <1.6  x  10~?,                 Spectroscopic
 Yttrotantalite, West Africa   <6  x  lO'jl                 Mass Spec.
 Iron-nickel Meteorite         <4.5  x  10~g                  Spectroscopic
 KReOa, pure          .  '   '   <8  x  10~,n  QQ              Spectroscopic
 Pitchblende (--50% uranium)    2.5  x  10"IU  *yTc (Ashizawa  and Kuroda, 1957)

 Brief Chemistry.
      There are  16 isotopes  and 6 isomers of technetium  presently known
 (Lavrukhina and Pozdnyakov, 1970).   Several technetium  isotopes are obtained
 by  the fissioning of nuclear  fuels of  uranium  in  the reactor.  The fission
                                                        99           101
 product technetium isotopes and their  yields  include:    Tc, 6.06%;    Tc,
 5.6%; 102Tc,  4.'3%;  103Tc, 3.0%; 104Tc, 1.8%;  105Tc, 0.9%; 107Tc, 0.19%
 (Katcoff, 1958).   According to Boyd  (1959), the technetium content of first
                                     QQ                               C
•cycle Purex waste is 41 mg/1.  Only    Tc is long  half-lived  (2.12 x 10  years)
 of  the several  fission product technitium  isotopes, and hence a potential
 hazard.
      Technetium exists in valence  states from  (VII) to  (-1).  The most stable
 state in aqueous  solutions  is heptavalent, usually  corresponding to the TcOT
 (pertechnetate) ion, which  other valences  encountered chiefly in complex

                                     3-204

-------
                                  +7         °
compounds.  The ionic radius of Tc   is 0.56 A, while its geochemical rela-
        +7          °
tive, Re  , is 0.46 A.  The corresponding radius of the solvated TcOd ion is
    o                                                               ^
4.6 A (Smith et al., 1953).  In weakly acid, neutral and alkaline solutions,
Tc(IV), the second most stable state of technetium, is oxidized by atmospheric
oxygen to Tc(VII).  The solubility of KTc04 in water at 20°C is 21.3 g/1
(Busey and Larson, 1958).  Technetium (IV) and Tc(V) form halogen complexes
                2
such as (TcClg)   , and Tc(II), (III) and (V) form diarsine complexes and
cyanide complexes.
Solid Phase and Solution Equilibria
     Baes and Mesmer  (1976) reported that technetium (VII) forms strong peracids
(HMOA), and that  its oxides are very soluble.  No information is available con-
cerning the soluble complexes of technetium or the solubility of its salts
(Pourbaix, 1966).  Pertechnetate ion, TcOT, is the most stable species of the
element in aqueous solutions (Boyd, 1959).
Experimental Adsorption Results
     Wildung et al. (1975) used a batch equilibrium technique to determine
pertechnetate ion adsorption on soils.  The range of the 22 soil type proper-
ties used in the  study are given in Table 3-88.  The Kd values for pertechneta-te
ranged from 0.007 to 2.8.  The Kd values were positively correlated with soil
organic carbon content and negatively correlated with soil pH.  The positive
charge on soil organic colloids is probably an important factor in technetium
adsorption by soils.  The soil carbon-pertechnetate ion removal correlation
was the only one  significant at the 1% confidence level.  Cation exchange
capacity-pertechnetate ion removal also was positively correlated but at a 5%
confidence level.  Negatively correlated at the 5% confidence level was pH-
pertechnetate ion removal, probably because as pH decreases, positive charge
increases with decreased ionization of acidic groups of soil organic colloids
and increased protonation of basic groups.  A regression equation was given
for pertechnetate distribution coefficient determination within the range of
significant independent variables.  This was Y-j = 0.08X3 - 0.09X4, where Y-j is
the pertechnetate distribution coefficient (based on the ranges shown in
Table 3-88), X3 is the soil organic carbon content and X, is the soil-solution
pH.  The soil silt content (X,) and clay content (Xg) correlations with Kd

                                    3-205

-------
 were  not  statistically significant at the 5% confidence level and did not,
 therefore, appear  in  the regression equation.

TABLE 3-88.   RANGE OF PROPERTIES OF 22 SURFACE SOILS USED  IN THE PERTECHNETATE
             ADSORPTION STUDIES (WILDUNG ET AL., 1975)

  Cation Exchange             	wt%	
Capacity, meg/100 g     pH     C03   Organic C     Sand       Silt       Clay
     5.5-90.0        3.6-8.9  0-6.5  0.23-28.8  14.1-73.1  17.6-58.0  3.8-46.6
     Gast  (1975) studied several Minnesota soils and their abilities to adsorb
 pertechnetate ions.  The soil characteristics are outlined in Table 3-89.
 From 2 to  5 weeks of contact time between soil and solution were required for
 equilibrium to be established.  Nicollet C and Zimmerman soils, both low in
                                             99
 organic carbon content, adsorbed very little   Tc, while up to 98% of the
 technetium was adsorbed from solution by the remaining soils corresponding to
  99
 a   Tc Kd value of 857.5 ml/g.  When the organic carbon was removed by hydrogen
 peroxide treatment from two soils originally high in carbon content, the adsorp-
 tion of technetium was greatly reduced.  Gast also mentioned that anaerobic
 conditions .prevailed in the soil-solution environment which may have led to
 the precipitation of technetium heptasulfide, rather than technetium,removal
 by adsorption processes.

            TABLE 3-89.  PROPERTIES OF SOILS USED IN THE TECHNETIUM
                         ADSORPTION STUDIES (GAST, 1975)
Soil
Bearden
Hegne
Ribbing
Nicollet A
Nicoll«t C
Omega
Bergland
Arveson
Uaukegan
Zimmerman
Peat

Sand
9.1
1.9
3.3
24.4
23.9
61.5
14.0
47.1
10.0
69.4
—
wt%
Silt
63.7
37.6
79.5
42.9
48.6
30.7
25.1
24.6
69.2
24.9
~

Clay
27.2
60.5
17.2
29.7
27.5
7.9
60.9
28.3
20.7
5.7
--
pH,
7
8
5
5
8
7
6
7
6
5
7
H?0
.68
.00
.48
.91
.40
.87
.35
.70
.25
.74
.83
Organic
C, wt%
5.39
2.20
2.29
2.39
0.12
1.26
5.67
2.80
2.36
0.80
45.95
wt% % Fe203
16
12


15
1

15


7
.29
.79
—
--
.21
.21
—
.58
—
—
.58
0
0
1
0
1
1
2
0
1
0
0
.09
.29
.40
.86
.35
.20
.39
.21
.03
.47
.40
C.E.C.,
meg/ 100 g
16.9
36,1
11.3
19.3
15.7
6.0
32.3
14.9
15.1
2.7
. 50.9
3-206

-------
     Routson et al.  (1976, 1977) determined distribution coefficients on tech-
netium with Washington and South Carolina soils.  Soil  properties are given in
Table 3-90.  The organic carbon content was not given.   The technetium Kd
values obtained were very low, as seen in Table 3-91, when competing bicarbonate
ions were present in the system.  All Kd values were essentially zero.
                   TABLE 3-90.  PROPERTIES OF SOIL SAMPLES
                                (ROUTSON ET AL., 1977)
                                       wt%
Soil
Washington
South Carolina
CaCO-3, mq/q
0.8
<0.2
Silt
10.1
3.6
Clay
0.5
37.2
CEC, meq/100 q
4.9
2.5
pH, H?0
7.0
5.1
       TABLE 3-91.
TECHNETIUM Kd AS A FUNCTION OF NaHCOs CONCENTRATION
FOR A SOUTH CAROLINA SOIL CHARACTERIZED IN TABLE 3-90
(ROUTSON ET AL., 1977)
                    NaHCOs, M
                      0.002
                      0.008
                      0.020
                      0.200
                   Average Kd,  ml/q
                      0.019 ± 0.06
                      0.0   ± 0.01
                    '  0.0   ± 0.01
                      0.010 ± 0.04
Migration Results
     Field Studies-
     Brown (1967).traced the movement of technetium discharged to the ground
near the chemical separations plant at Hanford via an extensive network of
monitoring wells.  Based on analytical results, a map of the contamination
pattern at the surface of the groundwater was given for 1966 at limits defined
                99     3
by the 0.01 pCi   Tc/cm  isoconcentration contour.  The contamination pattern
was very similar in shape and size to that for tritium.  Technetium-99 was
reported to be moving at essentially the same rate as the groundwater, and was
detectable in the groundwater at below public drinking water limits all the
way to the Columbia River.
                                    3-207

-------
     Brookins  (1976)  pointed  out that the technetium formed as a result of
nuclear fission of the deposit of uranium at Oklo, Gabon, had at least par-
tially migrated from  the  shale location in which it was formed.
     Laboratory Studies—
     Gast  (1975) reported the extractability of technetium from Bergland and
Arveson soils.  The characteristics of these soils are given in Table 3-92.
Two-gram samples of air-dried Bergland and Arveson were loaded with 0.06 uCi
   QQ                          QQ
of   Tc.   Less than 2% of the   Tc still  remained in the equilibrating solu-
                                                                        QQ
tion after 1 month of contact.  These 2-g samples loaded with 0.06 uCi   Tc
were used  in the extraction studies.   Twenty-five milliliters of extractant
(IN NaOH or IN HCIO^) were added to the 2-g samples in sequence and shaken for
24 hr with the results shown  in Table 3-93.
         TABLE 3-92.   PROPERTIES  OF BERGLAND AND ARVESON SOILS USED
                       IN THE  99fc EXTRACTION WORK (GAST, 1975)
                       wt%
Bergland
Arveson
Sand
14.0-
47.1
Silt
25,1 _
24.6
Clay
60.9
28.3
Organic I
Carbon, %
5.67
2.80
" %
-
15.58
pH,
HoO
6.35
7.70
CEC,
meq/lOOg
32,3 .
14.9
           TABLE 3-93.   EXTRACTABILITY OF   Tc FROM BERGLAND AND
                         ARVESON  SOILS (GAST, 1975)
                    Soil   Extractant  Extraction   99Tc. «    Total 99Te.i
                   Bergland    NaOH
                   Arveson    NaOH
                   Bergland    HC104
                   Arveson
1
2
3
4
1
Z
3
4
1
2
3
4
1
2
3
4
77.4*3.25
8.69*1.97
1.82±0.16
0.83+0.10
83. 68* 1.05
4.77t0.07
2.39*0.11
0.98±0.07
11.95±0.14
6.90*0.31
4.46t0.07
3.52±0.01
7.02*0.01
2.96±0.04
2.16±0.04
1.50±0.01

88.74*1.02



91.82±0.98



26.83±0.60



13.63±0.02


                                     3-208

-------
     The concomitant removal of soil organic matter suggested that the
ion may be associated with it.  However, Gast showed that large amounts of
                            p                                          qq
chloride ions (KC1) or HP04   from K2HP04 had no significant effect on   Tc
adsorption by Bergland and Arveson soils so that simple anion exchange prob-
                                                  qq
ably was not involved in the adsorption.  The low   Tc adsorption shown by
low organic soils, and the effects of prior hydrogen peroxide removal  of
                                          99
Bergland organic material on lowering the   Tc adsorption capacity, strongly
suggest that the technetium is adsorbed on the soil organics.  The presence
of FLS reported from certain of the adsorption experiments suggests the reduc-
tion of Tc(VII) to Tc(IV) and precipitation of TCgSy as well.  The NaOH removes
both adsorbed and precipitated technetium from the soil, while the perchloric
acid is partially consumed by reactions with the organic fraction and  is
therefore not as an effective technetium removal agent as NaOH.
Summary
     Pertechnetate ion, TcOT, is the most stable species of technetium in
aqueous solutions (Boyd, 1959).  A major portion of the ion exchange capacity
of soils and sediments is cation exchange capacity at the usual near-neutral
pH conditions.  Therefore a negatively charged TcOT would hardly be exchanged,
and hence show littTe adsorption by soils and rocks (Routson et al., 1976,
1977).  Gast (1975) reported high Kd values (800 to 900) for soils with high
organic matter.  He also reported the presence of H2S in his equilibrating
samples, and suggested that the high Kd values were due to precipitation of
technetium as Tc^Sy in the presence of H^S (Kotegov, 1968).  In conditions
where the soil contains appreciable organic matter, the Tc(VII) may be reduced
to Tc(IV) and adsorbed (Wildung et al., 1975).  Tc(IV) can coprecipitate with
ferric hydroxide (Anders, 1960).  With the exception of some sedimentary
rocks, most rocks contain very little organic matter.  Hence, technetium Kd
values of close to zero would normally be encountered in oxidizing conditions
and in rocks and soils that are relatively low in organic matter.  More study
is required to determine the nature of the organic matter-pertechnetate ion
reaction, and the influence of redox conditions on adsorption of technetium
by soils and rocks.
                                    3-209

-------
References

Anders, E.  1960.  The Radiochenristry of Technetium.   Nuclear Sci.  Ser.  Nat.
Acad. Sci. - Nat. Res. Council.  NAS-NS 3021.
Ashizawa, F. T. and P. K. Kuroda.  1957.  The Occurrence of the Short-Lived
Iodine Isotopes in Natural and in Depleted Uranium Salts.  J. Inorg.  Nucl.
Chem.  5:12-22.
Baes, C. F., Jr. and R. F. Mesmer.  1976.  The Hydrolysis of Cations.  John
Wiley and Sons, New York.
Boyd, G. E.  1959.  Technetium and Promethium.  J. Chem. Educ.  36:3-14.

Boyd, G. E. and Q. V. Larson.  1956.  Report on the Occurrence of Technetium
on the Earth's Crust.  J. Phys. Chem.  60:707-715.
Brookins, D. G.  1976.  Shale as a Repository for Radioactive Waste:   The
Evidence from Oklo. Environmental Geology.  1:255-259.
Brown, D. J.  1967.  Migration Characteristics of Radionuclides Through Sedi-
ments Underlying the Hanford Reservation.  ISO-SA-32.
Busey, R. H. and Q. V. Larson.  1958,  Chemistry Division Annual  Progress
Report for Period Ending June 20, 1958.  ORNL-2584.
Gast, R. G.  1975.  The Behavior of Technetium-99 in  Soils and Plants.
Progress Report.  COO-2447-1.
Katcoff, S.  1958.  Fission-Product Yield from U, Th,  and Pu.  Nucleonics.
16:78-86.
Kotegov, K. V., 0. N. Pavlov and V. P. Shvedov.  1968.   Technetium.   IN^;
Advances in Inorganic Chemistry and Radiochemistry, vol.'IT, pp.  1-90,
Academic Press, NY.
Lavrukhina, A. K. and A. A. Pozdnyakov.  1970.  Analytical Chemistry  of
Technetium, Promethium, Astatine, and Francium.  Translated by R.  Kondor.
Ann Arbor-Humphrey Science Publishers.
Pourbaix, M.  1966.  Atlas of Electrochemical  Equilibria in Aqueous Solutions.
Pergamon Press, Oxford, England.
                                                      99    237        241
Routson, R. C., G. Jansen and A. V. Robinson.   1976.     Tc,    Np,  and    Am
Sorption on Two Subsoils from Differing Weathering Intensity Areas.   BNWL-2000.
Pt. 2, pp. 50-52.
                                                      241    237         99
Routson, R. C., G. Jansen and A. V. Robinson.   1976.      Am,    Np, and  Tc
Sorption on Two United States Subsoils from Differing  Weathering  Intensity
Areas.  BNWL-1889.

