• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Disclaimer
 Perspective
 Executive summary
 Table of Contents
 List of Figures
 List of Tables
 Acknowledgement
 Introduction
 Methods
 Results: Chemical data
 Radiochemical systems
 Sequential-data analysis
 Summary and conclusions
 Reference
 Lithologic logs of cores associated...
 Raw chemical data






Group Title: Radiochemistry of uranium-series isotopes in groundwater : : chemical fate of uranium-daughter radionuclides in recharge wells, central Florida phosphate district
Title: Radiochemistry of uranium-series isotopes in groundwater
CITATION PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/AM00000213/00001
 Material Information
Title: Radiochemistry of uranium-series isotopes in groundwater chemical fate of uranium-daughter radionuclides in recharge wells, central Florida phosphate district
Physical Description: 1 v. (various pagings) : ill. ; 28 cm.
Language: English
Creator: Upchurch, Sam B.
Southwest Florida Water Management District (Fla.)
Florida Institute of Phosphate Research
Publisher: The Institute
Place of Publication: Bartow Fla
Publication Date: 1991]
 Subjects
Subject: Uranium -- Environmental aspects   ( lcsh )
Radioactive pollution of water -- Florida   ( lcsh )
Floridan Aquifer   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
 Notes
Bibliography: Includes bibliographical references.
Statement of Responsibility: by Sam B. Upchurch ... et. al ; in cooperation with Southwest Florida Water Management District ; submitted to Florida Institute of Phosphate Research.
General Note: "Submitted: November 28, 1988, revised: March 14, 1991."
General Note: "Publication No. 05-022-092."
General Note: "May 1991"--Cover.
 Record Information
Bibliographic ID: AM00000213
Volume ID: VID00001
Source Institution: Florida A&M University (FAMU)
Holding Location: Florida A&M University (FAMU)
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 23860995

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Title Page
    Disclaimer
        Unnumbered ( 4 )
    Perspective
        Unnumbered ( 5 )
        Unnumbered ( 6 )
    Executive summary
        Page i
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
        Page vii
        Page viii
    List of Figures
        Page ix
        Page x
        Page xi
        Page xii
    List of Tables
        Page xiii
        Page xiv
    Acknowledgement
        Page xv
        Page xvi
    Introduction
        Page A 1
        Page A 2
        Page A 3
        Page A 4
        Page A 5
        Page A 6
        Page A 7
        Page A 8
        Page A 9
        Page A 10
        Page A 11
        Page A 12
        Page A 13
        Page A 14
        Page A 15
        Page A 16
        Page A 17
        Page A 18
        Page A 19
        Page A 20
        Page A 21
        Page A 22
        Page A 23
        Page A 24
        Page A 25
        Page A 26
    Methods
        Page B 1
        Page B 2
        Page B 3
        Page B 4
        Page B 5
        Page B 6
        Page B 7
        Page B 8
        Page B 9
        Page B 10
        Page B 11
        Page B 12
        Page B 13
        Page B 14
        Page B 15
        Page B 16
        Page B 17
        Page B 18
    Results: Chemical data
        Page C 1
        Page C 2
        Page C 3
        Page C 4
        Page C 5
        Page C 6
        Page C 7
        Page C 8
        Page C 9
        Page C 10
        Page C 11
        Page C 12
        Page C 13
        Page C 14
        Page C 15
        Page C 16
        Page C 17
        Page C 18
        Page C 19
        Page C 20
        Page C 21
        Page C 22
        Page C 23
        Page C 24
        Page C 25
        Page C 26
        Page C 27
        Page C 28
        Page C 29
        Page C 30
        Page C 31
        Page C 32
        Page C 33
        Page C 34
    Radiochemical systems
        Page D 1
        Page D 2
        Page D 3
        Page D 4
        Page D 5
        Page D 6
        Page D 7
        Page D 8
        Page D 9
        Page D 10
        Page D 11
        Page D 12
        Page D 13
        Page D 14
        Page D 15
        Page D 16
        Page D 17
        Page D 18
        Page D 19
        Page D 20
        Page D 21
        Page D 22
        Page D 23
        Page D 24
        Page D 25
        Page D 26
        Page D 27
        Page D 28
        Page D 29
        Page D 30
        Page D 31
        Page D 32
        Page D 33
        Page D 34
        Page D 35
        Page D 36
        Page D 37
        Page D 38
        Page D 39
        Page D 40
        Page D 41
        Page D 42
        Page D 43
        Page D 44
    Sequential-data analysis
        Page E 1
        Page E 2
        Page E 3
        Page E 4
        Page E 5
        Page E 6
        Page E 7
        Page E 8
        Page E 9
        Page E 10
        Page E 11
        Page E 12
        Page E 13
        Page E 14
        Page E 15
        Page E 16
    Summary and conclusions
        Page F 1
        Page F 2
        Page F 3
        Page F 4
    Reference
        Page G 1
        Page G 2
        Page G 3
        Page G 4
        Page G 5
        Page G 6
        Page G 7
        Page G 8
        Page G 9
        Page G 10
        Page G 11
        Page G 12
    Lithologic logs of cores associated with monitor wells
        Page H 1
        Page H 2
        Page H 3
        Page H 4
        Page H 5
        Page H 6
        Page H 7
        Page H 8
        Page H 9
        Page H 10
        Page H 11
        Page H 12
    Raw chemical data
        Page I 1
        Page I 2
        Page I 3
        Page I 4
        Page I 5
        Page I 6
        Page I 7
        Page I 8
        Page I 9
        Page I 10
        Page I 11
        Page I 12
        Page I 13
        Page I 14
        Page I 15
        Page I 16
        Page I 17
        Page I 18
        Page I 19
        Page I 20
        Page I 21
        Page I 22
        Page I 23
        Page I 24
        Page I 25
        Page I 26
        Page I 27
        Page I 28
        Page I 29
        Page I 30
        Page I 31
Full Text


Publication No. 05-022-092



RADIOCHEMISTRY OF
URANIUM-SERIES ISOTOPES IN
GROUNDWATER


FIPR

Florida Institute of
Phosphate Research


Prepared By
Departments of Geology and Physics
University of South Florida
Under a Grant Sponsored by the
Florida Institute of Phosphate Research
SBartow, Florida


May 1991


WE









The Florida Institute of Phosphate Research was created in 1978 by
the Florida Legislature (Chapter 378.101, Florida Statutes) and
empowered to conduct research supportive to the responsible
development of the state's phosphate resources. The Institute has
targeted areas of research responsibility. These are: reclamation
alternatives in mining and processing, including wetlands
reclamation, phosphogypsum storage areas and phosphatic clay
containment areas; methods for more efficient, economical and
environmentally balanced phosphate recovery and processing;
disposal and utilization of phosphatic clay; and environmental
effects involving the health and welfare of the people, including
those effects related to radiation and water consumption.

FIPR is located in Polk County, in the heart of the central Florida
phosphate district. The Institute seeks to serve as an information
center on phosphate-related topics and welcomes information
requests made in person, by mail, or by telephone.




Research Staff


Executive Director
Richard F. McFarlin



Research Directors

G. Michael Lloyd Jr.
Gordon D. Nifong
Steven G. Richardson
Hassan El-Shall
Robert S. Akins


-Chemical Processing
-Environmental Services
-Reclamation
-Beneficiation
-Mining


Florida Institute of Phosphate Research
1855 West Main Street
Bartow, Florida 33830
(813) 534-7160












RADIOCHEMISTRY OF URANIUM-SERIES ISOTOPES


IN GROUND WATER


(Chemical Fate of Uranium-Daughter Radionuclides in Recharge
Wells, Central Florida Phosphate District)



by

Sam B. Upchurch', C. R. Oural2
D. W. Foss1, and H. Ralph Brooker2


Departments of Geology' and Physics2
University of South Florida
Tampa, Florida 33620







in Cooperation with
Southwest Florida Water Management District
Brooksville, Florida



S Submitted To:
Florida Institute of Phosphate Research
Bartow, Florida






Submitted: November 28, 1988
Revised: March 14,. 1991


























DISCLAIMER


The contents of this report are reproduced as received from
the contractor.