Smith, W. T., J. W. Cobble, and G. E. Boyd.  1953.  Thermodynamitf Properties  of
Technetium and Rhenium Compounds.  I. Vapor Pressures  of Technetium Heptoxide,
Pertechnic Acid and Aqueous Solution of Pertechnic Acid.  J. Am.  Chem.  Soc.
75:5773-5776.
Wildung, R. E., R. C. Routson, R. J. Seme, and T. R.  Garland.  1975.  Pertech-
netate, Iodide and Methyl Iodide Retention by Surface  Soils.  BNWL-1950.
Pt. 2, pp. 37-40.
                                     3-210

-------
THORIUM
Natural Soil and Rock Distributions
     The content of thorium in rocks and soil  is given in Table 3-94.   The
thorium in sedimentary rocks may be due either to the selective.adsorption of
thorium on clays or its retention in heavy resistate minerals such as  monazite.
Concentrations of thorium in metamorphic rocks are highly variable. The
thorium content of igneous rocks increases from basalts to granites.

           TABLE 3-94.  THORIUM CONTENT OF COMMON ROCKS AND SOILS
          Rock Type
    Granites
    Intermediate
    Basalt and Gabbros

    Shales, North America
    Bauxites
    Bentonites
    Limestones
    Sandstones
 Thorium Average
 or Range, ppm
  Igneous Rocks
      10-20
       2-10
     0.5-2
Sedimentary Rocks
      10-15
         49
         24
        1.1
        1.7
      Reference
Rogers, 1964
Heier and Carter, 1964
Heier and Rogers, 1963

Adams and Weaver, 1958
Adams and Richardson, 1960
Adams and Weaver, 1958
Adams and Weaver, 1958
Murray and Adams, 1958
    Soils

    Marble
    Slate
    Phyllite
    Schist
    Gneiss
Metamorphic Rocks
        0.03
        7.5
        5.5
        7.5
       13.1
Vinogradov, 1959

Pliler, 1956
Pliler, 1956
Pliler, 1956
Pliler, 1956
Billings, 1962
     Thorium adsorption studies were performed by several  investigators.
Holland and Kulp (1954) found red clay, globigerina ooze and green clay
readily adsorbed thorium.  They concluded that ion exchange was the adsorption
                                   3-211

-------
 mechanism.  Adams et al.  (1959) suggested that thorium is concentrated by clay
 minerals.  Up to 50 ppm thorium in the aluminum hydroxide and resistate min-
 erals in bauxite was reported by Adams and Richardson (1960).
 Brief Chemistry
      There are 13 isotopes of thorium with six of them found in nature.  Of
 the six natural thorium isotopes, five are relatively quantitatively unimpor-
 tant members of the 238U, 235U or 232Th decay series.  Thcrium-232 is the
 major isotope, with a half-life of 1.39 x 1010 years (Ryabchikov and Golbraikh,
 1969).  Thorium radionuclide data of interest in waste disposal  operations
 are in given in Table 3-95.
          v  TABLE 3-95.  THORIUM RADIONUCLIDE DATA (RYABCHIKOV
                          AND GOLBRAIKH, 1969)
Isotope
227Th
228
^bTh
229Th
230Jh
231 Th
232Th
234Th
Half -Life
18.5 days

1.913 years
7340 years
80,000 years
25.5 hours
1.41 x 1010 years
24.1 days
Decay Mode
a

a
a
a
6"
.a
e"
     Although other oxidation states of thorium are known in the laboratory,
only Th(IV) is found in nature. Th(IV) is found as Th+ .   The atomic radius
of Th+4 is 0.99 A (Ahrens, 1952).
     Common insoluble thorium compounds include the hydroxide, fluoride and
phosphate.  Soluble compounds include the chloride, nitrate and sulfate.
Thorium in solution is a small, highly charged ion that undergoes extensive
interaction with water and many anions.  The solution chemistry of thorium is
largely a-study of its complex ions.  Common anions that form strong complexes
with thorium include fluoride, chloride, nitrate, phosphate and sulfate.  At
pH values above 3, thorium undergoes hydrolysis in aqueous solutions.  During
the sedimentary cycle, thorium usually becomes separated from uranium because
.the uranium tends to mobilize in its U(VI) oxidation state until encountering
a reducing environment to become immobilized U(IV).  Thorium does not undergo
a comparable oxidation state change.
                                   '3-212

-------
Solid Phase and Solution Equilibria
                                           4+
     Figure 3-24 relates the activity of Th   to pH under an assumed weathering
environment in equilibrium with various thorium solid phases.  The thermo-
dynamic data for Th02(s) were selected from Baes and Mesmer  (1976).  The data
for the other compounds were selected from Si 11 en and Martell  (1964).  Under
the assumptions outlined in Figure 3-24, all thorium compounds except ThF4 can
be arranged in an increasing order of stability throughout the pH range as
follows:  Th(HP04)2, Th3(P04)4, Th(OH)4, and
in a pH of approximately >7 and most stable approximately pH <4.
            -4
ThF4 would be least stable
         fS
         S1
                                           (OCP + CaC03AT
                                            pCO,.3.52)
            -16-
            -18
Figure 3-24.  The relative stability of various thorium solids in equilibrium
              with Variscite and Gibbsite (V & G) Dicalcium Phosphate Dihydrate
              (OCPD) and Octacalcium Phosphate (OCP)

     The relative activity of solution species of thorium in equilibrium with
Th02(s) at assumed activities of various ions is plotted in Figure 3-25.  The
thermodynamic data for all the hydrolysis species except Th(OH)I were obtained
from Baes and Mesmer (1976).  The data for all the other species and Th(OH)"
                                    3-213

-------
were obtained from Sillen and Mar-tell (1964).  In general, the total concen-
tration of thorium in solution decreases with an- increase of pH from zero to
5.  Above 5, pH does not affect thorium concentration in solution due to the
formation of Th(OH)4<  The activity of all positively charged species decreases
with an increase in pH, while the activity of the negatively charged species
increases with the increase in pH.  Under the conditions assumed for Fig-
ure 3-25, the total activity of thorium in solution would be expected to be
approximately 10
              -6
                -9.6
moles/liter above pH 5.
           8*
                                                                10
Figure 3-25.  Activity of various thorium species in soil  in equilibrium with
              ThOgCs), pCT » pN03- = pS042" = 3.0, pF" = 4.5 and pH2P04~ =
              o. u •
                      -                                          23
     In addition to OH , thorium forms various complexes with SO*  ,  PO,   ,
Cl~, N0^~, and F".  Various anions in increasing order of their importance to
                           -    -             2
forming complexes are:  NO^ , Cl , ^^^ ,  SO^  , and F .  Figure 3-25 shows
that thorium exists as Th   only in very acidic solutions (pH < 3).   Above
        4+
pH 3, Th   hydrolyzes very .rapidly and it does not contribute significantly
                                    3-214

-------
 to  the  total  thorium  concentration.   Under  the conditions assumed  for  Fig-
 ure 3-25,  ThF *  would be  a  dominant  solution  species  at  pH  <  5,  and  Th(OH)4°
 would be a dominant solution  species  at  pH  >  5.   If fluoride  ion is  absent
                                                                 ?+
 from the solution,  or its concentration  is  extremely  low, Th(OH)2    would
 mainly  control the  thorium  concentration in solution  at  pH  <  5.

Experimental Adsorption Results
      Schulz (1965)  found the thorium in soils to be strongly adsorbed  by clay
particles or present as insoluble oxides and hydroxides.   Rubtsov (1966, 1972)
found thorium to  associate with the fine-grained  particles during soil  weather-
ing.  Katsurayama (1968) determined the distribution coefficient of thorium
but data are not  presented in the available abstract.
      Nishiwaki et al. (1972) spiked seawater and seawater distilled water mix-
tures with Th   and measured the adsorption on a  medium sand, very  fine sand
and silt-clay.  Twenty grams of soil  were contacted with 4 liters of spiked
water and mixed until  equilibrium was reached.  The Kd for thorium  increased
as the particle size of the soil decreased.   Chlorosity of the water did not
appear to consistently affect the thorium Kd for the fine sand or silt-clay.
The Kd for the medium sand increased as the ch'lorosity of the water decreased.
The chlorosity effect was compounded by a variable pH of the various salt solu-
tions, so that the exact cause of the trend was not determineble.  Kd values
for the medium sand, very fine sand and silt-clay were 40 to 130, 310 to 470,
and 2700 to 10,000 ml/g, respectively.
      Rancon (1973)  measured the thorium Kd for a soil developed on a schist
consisting of quartz and clay with no calcite or organic matter, for a  mixed
quartz-clay-calcite-organic matter soil and for illite with 100 mg  Th/1 versus
solution pH.  For the quartz-clay soil, at pH 6 the Kd was 5 x 10  ml/g, at
pH 4 the Kd was about 1 x 10  ml/g and at pH 2 the Kd was about 5 ml/g.  The
mixed quartz-clay-calcite-organic matter soil  could not be lowered  in pH with-
out removal of soil  calcite, but above pH 8, the thorium Kd dropped from
10  ml/g to 100 ml/g at pH 10.  Dissolution of humic acids in the soil  probably
resulted in thorium complexation and a decreased  Kd with rising pH.  Illite
behaved similarly to the quartz-clay soil, but the thorium Kd at pH 1 was about
500 ml/g and about 1 x 10  ml/g at pH 6.5.  For soils without calcite or organic
material, the thorium Kd decreased as the thorium concentration in  solution
                                    3-215

-------
initially increased.  Calcareous soils neutralized even high strength thorium
solutions to precipitate Th(OH)4.  The quartz-clay soil and illite, for example,
gave thorium Kd values of 8 ml/g and 120 ml/g, respectively, in a 1 g Th/1
solution, and 60 ml/g and 1000 ml/g, respectively, in a 0.1 g Th/1  solution.
The drop in thorium Kd was caused by saturation of available exchange sites as
a result of increased thorium concentration.  There was evidence for the con-
centration dependence of the thorium Kd down to 1  mg Th/1 in the initial con-
tacting solution.  In general, three types of soil-thorium adsorption reactions
were found:  1) Th(OH)4 precipitation as a result of soil calcite buffering,
2) strong adsorption on clay-containing soils and dilute thorium (<1 g/1) solu-
tions at a pH above 2, and 3) strong adsorption on organic-containing soils at
the neutral to acid pH range, but diminishing adsorption into the alkaline pH
range.
      Bondietti (1974) studied the adsorption of hydrolyzed thorium from waters
at pH 6.5 by calcium saturated reference clays (montmorillonite and kaolinite)
and calcium humate and found 95% and 99.9+% adsorption, respectively.  Desorp-
tion studies utilizing calcium citrate removed 10 to 30% of the thorium from the
clays but only 1% from the humate.  Stronger complexers (DTPA and EDTA) removed
20 to 30% of the thorium from the humate.  A mixed organic-hydroxy complex was
proposed for the reaction of thorium with humic substances.
Migration Results
      Field Studies—
                          i
      The thorium content of groundwater was reported by Dementyev and
Syromyatnikov (1965) to be highest in low salinity, low hardness, low pH, high
organic content groundwaters.  These characteristics suggest transport of
thorium as colloidal suspensions and anionic complexes involving soil acids.
      From a fresh granodiorite containing 9.3 ppm thorium and 2.5 ppm uranium,
the first stages of weathering resulted in apparent removal of 25% of the
thorium and 60% of the uranium (Hansen and-Huntington, 1969).  An acid leach
of the fresh rock removed 90% of the thorium and 60% of the uranium indicating
that most of the thorium and uranium are in acid soluble or interstitial mate-
rials.  After an initial drop in concentration, the total uranium and thorium
content of the weathered rock increases by at least a factor of 4 in the upper-
most weathered material.  Leaching studies showed that thorium was associated
with clays formed during weathering and with accessory minerals such as zircon.
                      '              3-216

-------
      Hansen (1970) reported that when freed from minerals by weathering,
thorium was leached comparatively slowly.  From a fresh granodiorite containing
9.3 ppm thorium, the first stages of weathering resulted in apparent removal of
25% of the thorium.  An acid Teach of the fresh rock removed 90% of the thorium
indicating that most of the thorium was in acid soluble or interstitial mate-
rials.  The thorium content of the weathered rock increased by a factor of 4
due mostly to association with the clays formed during weathering.
      Laboratory Studies—
      Desai and Ganguly (1970) showed humic acids from a coastal marine sedi-
ment solubilized 100% of the thorium added to an ammonia solution (2.5N).
Thorium in this solution without humic acid was observed to predominantly
precipitate (95%).   The humic acid-thorium complex was noncationic.  In an
identical experiment, fulvic acid extract was shown to solubilize 59% of
thorium added to an ammonia solution.  Again, the solubilized organic-thorium
fraction was noncationic.
Summary
      Under alkaline conditions, Th(OH). and ThO^ maintain low activities in
soil solutions (Figures 3-24 and 3^25) and these compounds could form and
govern thorium concentration.  Thorium' hydrolyzes readily even in moderately  •
acidic environments (Figure 3-25) so that Th   would be present only in very
acidic solutions.  Laboratory studies also show that thorium tends  to precipi-
tate as thorium hydroxide and hydrated thorium oxide in soils (Schulz, 1965;
Rancon, 1973).
      An increase in thorium content with increase in CaCO^, phosphate and
humus content of soils and sediments has been reported (Kuznetsov et al., 1968;
Pashneva et al., 1965; Menzel, 1968; Yakobenchuck, 1968; Hansen and Huntington,
1969; Pokidin et al., 1972).  However, Tyuryukanova and Kalugina (1971) reported
low thorium concentrations in high humus soils (peats and forest podzols) com-
pared with alluvial soils.  Thorium adsorption increases with increase in pH
(Rancon, 1973) and decrease in soil particle size (Hansen and Huntington, 1969;
Hansen, 1970; Nishiwaki et al., 1972; Rubtsov, 1966, 1972; Bondietti, 1974).
Strong humic and fulvic acid complexes with thorium occur in the neutral to
acidic range (Rancon, 1973; Bondietti, 1974) which are noncationic  (Desai and
Ganguly, 1970) and mobile.  It has been reported also that thorium  migrates
primarily in the colloidal form (polymeric) in the natural environment
(Baranov et al., 1956; Lazarev et al., 1961; Kimura et al., 1968).
                                    3-217

-------
References

Adams, J. A. S., J. K. Osmond, and J. J. W. Rogers.  1959.   The Geochemistry
of Thorium and Uranium.  IN;  Physics and'Chemistry of the Earth.   3.
L. H. Ahrens, F. Press, K. Rankama, and S. K. Runcorn (eds.), pp.  298-348.

Adams, J. A. S. and K. A. Richardson.  1960.  Thorium, Uranium, and Zirconium
Concentrations in Bauxite.  Econ. Geol. 55:1653.

Adams, J. A. S. and C. E. Weaver.  1958.  Thorium-to-Uranium Ratios as Indi-
cators of Sedimentary Processes — An Example of Geochemical Facies.  Bull.
Am. Assoc. Petrol. Geol. 42:387.

Ahrens, L. H.  1952.  The Use of lonization Potentials.   Part I.  Ionic Radii
of the Elements.  Geochim. et Cosmochim. Acta. 2:155. -

Baes, C. F., Jr. and R. F. Mesmer.  1976.  The Hydrolysis of Cations.   John
Wiley and Sons, New York.

Baranov, V. I., A. B.  Ronoy, and K. G. Kinashova.   1956.   On Geochemistry of
Dispersed Thorium and Uranium in Clays and Carbonate Rocks  of Russian  Platform.
Geokhimiya No. 3:3-8.

Billings, G. K.  1962.  A Geochemical Investigation of the  Valley  Spring Gneiss
and Packsaddle Schist, Llano Uplift, Texas.  Texas J. Sci.  9:328.

Bondietti, E. A.  1974.  Adsorption of Pu(IV) and  Th(IV)  by Soil Colloids.
Agronomy Abstracts:

Dementyev, V. S. and N. G. Syromyatnikov.  1965.   Mode of Occurrence of Thorium
Isotopes in Ground Waters.  Geokhimiya. No. 2:211-218.

Desai, M. V. M. and A. K. Ganguly.  1970.  Interaction of Trace Elements with
Organic Constituents in the Marine Environment.  BARC-488,  p. 102.

Hansen, R. 0.  1970.  Radioactivity of a California Terrace Soil.   Soil
Science.  110:31-36.

Hansen, R. 0. and G. L. Huntington.  1969.  Thorium Movements in Morainal
Soils of the High Sierra, California.  Soil Science.  108:257-265.

Heier, K. S. and J. L. Carter.  1964.  Uranium, Thorium,  and Potassium Contents
in Basic Rocks and Their Bearing on the Nature of  the Upper Mantle.  IN;
J.A.S. Adams and W. M. Lowder (eds.).  The Natural  Radiation Environment,  p.  75.

Heier, K. S. and J. J. W. Rogers.  1963.  Radiometric Determination of Thorium,
Uranium, and Potassium in Basalts and in Two Magmatic Differentiation  Series.
Geochim. et. Cosmochim. Acta. 27:137.

Holland, H. D. and J.  L. Kulp.  1954.  The Mechanism of Removal  of  Thorium
and Radium from the Oceans. Geochim. et Cosmochim.  Acta.  5:214-224.
                                   3-218

-------
Katsurayama, K.  1968.  Accumulation of Radioactivity in Rice Fields.   Annu.
Rep. Res. Reactor Inst. Kyoto University.  1:293-97.

Kimura, Y., H. Morishuma, T. Koga, H. Kawai  and Y.  Honda.  1968.   Studies on
the Behavior and Distribution of Radioactive Substances in Coastal  and
Estuarine Waters.  Kinki Daigaku Genshiryaku Kenkyusho Nonpo. 7:21-31  (in
Japanese).

Kuznetsov, Y. V., Z. N. Simonyak, A. P. Lisitsyn, and M. S. Frenklikh.  1968.
Thorium Isotopes (230ih, 232fh) in the Surface Layer of the Indian  Ocean
Sediments.  Geochem. Int. 5:169-77.

Lazarev, K. F., D. S. Nikolaev, and S. M.  Grashchenko.  1961.  Concentration
of Thorium Isotopes in Sea Water.  Radiokhimiya.  3:623-35 (in Russian).

Menzel, R. G.  1968.  Uranium, Radium, and Thorium Content in Phosphate Rocks
and Their Possible Radiation Hazard.  J. Agr. Food Chem. 16:231-34.

Murray, E. G. and J. A. S. Adams.  1958.  Amount  and Distribution of Thorium,
Uranium, and Potassium in Sandstones.  Geochim. et.  Cosmochim.  Acta. 13:260.

Nishiwaki, Y., Y. Honda, Y. Kimura, H. Morishima, T. Koga, Y. Miyaguchi, and
H. Kawai.  1972.  Behavior and Distribution  of Radioactive Substances  in
Coastal and Estuarine Waters.  IN: Radioactive Contamination of the Marine
Environment.  IAEA-SM-158/11, pp7 177-193.

Pashneva, G. E., T. P. Slavnina, and V. V. Serebrennikov.  1965.  Rare Earth
and Thorium Content in Soils of Tomsk Region.  Izv.  Sibirsk. Otd. Akad.  Nauk
SSSR, No. 4 Ser. Biol.-Med. Naulc. No. 1:48-52 (in Russian).  •

Pliler, R.  1956.  The Distribution of Thorium and Uranium in Sedimentary
Rocks and the Oxygen Content of the Precambrian Atmosphere.  Thesis.  Rice
Institute.

Pokidin, V. K., Y. V. Kuznetov, E. A. Prozorovich,  and F. A. Asadullaeva.
1972.  Radioactivity and Rate of Sediment Formation in the Caspian  Sea.
Geokhimiya. No. 7:834-43 (in Russian).

Rancon, D.  1973.  The Behavior in Underground Environments of Uranium and
Thorium Discharged by the Nuclear Industry.   IN;  Environmental  Behavior of
Radionuclides Released in the Nuclear Industry.  IAEA-SM-172/55,  pp. 333-346
(in French).

Rogers, J. J. W.  1964.  Statistical Test of the  Homogeneity of the Radio-
active Components of Granite Rocks.  IN:  J.A.S.  Adams and W." M.  Lowder (eds.),
The Natural Radiation Environment, p.~Tl.

Rogers, J. J. W. and J. A. S. Adams.  1974.   Thorium Solubilities in Water.
Handbook of Geochemistry, K. H. Wedepohl, Ed., Vol.  II-4, p. 90-H-l, Springer-
Verlag, New York.
                                   3-219

-------
 Rubtsov,  D. M.   1966.  Distribution of Thorium in Various Soils.
 Pochvovedenie.   No. 3:55-67  (in Russian).

 Rubtsov,  D. M.   1972.  Thorium and Radium Content in the Silt Fraction of the
 Podzolic  Mountain Soils of Thin Forests.  Radioekologicheskie Issledavaniya
 Prirodnykh Biogeotsenozakh.   IN: Verkhovskaya (ed.) Izdatel'stov Nauka,
 pp. 42-53.

 Ryabchikov, D.  I. and E. K. Golbraikh.  1969.  Analytical Chemistry of
 Thorium.  Translated by A. Aladjem.  Ann Arbor-Humphrey Science Publishers.

 Schulz, R. K.   1965.  Soil Chemistry of Radionuclides.  Health Physics.
 11:1317-24.

.Si 11 en, L. G. and A. E. Martell.  1964.  Stability Constants of Metal-Ion
 Complexes.  Special Publication No. 17.  The Chemical Society, London.

 Tyuryukanova, E. B. and V. A. Kalugina.  1971.  The Behavior of Thorium in
 Soils.  Soviet  Journal of Ecology.  2:467-469.

 Vinogradoy, A.  P.  1959.  The Geochemistry of Rare and Dispersed Chemical
 Elements  in Soils.  Consultants Bureau, Inc.

 Yakobenchuk, V.  F.  1968.  Radioactivity and Chemical Properties of Sod-Pod-
 zolic Soils in  the Ukrainian Western Polesie.  Visn. Sil's1 Kogosped. Nauki.
 11:45-50  (in Ukranian).
                                    3-220

-------
TRITIUM
Natural Soil and Rock Distributions
     Natural tritium concentrations vary geographically from about 10 T.U.  for
northern and southern midlatitude precipitation to 1  T.U.  for equatorial  pre-
cipitation.  The background tritium content of deep ocean  and groundwater is
essentially zero (Fairbridge, 1972).  Hence, the tritium content of rocks,
minerals and deeper sediments also is nearly zero unless contacted by tritium-
contaminated groundwater.
Brief Chemistry
     Tritium ( H or T) is an isotope of hydrogen with a half-life of 12.26  years.
Tritium occurs naturally in the upper atmosphere, chiefly by the interaction
of fast neutrons with nitrogen (Kaufman and Libby, 1954).   The amount of  natu-
ral tritium is about one part in 10   parts of normal hydrogen ( H) (Cotton
and Wilkinson, 1962).
     It was a fairly recent discovery that tritium also is a product of uranium
fission (Albenesius, 1959; Albenesius and Ondrejcin,  1960).  Several years
prior to the identification of tritium as a uranium fission product, the
Environmental Monitoring Group at. Savannah River plant recognized tritium as
occurring in Purex separations process wastes, and was using it as a ground-
water tracer (Norton and Ross, 1960).  Tritium yields of 3.8 to 20.3 curies/
ton uranium were obtained for exposures between 300 and 1600 megawatt-days/
ton uranium.
     Tritium oxidizes rapidly to HTO, existing essentially as water, and  its
distribution in ground and surface waters is controlled by the operation  of
the hydrological cycle.  Tritium is measured in tritium units, T.U., where
           18
T.U. = T/10  H, or expressed as picocuries/liter.  Natural background levels
have risen since weapons testing of nuclear devices from 1 to 10 T.U. to
several hundred T.U. up to the 1963 atmospheric detonation moratorium. Most
of the tritium added to the hydrological cycle by fusion devices was due  to
residual tritium H5.7 x 10  Ci  H/megaton fusion device,  Stead, 1963).
                                   3-221

-------
Solid Phase and Solution Equilibria
     Tritium (H ) is a radioactive isotope of hydrogen.  Thus, tritium behavior
in soils would be expected either to be similar to hydrogen or to exist as an
ion, gas and liquid (tritiated water).  Except for the slight differences in
vapor pressure, tritiated water behaves the same as ordinary water (Jacobs,
1974).  Although no discrete solid phases of tritium are expected in soils,
it could associate itself with soil organic components containing hydrogen or
with some soil  minerals as mobile water, water of hydration, or as part of
structural hydroxyl groups (Jacobs, 1974; Stewart, 1967).
     Tritium rapidly travels at about the same velocity as the soil water or
groundwater in the form of HTO.  Some replacement of nontritiated water on
clays and other hydrated soil constituents occurs, but the reaction is of
marginal value as far as tritium retention is concerned.  Tritium can move in
a vapor phase through the soil under certain conditions as well.
Experimental Adsorption Results
     Tritium ultimately exists in the soil as a tritiated water molecule.
Theoretically,  tritium ions are capable of exchanging for hydrogen and other
ions on soil.  However, the. tritium is usually in an aqueous solution before
contacting soils, so that isotopic exchange and replacement of hydroxyl 'water
molecules is the most common mechanism of tritium removal.  Little quantita-
tive work on tritium distribution has been reported.  Rabinowitz et al. (1973)
have used an electrodialysis technique to speed up tritium hydroxyl exchange
in clays.  They reported that tritium hydroxyl  exchange was proportional to
cation exchange capacity of the clays used in the study:  illite, 35 meq/100 g
and 93 pCi3H/ml; kaolinite No. 4, 10 meq/100 g  and 36 pCi3H/ml;  kaolinite No. 9,
2 meq/100 g and 12 pCi H/ml.. Most of the field studies indicated that tritium
distribution coefficients were very low because tritium migration velocities
were essentially the same as those of the .accompanying groundwater (Brown and
Haney, 1964; Brown, 1967).
                                   3-222

-------
Migration Results
     Field Studies—
     Haney, Brown and Reisenauer (1962) and Brown and Haney (1964) estimated
by studying the rate and direction of groundwater flow that movement of tri-
tium disposed to the ground at the Hanford Purex Plant would require 6 or
7 years travel time to the Columbia River, approximately 17 miles from the
Purex Plant.  The half-life of tritium allowed about 70% of the total tritium
discharged to reach the river.  The tritium content of Columbia River water
was not expected to rise over 50% higher than the present background because"
of tritium dispersion during travel and the large dilution represented by the
Columbia River (80,000 to 200,000 ft3/sec).
     Brown (1967) reviewed the hydrology and geology of the Hanford area, and
showed the 1966 limits of the areas defined by 3000 pCi/cm  and 10 pCi/cm
                                                              3    3
of tritium based on well water analytical results.  The 10 pCi H/cm  area was
3 km from the Columbia River in 1966.  The concentration of tritium added at
                                             3    3
the Purex Plant disposal sites was 10,000 pCi H/cm .  Tritium moved at essen-
tially the same rate as the groundwater.
    ' Kline and Jordan (1968) reported a field experiment on a clay Puerto
Rican soil where 1 liter of 20 mCi/1 tritiated water was applied to a plot
              2
with about 1 m  area, and sampled via a lysimeter installed horizontally,
without disturbing the plot surface soil, 18 cm below the soil surface.  The
sampling program lasted 210 days during which time 137 cm of rain fell on the
plot.  Runoff surface water was collected as well.  The soil remained satu-
rated, or close to saturation for the sampling period.  Most of the tritium
passed through the profile as a peak or front 16 days after application, and
declined in concentration exponentially during the remainder of the experi-
ment.  The surface runoff peaked in 2.9 days, with another change in decline
of specific activity at 35.6 days.  Because of the loss rate curve of tritium
in the free soil water, the authors suggested that tritium movement through
soils must be modified to allow for the existence in clay soils of isolated
compartments of immobile water that does not have free interchange with the
more rapidly moving water.
                                   3-223

-------
     Jordan et al. (1971) monitored tritium movement through a soil near
Argonne, Illinois.  They concluded from the results that a pulse of tritium
moves downward at a rate determined by precipitation entering the top of the
soil column.  The tritium peak broadens and flattens from diffusion and will
become immobilized if the upper soil layers dry out.  Loss of tritium occurred
both by evapotranspiration and deep drainage.  Deviation of the field results
from a model of tn'tiated water movement proposed by Sasscer et al. (1971)
was explained as due to parameters that were difficult to quantify such as
root holes and hydroxyl exchanges.
     Ehhalt (1973) pointed out that tritiated molecular hydrogen in the atmo-
sphere was oxidized by soil  microorganisms.  He also estimated from the tritium
content of H, of 2 x 10  T.U., steady since 1962, that the HT input rate was
                    2
about 5.4- T atoms/cm  sec into the soil from the action of soil microorganisms.
     Purtymun (1973) investigated the underground movement of tritium from
solid waste storage shafts in rhyolite tuff at Los Alamos.  Levels of 100 pCi
 H/ml had moved a distance of 105 ft in 4 years.  Core samples of tuff were
collected, water distilled from them and tritium in the water determined.
Asphalt coatings on containers and shaft walls was suggested to control
tritium migration from the shafts.
     In contrast, Wheeler and Warren (1975) reported that a later comparison
between asphalted and nonasphalted shafts showed that the asphalt containment
techniques used were ineffective in slowing tritium migration from the shafts.
Double containment and complete encapsulation of the 210-liter disposal drums
in roofing asphalt was suggested to slow tritium movement from the storage
shafts.
     Laboratory Studies—
     Because soil water and groundwater systems are similar, it would be
expected that tritium oxide would move through these systems at essentially
the same rate as light water.  However, in comparing chloride ions arid tritium
breakthrough curves for various soil types, the tritium curve lags behind
that for chloride ions in soils containing clays and silt (Kaufman and Orlob,
1955).  This behavior was reported as caused by the tritiated water molecule
entering the clays and replacing nontritiated water adsorbed on clay surfaces
or inter!ayers.
                                   3-224

-------
     Haney (1963, 1964) reported a laboratory study where actual  groundwater
containing complexed    Ru and tritium was passed through several Hanford
project soils.  In silty soils, the ruthenium breakthrough curve was slightly
behind that for tritium, but in sandy soil columns, the breakthrough curves
were essentially identical.  Several  retention mechanisms of tritium on clays
are possible.  For example:  1) exchange of tritium ions for hydrogen ions in
exchange positions, 2) tritiated water molecules exchanged for hydroxyl water
or water of crystallization, 3) exchange of tritium ions for exchangeable
cations other than hydrogen, and 4) replacement of aluminum in lattice sites
by tritium, with the aluminum ions moving into exchangeable cation positions.
In the last case, the tritium may be more or less "fixed" in its  octahedral
position.  This "fixation" has been considered to be minimal by some investi-
gations (Higgins, 1959; Halevy, 1964; Corey and Morton, 1968; Teller et al.,
1968).
     It has been hypothesized by others (Corey and Fenimore, 1968) that the
chloride ions are repelled from negatively charged soil particles into the
central region of the soil pore where groundwater velocity is greatest (anion
exclusion).  This would be most effective in high cation exchange capacity
soils which do not retain anions.  Also, Corey and Fenimore (1968) showed that
chloride ions lag far behind tritium in acid kaolinitic soil due  to anion
exchange of the chloride ions.
     Corey and Morton (1968) reported no appreciable tritium "fixation" upon
investigation of the movement of tritiated and deuterated water through
acidic, kaolinitic soil.  The work of other investigators, however, indicates
that hydrogen isotopes may be selectively fixed (Rosenqvist, 1963; Koranda,
1965; Clayton et al., 1966; Savin, 1967; Stewart, 1967).  The work of
Rabinowitz (1969) and Rabinowitz et al. (1973) with forced exchange of tri-
tiated water on clays also supports the view that part of the tritium can be
"fixed" on clays.  The isotopic exchange and fixation was forced  with a mild
form of electrodialysis (2 V/cm) resulting in an increase in the  reaction
rate of about five times the rate without electrodialysis.  Aluminum and
other basing cations were determined after each experiment.  The  presence of
               +3
exchangeable Al   correlated with tritium loss from the water resulting from
tritium "fixation".  A mass balance on the tritium also was determined for each

                                   3-225

-------
experiment.  The analytical  data for tritium facilitated the differentiation of

isotopic exchange and the "fixation" of tritium by kaolinite and ill He.