The opinions, findings and conclusions expressed herein are
not necessarily those of the Florida Institute of Phosphate
Research, nor does mention of company names or products constitute
endorsement by the Florida Institute of Phosphate Research.








PERSPECTIVE


GORDON D. NIFONG, Ph.D.

FLORIDA INSTITUTE OF PHOSPHATE RESEARCH


It has long been known that elevated levels of uranium occur
naturally associated with the sedimentary phosphate deposits found
in central Florida. Mainly because of its low solubility, uranium
is not generally considered to be a major environmental hazard, but
many of the members within the uranium decay series are more of a
cause for concern. These would include radium-226, a radioactive
element chemically similar to calcium; radon-222, a gas that is
chemically inert but radioactive; several "short-lived" daughter
products of radon; and finally two longer-lived decay products --
lead-210 and polonium-210. All of the above are naturally
occurring radioactive materials that are ubiquitous in the
environment but tend to be elevated in phosphate-related materials.

Since its inception, the Florida Institute of Phosphate
Research has been interested in the environmental aspects of the
phosphate industry. It is believed that all phases of ore mining,
minerals processing and land reclamation can be accomplished in an
environmentally acceptable manner. Because of the array of radio-
nuclides found in phosphate ores, much of that concern for the
environment has been focused on the issue of radiation. Well over
a dozen projects have been conducted or sponsored that directly
address the topic of radiation, and numerous other projects have
had radiological components as secondary issues. Strong interest
exists not only in characterizing natural radionuclides as to their
nature, extent and magnitude, but also in determining their effects
on the population that lives and works in the phosphate region.
The Institute has addressed both concerns.

One of the great natural resources available to the citizens
of the State of Florida, but one perhaps too often taken for
granted, is a source of plentiful and clean water for domestic
consumption and other uses. For most of the people living in
central Florida, this source is groundwater, often obtained from
shallow, private wells,but more often obtained from public,
community wells tapping the Floridan aquifer. The supply is not
infinite, however, and continued growth and development of the area
depend in part on proper management of this resource, including
especially conservation measures and avoidance of pollution. The
central Florida phosphate industry requires a great deal of water
in the mining and processing of phosphate rock, even though great
strides have been made in recent years in the recycling of process
water. To mitigate water withdrawals from the Floridan aquifer,
the industry has frequently utilized recharge wells, designed to
transfer water from surficial zones to the deeper aquifer. Not
only does this lessen the overall withdrawal impact on the
Floridan, but it aids in the dewatering of land to be mined.









Assessing the quality of water has been a goal of several
Institute-sponsored studies. In 1981 the Institute sponsored a
study by the state Department of Health and Rehabilitative Services
to study radiochemical contamination in shallow drinking water
wells in the phosphate region. Later this study was expanded to be
statewide in scope. Further water quality studies, done mainly at
the University of South Florida and at Florida State University,
have looked in detail at the radiological components of
groundwater.

In earlier work, the results of which were published by the
Institute in 1985, Dr. Upchurch and his co-workers at South Florida
found that polonium-210, rather than uranium or radium, was the
dominant radionuclide in much central Florida groundwater,
especially that from surficial aquifers. That and other findings
in his study suggested that a more in-depth chemical and
radiological analysis of water from inter-aquifer connector
"recharge" wells, and from shallow and upper Floridan aquifers
monitoring wells, might provide a basis for better understanding
the chemistry of the waters. This in turn could lead to better
location and management of recharge wells, perhaps increasing their
acceptability. Thus, this current study was initiated.

Dr. Upchurch has now confirmed that polonium does indeed
account for most total alpha radiation in groundwater, and that
mostly it is found in surficial sources. Polonium in recharge
wells is seasonally related to water table fluctuations. High
levels of alpha activity are spatially related to fracture traces
in the earth's crust. Thus, recharge wells should be more
acceptable for use in the central Florida phosphate district if
several precautions are taken. They should be located to avoid
fracture traces or areas where leached zone material is thick and
the water is acidic. Sampling protocols should be designed to
avoid contamination from well casing slimes.

Chemical modeling and raw data clearly indicate that
conditions for transport of polonium and lead exist in the
surficial aquifer, but not in the Floridan aquifer. Therefore, it
appears that, with respect to the parameters measured, the quality
of receiving Floridan aquifer waters is not jeopardized by the
practice of interaquifer connector (recharge) wells.

A central theme that runs through much of the Institute's
environmental research is the evaluation of human exposure to
radiation dose as contributed by some phase of the natural
environment. This work seems consistent with the societal goal of
keeping radiation exposures to "as low as reasonably achievable."








EXECUTIVE SUMMARY


Chemical analysis of water from interaquifer connector wells
("recharge wells") and associated monitor wells in the Central
Florida Phosphate District has provided a basis for understanding
the chemistry of the Surficial and upper Floridan aquifers and of
recharge wells. The data are particularly important because they
suggest chemical controls on the mobility of uranium and uranium
progeny. Time-series and nearest-neighbor analyses of historical
data suggest hydrologic conditions that cause mobilization of iron,
sulfur, and radionuclides in ground water and provide a basis for
recharge-well location and management to minimize risks of failure
of water-quality criteria for gross-alpha radioactivity, iron, and
color.

Specific findings can be summarized as follows:

1. Twenty three recharge wells were sampled. A total of
51 water samples from recharge wells were analyzed for 31
chemical variables, including uranium-238, uranium-234,
radium-226, lead-210, and polonium-210. Three recharge
wells had associated Surficial and Floridan aquifer
monitor wells. Eighteen samples were analyzed from the
monitor wells.

2. Water of the Surficial aquifer is acidic, reducing,
and organic rich. Eh-pH data show that the aquifer water
is near the hydrogen sulfide-sulfate stability boundary,
and sulfides are the major sulfur species in the water.
Ferrous iron is the stable iron species. Polonium and, to
some degree, radium may be present in activities
sufficient to be of concern.

3. Floridan aquifer water is neutral to slightly alkaline
and slightly reducing. Sulfate is the dominant sulfur
species, and ferrous iron is stable. Activities of
radionuclides are low in the Floridan aquifer where
sampled, although radium has been shown to be a problem
in other studies.

4. Surficial aquifer water is highly variable and ranges
from a sodium-sulfate to sodium-calcium-bicarobonate-
sulfate compositions. Upper Floridan aquifer water
compositions do not vary significantly and the water is
a calcium-magnesium-bicarbonate solution.

5. In recharge wells, sulfur is oxidized to sulfate and
iron may be oxidized to the ferric state. Ferric
hydroxides precipitate if the system is oxidizing, but
most of the ferric hydroxide precipitate appears to form
on well casings and screens, not in the turbulent flow of
the well water. In general, recharge-well water closely









reflects the chemistry of Surficial-aquifer water, from
which it is derived.

6. Sampling trials indicate that much of the ferric
hydroxide and biomat that causes color and turbidity
violations in water quality is adherent on well casings.
Sampling with a tripod to prevent contact of the sampling
vessel with the walls of the well avoids contamination of
the sample with color- and turbidity-causing casing
slimess".

7. Sulfate reduction is initiated at reduction-oxidation
redoxx) potentials less than -150 mv.

8. Gross-alpha radioactivity measurement techniques are
not reliable measurements of true alpha activity because
of short-lived radon daughters that may be present in the
sample. Gross-alpha radioactivity should only be used as
a screening technique to determine if there is a
potential for radiation problems.

9. Uranium was low in all samples and is not a problem.
Uranium is mobilized as uranyl ion in oxidizing
conditions.

10. Radium was not a problem in most samples. Four
recharge wells had samples in excess of the 5 pCi/l
standard. Radium appears to be mobilized as a sulfate
complex. Strontium, and perhaps barium, appears to
enhance nucleation of radiocolloids of radium sulfate.
All samples were well below chemical equilibrium with
respect to particulate radium sulfate.

11. Lead-210 is present in all systems in minor
quantities. Soluble lead-210 does not support its
daughter polonium-210, so the aqueous polonium is derived
from lead sorbed or precipitated on aquifer materials.
Isotherms indicate that lead is strongly adsorbed on
clays and is precipitated in contact with ferruginous
quartz sand and limestone. Lead is slightly mobile in the
Surficial aquifer owing to the low pH of the water. In
the Floridan aquifer it is controlled by precipitation of
lead carbonate (cerussite). Lead-210 activity is not
correlated to total lead concentration.

12. Polonium-210 is most abundant in the Surficial
aquifer where it constitutes most of the gross-alpha
radioactivity. Polonium mobility is controlled by
solubility of the radiocolloids formed by completing of
polonium with hydroxyl. In acid, low OH"1 water, polonium
is mobilized, while in alkaline, high OH'' waters, it
forms the radiocolloid. The polonium-hydroxide








radiocolloid coexists and co-precipitates with iron-
hydroxy complexes. Movement of the radiocolloid and iron-
hydroxy colloids is mechanically inhibited in aquifer
systems.

13. High gross-alpha radioactivity is spatially related
to fracture traces. Fracture traces are characterized by
enhanced vertical leakance, thickened "leached zone", and
extreme temporal variability in saturated-zone thickness.
Recharge wells within 0.3 km of a fracture trace are most
prone to violation of the 15 pCi/l MCL for gross-alpha
radioactivity. Therefore, placement of recharge wells
over 0.3 km from fractures should minimize water-quality
violations.

14. Gross-alpha (polonium) radioactivity in recharge
wells is seasonally related to water-table fluctuations
(as reflected in precipitation data at a nearby gage).
High recharge and water-table conditions lead to dilution
of the radioactivity on the scale of weeks to a month. On
the longer term, it appears that years with maximum
unsaturated zone development (low water table elevations
or wide variations in water-table elevation) have low
gross-alpha radioactivity in recharge wells. This is
because the particulate, polonium-hydroxide complexes
(and other metal-iron-hydroxide complexes) are stable,
filterable, and, therefore, not transported. In wet
years, when the saturated zone is thick (high water
table, little variation in water-table elevation), acid
conditions develop that break up the hydroxy-complexes
and polonium is mobilized to produce high gross-alpha
radiation in the wells.

15. It appears that recharge wells can be utilized in the
Central Florida Phosphate District if simple precautions
are taken. First, they should be located to avoid
fracture traces or other locations where the "leached
zone" is thick and water-table conditions favor acid
water. Second, sampling protocols should be carefully
designed, in conjunction with regulatory agencies, to
avoid contamination from casing slimes.

16. Chemical modeling and raw data clearly indicate that
conditions for transport of polonium and lead exist in
the Surficial aquifer, but not in the Floridan aquifer.
Therefore, it appears that, with respect to the
parameters measured, the quality of receiving Floridan
aquifer waters is not jeopardized by the practice of
interaquifer connector (recharge) wells.


iii











TABLE OF CONTENTS


Page


Executive Summary...................................
Table of Contents..................................
List of Figures.....................................
List of Tables......................................
Acknowledgements....................................
I. Introduction.........................................
Statement of Problem..........................
Purpose of This Report....................
Background................................
Previous Works.................................
Equilibrium Studies.......................
Radioactive Equilibrium...................
Uranium Chemistry and Migration...........
Radium Chemistry and Migration............
Radon Chemistry and Migration.............
Lead Chemistry and Migration..............
Polonium Chemistry and Migration..........
Iron Chemistry and Migration..............
Sulfur Chemistry and Migration............
Regional Hydrogeology ..........................
Surficial Aquifer System..................


Intermediate aquifer System and
Floridan Aquifer System........
II. Methods..................................
Sample Collection...................
Well Selection.................
Monitor Wells..................
Sampling Protocols.............
Analytical Methods..................
Field Methods ........... ......
Laboratory Methods.............
Isotherms.......................
Chemical Models......................
Equilibrium models.............
Statistical models.............
Sequential-Data Analysis............
Time-series analysis...........
Spatial-analysis methods.......
III. Results Chemical Data..................
Data Evaluation.....................
Monitor Well Data..............
Introduction............
Water Type................
Surficial aquifer....
Floridan aquifer.....
Sulfur Species............
Surficial aquifer....
Floridan aquifer.....


Confining
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
...........
.......o....
...........
...........
...........
eoe


eeeeeeeooe

eo ee oe e


eoeeee


eeoee

eeeee
eeeeoe ee
eeeeeee
eeeeeeeeeee
ee ee e e
ee eo ee
.e ee
,eeeooeoee
eoe eee


.......... i
......... iv
........vii
..........x
.........xi
..........I-
.........I-i
.........1-2
.........1-3
.........1-4

.........1-5
.........1-5
.........1-6


.........1-13

.........1-14
.........1-19
.........1-20
.........1-20
.........1-20


......... 1-23
........ 1-21
Unit.....I-22
........ I1-2
........II-1
........II-1
........II-1
........II-1
........11-2
........11-4
........11-4
........11-4
........11-7
........11-8
........11-8
....... II-10
....... II -11
....... II -11
........I1-12
.......III-1
.......III-1
.......III-1
.......III-1
.......III-1
.......III-1
.......11 -2
.......111-2
.......111-2
.......111-3









Iron Species................................III-3
Surficial aquifer......................III-4
Floridan aquifer.......................III-4
Organics.....................................III-4
Surficial aquifer ......................III-4
Floridan aquifer.......................III-4
Radionuclides.... ...........................111-4
Surficial aquifer......................III-5
Floridan aquifer.......................III-5
Recharge Wells....................................111-6
Introduction.... .......................... .III-6
Water Type................................... 11-6
Sulfur Species..............................III-6
Iron Species.................................III-7
Organic Carbon..............................III-7
Radionuclides.................................III-7
Changes in Water Chemistry in the Recharge Well
System...........................................III-8
Reliability of Monitor Well Data..................III-8
Chemical Changes During Recharge.................III-8
Eh-pH Changes............................... II -8
Reaction Paths..............................III-9
Saturation State of Water........................III-9
Chemical Speciation..............................III-10
Introduction...............................III-10
Sulfate Chemistry...........................III-11
Relationship to Ionic Strength..........III-11
Relationship to Eh-pH..................III-12
Iron Chemistry..............................III-13
Organic Chemistry...........................III-14
IV. Radiochemical Systems.......................................IV-1
Uranium Systems.............................. ....... ...IV-1
Uranium Abundance.................................IV-l
Uranium Correlations.............................IV-l
Radium Systems..........................................IV-2
Radium Abundance..................................IV-2
Radium Correlations..............................IV-3
Lead Systems..........................................IV-5
Lead-210 Abundance.................................IV-5
Lead-210 Correlations............................IV-5
Lead Isotherms....................... ............. IV-6
Polonium Systems..................................IV-7
Polonium-210 Abundance............................IV-7
Correlations with Polonium-210....................IV-8
Co-precipitation with Iron Hydroxides..............IV-10
Gross-Alpha Radiation..................................IV-I1
Analytical Methods...............................IV-11
Gross-Alpha Radioactivity Distribution............IV-11
Source of Gross-Alpha Radioactivity..............IV-12
V. Sequential-Data Analysis.....................................V-l
Spatial Analysis.......................................V-1
Distribution of Wells That Exceed the MCL..........V-1