Summary

     Tritium substitutes readily for the hydrogen in water and thus becomes  a

part of the hydrological cycle (Jacobs, 1974).   Migration of tritium through

the groundwater takes place at the same velocity as the groundwater through

sandy soils (Haney 1963, 1964; Haney et al., 1962; Brown and Haney, 1964;
Brown, 1967).   Some investigators have reported the selective fixation of

tritiated water on clays and other hydrated  minerals (Rosenqvist, 1963;

Koranda, 1965; Clayton et al., 1966; Savin,  1967; Stewart, 1967; Rabinowitz,

1969; Rabinowitz et al., 1973) while other investigators have considered tri-
tium fixation  to be minimal  (Higgins, 1959;  Halevy, 1964; Corey and Horton,
1968; Teller et al., 1968).   However, all  of the field studies have indicated
that tritium movement is synonomous with water  movement (Brown, 1967).

References

Albenesius, E. L.  1959.  Tritium as a Product  of Fission.  Physics Review
Letters.  3:274.

Albenesius, £. L. and R. S.  Ondrejcin.  1960.   Nuclear Fission Produces
Tritium.  Nucleonics.  18:100.

Brown, D. J.  1967.  Migration Characteristics  of Radionuclides Through
Sediments Underlying the Hanford.Reservation.   ISO-SA-32.

Brown, D. J. and W. A. Haney.  1964.  Chemical  Effluents Technology Waste
Disposal Investigations July-December, 1973  - The Movements of Contaminated
Ground Water from the 200 Areas to the Columbia River.  HW-80909.

Clayton, R. N., I. Friedman, D. L. Graaf,  T. K. Mayeda, W. F. Meents and
N. F. Shimp.  1966.  The Origin of Saline Formation Waters.  I. Isotopic
Composition.  J. Geophys. Res., 71, 3869-3882.

Corey, J. C. and J. W. Fenimore.  1968.  Tracing Groundwater with Chloride
Ions and Tritium Through Acid Kaolinitic Soil.   Int. J. Appl. Rad.  Isotopes.
19:741-746.

Corey, J. C. and J. H. Horton.  1968.  Movement of Water Tagged with 2H, 3H
and  '°0 Through Acidic Kaolinitic Soil.  Soil Sci. Soc. Am.,  Proc.  32:471-475.

Cotton, F. A.  and G. Wilkinson.  1962.  Advanced Inorganic Chemistry.   Inter-
science Publishers.

Ehhalt, D. H.   1973.  On the Uptake of Tritium  by Soil Water and Ground Water.
Water Resources Research.  9:1073-1074.

                                     3-226

-------
Fairbridge, R. W. (ed.).  1972.  The Encyclopedia of Geochemistry and Environ-
mental Sciences.  Van Nostrand Reinhold Company.

Halevy, E.  1964.  The Exchangeability of Hydroxyl Groups in Kaolinite.
Geochim. et Cosmochim. Acta.  28:1139-1145.

Haney, W. A.  1963.  Fission Product Tritium in Fuel Reprocessing Wastes.
HW-SA-3303.

Haney, W. A.  1964.  Consequences of Activity Release.  Nuclear Safety.
5:399-403.

Haney, W. A., D. J. Brown, and A. E. Reisenauer.  1962.  Fission Product
Tritium in Separations Wastes and in the Ground-Water.  HW-74536.

Higgins, G. H.  1959.  Evaluation of the Ground-Water Contamination Hazard
from Underground Nuclear Explosions.  J. Geophys. Res.  64:1509-1520.

Horton, J. H. and D. I. Ross.  1960.  Use of Tritium from Spent Uranium Fuel
Elements as a Ground-Water Tracer.  Soil Science.  90:267-271.

Jacobs, D. C.  1974.  Impacts in Groundwater of Effluents Arising in  the
Nuclear Industry.  NP-20456, pp. vp., Paper 16.

Jordan, C. F., J. R. Kline, and D. S. Sasscer.   1971.  Tritium  Movement in an
Old-Field Ecosystem Determined Experimentally.   IN:  Radionuclides in Eco-
systems.  CONF-710501-P1, pp. 199-203.

Kaufman, S. and W. F. Libby.  1954.  The-Natural Distribution of Tritium.
Phys. Rev.  93:1337.

Kaufman, W. J. and G. T. Orlob.  1955.  Measuring Ground Water  Movement with
Radioactive and Chemical Tracers.  Paper presented at the Annual  Meeting of
the California Section of AWWA, Sacramento.

Kline, J. R. and C. F. Jordan.  1968.  Tritium Movement in Soil  of Tropical
Rain Forest.  Science.  160:550-551.

Koranda, J. J.  1965.  Preliminary Studies of the Persistence of Tritium and
14c.in the Pacific Proving Ground.  Health Phys., 11, 1445-1457.

Purtymun, W. D.  1973.  Underground Movement of Tritium from Solid-Waste
Storage Shafts.  LA-5286-MS.

Rabinowitz, D. D.  1969.  Forced Exchange of Tritiated Water with Natural
Clay.  MS Thesis, N. Mex. Inst. of Min. and Tech., Socorro, NM.

Rabinowitz, D. D., C. R. Holmes and G. W.  Gross.  1973.  Forced Exchange of
Tritiated Water with Clays.  IN;  Trftium, Messenger Graphics Pub., Phoenix
and Las Vegas, 471-485.

Rosenqvist, I. T.  1963.  Studies in Position and Mobility of the H Atoms in
Hydrous Micas.  IN_:  Clays and Clay Minerals, Monograph No. 13, p. 11-135,
Macmi11 an Co., NY.


                                     3-227

-------
Sasscer, D. S., C. F. Jordan, and J. R. Kline.  1971.  Mathematical  Model  of
Tritiated and Stable Water Movement in an Old-Field Ecosystem.   IN:   Radio-
nuclides in Ecosystems.  CONF-710501-P1, pp.  915-923.

Savin, A. M.  1967.  Oxygen and Hydrogen Isotope Ratios in Sedimentary Rocks
and Minerals.  PhD Dissertation, Calif. Inst. of Tech. Pasadena, CA.

Stead, F. W.  1963.  Tritium Distribution in Ground Water Around Large Under-
ground Fusion Explosions.  Science.  142:1163-1165.

Stewart, G. L.  1967.  Fractionation of Tritium and Deuterium in Soil Water.
IN:  Isotope Techniques in the Hydrologic Cycle, Geophys. Monograph No. 11,
139-168.

Teller, E., W. K. Talley and G, H. Higgins.   1968.  The Constructive Uses  of
Nuclear Explosives.  McGraw-Hill, New York.

Wheeler, M. L. and J. L. Warren.  1975.  Tritium Containment After Burial  of
Contaminated Solid Waste.  Trans. Am. Nucl.  Soc. 22:741.
URANIUM

Natural Soil and Rock Distributions

     The range in abundance of uranium in natural rocks is given in Table 3-96.

Vinogradov (1959) reported 1 ppm uranium as the average content of uranium in

soils.  The oxidation of organic matter in black shales tends to precipitate

U(IV).  Consequently, the black shales usually contain more uranium than red,

green or gray shales.

Brief Chemistry

     There are 14 known isotopes of uranium from    U to    U and one uranium
        235
isomer (   U) (Fairbridge, 1972).  The half-lives of the uranium isotopes vary

from a few minutes to over 4 billion years.  Only three of the isotopes occur
                             235      238
naturally, and two of these (   U and    U) are parents of series that end in

lead isotopes.  The nuclear data on the three natural uranium isotopes are
given in Table 3-97.  The uranium isotopes of interest in waste disposal are

given in Table 3-98.  The common oxidation states of uranium are U(III), U(IV),

U(V) and U(VI) (Udaltsova, 1963).  U(III) is easily oxidized in air to U(IV).

                                     3-228

-------
           TABLE 3-96.  ABUNDANCES OF URANIUM IN NATURAL  MATERIALS
                         (ADAMS ET AL.,  1959; CLARK ET AL., 1966)
                   Material
                                               U Concentration Range, ppm
                                   Igneous Rocks
         Dunites and Peridotites
         Gabbro and Diabase
         Intermediate (diorite and
           quartz diorite)
         Sialic (granite, syenite,
           monaozite)
                                                0.001-
                                                  0.3-

                                                  0.1-1
    •0.8
    •3.4
    -11.0

0.15-21.0
                                 Sedimentary Rocks
         Black  shales
         Red, gray and green shales
         Orthoquartzite
         Limestone and dolomite
         Bentonite
         Bauxite
         Halite
         Anhydrite
                                 Metamorphic Rocks
         Marble
         Slate
         Phyllite
         Schist
         Gneiss
         Amphioolite
         Granulite
                                                 0.11-
                                                  1.2-
                                                  1.0-
                                                  1.8-
                                                  4.5-
                                                  2.6-
                                                  3.2-
    •0.24
    •6.1
    •2.7
    •2.9
    •15.0
    4.1
    7.0
Median 0.5

Median 1.7

Median 3.9
3.0-25.0
1.2-12.0
0.2-0.6
0.1-9.0
1.0-21.0
3.0-27.0
0.01-0.02
0.25-0.43
Median 8.0
Median 3.2
Median 0.45
Median 2.2
Median 5.0
Median 8.0
Median 0.013
Median 0.37
The U(IV)  state is fairly stable in aqueous solutions  if they are very acidic.
Uranium  (V)  disproportionates. to U(IV)  and U(VI):
+ 6H20  (Udaltsova, 1963).   U(VI), as UO*2 (Uranyl) at  pH < 2.5, is  the  most
                                                              40"
                                                                               H
                                     3-229

-------
             TABLE 3-97.   NUCLEAR PROPERTIES OF NATURAL URANIUM
                          ISOTOPES (FAIRBRIDGE, 1972)

                                       Half-Life,         Decay Series
    Isotopes        % Abundance           yr                Parent
      234                                    •   R
      "4U             0.0056          2.48 x 10b
      235U             0.72            7.13 x 108            4N +  3
      238U            99.27            4.51 x 109            4N +  2
                   TABLE  3-98.   URANIUM  RADIONUCLIDE  DATA
                                (WEAST,  1976)
                  Isotope       Half-Life         Decay  Mode
                   232
                   "ni       73.6 years            a, SF
                     U       162,000 years         a
                   27A
                   "\l       247,000 years         a
                   235U       7.13 x  108  years      a, SF
                 -  236U       2.39 x  107  years      ct, SF
                   237U       6.75 days    .         6"
                   238U -'    4.51 x  109  years      a, SF
stable state of uranium in aerated aqueous  solutions  (Seaborg and Katz, 1954).
At higher pH values, hydrolyzed uranyl  ions predominate.   Approximately
103 uranium minerals have been confirmed.   Uraninite  (ideally U02) is the
primary ore mineral of uranium, but other  secondary minerals include carbonates,
sulfates, molybdates, phosphates,  vanadates, silicates  and multiple oxides.
Uranyl ions readily form many complexes with anions ordinarily found in soil-
water environments such as carbonate,  sulfate and fluoride.

Solid Phase and Solution Equilibria
     The thermodynamic data for 1) U03> Na2U04 and U02C03  were obtained from
Garrels and Christ (1965), 2) U02(OH)2> U02(OH)2H20 and Na4U02(C03)3 were
obtained from Sillen and Martell (1964), and 3) the remaining species reported
in Figure 3-26 were obtained from Palei (1970).  The  relative stability of
several uranium solid phases, in terms  of  the uranyl  ion  activity produced by
each, is shown in Figure 3-26.  Since  all  the compounds shown in Figure 3-26

                                   3-230

-------
                                                  2+
are of U(VI), and are plotted as a function of U02  , their curves would not
move with changes in oxidation-reduction conditions.  U(IV) compounds, such
as UCL, are very soluble in an oxidizing environment and fall outside the
boundary of Figure 3-26.  With an increase in reducing conditions, U(IV) com-
pounds would become more stable.  At a pO« of >71, the U02 curve would fall
just below the UOJW.PO. curve.  U(VI) compounds are stable in an oxidizing
environment and U(IV) compounds are stable in a reducing environment.  Consis-
tent with the thermodynamic data is the observation that carnotite, a U(VI)
mineral is found in the oxidized zones of uranium ore deposits and uraninite,
a U(IV) mineral is a primary mineral  in reducing ore zones.
                -2
             f
                -8
                -10
                -12
                                                           10
                                      PH
Figure 3-26.
              The relative stability of various uranium solids in an oxidizing
              soil environment [pOgtg) = 0.68 atm], PC02(g) = 3-52 atm, pK+ =
              pNa* = PNH4"1" = 3.0 and phosphate levels in equilibrium with "
              Variscite and Gibbsite.
     The thermodynamic data for UO^F  and UO^PO*  were selected from Palei
 (1970) and selected data from Si 11 en and Martell (1964) were used for the
 remaining species in Figure 3-27.  All the species except U02  shown in Fig-
 ure 3-27 are of U(VI).  Since equilibrium is assumed with Na2U04, a U(VI) com-
 pound, a change in oxidation-reduction conditions will affect the position of
 the curves.  In an oxidizing environment (p02 = 0.68 atm), the U(IV) species
 are very low in concentration (log a^ <38) and fall outside of the boundaries
of Figure 3-27.
                                     3-231

-------
     CO
     s*
Figure 3-27.  Activity of various uranium species in equilibrium with
              in an oxidizing soil environment [p02(q) = 0.68 atm], pCOpfn)
           -   3.52 atm, pCl" = pS042- = 3.0, pF~ = 4T5 and pH2P04 = 5.0.
     Figure 3-27 shows that U (VI) species will  control  the solution concen-
                                         2+
tration in an oxidizing environment.  UO?   is  the predominant solution species

up to a pH of approximately 6.   The predominant solution species over pH

ranges of from 6 to 8 and >8 are UOgtOHK" and  U02(CO,),  ,  respectively.

Experimental Adsorption Results

     Uranium adsorption studies have been performed by  several  investigators.