Nearest-Neighbor Analysis..........................V-1
Discussion of Spatial-Analysis Data................V-1
Time-Series Data.......................................V-2
Data Trends........................................ V-2
Short-Term Variations........................V-3
Long-Term Variations..........................V-4
Discussion of Time-Series Data.....................V-5
Causes of Spatial and Temporal Variability..............V-8
VI. Summary and Conclusions....................................VI-1
VII. References Cited...........................................VII-1
A. Appendix A Lithologic Logs of Monitor-Well Cores...........A-1
B. Appendix B Chemical data ...................................B-1
1. Surficial Aquifer Monitor Wells............... ....B-1
2. Floridan Aquifer Monitor Wells.......................B-7
3. Monitored Recharge Wells.............................B-14
4. Other Recharge Wells.................................B-21








LIST OF FIGURES


Figure Page

I-i Uranium-238 decay series............................ 1-25
I-2 Diagram showing the design of recharge well KR-98B
and the hydrostratigraphy of a typical
recharge well.....................................1-26
III-1 Piper diagram showing major-element compositions of
Surficial aquifer, recharge well, and Floridan
aquifer waters................................III-20
III-2 Eh-pH diagram showing stability fields of major
sulfur species ...................... ...... ..... 11I-21
III-3 Eh-pH diagram showing stability fields of major
iron species......... ..........................111-22
III-4 Scatter diagram showing the relationship of
calculated sulfate activity to ionic
strength..................................... III-23
III-5 Scatter diagram showing the relationship of
calculated sulfate activity to pH.............III-24
III-6 Scatter diagram showing relationship of sulfate
activity to redox potential (Eh)..............III-25
III-7 Scatter diagram showing relationship of pH to
ionic strength................................ III-26
III-8 Scatter diagram showing the relationship of
strontium to log sulfate concentration.......III-27
III-9 Scatter diagram showing the relationship of total
iron concentration to pH.....................III-28
III-10 Scatter diagram showing relationship of total
iron concentration to redox potential (Eh)...III-29
III-11 Scatter diagram showing the relationship of total
organic carbon (TOC) to pH....................III-30
III-12 Scatter diagram showing the relationship of total
organic carbon (TOC) to redox potential (Eh).III-31
III-13 Scatter diagram showing relationship of total
organic carbon (TOC) to sulfate concentration
............... .............................. III-32
III-14 Scatter diagram showing relationship of total
organic carbon (TOC) to total iron
concentration................................III-33
IV-1 Scatter diagram showing relationship of radium-
226 activity to ionic strength................IV-18
IV-2 Scatter diagram showing relationship of radium-
226 activity to calculated sulfate activity...IV-19
IV-3 Scatter diagram showing relationship of radium-
226 activity to strontium concentration.......IV-20
IV-4 Scatter diagram comparing calculated maximum
radium activity, based on saturation with
respect to radium sulfate, and measured
radium activity to sulfate activity...........IV-21
IV-5 Scatter diagram comparing calculated maximum radium
activity, based on saturation with respect to









radium sulfate, and measured radium activity to
ionic strength................................IV-22
IV-6 .Scatter diagram illustrating the relationship of
lead-210 activity to pH.......................IV-23
IV-7 Scatter diagram showing the relationship of lead-
210 activity to calculated sulfate activity...IV-24
IV-8 Scatter diagram showing the relationship of lead-
210 to ionic strength.........................IV-25
IV-9 Scatter diagram illustrating the relationship of
lead-210 to total lead concentration..........IV-26
IV-10 Freundlich isotherms for the three aquifer
materials...... ......... ...........IV-27
IV-11 Scatter diagram showing the relationship of
polonium-210 activity to pH...................IV-28
IV-12 Scatter diagram illustrating the relationship of
polonium-210 to sulfate activity..............IV-29
IV-13 Scatter diagram illustrating the relationship of
polonium-210 to ionic strength................IV-30
IV-14 Scatter diagram showing the relationship of
polonium-210 to redox potential (Eh)..........IV-31
IV-15 Scatter diagram showing the relationship of
polonium-210 to total iron concentration......IV-32
IV-16 Scatter diagram showing the relationship of
polonium-210 to total organic carbon (TOC)....IV-33
IV-17 Scatter diagram showing the ratio of actual
polonium-210 activity to the maximum polonium
activity possible assuming equilibrium with
polonium hydroxide as a function of calculated
hydroxyl activity.............................IV-34
IV-18 Scatter diagram showing the relationship of ferrous
iron activity to polonium-210 radioactivity...IV-35
IV-19 Scatter diagram showing the relationship of ferric
iron activity to polonium-210 radioactivity...IV-36
IV-20 Scatter diagram comparing polonium-210 activity
with Fe(OH)+ activity.......................IV-37
IV-21 Scatter diagram showing the relationship of
polonium-210 activity to Fe(OH)2 activity.....IV-38
IV-22 Scatter diagram showing relationship of polonium-
210 radioactivity to Fe(OH)3 activity.........IV-39
IV-23 Scatter diagram showing relationship of polonium-
210 radioactivity to Fe(OH)4-1 activity.......IV-40
IV-24 Scatter diagram showing the relationship of
polonium-210 radioactivity to aqueous ferric
hydroxide activity........................... IV-41
IV-25 Scatter diagram showing relationship of polonium-
210 radioactivity to ferric hydroxide
saturation index. .............................IV-42
IV-26 Relationship of gross-alpha radioactivity to
radium-226 and polonium-210. A. Relationship
to radium-226. B. Relationship to polonium-
210. C. Relationship to the sum of radium-226
and polonium-210 activities....................IV-43








V-1 Fracture-trace map of the Keysville Quadrangle
showing locations of recharge wells and
the precipitation gage at the New Wales
chemical plant................................V-10
V-2 Scatter diagram showing the percentage of samples
that exceeded the 15 pCi/l MCL for gross-alpha
radiation as a function of distance from
the nearest fracture trace.....................V-ll
V-3 Time-series diagrams showing the temporal variation
and progressions of gross-alpha radiation as a
function of precipitation for 4 selected
recharge wells in 1980-1982....................V-12
V-4 Time-series diagrams showing-the temporal variation
and progressions of gross-alpha radiation as a
function of precipitation for 3 selected
recharge wells in 1983-1985....................V-13
V-5 Temporal variation of monthly precipitation during
the period of study at the New Wales chemical
plant, Keysville Quadrangle.....................V-14
V-6 Time-series response of residuals from linear
regressions on gross-alpha radiation and
precipitation..................................V-15
V-7 Correlogram showing the cross-correlation of gross-
alpha radiation and precipitation of a typical
recharge well.................................. V-16