Szalay (1954, 1957) showed that uranium adsorption by decomposing plant debris,
peat, lignite and brown coal  is quite high.   He determined that humic substances
                                   3-232

-------
in these materials were responsible for the adsorption which was described as
an ion exchange-like sorption.  Adsorption isotherms for the humic acid were
measured.
     Goldsztaub and Wey (1955) determined that 7.5 g of uranium was adsorbed
from a 1% uranyl nitrate solution per 100 g of calcined montmorillonite and
2.0 g of uranium per 100 g of calcined kaolinite.
     Manskaya et al. (1956) described the adsorption of uranium on fulvic
acids as a function of pH.  The curve of percent adsorption versus pH showed
a maximum at pH 6 of about 90% uranium removal.  At pH 4 and pH 7, uranium
adsorption was down to 30%.
     Starik et al. (1958) showed a similar adsorption curve of pH versus
uranium adsorption on ferric hydroxide.  The maximum of the adsorption curve
was at pH 5 with about 50% uranium adsorption, and rapidly decreased above
and below pH 5.
     Rozhkova et al. (1959) showed similar uranium adsorption curves versus
pH for lignite and humic acids.  The curve maxima  occurred between pH 5 and
6;  Dementyev and Syromyatnikov (1968) showed that these adsorption curve
maxima result because the pH 6 region is a-boundary between anionic and
cationic uranium forms and corresponds to:  [UO^2] + LUOgOH"1"] ~ [UO^CO^2],
an equality between cation and anion uranium forms in solution.
     Horrath (1960) measured the enrichment factor for the adsorption of
uranium by peat and obtained an average of 200 to  350.  Although it is not
possible to determine what methods or exactly what the enrichment factor means
from the English translated abstract, it is assumed that the data represent a
uranium Kd of 200 to 350 by weight or volume.
     Kovalevskii (1967) found the uranium content of noncultivated soils in
western Siberia increased with the clay content of the soils.  Clay soils con-
tained at least three times as much uranium as sands.  Yakobenchuck (1968)
correlated the total uranium content in Russian sodpodzilic soils from the
Ukraine with other soil constituents.  Uranium showed correlation with the
oxidizes of silicon, iron, and aluminum suggesting coprecipitation or
inclusion.
                                   3-233

-------
     Masuda and Yamamoto (1971) studied the adsorption of uranium (1 to
100 ppm U) dissolved in water onto volcanic ash, alluvial, and sandy soils.
The uranium was almost completely adsorbed on each of the soils.  Uranium
desorption with salt solutions was extremely difficult especially for the
volcanic ash.  Similar studies by Yamamoto et al. (1973) on the three soils
                                                                    _2
using uranium (1 to 50 ppm U) and carbonated waters (4 to 109 ppm COg )
showed approximately 100% adsorption and less than 2% desorption.
     Rubtsov (1972) determined the uranium content in forest podzolic mountain
soils and found a relatively high level of uranium in the <1 u particle size
fraction of the podzolic k* horizon.  In general, for the soils studied 58%
of the total uranium was found in the <1 u soil  fractions.  Ritchie, Hawks,
and McHenry (1972) found the uranium content of sediments from the Little
Tall ahatchie River to increase with decreasing particle size.
     Rancon (1973) studied the adsorption of uranium using four soils described
as follows:  1) .a river sediment containing a mixture of quartz, clay, calcite
and organic matter, 2) a river peat, 3) a sediment from Cadarache containing
a mixture of quartz, clay and calcite with no organic matter, and 4) a soil
developed on an altered schist from near laHague containing a mixture of
quartz and clay but no calcite or organic matter.  The"first two soils were
equilibrated with their river waters containing  10 ppm uranium and the last
two soils were equilibrated with their respective groundwaters also containing
10 ppm uranium.  The resulting uranium distribution coefficients are shown in
Table 3-99, which also includes the Kd values on pure quartz, calcite and
illite.  The clay minerals in Soils 1, 3 and 4 were not identified or the
soils further characterized.  Rancon also examined the effects of initial
uranium concentration on Kd values.  Both the uranium concentration and solu-
tion pH changed as uranium was added to the solution.  At 0.1 mg U/l, the pH
was 7.6, for example, and at 1.0 g U/l, the pH was 3.5.   Because the pH
changes are a function of uranium concentration  changes, the results are not
easily interpreted.  In addition, the Kd concept, is invalid above the trace
uranium concentration KI.O mg U/l).  Uranium adsorption data at 1  ppm versus
Kd also are presented.  For Soil  4, three peaks  were observed:  Kd 300 ml/g
at about pH 5.5, Kd 2000 ml/g at pH 10 and Kd 270 ml/g at pH 12.
                                   3-234

-------
               TABLE 3-99.  URANIUM Kd VALUES (RANCON, 1973)

             	Soil	                    Kd, ml/g
             1 - River Sediment                           39
                 (clay, CaC03, OM)
             2 - River Peat                               33
             3 - Sediment (clay, CaC03)                   16
             4 - Altered Schist (clay)                   270
             Quartz                                        0
             Calcite                                       7
             Illite                                      139

     Rancon believed that the adsorption maxima  represented by the three
peaks also represent electrokinetic potential  maxima.   Quartz was  charac-
terized as inert, calcite was a poor uranium adsorber  and clays were  the
best adsorbers of uranium from solution.  Acid,  organic-rich soils show
much higher uranium Kd values than the alkaline  peat (Soil  2) of this study.
Migration Results
     Field Studies—
     A study of granitic rock weathering by Harriss and Adams (1966)  included
autoradiographs of fresh and weathered samples of several granitic rocks.
There was a definite increase in the density of concentrated radioactive mate-
rials with weathering.  However, analyses indicated a  small loss for  uranium.
The increased density, therefore, must be due to losses of other materials
(alkalies and alkaline earths) during weathering.  From a fresh granodiorite
containing 2.5 ppm uranium, the initial weathering resulted in losses of 60%
of the uranium.  An acid leach of fresh rock also removed 60% of the  uranium
indicating that most of the uranium was in acid soluble or interstitial mate-
rials.  After an initial drop in concentration,  the total uranium  content  of
the weathered rock increased by at least a factor of 4 in the uppermost
material.
     Laboratory Studies—
     Schulz (1965) suggested that uranium may be present in the soil  as the
                       2+
divalent uranyl ion, UO^.  , and will be mobile in soils if present  as  the uranyl
ion.
                                    3-235

-------
     Masuda and Yamamoto (1971) examined the desorption of uranium from
alluvial, sand and volcanic ash soils.  The cation exchange capacities were
13.7, 7.7 and 33.0 meq/100 g, respectively, for the alluvial sand and vol-
canic ash soils.  Strong salt solutions and distilled water were used as
leachates.  Loads of more than 2000 yg U/g of soil were required before desorp-
tion by distilled water was 1% of the uranium on the soils.  The uranium was
amended to the soils as uranyl nitrate before the desorption work.  Desorption
of uranium was higher with 0.5M (NH4)2S04, 1.34M KC1 and 1.44M <2HP04 salt
solutions, but reached 50% removal only for the alluvial soil with a high
adsorbed uranium content in 1.44M KpHP04 solution.  The volcanic ash soil did
not attain 5% uranium desorbed in any of the salt solutions.
     The effects of carbonate ions on uranium desorption from soils was
examined by Yamamoto et al. (1973).  They used an alluvial soil, a volcanic
ash soil and a sandy soil containing up to 500 ug U/g air-dried soil in their
desorption experiments.  The desorption of uranium declined as a power func-
tion of the amount of uranium on the soil.  Desorption results are shown in
Table 3-100.  Ten grams of soil were magnetically stirred with 100 ml of potas-
sium carbonate solution for 30 min and stood overnight before filtering and
a fluorometric uranium analysis.  As can be seen in Table 3-92, uranium desorp-
tion was very low in the presence of low to moderate environmental carbonate
concentrations.

     TABLE 3-100.  DESORPTION OF URANIUM FROM SOILS WITH DISTILLED WATER
                   AND CARBONATE SOLUTIONS (YAMAMOTO, 1973)
Soil
Alluvial
Sandy
Volcanic Ash
U Content,
yg/g
7.1
485.0
10.1
488.2
8.3
500.0
Distilled
Water, %
- 0.31
0.22
0.14
0.12
0.18
0.09
Carbonate
4.3mg
C03-2/l
0.62
0.25
0.27
0.13
0.57
0.10
Solution, %
43.4mg
CO,-2/T
— o — ! —
1.20
0.41
0.90
0.46
1.15
0.20
     Dall'Aglio et al.  (1974) discussed some of the geochemical  processes
responsible for precipitating secondary uranium minerals.   There are a series

                                    3-236

-------
of processes capable of bringing about a very effective separation and con-
centration of uranium.  The most important process is the attainment of high
  +2
U09  activity in circulating waters because of low concentration or depletion
     -2
of COg  which is the most effective uranyl complexing agent.  The micro-
organisms involved were soil microflora and bacteria of mine-waters and
granites.  Some species were identified.  Batch cultures were used to study
uranium insolubilization involving biodegradation of uranium complexing
organic compounds.  The authors suggested that the redeposition of uranium by
bacteria may be the origin of some uranium deposits.
     Uranium solubilization and insolubilization from granites by heterotrophic
bacteria were investigated by Magne et al . (1974).  Microbial activity
increased the solubilization of uranium from 2 to 97 times by biosyntheses of
complexing and chelating compounds.
     Grandstaff (1976) arrived at a rate expression for the effects of surface
area and uraninite composition, oxygen content of the solution, carbonate
content, organics content, pH and system temperature on uraninite dissolution.
R, the rate of uraninite (U02) dissolution = "d^an^ s 1020'25 (SS)(RF)"]
(1Q3. 38-10.8 NOC} (aZcQ) (D>Q>) (aj exp (_7045/T)day-l , where SS is the
specific surface area (cm2/g), RF is an organic retardation factor, NOC is. the
mole fraction of nonuranium cations in the uraninite, D.O.  is the dissolved oxy-
gen content of the water (ppm),  CC^ is the total  dissolved carbonate and T is
the absolute temperature.  The rate expression was used to  predict UOg dis-
solution rates under varying conditions in the absence of organics in the
contacting water with good results.  Dissolved inorganic species in artificial
seawater other than hydrogen ion activity (pH), total carbonate and dissolved .
oxygen did not appreciably affect the U02 dissolution rate.  The organic
retardation factor must be determined experimentally for each organic-
containing solution by comparing the calculated U02 dissolution rate without
organics i-n the environment with the observed dissolution rate, or RF = R
calculated/R observed.  Retardation values of up to 420 were obtained.  The
magnitude of the affects of the several factors in the expression on the
oxidation rate of U(IV) to U(VI) could be of assistance in  understanding
uranium mobility in soil and rock environments.
                                    3-237

-------
Summary
     The common oxidation states of uranium are U(III),  U(IV),  U(V)  and  U(VI)
(Udaltsova, 1963).  However, in the geologic environment U(IV)  and U(VI)  are
the most important oxidation states.  In oxidizing environments U(VI)  compounds
such as KU02V04 (Garrels and Christ, 1965), U02NH4P04, Na2U04 and U02KP04
(Figure 3-26) are stable and can precipitate.  U(IV)  would precipitate as U02
in a reducing environment.  U(VI) solution species govern uranium concentra-
tions and movement in oxidizing environments (Figure  3-27).   Uranium retention
by soils and rocks in alkaline conditions is poor because the predominant
uranium species at pH > 6 in oxidizing environments (Figure 3-27) are either
neutral or negatively charged.  An increase in CCL pressure in  soil  solutions
reduces uranium adsorption and can increase uranium concentration.  The  cation
exchange properties of soils could contribute to the  adsorption of uranium in
                                                        2+
the neutral to acidic pH range due to the presence of UCL .   Oxidation-
reduction conditions and pH would be important parameters of uranium mobiliza-
tion and immobilization.
     The above theoretically based predictions are substantiated by  experi-
mental results.  Uranium has been reported to be solubilized and highly  mobile
in .carbonate-containing waters (Brown and Keller, 1952;  Naumov, 1961;
Ermolaev et al., 1965; Legin et al., 1966; Haglund, 1968, 1969).   Soluble
uranium [U(VI}] can:
 1.  precipitate in the presence of phosphorus as evidenced  by  a direct  cor-
     relation of uranium and phosphate content in soils  and  rocks (Bell,  1960;
     Sakanoue, 1960; Habashi, 1962; Kuznetsov et al., 1968;  Menzel,  1968;
     Mihalik, 1968), and
 2.  be adsorbed by the soil organic component and/or reduced to U(IV) fol-
     lowed by precipitation (Breger et al., 1955; Kolodny, 1969; Kolodny and
     Kaplan, 1970; Baturin, 1971; OalTAglio, 1971; Dorta and Rona,  1971;
     Gabelman, 1971; Baturin and Kochenov, 1973; Mo et al.,  1973).
     An increase in uranium content with a decrease in soil  or  sediment  par-
ticle size was reported by several workers (Kovalevskii, 1967;  Mizuno  and
Mochizuki, 1970; Ritchie et al., 1972).
                                    3-238

-------
     The uranyl ion can be adsorbed on clay minerals (Goldsztaub and Wey,  1955;
Kovalevskii, 1967; Rubtsov, 1972; Ritchie et al.,  1972; Rancon,  1973) and
other adsorbent materials (Masuda and Yamamoto,  1971;  Yamamoto et al., 1973),
but also is inclined to form complexes with anions, such as carbonate, that
are commonly found in the soil solution (Figure  3-27).   Uranyl salts also
                                    2+
have been shown to substitute for Ca   during replacement of calcite by
                            2+                 2+
apatite (Ames, 1960), and Ca   competes with UO^  for  available  sites during
ion exchange reactions on inorganics such as calcite (Rancon, 1973)  resulting
in low uranium Kd values.  However, uranyl ion adsorption on organic mate-
rials and humic substances is quite high (Szalay,  1954, 1957; Manskaya et  al.,
1956; Rozhkova et al., 1959; Rancon, 1973) especially  at acid pH values.   The
most important parameters of uranium migration and retention are system Eh
and pH.
References
Adams, J. A. S., J. K. Osmond, and J. J. W. Rogers.  1959.  The  Geochemistry
of Uranium and Thorium.  Phys. Chem. Earth.  3:298.
Ames, L. L., Jr.  1960.  Some Cation Substitutions During the Formation of
Phosphorite from Calcite.  Economic Geology.  55:354-362.
Baturin, G. N.  1971.  Uranium in Solutions of Ooze Depo.sits in  the  South-.
eastern Atlantic.  Dokl. Akad. Nauk. SSSR.  198:1186-1188 (in Russian).
Baturin, G. N. and A. V. Kochenov.  1973.  Uranium in  Interstitial Waters  of
Marine and Oceanic Sediments.  Geokhimiya.  No.  10:1529-1536 (in Russian).
Bell, K. G.  1960.  Deposition of Uranium in Salt-Pan  Basins. -U.S.G.S.
Professional Paper No. 354-6:161-169.
Breger, I. A., M. Deul, and S. Rubinstein.  1955.   Geochemistry  and  Mineralogy
of a Uraniferous Lignite.  Econ. Geol.  50:206-226.
Brown, J. H. and W. D. Keller.  1952.  Uranium in  the .Clay of a  Black Radio-
active Shale.  Science.  116:632-633.
Clark, S. P., Jr., Z. E. Peterman and K. S. Heier.  1966.  Abundances of •
Uranium, Thorium, and Potassium.  IN:  Handbook  of Physical Constants.
S. P. Clark (ed.).  Geol. Soc. Am."Mem. 97.
Dall'Aglio, M.  1971.  Study of the Circulation  of Uranium in the Supergene
Environment in the Italian Alpine Range.  Geochim. et  Cosmochim. Acta. 35:47-59.
Dall'Aglio, M., R. Gragnani, and E. Locardi.  1974. Geochemical Factors Con-
trolling the Formation of the Secondary Minerals of Uranium.  IAEA-SM-183/21.
pp. 33-48.
                                    3-239

-------
Dement'yev, V. S. and N. G. Syromyatnikov.  1968.  Conditions of Formation of
a Sorption Barrier to the Migration of Uranium in an Oxidizing Environment.
Geochemistry International.  5:394-400.