LIST OF TABLES


Table Page

II-1 Recharge wells sampled for chemical and
radiological analysis in this study............11-14
II-2 Variables measured and methods used in the chemical
study.......................................... II-15
II-3 Conditions of the lead isotherm determinations......II-16
II-4 Sources of historical gross-alpha radiation data
utilized for time-series analysis...............II-17
III-1 Means and standard deviations of raw data from
monitor wells ...............-................... 11I-16
III-2 Means and standard deviations of raw data from
recharge wells................................ III-17
III-3 Activity ratios of aquifer waters as compared to
recharge well water............................III-18
III-4 Average log-saturation indices of aquifer and
recharge-well water with respect to selected
mineral species.............................. III-19
IV-1 Significant Pearson product-moment correlations of
chemical variables with uranium................IV-13
IV-2 Significant Pearson product-moment correlations of
chemical variables with radium.................IV-14
IV-3 Average activity ratios for radium-226, lead-210,
and polonium-210 by environment................IV-15
IV-4 Significant Pearson product-moment correlations of
chemical variables with respect to lead-210....IV-16
IV-5 Significant Pearson product-moment correlations of
monitor-well data with polonium-210............IV-17


xiii








ACKNOWLEDGEMENTS


The list of persons who contributed to this study and should
be acknowledged too extensive to undertake. We thank you all with
deepest appreciation. We especially thank Dr. Gordon Nifong and
the staff at the Florida Institute for Phosphate Research for
their support (and patience). IMC, W.R. Grace, Mobil, AMAX and
other companies gave us access to their recharge wells and data.
Bob Swanson (AMAX), Jay Allen and Jerry Tanner (IMC), Joel Butler
and Jim Boyd (W.R. Grace), and Frank Eisenhardt and Terry Snyder
(Mobil) were most gracious and helpful. IMC also funded
installation of the monitor well complex at KR-98B. The
Underground Injection Control group of the Florida Department of
Environmental Regulation supplied geophysical logging for us, and
they installed the monitor wells at KR-127 and KR-83. We thank
Rich Deuerling for logging the wells and Bill Davis and Paul
Spears for installing the wells. Finally, Diane Bloomberg, Alex
Fay, Chuck Heintz, Pete Kwiatkowski, Tauna Murphy, Sonny
Mikulencek, and Sandy Springer helped with the sampling and
analytical work.

Drs. Rodney Dehan and Bill Burnett reviewed the report and
made many helpful comments and criticisms.








I. INTRODUCTION


STATEMENT OF PROBLEM

Radiation in ground water is a problem of major concern in
central Florida. The radionuclides of concern are the daughters
of uranium-235, uranium-238, and thorium-232 (Figure I-1), all of
which are naturally occurring, and are found throughout Florida.
These radionuclides are relatively more abundant in phosphatic
sediments (Osmond, 1964), including commercially-important
phosphatic strata of the Hawthorn Group, than in most other
strata in Florida, and they present potential water-quality
problems to the phosphate industry as a result.

Uranium-238 is the most abundant of these and its decay
products include several environmentally-important radionuclides.
Guimond and Windham (1975) report that uranium-238 and its
daughters constitute over 90% of the alpha activity in Florida
phosphate deposits. Daughters of principal concern include
radium-226, radon-222, lead-210, and polonium-210. These
daughters have been found in ground waters in mined and unmined
phosphate deposits (Kaufmann and Bliss, 1977; Texas Instruments,
Inc., 1977; Upchurch et al., 1984; Humphreys, 1984; and others),
and in coastal areas remote to mining or commercial deposits
(Kaufmann and Bliss, 1977; Sutcliffe and Miller, 1981; King,
1983; Miller and Sutcliffe, 1985).

There have been several large-scale environmental
assessments of radionuclide distributions in the central Florida
phosphate-mining district, but none have related the
radionuclides to the specific chemical systems of the aquifers
present, and most have omitted several of the most important
radionuclides (e.g., polonium-210 and lead-210). If valid
judgments are to be made about the hazards and management of the
radioisotopes in ground water in the district, it is necessary to
understand the specific behavior of these isotopes in the aquifer
chemical systems, and that requires detailed knowledge of ground-
water pH, reduction-oxidation potentials, and chemical equilibria
with rock materials. This report addresses these chemical
systems, especially with respect to recharge wells.

The use of recharge wells by the phosphate industry is one
of the areas of recent concern with respect to the environmental
impact of phosphate mining. Some of the water utilized in the
recharge process contains radionuclides in excess of state and
federal drinking-water standards. In particular, recent data show
that some recharge wells contain radium-226 activities in excess
of the state and federal 5 pCi/l standard and/or gross-alpha
activities in excess of the 15 pCi/1 standard (Polk County Health
Department, Office of Radiation Control, 1983). In many of the
wells, radium-226 activities meet standards, but gross-alpha


I-1









activity does not. The gross-alpha radioactivity that is
unsupported by radium has been called the "gross-alpha anomaly"
(Oural et al., 1986). Oural et al. (1986) have shown that the
excess gross-alpha radiation is, in part, a result of excess
polonium-210, a radon-222 daughter. Oural et al. (1987, 1989)
have shown that uncertainties inherent to the gross-alpha
analytical method may also contribute to the "gross-alpha
anomaly".

Recharge wells are an important and cost-effective water-
management tool because they allow dewatering of the overburden
(the Surficial aquifer) for mining, and mitigate water
withdrawals from the deeper Floridan aquifer system. Because of
popular and regulatory pressures, and in the absence of reliable
data on the chemistry of well-water environments and the fates of
potentially hazardous radionuclides, most of the phosphate
companies have abandoned or temporarily closed existing wells and
postponed permit applications for new ones until the radiation
issues can be resolved to insure that environmental hazards are
minimized. Because significant losses may accrue (1) to the
environment if there is a radiation hazard, and (2) to the
phosphate industry if recharge well use is deemed hazardous, it
is important that the chemical behavior of uranium and its
daughters be thoroughly understood.


Purpose of This Report

This document reports the results of a two-year study to
investigate the chemical and physical environment of uranium
daughters in recharge wells, with particular emphasis on
polonium-210 and lead-210. The study also addresses the
district-wide distribution of uranium decay products, major
elements, and certain minor or trace constituents of waters in
recharge wells. This report discusses the chemical causes of
radionuclide mobilization in recharge wells and related aquifer
systems. It also deals with a number of other, chronic water-
quality problems that such wells experience. These chronic
problems include violations of water-quality standards for color,
turbidity, and iron.

In order to understand the reasons for radioactivity
mobilization in the Surficial aquifer and recharge wells, a
detailed study of the chemical environment of recharge wells, the
Surficial aquifer, and upper Floridan aquifer was undertaken.
This study has shown that specific radioactive isotopes (isotopes
of uranium, radium, lead, and polonium) are mobilized under
unique chemical conditions, some of which may be controllable.
Spatial and time-series analyses of existing data from area
recharge wells were conducted to determine the geologic and
hydrologic conditions under which those unique chemical


I-2








conditions exist. A protocol for recharge-well location was
developed to minimize radionuclide mobilization. As part of the
development of the analytical procedures utilized in this report,
significant modifications of the gross-alpha analytical
procedure, and of well sampling methodology were developed.

Water-quality problems associated with color, turbidity, and
iron are related to precipitation and entrainment of iron oxides
in association with bacterial mats on the well casings and
screens. Clogging of well screens by microbes and ferric
hydroxides is also a widespread problem (LaMoreaux & Associates,
Inc. 1978). The chemical processes associated with iron are
included because they impact radionuclides -in the wells.

To accomplish the goals of this project, the study was
organized into five tasks. These tasks were:

Task 1 Survey radionuclide activities in a
regionally distributed subset of the 50 or so
recharge wells that remain available in the
Central Florida Phosphate District;

Task 2 Obtain detailed chemical analyses on
wells selected to represent the range of
conditions in the wells sampled in Task 1;

Task 3 Obtain detailed chemical analyses through
time from recharge wells that have associated,
aquifer-specific monitor wells;

Task 4 Determine adsorption isotherms and
precipitation reactions that fix uranium
daughters; and

Task 5 Model the aqueous chemical reactions,
rock-water interactions, and dilution/dispersion
that affect radionuclides in recharge wells and
Surficial and Floridan aquifer waters.