Dorta, C. C. and E. Rona.  1971.  Geochemistry of Uranium in the Cariaco
Trench.   Bull. Mar. Sci.  21:754-765.

Ermolaev, N. P., A. P. Zhidikora, and V. A. Zarinski.  1965.  On the Silicate
Form of Uranium Transport in Aqueous Solutions.  Geokhimiya No. 7:813-826
(in Russian).

Fairbridge,. R. W. (ed.).  1972.  The Encyclopedia of Geochemistry and Environ-
mental Sciences.  Van Nostrand Reinhold Company.

Gabelman, J. W.  1971.  Sedimentology and Uranium Prospecting.  Sediment.
Geol.  6:145-186.

Garrels, R. M. and C. L. Christ.  1965.  Solutions, Minerals, and Equilibria.
Harper and Row, New York.

Goldsztaub, S. and R. Wey.  1955.  Adsorption of  Uranyl  Ions by Clays.   Bull.
Soc. Franc. Mineral. Crist. 78:242.

Grandstaff, D. E.  1976.  A Kinetic Study of the  Dissolution of Uraninite.
Econ. Geol. 71(8):1493-1506.

Habashi, F.  1962.  Correlation Between the Uranium Content of Marine Phos-
phates and Other Rock Constituents.  Econ, Geol.  57:1081-1084.

Haglund, D. S.  1968.  The Distribution of Uranium in Recent Carbonate Sedi-
ments and Skeletons of Organisms and the Effect of Diagenesis on Uranium
Redistribution.  Rensselaer Polytechnic Institute.  Thesis.

Haglund, D. S., G. M. Friedman, and D. S. Miller.  1969.   The Effect of Fresh-
water on the Redistribution of Uranium in Carbonate Sediments.  J. Sediment.
Petrol.  39:1283-1296.

Harriss, R. C. and J. A. S. Adams.  1966.  Geochemical  and Mineralogical
Studies on the Weathering of Granitic Rocks.  Am. J. Science. 264:146-173.

Horrath, E.  1960.  Investigations of Uranium Adsorption  to Peat in^Natural
Waters Containing U-Traces.  Magyar Tudomanyos Akad.  Atommag Kutato Intezete,
KOzlemenyek.  2:177-183 (in Hungarian).

Kolodny, Y.  1969.  Studies in "Geochemistry of Uranium and Phosphorites.
University of California.  Thesis.

Kolodny, Y. and I. R. Kaplan.  1970.  Deposition  of Uranium in the Sediment
and Interstitial Water of an Anoxic Fjord.  CONF-700965-1.
                                    3-240

-------
Kovalevskii, A. L.  1967.  Dependence of the Content of Some Trace Elements on
the Clayiness of Soils.  Mikroelem. Biosfere Ikh Primen.   Scl.  Khoz.  Med.  Sib.
Dal'nego Vostoka, Dokl. Sib. Knof., 2nd.  1964.   0.  V.  Makew.   Buryat.  Khizhn.
Izd. Ulan-Ude, USSR.

Kuznetsov, Y. V., Z. N. Simonyak, A. P.  Lisitsyn, and M.  S.  Frenklikh.   1968.
Thorium Isotopes (230Th, 232TH) in the Surface Layer of the  Indian Ocean
Sediments.  Geochem. Int.  5:169-177.

Latimer, W. M.  1952.  The Oxidation States of the Elements  and Their Potentials
in Aqueous Solutions.  Second Edition.  Prentice-Hall,  Inc., Englewood  Cliffs,
NO.

Legin, V. K., Y. V. Kuznetsov and K. F.  Lazarev.  1966.  Uranium Occurrence
in Marine Sediments.  Geokhimiya No. 5:606-608 (in Russian).

Magne, R., J. R. Berthelin, and Y. Dommergues.  1974.  Solubilization and
Insolubilization of Uranium from Granites by Heterotrophic Bacteria.   IN;
Formation of Uranium Ore Deposits.  IAEA-SM-183/18,  p.  73.

Manskaya, S. M., G. V. Drozdora, and M.  P. Yemel'yanova.   1956.  Fixation  of
Uranium by Humic Acids and Melanoidins.   Geokhimiya.  No. 4.

Masuda, K. and T. Yamamoto.  1971.  Studies on Environmental Contamination by
Uranium.  II.  Adsorption of Uranium on  Soil and Its Desorption.  0.  Radiat.
Res.  12:94-99.

Menzel, R. G.  1968.  Uranium, Radium, and- Thorium Content in  Phosphate Rocks
and Their Possible Radiation Ha.zard.  J. Agr. Food Chem.   16:231-34.

Mihalik, P.  1968.  Uranium Compounds in the Dominion Reefs, and Their  Associa-
tion with Phosphorus Compounds.  NIM-415.

Mizuno, A. and T. Mochizuki.  1970.  Distribution of Uranium in the Sediments
from San'in Off-Shore, Southwestern Japan.  Chishitsu Chosasho  Geppo.
21:287-292  (in Japanese).

Mo, T., A. D. Suttle, and W. M. Sackett.  1973.   Uranium Concentrations in
Marine Sediments.  Geochim. et Cosmochim. Acta.   37:35-51.

Naumov, G. B.  1961.  Some Physico-Chemical Peculiarities of Uranium Behavior
in Hydrothermal Solutions.  Geokhimiya.   No. 2:115-132 (in Russian).

Palei, P. N. (comp.).  1962.  Analytical Chemistry of Uranium.   Trans,  by
N. Kaner.  1963.  Israel Program for Scientific  Translations Ltd. 421 p.

Racon, D.  1973.  The Behavior in Underground Environments of Uranium and
Thorium Discharged by the Nuclear Industry.  H^:  Environmental Behavior of
Radionuclides Released in the Nuclear Industry.   IAEA-SM-172/55.  pp. 333-346
(in French).

Ritchie, J. C., P. H. Kawks, and J. R. McHenry.   1972.   Thorium, Uranium,  and
Potassium in Upper Cretaceous, Pateocene, and\Eocene Sediments  of the Little
Tallahatchie River Watershed in Northern Mississippi.  Southeast Geol.
14:221-231.
                                    3-241

-------
Rozhkova, Ye.V., Ye. G. Razumnaya, M. B. Serebryakova and 0. V. Shchebak.
1959.  Role of Sorption in Concentration of Uranium in Sedimentary Rocks.
Tr. II.  Mezhdunar, knof. po miro nmu ispol'z. atom, energii. 3.

Rubtsov, D. M.  1972.  Thorium and Uranium Content in the Clay Fraction of
Podzolic Mountain Soils of Thin Forests.  Radioekol. Issled Prir.
Bio'geotsenozakh.  pp. 53-66 (in Russian).

Sakanoue, M.  1960.  Geochemical Studies on the Radioactive Sediments.  II.
Uranium Content of Natural Waters from the Ningyo-Pass Mining Area.  Nippon
Kagakii Zasshi 81:896-898.

Sakanoue, M.  1960.  Geochemical Studies on the Radioactive Sediments.  III.
Uranium, Phosphorus, and Arsenic in the Sedimentary Bed at Ningyo Pass.
Nippon Kagaku Zasshi 81:898-902.

Schulz, R. K.  1965.  Soil Chemistry of Radionuclides.  Health Physics.
11:1317-1324.

Seaborg, G. T. and J. J. Katz (ed.).  1954.  The Actinide Elements.  McGraw
Hill Book Company, Inc.

Sillen, L. G. and A. E. Martell.  1964.  Stability Constants of Metal Ion
Complexes.  Second ed.  Spec. Pub!. No. 17.  The Chemical Society, London.

Starik, I. Ye., F. Ye Starik and A. N. Apollonova.  1958.  Adsorption of
Traces of Uranium on Iron Hydroxide and Its Desorption by the Carbonate
Method.  Zh. Neorgan. Khimii. 3(1).

Szalay, A.  1954.  The Enrichment of Uranium in Some Brown Coals in Hungary.
Acta Geol. Acad. Sci. Hungary.  2:299-311.

Szalay, A.  1957.  The Role of Humus in the Geochemical  Enrichment of U in
Coal and Other Bioliths.  Acta Phys. Acad. Sci. Hungary.  8:25-35.

Udaltsova, N. I.  1963.  General Information on Uranium.  IN:  Analytical
Chemistry of Uranium.  Israel Program for Scientific Translations,  pp. 3-7.

Vinogradov, A. P.  1959.  The Geochemistry of Rare and Dispersed Chemical
Elements in Soils.  Consultants Bureau.

Weast, R. C., Editor.   1976.  Handbook of Chemistry and  Physics.  The Chemical
Rubber Co., Cleveland, Ohio.  pp. B344-B345.

Yakobenchuk, V.  F.  1968.  Radioactivity and Chemical  Properties of
Sod-Podzolic Soils in the Ukram'an Western Polesie.  Visn. Sil's Kogosped.
Nauki.   11:45-50 (in Ukranian).

Yamamoto, T., E. Yunoki, M.  Yamakawa, and M. Shimizu.   1973.  Studies on
Environmental  Contamination  by Uranium.   3.   Effects of  Carbonate Ion on
Uranium Adsorption to and Desorption from Soils.   J. Radiat. Res.  14:219-224.


                                   3-242

-------
ZIRCONIUM
Natural Soil and Rock Distributions
     The zirconium distribution and content in igneous and sedimentary rocks
are shown in Table 3-101.   It should be noted that much of the zirconium con-
tained in sedimentary rocks is present as resistate material  such as zircon
(ZrSiO^), which is very slightly soluble in any aqueous solution.  The average
content of worldwide soils is given by Vinogradov (1959) as 300 ppm zirconium.

            TABLE 3-101.  THE AVERAGE ZIRCONIUM CONTENT OF ROCKS
                          (TUREKIAN AND WEDEPOHL, 1961)

                 Rock Type	                        Zr, ppm
                                   Igneous Rocks
             Ultramafic                                   45
             Basaltic                                    140
             Granitic, high Ca                           140
             Granitic, low Ca                            175
             Syenite                                     500
                                 Sedimentary Rocks
             Shale             '                          160
             Sandstone                                   220
             Limestone-dolomite                           19
Brief Chemistry
     Eighteen isotopes and two isomers of zirconium are known.   The naturally
                                                                     nc
occurring isotopes are listed in Table 3-102.   All  are stable except   Zr which
has a half-life listed as >3.6 x 10   years (Weast, 1976).   For all  practical
          96
purposes,   Zr also is a stable isotope.
                                                                95
     The immediate concern in radioactive waste disposal  is for   Zr with a
relatively short half-life of 65 days.  Also obtained in  much smaller yields
                                         QQ                      c
as a fission product is the longer-lived   Zr (half-life  1.5 x  10  years)
                                                             95
which is of long-term interest (Schneider and Platt, 1974).    Zr decays  by
 -             95                       -                      95
6  emission to   Nb which also emits a 6  and decays to stable    Mo.
                                    3-243

-------
            TABLE 102.  NATURAL ISOTOPES OF ZIRCONIUM AND THEIR
                        ABUNDANCES (WEAST, 1976)

                Isotope                          Abundance, %
                  90Zr                             51.46
                  91Zr                             11.23
                  Q2
                  y^Zr                             17.11
                  94
                  *^Zr                             17.40
                  96Zr                              2.80
     The most stable and common form of zirconium in rocks is as the simple
orthosilicate zircon, ZrSiO*.  The only oxidation state of zirconium of any
                                                                   +4
importance in aqueous solutions is Zr(IV), though usually not as Zr   cations,
                                                       o
Due to the high charge and a small atomic radius (0.72 A in six-fold coordi-
             o
nation, 0.84 A in octahedral coordination, Shannon and Prewitt, 1969), zir-
conium ions are readily hydrolyzed in aqueous solutions.  When an acid solu-
tion is neutralized with a base such as NH.OH, hydrous zirconium oxide is
precipitated rather than zirconium hydroxide (Blumenthal, 1958).  Zirconium
compounds, such as ZrCl., react violently with water, producing high acidity
solutions.  A zirconyl salt (ZrOCl^'SHpO) is produced with liberation of
hydrogen ions.  A solution 0.05M in ZrOCl9-8H«0, for example, has a pH of 1.
                                 +4                4-n     +
The equilibrium established is Zr   + nH20 -»• Zr(OH)p   + nH  and depends on
the solution acidity.
     According to Connick and McVey (1949), in 2.0 to 0.2M perchloric acid at
25°C, zirconium exists as Zr+  and Zr(OH)   ions.  At concentrations of per-
                                                                         +2
chloric acid of less than 0.2M, the zirconium species was probably Zr(OH)2 .
At or above 2.0M HC104, zirconium exists exclusively as Zr   ions (Larsen and
Wang, 1954).  When zirconium hydroxides are acidified with HC1, oxo cations
    +2
(ZrO  ) first appear, and not until the solution is 2N in HC1 or higher, do
                         +4
appreciable amounts of Zr   ions appear (Lister and McDonald, 1951).
     Zirconium forms very stable complexes with halogens of the type M-ZrGg
where M is a univalent metal and G is a halogen.  Zirconium also forms a
series of complexes with fluoride phosphate and sulfate ions and many of the
organic ligands.  Electrolysis of a solution of ZrfSO^^HgO, for example,
produces hydrogen at the cathode and zirconium sulfate at the anode.  Hence,

                                   3-244

-------
the compound was H2 [ZrO(S04)2], with zirconium in the complex anion
[ZrO(SOA)9]  .   Since most of the waste solutions containing zirconium are
                             +4
near-neutral solutions, no Zr   ion should exist in solution.  There are only
charged to neutral polymers and complexes of zirconium existing  in  the pH
range of most radioactive waste solutions.
Solid Phase and Solution Equilibria
     The zirconium minerals in increasing order of stability are:   Zr(OH)4,
Zr02, and ZrSiO^, as shown in Figure 3-28.  The thermodynamic data  for Zr(OH)4,
Zr02 and ZrSiCL were obtained from Sillen and Martell (1964), Baes  and Mesmer
(1976) and Wagman et al. (1971), respectively.  The most stable mineral
throughout the pH range is zircon (ZrSiO.).  This is consistent with the
empirical observations of many geologists and soil scientists who find that
zircon (ZrSiO*) is ubiquitous in soils and is very stable even compared to
other common soil minerals.
                    -8
                   -10
                   -12
                   -14
                 S1
                   -16
                   -18
                   -20
                   -22
                                       6
                                       PH
       Figure 3-28.  The stability of zirconium  solids at 25°C and  in
                     equilibrium with soil silica  (pHUSiCty  = 3.1).
                                    3-245

-------
     The activity of various solution species with the change in pH and in
equilibrium with zircon (ZrSiO^) and soil Si02 is given in Figure 3-29.  The
thermodynamic data for all the hydrolysis species were selected from Baes and
Mesmer (1976).  Other solution species data were selected from Si 11 en and
MarteH (1964).  Zirconium forms complexes with the common soil am'ons (F~,
  -     2-     -
Cl", SOA  , NO,").  The significance of complexation of zirconium with these
       *      **                                  2-
anions decreases in the following order:  F , SO,  , Cl , NO, .  Although
                                                      3+
zirconium forms complexes with common soil anions, ZrF   complex would be the
only one which could contribute significantly in acidic environments (pK 3.5)
to the total zirconium in solution.  It can be seen from the diagram that the
uncomplexed zirconium (Zr  ) would exist in significant amounts only in very
acidic solutions (pH <1).  The zirconium ions are hydrolyzed in solutions with
pH values of greater than 1.0.  The most dominant zirconium solution species
                                                  3+         °
with the conditions assumed in Figure 3-29 are ZrF  , Zr(OH)4 , and Zr(OH)5"
in pH ranges of <3.5, 3.5-6.25, and >6.25, respectively.
                  s
                    -2
                    -16
                    -20
                    -24
                    -23
                                   ZrtOHI.
                     Z    3     4    5    6
                                      PH •  '
                                              D
Figure 3-29.  The activity of various zirconium species in equilibrium with
              zircon [ZrSiO^s)] and soil solution silica (pfyS^ = 3.1) at
              pCT = pS042~ = 2.5, pN03~ = 3.0 and pF" = 4.5.
     If a waste solution containing zirconium is assumed to be in equilibrium
with zircon and the conditions assumed in Figure 3-29, then the total  activity
                                               -14
of zirconium in solution would be lower than 10    over a pH range of 3.5 to 8.
                                    3-246