Background

The Florida Institute of Phosphate Research funded a study
by us (Oural et al., 1986) in mid-1983 to determine the cause of
the excess gross-alpha radiation in recharge wells, and to
investigate the suspicions of several phosphate-company
representatives that sampling techniques, rather than water
quality, cause the reported water-quality problems. Oural et al.
(1986) focused on one recharge well so that water quality could
be held relatively constant throughout the different tests (Oural
et al., 1988). This well was located at the Kingsford Mine of the


I-3








International Minerals and Chemical Corporation (IMCC). The
results of that study indicated that the problem results from a
combination of inconsistent sampling procedures and relatively
high polonium-210 activities in the sampled waters.

Oural et al. (1986) recommended sampling procedures to
standardize data reporting. Adoption of standard sampling
protocols will significantly reduce the apparent variability in
radium and gross-alpha analyses and provide a basis for better
management of water quality in the recharge wells. Some of the
high gross-alpha activity values appear to be artifacts of the
sampling and analytical procedures. For example, Oural et al.
(1986) found that gross-alpha analyses are,.at best, estimates of
alpha activity. Polonium can be volatile and some appears to be
lost in the typical gross-alpha analytical procedures (Oural et
al., 1986). Gross-alpha activity measurements are also sensitive
to the radon daughters remaining in the sample at the time of
counting (Oural et al., 1989). Procedure standardization makes it
easy to compare analyses and determine behavior of the
radioisotopes, but it does not eliminate the problem of high
polonium activities. High polonium activities remain a problem
that needs to be properly evaluated before irreversible
management policies are set.

The problem, however, may not be as serious as it seems, and
it can possibly be mitigated because the activity of polonium can
be significantly reduced in the well by either (1) dilution, (2)
adsorption on rock material, (3) precipitation, or, possibly, (4)
metabolic activity of microbial populations in the well. Similar
reactions involving parent isotopes (e.g., lead-210) also govern
polonium availability in the aquifers. To evaluate the potential
for transport of polonium, and other uranium-daughter isotopes,
into the Floridan aquifer by recharge wells, it is necessary to
(1) identify the chemical and physical reactions that govern
polonium mobility, and (2) determine if the results of Oural et
al. (1986) are valid throughout the phosphate district.
Management practices must include knowledge of the ultimate fate
of these uranium daughters. If it can be shown that lead,
polonium, and other radionuclides are immobilized outside of the
recharge-well environment in Surficial and Floridan aquifers,
then a valid basis for operating the recharge wells exists.


PREVIOUS WORKS

Previous work on the chemistry and distribution of radiation
in aquifers in the phosphate district is scant and, for the most
part, descriptive in nature. Few researchers have undertaken to
model the chemistry of the aquifers involved and relate chemical
conditions in those aquifers or in recharge wells to radionuclide
mobility. Oural et al. (1986) give a detailed discussion of these


I-4








previous studies.


Equilibrium Studies

There are few studies that have attempted to work out the
chemical interactions in Florida's aquifers. Back and Hanshaw
(1970) and their colleagues at the U.S. Geological Survey have
modeled rock-water interactions in the Floridan aquifer on a
regional scale. Their work is too generalized to allow specific
statements about chemical conditions and causes of radionuclide
mobility in the district's ground water. Upchurch et al. (1984)
modeled rock-water interactions in the Floridan aquifer in
Manatee County with the goal of determining the causes of release
of gross-alpha radioactivity when treated water is injected into
a potable aquifer horizon. They suggested that desorption from
rock surfaces as ionic strength increases is the most probable
cause of the increase in gross-alpha radiation. Upchurch and
Lawrence (1984) modeled the interactions of water with aquifer
materials that occur with recharging of the Floridan aquifer in
north Florida.


Radioactive Equilibrium

For any particular radionuclide in a decay series, there is a
rate of production via decay of its immediate parent and a rate
of loss via decay into the daughter. A condition of equilibrium
is established if the entire series is contained in a chemically-
and physically-closed system over a sufficiently long period of
time. This decay equilibrium is maintained only if the system
remains undisturbed.

In ground-water systems, chemical reactions, such as ion
exchange, chemical completing, and equilibration reactions with
minerals, and physical processes, such as alpha recoil (Osmond
and Cowart, 1976), can cause considerable departures from decay
equilibrium. Because of their greater mobility, uranium and
radium-226 are usually depleted relative to thorium-230 on the
surfaces of soil or rock particles (e.g., minerals and
particulate organics. Uranium-238, -234, and radium-226 are,
therefore, present in greater abundance in ground water than
thorium-230 (Bolch, 1979). Radon-222 enters water by alpha recoil
and/or leaching (Osmond and Cowart, 1976), and, because radon is
a chemically inert gas, it migrates. Polonium-210 and lead-210
also migrate to a lesser extent, and they show disequilibrium
relationships with other isotopes in the uranium-238 series.


I-5









Uranium Chemistry and Migration


Uranium is an important trace constituent in marine
phosphorite deposits. It is co-precipitated with carbonate
fluorapatite (a k a francolite [ Cas(PO4,CO),3F ], the predominant
mineral in marine phosphorite deposits) in low-oxygen, reducing
environments. Uranium is both incorporated within the crystal
lattice of the phosphate mineral (Altschuler et al., 1958) and as
a sorbed or chemically-complexed phase on clay minerals and
organic. Upon weathering in ground-water environments, uranium
is generally conserved (Altschuler et al., 1956) and concentrated
in the weathering profile. Thus, the so-called "leached zone", a
weathering profile on the upper surface of the phosphatic strata
in central Florida, contains elevated uranium concentrations
(Cathcart, 1956).

Uranium, which is polyvalent, has complex solubility
relationships depending on reduction-oxidation potential and
water chemistry (Langmuir, 1977). In reducing waters, at 25"C and
with compositions similar to Florida aquifer waters, uranous
( U'4 ) ion forms complexes with fluoride below pH's of 3-4. At
higher pH's uranous-hydroxy complexes form. However, the
solubility of uranium-oxide and -silicate minerals is low (U<104
pg/l), so uranium remains below normal detection limits. In
oxidized waters, uranium is hexavalent ( U*6 ). At pH's below

about 5, uranyl ion ( (U6'02)2* ) and uranyl fluoride predominate.

At pH's of 4-7.5, UO2(HPO4) is important. At higher pH's
uranyl di- and tricarbonate species are important. Other uranyl
complexes with fluoride, phosphate, sulfate, and carbonate form
under these conditions, as well. Thus, formation of uranyl
greatly increases uranium solubility.

At pH's between 5 and 8.5, uranyl minerals are at minimum
solubility and uranyl complexes are readily sorbed on colloidal
particles. Preferred sorption sites include organic, ferric
hydroxides, Mn- and Ti-oxyhydroxides, and clays, in approximate
decreasing order of sorption. Therefore, in environments with
high particulate surface area to water volume ratios, such as in
ground-water systems, uranyl complexes should tend to obey a
sorption-controlled equilibrium with the water.

Radioactive decay influences the sorption-controlled
equilibrium somewhat. While uranium chemical activity is
determined by ionic strength, chemical species, Eh, and pH, the


I-6








distribution of isotopes of uranium is controlled by production
or loss through radioactive decay, and decay events. Uranium-238
decays by alpha emission.to thorium-234, which has a half life of
24.1 days. Thorium-234 decays to uranium-234 by means of two
beta-particle emissions (234Th-- -> 234Pa-- -> 234U). When a
sorbed uranium-238 decays, therefore, there is a probability that
the alpha particle will emerge towards the sorption surface.
Energy transfer to the sorption surface by the massive alpha
particle can disrupt the sorption binding energy and cause the
daughter thorium-234 to be ejected into the water or air in the
adjacent pore space. Therefore, with time, water should become
relatively enriched with uranium-234 as compared to uranium-238.
The activity ratio (A.R. = 34U/238U) is-a measure of this relative
maturity. The activity ratio has been utilized by Osmond et al.
(1974), Osmond and Cowart (1977), Cowart et al. (1978) and others
to characterize Floridan aquifer flow systems and mixing
relationships.


Radium Chemistry and Migration

Of the 16 known isotopes of radium, only two are important
in Florida systems. These are radium-226, which is a decay
product in the uranium-238 series, and radium-228, which occurs
in the thorium-232 decay series. Radium-228 is rare owing to the
scarcity of thorium in marine sediments and its relatively short
half life. The Polk County Health Department, Office of Radiation
Control (1983) analyzed for radium-228 in recharge wells and
found that out of 71 wells, 76% had less than 1 pCi/l 228Ra, 22.5%
had 1-5 pCi/l, and 1.5% (1 well) exceeded 5 pCi/l. Radium-226 is
present in larger quantities and is more widespread (see below)
than is radium-228. Therefore, radium-226 is the isotope of
principal interest.

Radium-226 forms by the alpha decay of thorium-230, and it
has a half life of 1,620 years. The immediate decay product of
radium-226 is radon-222. The radium ion is divalent and behaves
somewhat like calcium or barium, so it can substitute for calcium
in bone tissues. The combination of substitution in calcified
tissues and alpha emission causes radium-226 to be carcinogenic
and of major health concern. Radium forms strong bonds with
sulfate and carbonate. Solubility of radium sulfate is quite low,
and limits radium mobility in most systems. There is some
indication that radium sulfate does not readily nucleate in-the
absence of barium (Gilkeson et al., 1983). When barium is
present, it co-precipitates with radium sulfate and controls
radium solubility through nucleation kinetics. The solubility-
product constants of radium sulfate and barium sulfate baritee)
at 25"C are similar, with a K of 10-10.4 for radium sulfate and
10'10 for barite (Sillen and Martell, 1964; Gilkeson et al.,
1983). Radium can also complex with soluble sulfate and carbonate


I-7










species. Based on data presented in this study, RaSO appears
to be the most probable complex in the aquifer systems studied.

Radium can enter aqueous systems by ion exchange, leaching,
alpha recoil, and possibly microbially induced sulfate
reduction'. Numerous studies have documented leaching of radium
from weathering profiles (e.g., Cathcart, 1956; Burnett and
Deetae, 1987). Changes in ionic strength of ground water in the
coastal, salt-water/fresh-water transition zone of Florida have
been suggested to cause competition for sorption sites on aquifer
walls and encourage release of sorbed radium into ground waters
(Upchurch, 1987). Microbial sulfate reduction has been documented
in (Ba,Ra)SO, precipitates produced to remove radium from
uranium-mine wastes (Fedorak et al., 1986). Barium, radium, and
sulfide were products of the resulting reactions.

There have been many studies of the occurrence of radium in
the aquifers of the Central Florida Phosphate District. Kaufmann
and Bliss (1977) compared radium in ground waters in coastal
areas of Florida with that of the district. Upchurch et al.
(1979) compared major element chemistry of waters at the
Kingsford Mine with radium-226 activities. Their study included a
number of recharge wells. King (1983) and the Polk County Health
Department, Office of Radiation Control (1983) included analyses
of radium and gross-alpha radiation in ground waters in recharge
wells in the district. Kimrey and Fayard (1984) analyzed several
recharge wells in the district for radionuclides and other
pollutants. Humphreys (1984) and Osmond et al. (1984) have also
discussed mobility of radium in waters of the district, including
recharge wells.


Radon Chemistry and Migration

Radon is a noble, gaseous element incapable of sorption or
equilibration with-mineral phases, and considerably greater
disequilibrium occurs with respect to its precursors and its
daughters than for most other isotopes in the uranium-238 decay
series. Because of the great discrepancy between radon activities
and the activities of both progeny and parent, radon was not
determined in this study. However, the presence of radon in the
aquifers is important to this study for three reasons:

1. It is highly mobile, especially with respect to
transfer from the aquifer matrix to pore water or soil
atmosphere,

2. It a primary means by which decay products within
the decay series are redistributed in nature, and


I-8








3. It decays to lead-210 and polonium-210, which have
relatively long half lives and are the are progeny of
interest.

Migration of radon-222 determines the reservoirs of radon
daughters from which lead-210 and polonium-210 are derived.
Activities of radon in ground-water systems vary from a few
picoCuries per liter to well over ten thousand. Therefore, the
chemical and physical properties of radon are included below in
order to assist in explaining the behavior of polonium.

There are numerous factors that affect the emanation of
radon into ground water and soil atmospheres (Rama and Moore,
1984). At the rock or soil pore-space scale, radon activity in
the pore water may be approximately equal to activity in the
adjacent mineral phases. The reported activity of radon in pore
environments is a few disintegrations per minute (dpm) per
milliliter of pore fluid. This activity is comparable with that
found in an equivalent volume of solids in the aquifer. Thus,
ground water receives radon-222 effectively from a volume of
solids roughly equal to that of the pore water. It has been
further observed that the level of non-gaseous uranium- and
thorium-series isotopes in ground water is three to five orders
of magnitude lower than that of radon-222.

Belin (1959) related radon solubility to pH and residence
time of water in contact with rock sources. Andrews and Wood
(1972) have related radon in ground water to rainfall. They
suggested that soil moisture dissolves radon, which is then
transported in the direction of flow. Panik and Burnett (1987)
documented losses of radon in phosphatic rocks of Florida. Dry
phosphate rock losses of 3-74 dpm/g were detected. Average loss
from dry rock was 18.5 dpm/g or 12.9% of the radium-226-generated
radon. In wet rock losses were three times that of dry rock, with
a mean of 43.2 dpm/g, or 31.5% of radium-226-supported radon.
Therefore, leaching may be an important mechanism for
introduction of radon to pore waters.

Rama and Moore (1984) attempted to understand the mechanism
of release of radon and other uranium- and thorium-series
isotopes into ground water. Tanner (1964) showed that
approximately 20% of the radon-222 generated in the solids under
investigation emanated into the water. Since the diffusion of
radon out of the crystal lattice is negligible (Tanner, 1964),
Rama and Moore concluded that radon must come out of the mineral
phase in other ways. They proposed that rock or soil mineral
grains are permeated with "nanopores", which have opening widths
at the grain surface that are less than 1 micron. Radon, they
suggested, is being released into these body pores and diffusing
out into the intergranular pores, which have dimensions of tens
to hundreds of microns. Radon-222 and other, non-gaseous isotopes


I-9









are introduced into the nanopores by alpha recoil from the pore
walls in approximately equal quantities. Since the nanopore wall
area-to-volume ratio is very high, the ionic isotopes are quickly
sorbed within the pores, while the radon can pass outward into
intergranular pores.. Migration of radon-222 and its progeny
through nano-, micro-, and macropores is not well understood. The
process, however, explains the accumulation of inert radon in
pores in relatively high concentrations.

Radon can migrate from pore to pore under certain conditions
and additional accumulation may result. In unsaturated soils, the
radon is free to migrate by diffusion or advection, but the short
half life of the radon generally limits migration to a few
meters. Ogden et al. (1987) showed that radon emanation into
homes increases with thickness of the unsaturated zone. In water-
saturated soils, radon accumulates in the water because diffusion
and degassing are restricted, and transport is limited to flow
rates of the water and degassing at the water-table interface.
Fluctuations in radon emanation rates were attributed by LeGrand
(1987) to pumping action produced by fluctuations in water-table
elevation. These processes are exacerbated when radium-bearing
aquifer materials are fine grained (Andrews and Wood, 1972).
Several workers (Tanner, 1964; King et al., 1982; Krishnaswami et
al., 1982) have observed very large amounts of unsupported
radon-222 gas in ground waters. The mechanisms of radon migration
have been extensively studied because of their impacts on human
health and uses in earthquake prediction (Chiang, 1977).