-------
     The Zr   ion becomes increasingly hydrolyzed above pH 1  (Figure 3-29).
      4-f
The Zr '  content of the soil solution falls rapidly, as does  the content of
fluoride complex ZrF  , in favor of hydrolyzed species at higher pH values.
Only at relatively low pH values of less than 4 do positively charged species
of zirconium predominate.  Because most soils and rocks buffer waters contact-
ing them to pH 5 or higher, ion exchange probably plays a relatively minor
role in zirconium retention by soils and rocks.  Because of the formation of
negatively charged complexes in solutions of pH >6.5 (Figure  3-29), zirconium
adsorption would be expected to decrease with an increase in  pH above 6.5.
Experimental Adsorption Results
     Rhodes (1957) determined the effect of pH on the distribution coefficient
of zirconium-niobium on a Hanford soil.  The soil was. made up of 2% CaC03 with
>2 mm diameter material, 6%; 2 to 0.2 mm diameter material, 67%; 0.02 to
0.2 mm diameter material, 19%; 0.02 to 0.002 mm diameter material, 6%; and
<0.002 mm diameter material, 2%.  The cation exchange capacity was 5 meq/100 g
and the water paste pH was 8.6.  The bulk of the soil consisted of quartz,
feldspars, augite, olivine and mica.  The clay fraction contained mainly mont-
morillonite.  The Kd values for zirconium as a function of pH are given in
Table 3-103.
        TABLE 3-103.   ZIRCONIUM-NIOBIUM Kd VALUES AS A FUNCTION OF pH
                      (RHODES, 1957)
fitL.
0
i
2.6
4.1
6.4
7.7
Kd, ml/g
82
1028
1340
>1980
>1980
>1980
PH
8,3
9.6
10.2
11.0
12.2
' 14.0
Kd, ml/q
282
90
104
180
>1980
>1980
     Rhodes considered that the dips in zirconium Kd for pH 8 to 12 were due to
a charge alteration of the zirconium polymer or colloid within this pH range.
     Dlouhy (1967) gave Kd values for zirconium on Casaccia soil and Casaccia
tuff from CNEN, Italy.  The soil and tuff were not characterized.  The range of
zirconium Kd values given for the soil was 130 to 150 ml/g and for the tuff,
260 to 350 ml/g.
                                    3-247

-------
     Benson (1960) reported a visible, white precipitate in systems containing
      -2                                  95
3 x 10  M zirconium.  The uptake of trace   Zr was not appreciably different
from that noted in the presence of carrier indicating that precipitation also
occurred at the trace concentration.  Zirconium was not appreciably affected
by the presence of other cations unless the pH was below 2.  Benson also
reported that carbonate, oxalate and citrate ions inhibited zirconium uptake
on soil over a wide pH range, probably because of the formation of complex
anionic species.
     Sorathesn et al. (1960) stated that 95Zr behaved like a colloid at pH 6
to 9.  They reported the zirconium Kd on illite at 7 days equilibration time
to be 24,470 ml/g.
                                                          -3      -2
     Prout (1959) reported that tests made at pH 4 with 10  M C^O^   showed
adsorption of zirconium-niobium on soil to be inhibited, but at pH 8, the
oxalate did not seriously affect the removal of zirconium-niobium.  Schulz
(1965) reported that zirconium and niobium were two of the elements that were
adsorbed by soil in a very immobile form.
     Arnold and Grouse (1971).determined the adsorption of fission products on
copper ore produced by nuclear fracturing.  The copper leach' solution was
pH 1.2 to 1.8 with FLSO, and at 85°C.  The copper ore was in 2.5 kg, 4-1-n.
columns with 2000 g of radioactive rubble from a nuclear shot added between
500 g layers of minus 1/2-in. copper ore.  Sulfuric acid was added during the
run to assure that pH values stayed below 2.  Almost none of the   Zr-  Mb
was found in the column effluent because it was strongly adsorbed by the ore.
                                                                95   95
When a second run'was made with the fission products, including   Zr-  Mb
added in soluble form, the adsorption results were the same.  Zirconium Kd
values of 50 to 60 ml/g were obtained on batch equilibria from pH 1.5 to
pH 2.8.
Migration Results
     Field Studies—
     Spitsyn et al. (1958) used an alkaline solution (4 to 8 g NaOH/1, 200 g
NaN03/l) and an acidic solution [6 to 8 g HN03/1, 200 g A1(N03)3/1] in labora-
tory and field studies of radionuclide migration rates.  Zirconium was
                                    3-248

-------
adsorbed and retained by a sand in field tests near the point of injection of
the waste solutions for both the acidic and alkaline solutions.  Little move-
ment was indicated even under acidic conditions.
      Magno  et  al.  (1970)  investigated  the  radionuclides  from the Nuclear  Fuel
 Services  plant in  western New York  that were  migrating through the  plant
                                                                        95
 effluent  lagooning system.   They  estimated that greater  than 90% of the   Zr
 discharged  from the plant was deposited by sedimentation in  the lagoon system.
      Brookins  (1976)  reported that  fission product zirconium was retained in
 the shale very near to  areas of generation in the  natural  reactor at Oklo.
      Laboratory Studies-
      Fallout material on  soil  from  an  underground  nuclear  explosion was used
 by Essington et al.  (1965)  to determine the leaching  effects of water, HC1 and
 chelating agents.   The  soil  was suspended  in  25 ml  of solution and  equilibrated
                            -4
 for up to 106  days.   The  10   molar chelating agents  in  solution included
 sodium-diethylenetriaminepentaacetate,  sodium-cyclohexane-1,2-diaminetetra-
 acetate,  sodium-ethylenediamine di(o-hydroxyphenylacetate) and 0.1M HC1.
   Zr-  Nb was  identified  in the original soil.   Chelating  agents, HC1 and
 water equilibrations  did  not appreciably increase   Zr-  Nb  solubility.   In
 column studies with Yolo  soil  (pH 7.9), Sorrento soil  (pH  7.8)  and  Hanford
 soil  (pH  6.6), water  and  calcium-diethylenetriaminepentaacetate (10" M) leaches
 of 7 days produced no detectable    Zr-  Nb in the  effluent solution.  The domi-
 nant clay minerals in the Hanford,  Sorrento and Yolo  soils were illite, kao-
 linite and  montmorillonite.
      According to  Schulz  (1965),  the zirconium in  soils  is either very strongly
 adsorbed  by the clay  particles or is present  as an insoluble oxide  or hydroxide.
 In either case, the zirconium in  the soil  is  immobile.
      Eichholz  et al.  (1967)  investigated the  partitioning  of dissolved radio-
 nuclides  between suspended sediment particles and  aqueous  solutions.  Several
 natural water  samples were characterized and  used  in  the adsorption studies  as
 suspended solids sources  (see Table 3-30).  A fission product mixture was
 added to  the natural  water samples, the system equilibrated  and the water
                                            95                95
 recovered and  filtered.   The adsorption of   Zr and daughter   Nb are shown
                                     3-249

-------
in Table 3-104.  A large portion of the zirconium was associated with the par-
ticulate matter in a highly concentrated form.

       TABLE 3-104. 'ADSORPTION OF 95Zr AMD 95Nb ON SUSPENDED SOLIDS
                     IN NATURAL WATERS (EICHHOLZ ET AL., 1967)
                                        95                 95
                          Suspended       Zr Adsorption,     Nb Adsorption,
   	Source	   Solids, mg/l,   	%	   	%	
   Colorado River            229              61.1               33.3
   Camp McCoy                 12              50.0                4.7
   Bayou Anacoco              24              80.5.               23.8
   Lodgepole Creek           965              81.3
   Chattahoochee River       131              88.4               18.9
   Billy's Lake                8              53.9                8.9

     Adsorption does not appear to be a direct function of suspended solids
content but has to include a consideration of dissolved solids content, pH and
solids composition before the adsorption results begin to make any sense.  The
     95
high   Zr adsorption on Chattahoochee River suspended solids was due to the
low dissolved solids content (31 ppm), which suggests that a positively charged
zirconium species was involved.
                                                             95   95
     Harrison (1969), using nuclear debris contaminated with   Zr-  Nb as a
source, approached equilibrium from the direction of starting with the radio-
nuclide on the solid.  The leaching produced   Zr-  Nb Kd values of 5,200 to
17,000 ml/g in distilled water as a function of particle size range (<4000 to
<62 urn), 790 to 2700 ml/g in synthetic seawater for the same particle size
range and 270 to 1200 ml/g in 1M ammonium acetate for that particle size range.
An implication of this work is that cation exchange is involved because the
95   95
  Zr-  Nb Kd values are sensitive to competing ion concentrations.
Summary
     Trace concentrations of zirconium are strongly adsorbed on soils at pH 1
or higher with a dip in Kd values .between pH 8 to 12 (Rhodes, 1957), probably
due to the prevalence of uncharged or anionic zirconium species over this pH
range (Figure 2-14).  There is some evidence for affects caused by competing
cations at near-neutral pH, suggesting at least partial removal by an ion
                                   3-250

-------
exchange mechanism of polymerized or colloidal  species (Eichholz et a!., 1967;

Harrison, 1969).  Zirconium is well-adsorbed by the soil  over the normal pH
range of 4 to 8 (Spitsyn et al., 1958; Schulz,  1965; Magno et al., 1970).

Carbonate, oxalate and citrate complexes of zirconium have been reported to

migrate rapidly through soil columns (Bensen, 1960).

References

Arnold, W. D. and D. J. Grouse.  1971.  Radioactive Contamination of Copper
Recovered from Ore Fractured with Nuclear Explosions.  ORNL-4677.

Baes, C. F., Jr. and R. F. Mesmer.  1976.  The  Hydrolysis of Cations.   John
Wiley and Sons, New York.

Benson, D. W.  1960.  Review of Soil Chemistry  Research at Hanford.  HW-67201.

Blumenthal, W. B.  1958.  The Chemical Behavior of Zirconium.  Van Nostrand
Company, Inc.

Brookins, D. G.  1976.  Shale as a Repository for Radioactive Waste:  The
Evidence from Oklo. Environmental Geology.   1:255-259.

Connick, R. E. and W. H. McVey.  1949.  The Aqueous Chemistry of Zirconium.
J. Am. Chem. Soc.  71:3182-3192.

Dlouhy, Z.  1967.  Movement of Radionuclides in the Aerated Zone.-  IN:
Disposal of Radioactive Wastes into the Ground.  IAEA-SM-93/18.

Eichholz, G. G., T. F. Craft, and A. N. Galli.   1967.  Trace Element Fractiona-
tion by Suspended Matter in Water. Geochim. et  Cosmochim. Acta.  31:737-745.

Essington, E. H., H. Nishita and A.  J. Steen.  1965.  Release and Movement of
Radionuclides in Soils Contaminated with Fallout Material from an Underground
Thermonuclear Detonation.  Health Physics.   11:689-698.

Harrison, F. L.  1969.  Radioactive Debris  from Underground Nuclear Explosions:
1.  Physical and Chemical Characteristics,  2. Biological  Availability of the
Radionuclides to Aquatic Animals.  UCRL-5Q596.

Magno, P., T. Reavy, and J. Apidianakis.  1970.  Liquid Waste Effluents from
a Nuclear Fuel Reprocessing Plant.  BRH-NERHL-70r2.

Larsen, E. M. and P. Wang.  1954.  The Ion  Exchange of Zirconium and Hafnium
in Perchloric Acid with Amberlite IR-120.  J. Am. Chem. Soc.  76:6223-6229.

Lister, B. and L. McDonald, 1951.  Some Aspects of the Solution Chemistry  of
Zirconium.  AERE-C/R-801.

Prout, W. E.  1959.  Adsorption of Fission  Products by Savannah River Plant
Soil.  OP-394.

                                    3-251

-------
Rhodes, D. W.'  1957.  The Effect of pH on the Uptake of Radioactive Isotopes
from Solution by a Soil.  Soil Science Society America, Proceedings.
21:389-392.

Schneider, K. J. and A. M. Platt.  1974.   High-level Radioactive Waste
Management Alternatives.  BNWL-1900, Vol. !.

Schulz, R. K.  1965.  Soil Chemistry of Radionuclides.   Health Physics.
11:1317-1324.

Shannon, R. D.  and C. T. Prewitt.  1969.   Effective Ionic Radii  in Oxides and
Fluorides.  Acta Cryst. B. 25:925.

Sillen, L. G. and A. E. Martell.  1964.  Stability Constants of Metal-Ion
Complexes.  Special Publication No. 17.  The  Chemical  Society, London.

Sorathesn, A.,  G. Bruscia, T.  Tamura, and E.  G. Struxness.   1960.   Mineral
and Sediment Affinity for Radionuclides.   CF-60-6-93.

Spitsyn, V. I., V. D. Balukova, A.  F. Naumova, V.  V. Gromov, F.  M. Spiridonov,
E. M. Vetrov, and G. I. Gravfov.  1958.  A Study of the Migration of  Radio-
elements in Soils.  IN:  Proc. Second Annual  U.N.  Conf. Peaceful Uses of
Atomic Energy.   18:439-448.

Turekian, K. K. and K. H. Wedepohl.  1961. Distribution of the Elements in
Some Major Units of the Earth's Crust.  Bull. Geol. Soc. Am. 72:175-192.
    •
Vinogradav, A.  P.  '1959.  The Geochemistry of Rare and Dispersed Chemical
Elements in Soils.  Consultants Bureau.

Wagman, D. D.,  W. H. Evans, V. B. Parker, I.  Halow, S.  M.  Bailey,  R.  H.  Schumm,
and K. L. Churney.  1971.  Selected Values of Chemical  Thermodynamic  Properties.
Tables for Elements 54 Through 61 in the  Standard  Order of Arrangement.   NBS
Technical Note  270-5.

Weast, R. C.  1976.  Handbook of Chemistry and Physics.  The Chemical  Rubber
Co.
                                     3-252

-------
                                 SECTION 4
                         CONCLUSIONS AND EVALUATION

     The available data on interaction of various elements with the geologic
media were collected and discussed in Section 3.  At the end of discussions on
each element a summary is presented.  The information presented in Section 3
will not be repeated here.  This section is concerned with general conclusions
based upon Section 3.  For more details the reader is directed to the previous
sections.
     The solution ions exist in dynamic equilibrium with soils, sediments, and
rocks.  Many factors that are inherent to the solid matrix and the solution
can influence the concentration of the element and its species in solution.
These factors include pH, Eh, CEC, type and amount of soil minerals, solid
phases of the element, complexing ligands, and competing ions.  In order to
determine the distribution of the element between solution and solid and its
reactions with the solid matrix, the identity and relative magnitude of the
various factors are required.  Such information would help predict the solu-
tion concentrations of the element and help extrapolate the results to other
situations.  However, the data presented in Section 3 indicate that there is
a general lack of systematic evaluation of various factors that determine
element-solid matrix interactions, and no information at present is available
to determine the magnitude of the various factors.  At best, the available
data suggest trends of the influence of some of the factors that control solu-
tion concentrations and interaction with solid matrices.  This type of informa-
tion would not be very useful for precisely predicting the general fate of
radionuclides in the environment.  Additional research required for general
understanding of the behavior of radionuclides is outlined at the end of this
section.
                                     4-1

-------
FACTORS INFLUENCING ADSORPTION OF RADIONUCLIDES ON GEOLOGIC MEDIA
     Some of the factors that have been reported to influence the adsorption
of elements by geologic media are listed for selected elements in Table 4-1.
An "X" indicates that a given factor was reported to influence adsorption of
that element (present as a radionuclide).  The results reported in Table 4-1
are qualitative only.  In most cases, the authors did not definitively show
that a specific factor was operative, but hypothesized that it was.  Since
these data cannot be used for quantitative predictions, they are mainly pre-
sented for use as guidelines in future experiments.  The probable principal
adsorption mechanisms were deduced from an examination of the factors reported
for each element.  For example, colloid formation is a good indication of pre-
cipitation, as pH also may be.  An adsorption sensitivity to soil cation
exchange capacity, competing ions and system pH usually indicates an ion
exchange adsorption mechanism.  There is, admittedly, a certain amount of
value judgment that went into the summation process because in some cases the
literatur-e was ambiguous and vague concerning adsorption-influencing factors,
and in other cases, the experimental work has yet to be done.  Neptunium and
technetium literature data could not be used to determine a probable adsorp-
tion mechanism..  Adsorption of'iodine by organic matter is listed as the
adsorption mechanism, but various authors gave conflicting results.  How-
ever, the majority of the workers reported organic matter to be the principal
adsorption media  for iodine.
     In addition to concentration, the nature of the solution species would
have, a tremendous effect on the radionuclide interaction with the geologic
media and on the  mobility of the radionuclides.   Very few data are available
in the literature delineating species that have been identified in natural
environments.  Thermodynamic data were used (Section 3) to develop solution
species diagrams  to predict the predominant solution species.  The limitations
and drawbacks of this approach are discussed in Section 2.  The predicted
nature of the predominant solution species of various elements are reported in
                                                                    •*
Table 4-2.  The environmental conditions assumed for Table 4-2 data include a
pH range of 4 to  9, p02 range of 0.68 to 80 atmospheres, pC02 of 1.52 to 3.52
atmospheres, pd" = pN03" = pSOJ£  = 3.0, pF" 4.5, and pH^O^ 5.0.  It should
be pointed out that a change in environmental  conditions, in addition to the

                                     4-2

-------
      TABLE 4-1.   FACTORS REPORTED TO EFFECT ADSORPTION OF  RADIOELEMENTS OVER THE
                   pH RANGE OF  4 TO 9
                                      Factors
Soil
Element pll Eh CEClO
Am
Sb
Ce
Cs
Co
Cm
Eu
I
Np
Pu
Pm
Ra
Ru
Sr
Tc
Th
3H
U
Zr
X
X
X
X
X

X


XX X
X
X
X
X X
X X

XX X
X


Competing Selectively Inorganic
. Ions Adsorbed On Llgands
X
Iron Oxides

Zeolites,
Micas
Illlte,
Iron Oxide

X
OM


X .
X Zeolites,
Bar He
OM
X • Calclte,
Zeolites
OM
H20
OM

X
X
X

X
X
X

X
X


X
X
X

X
X
Complex Ions
Organic
Constituents
X
X

X


X

X
X

X
X
X
X

X
X

Colloid Probable Adsorp-
Formatlon tlon Mechanlsmsl?)
IE
X PPT
X TE. PPT
IE
X IE, PPT
X PPT
IE, PPT
OM
X UNK
X IE, PPT
IE, PPT
IE
PPT
IE
UNK
X IE, PPT
NONE
PPT, IE
X PPT
1.   CEC  = Cation Exchange Capacity
2.    IE  = Ion exchange, OM = Organic matter adsorption, PPM =? Precipitation, UNK = Unknown.

-------
          TABLE 4-2.  PREDOMINANT SOLUTION SPECIES OF ELEMENTS
                                                              (a)
El ements
Am
Sb
Ce
Cs
Co
Cm
Eu
I
Np
Pu
Pm
Ra
Ru
Sr
Tc
Th
U
Zr
Little Affected by In An In A
Oxidation-Reduction Oxidizing Environment Reducing Environment
Am3*, AmS04*. AM(OH)2*
HSb02°, Sb(OH)3°, SbOF\ SbO*
Sb(OH)4-
Ce3*, CeS04*
Cs*
Co2*, Co(OH)2*
Cm3*, CmOH2*, Cm(OH)2*
Eu3*, EuS04*. Eu2P20?2*
r, io3" r
Np02*. Np02HP04', NpOH3*, Np4*
- Np02HC03
Pu022*, Pu02(C03)(OH)|', PuOH2*,.Pu3*
Pn,3* .
Ra2*
Ru(OH)2*, Ru04", Ru042" Ru04"
Sr2*
Tc04' Tc2*
ThF3*, Th(OH)3*
H*. 3H-0-H
UO|*. U02F*. U02(OH2)8, U02,*, UOH3*, U02*.
UO,(CO,)i~ UO,(CO,),4"
6 J J ' fc O J
Zr(OH)4°, Zr(OH)s-,
ZrF3*
(a)  In a pH 4 to 9,  pOa 0.68 to  80,  pCOe 1.52 to  3.52,  pCl"  =  pNOs"  =
    pS042" =3.0, pF~ 4.5 and pH2P04"  5.0 environment without  organic
    ligands.
                                   4-4

-------
ligands not considered in the above assumptions, may change the nature of the
solution species.  Depending upon the given chemical environment, several
species of an element can exist in solution.  Cesium, cobalt, promethium,
radium, strontium, and tritium would be expected to exist predominantly as
uncomplexed solution species as Cs , Co  , Pm  , Ra  , Sr   and  H+.   Certain
elements such as plutonium, neptunium and uranium can exist in acidic solu-
                                                                    +     2+
tions in an oxidizing environment as uncomplexed solution ions (PuO~ , PuCL ,
   +    2+
NpO-, UOg ).  Some of the elements such as curium, thorium and zirconium
hydrolyze very readily even in acidic environments, so they mainly exist as
hydrolyzed species [CmOH2*, Cm(OH)t, Th(OH)t, Zr(OH)!, Zr(OH)"].   Some of the
                       n     n    <-        J        *T        D
common soil ligands (COg » S0^~) form strong complexes with americium, curium,
europium, neptunium, plutonium and uranium so that these ligands  would
influence the concentration of these elements in solutions.  Oxidation-
reduction conditions would predominantly influence the nature of  the  solution
species of antimony, neptunium, plutonium and uranium.
     Based upon the nature of the predominant solution species, qualitative
predictions regarding the adsorption and movement of various elements can be
made.  Soils and sediments mainly show cation exchange.capacities (since
materials carry a large net negative charge) and to a limited extent,
anion exchange capacities.  Thus, most cations migrate through the soil or
rock column at speeds slower than the groundwater.  Relative to each  other,
the trivalent cations generally move the slowest, the divalent cations at
intermediate velocities and the monovalent cations most rapidly.   Tritium is
unique in that it readily substitutes for hydrogen in water and migrates,
therefore, at the same velocity as water.  Complicating factors include a
higher selectivity of a soil or rock component for a given cation or  a more
successful hydrogen ion competition with one cation relative to another.  The
relative mobilities of strontium and cesium at low pH is a good example of
the latter effect.  Strontium is much more mobile at low pH than is cesium.
     The simple anions tend to migrate through soils and rocks with little
reaction because usually a pH of less than 4 is required to activate  a signifi-
cant soil anion exchange capacity.  However, both I and Tc which  exist pre-
dominantly as anions (I", TcOT) undergo other reactions with organic  ligands
that can greatly slow their migration.

                                     4-5

-------
     The migration and retention of inorganic complex species (mononuclear and
polynuclear) would also be dependent upon the charge and size of the species.
Positively charged species would move slower than the negatively charged
species.  The highly charged metal ions tend to polymerize or form colloidal
size, charged precipitates.  The trivalent metal  ions (Am, Cm, Sb, Ru) and
quadrivalent or higher charged metal ions (Pu, Th, U, Zr) are known to form
polymers.  Initially, at lower pH, the polymers are positively charged and at
pH 8 become increasingly negatively charged.  The result is good adsorption
and very slow movement at acidic to neutral  pH, and greatly lessened adsorp-
tion and rapid migration from pH 8 upward.
     The behavior of organic complexed species of elements is difficult to
predict because of the lack of knowledge regarding the exact nature of the
organic ligands, a wide variation in amounts and types of organic ligands,
and the size and solubility of these organics.  These organic materials can
either retain the element or complex it in a form that migrates readily.  The
source of these organic ligands can be the radioactive waste solutions con-
taining the synthetic organic ligands such as TBP and/or the organic ligands
produced by soil flora.
     The metal ions that take part in replacement reactions depend upon the
size and charge of the species involved and the ability of the final mineral
product to accept that species into its growing structure.  Apatite, for
example, accepts into its structure a wide range of metal ions such as
strontium, radium, uranium and probably americium and cerium.  Cobalt replaces
calcium in calcite, and radium and strontium replace barium in barite.  These
metal ions are buried in the replacement product crystal structure and are
effectively retained..unless a changing chemical environment causes instability
and dissolution of the crystal.

ADDITIONAL RESEARCH NEEDS
     The areas that require more work to better understand radionuclide inter-
actions with soil and rock media include:
 1.  Determination of radionuclide adsorption mechanisms.
                                     4-6

-------
     Additional work is required in the area of radionuclide adsorption
mechanisms (ion exchange, precipitation, coprecipitation).   There are many Kd
values that include several  soil adsorption mechanisms and  elemental  species
operating simultaneously.  Definitive experimental  data that would allow
deduction of probable soil or rock adsorption mechanisms for different elements
are lacking in the literature.  In addition, the environment in which a given
adsorption mechanism predominates should be delineated as well.
 2.  Factors that influence radionuclide-geologic media interactions.
     The influence of different factors (such as pH, Eh, complexing ligands,
competing ions, CEC, type and amount of soil minerals, solid phases of ele-
ment) on the magnitude and extent of adsorption of radionuclides by the
geologic media need to be evaluated.

 3.  An  experimental .consideration  of  the  effects of redox potential on radio-
     nuclide migration.
     At  depth  (below the soil humic  zone), where geologic storage has been
proposed, the  environment is  considerably  different from that contributing to
radionuclide migration on the land  surface.  There has been minimal examina-
tion of  the effects on radionuclide  migration of the relatively low oxygen
partial  pressures that will be  encountered as a result of geologic waste
storage  at depth.  Many  of the  igneous rocks comprising a proposed waste
repository contain phases such  as magnetite, amphiboles or pyroxenes with
ferrous  iron as a constituent that  tends to  "buffer" the system at a rela-
tively low oxygen partial pressure  until it  is all oxidized to ferric iron.
The equilibrium partial  pressure of  oxygen for the reaction, 2Fe304  (magnetite)
+  1/2 02 g » 3Fe203 (hematite)  at 25°C and one atmosphere pressure is 10"68'2,
a  reducing environment.   Of the 19  elements  studied a large change in redox
potential would be expected to  affect  the  oxidation states of antimony,
neptunium, plutonium,  ruthenium, technetium  and uranium, henoa their solution
species  and ultimately their  migration rate  through the surrounding rock.

-------
 4.  Characterization and influence of organic ligands on radionuclide
     migration.
     The existing thermodynamic data on species are incomplete in many cases,
and of dubious quality in other cases.  The thermodynamic data should be con-
firmed by experimental evidence on radionuclide adsorption and migration.
Thermodynamic data on the radionuclide complexes with natural soil and water
organic components are essentially nonexistent.  Note in Table 4-1 that radio-
nuclide reactions with organic material were reported for 12 of the 19 radio-
nuclides reviewed.  Hence, what may prove to be a most important influence on
radionuclide adsorption and migration is one of the least understood.  The
sources of organic materials in the environment capable of forming complexed
species with radionuclides include the many industrial wastes from various
manufacturing processes disposed to the atmosphere and surface waters, the
organics included in municipal sewage treatment effluents also disposed to
surface waters, the various pesticides and herbicides used during farming
operations contained in the atmosphere and irrigation runoff, humic substances
and decaying plant and animal remains already present in the soil and rocks
along with the associated living soil microflora.  With such varied sources
and types of organic ligands, a large quantity of work remains to be done in
even identifying the metal ligand species involved in radionuclide migration
on the land surface.
 5.  Thermodynamic data determinations for several  solids and solution species.
     Thermodynamic data are badly needed for several  solids and complex solu-
tion species of different elements.  For example, the ruthenium complexes
RuNO(N02)2OH(H20)2 and RuNON02(N03)2(H20)2 were reported as very stable and
commonly found in nuclear waste solutions by several  authors.  Yet no firm
thermodynamic data for these two complexes were located in the literature.
When the data are determined, they may drastically change the order and
speciation for ruthenium solution species reported in Section 3.
                                     4-8

-------
 6.  The generation of Kd data that can be used in a comparative mode.
     As mentioned earlier, an attempt was made to evaluate and select the
best Kd values from the literature in writing Section 3.  However, the experi-
mental conditions, including soil, rock and solution characterization, were
only sketchily given, or not given at all in most cases.  Hence, the Kd values
in Section 3 are the best available, but evaluation of these data by any
objective methodology would result in discarding a large number of them.  The
requirements for determination of a Kd value that could be used in a compara-
tive mode were listed in Section 2.  A program to determine such Kd values is
now underway by the Office of Waste Isolation of the Department of Energy.
 7.  Maintenance of a bank of such Kd data.
     A bank of the Kd data should be maintained and kept current.  The data
could be computer-stored and made available to all personnel concerned with
waste management upon request.  The present system has resulted in using
unevaluated Kd values that are often not applicable to the case in point.
     If Kd values for biotite, quartz and orthoclase are known, a computed
resultant Kd value for a granite, for example, should be possible if com-.
parable surface areas are equated and if the biotite, quartz, and orthoclase
content of the granite are known.  If the granite is composed of 25% quartz,
65% orthoclase and 10% biotite, and surface areas of minerals and rocks are
similar, then (0.25) (quartz Kd) + (0.65) (orthoclase Kd) + (0.10) (biotite Kd)
should equal the granite Kd value.  There is no evidence in the literature
that the above hypothesis has ever been tested.  Until Kd determinations are
placed on a comparable experimental basis, there would be little point in
attempting to test and apply it.
 8.  Determination of radionuclide adsorption and desorption reaction kinetics
     with soils and rocks.
     Several authors that have equilibrated solutions containing radionuclides
with soils or rocks for several weeks have reported that a considerable period
of time was required to attain equilibrium.  Kinetic studies 1) could indicate
radionuclide adsorption mechanisms, 2) could help in separating adsorption
rates due to different mechanisms, and 3) would help in increasing the  confi-
dence in extrapolating the predicted results into the future.

                                     4-9

-------
                               ACKNOWLEDGMENTS

     The authors would like to acknowledge the following persons who particv
pated in the review of this report:
     Mr. H. J. Morton
     Mr. W. K. Summers
     Dr. K. A. Kraus
     Dr. D. Isherwood
     Mr. G. Grisak
     Mr. G. L. Meyer
     Dr. D. C. Hoffman •
     Mr. J. P. Hamric
Other support was given by the EPA Office of Research and Development and
the Office of Energy, Minerals and Industry.
     The authors also wish to thank the following Battelle staff members
who assisted in the report preparation:  S. I. Thoreson, D. J. Kennedy,
M. V. Heid, and S. R. Gano.
                                    Ack-1

-------