Israel and Bjornsson (1966) suggested that, in the vadose
zone, diffusion controls radon migration. Diffusion coefficients
may be as great as ixl102 cm2/sec, depending on porosity,
permeability, and soil moisture. They argued that in phreatic
(saturated) environments the diffusion coefficient is less than
1x10'5 cm2/sec., therefore radon would decay through many half
lives before it travelled 10 cm from the source. In phreatic
environments, therefore, they suggested that fluid flow and
changes in hydrostatic pressure control radon migration. Gingrich
(1976) found evidence of radon migration over a distance of 100 m
from uranium deposits, which suggests that mechanical transport
in flow systems is of importance. Banwell and Parizek (1988)
studied radon and helium (alpha particles) associated with
fracture traces. Radon activities in ground water were found to
be reduced in the vicinity of fractures because of enhanced
ground-water flow associated with higher transmissivities in
highly fractured rock. Radon activity is also correlated to H2S,
indicating two possible explanations. These are:

(1) In sulfide-rich, reducing waters radium-226, the parent
of radon-222, is somewhat soluble and, possibly, transported
to the vicinity of the well. Near the well, oxidizing


1-10








conditions prevail, and the radium is fixed by sorption on
oxyhydroxides. Thus, a reservoir of radium accumulates near
the well and H2S and radon might co-exist.

(2) Deep aquifer waters exist in low transmissivity rocks
and are somewhat stagnant. High H2S waters and radon could
co-exist in this environment and produce the observed
correlation.

Banwell and Parizek (1988) also monitored helium-4, which
comprises the alpha particles produced by uranium-series decay.
Since helium-4 is a stable, noble gas, transport over long
distances is possible, whereas radon-222 migration is limited by
decay. High helium concentrations in ground water delineate the
fractures, with highest concentrations above background within a
half width of 0.4 km for the fracture-trace center. Their data
clearly indicate that fractures have the potential for
accumulation of reservoirs of uranium-series isotopes.

Kovach (1945) related radon gas emanation from soils to
variations in atmospheric pressure. Rodgers (1954) and Chen et
al. (1973) found that increasing temperature decreases radon
solubility in water. Arndt (1953) described radon distribution in
springs and found that about 40% of the radon is lost to the
atmosphere within 4 ft. of the spring emergence.

Radon can diffuse upward through porous strata. The presence
of fractures or other pathways of increased porosity and/or
permeability in vadose environments encourages migration. Radon
detection has, therefore, been used to locate faults and
fractures (Stothart, 1948; Israel and Bjornsson, 1966, Chiang,
1977).

The effects of water flow on radon in systems influenced by
operating wells is unknown. LeGrand (1987) described a
conceptual model of radon-rich ground-water flow under in
influence of well pumpage. He argued for hydraulic pumping of
radon in the vadose atmosphere as the water table fluctuates, and
for transport of the gas in water moving towards well screens or
other discharge points. Therefore, there is good evidence that
radon gas can migrate significantly in ground-water systems
wherever a radon-concentration or hydraulic gradient exists.

Recharge wells may encourage migration and/or mechanical
accumulation of radon gas. There are several possible
consequences of recharge-well use that should be investigated.
First, by proper selection of the screened interval in a recharge
well, it may be possible to avoid naturally high radon, and
polonium, horizons of the aquifer. Second, production of a cone
of depression in the Surficial aquifer by artificial drainage,


I-11









alters a portion of the aquifer from phreatic to vadose. This
change in regolith pore filling (water to atmosphere) may have an
influence on the radon distribution and on the distribution of
sorbed, reactive parents and daughters of radon.

Most of the emphasis on radon and radon daughters at state
and national levels has been focused on radon in structures and
migration out of the soil. Radon has been found in homes in many
areas of Florida, and a state priority has been identified to
develop criteria for evaluation of radon risks from measurements
of soil radon prior to construction (Nagda et al., 1987). Early
studies on the potential for radon in structures in Florida
include Guimond et al. (1979), Roessler et al. (1980), and
others.

Roessler et al. (1980) and Roessler et al. (1983) studied
radiation in the Florida phosphate district. In both papers it
was concluded that "debris lands" produced by former mining have
the greatest potential for radon emanation into structures. More
recently reclaimed lands appear to be less prone to radon
hazards, but problem areas may exist. Methods of reclamation and
site treatment, such as replacement of overburden over clay
wastes and/or "leached zone" materials, were suggested to further
reduce hazards. Natural, undisturbed land was found to be at less
risk of high radon emanation rates than mined lands.

The most significant radon study in Florida was the
statewide radiation study of Geomet, Inc. (Nagda et al., 1987).
The Geomet study identified indoor radon risks based on both
population and geologic criteria. It identified, by U.S.
Geological Survey quadrangle area, regions where there is a
probability of hazardous levels of radon emanating into
structures. The study also made recommendations for dealing with
the problem, including a recommendation that methods of
characterizing soil radon be studied in anticipation of
development of predictive criteria for high radon risks prior to
construction. The study found that homes on reclaimed land in
Polk County have 50% higher indoor radon levels that homes on
mineralized, unmined lands and that soil radon levels are two
times those on unmined lands (Nagda et al., 1987, Table 4-23).
Correlation of soil radon with indoor radon for 2,749 homes was
0.52, which is statistically significant (a=0.01), but not strong
(R2 = .27, or 27% of total variability accounted for by the
analysis). The correlation was improved to 0.80 when the sample
set was limited to quadrangles with 4 or more homes tested.

In 1987, Roessler et al. compiled literature data on radon
potential in Florida. They concluded that reclaimed lands have
higher potential for radon hazards than do unmined lands based on
data from 17 sites on mined land and 61 sites on equivalent,
unmined sites. The Florida Department of Health and


1-12








Rehabilitative Services (1987) also compared mined and unmined
lands for radon potential. Differences in radon activities in
their data were less dramatic, and significant differences were
not evident.

Bolch et al. (1987) and Nagda et al. (1987) have done most
of the pioneering work on methods of characterizing soils for
radon potential prior to construction. Bolch et al. (1987)
studied radium-226 profiles in soils to a depth of 1.8 m. Near-
surface radium-226 was found to be higher in mined lands than in
undisturbed lands, where Plio-Pleistocene marine sands form
overburden. Clearly, radium-226 profiles in shallow soils provide
an indication of potential for radon emanation into structures
and indicate that one cannot assume that all lands have the same
potential for emanation.


Lead Chemistry and Migration

At the Eh and pH range of most natural waters, lead is
divalent (Garrels and Christ, 1964; Hem, 1970). If carbonate is
present, lead carbonate (cerussite, PbCO3) is relatively
insoluble and controls solubility of lead. Lead sulfate
(anglesite, PbSO4) and lead sulfide (galena, PbS) are also
relatively insoluble and are likely to precipitate in oxidizing
and reducing environments, respectively. Therefore, lead
concentrations in natural systems are expected to be in the order
of 2 gg/l or less.

Lead-210 is the longest-lived daughter of radon-222. It has
a half life of 19.4 years, and it forms by the alpha decay of
polonium-214. At the picoCurie activity level, lead-210 is
present in such minute quantities that solubility constraints are
minimal, unless other lead is present in the system. The longest
half life of the four radon daughters that precede lead-210 is 27
min. (214Pb), so lead-210 is quickly made available for transport
by radon release into the water, and it may exist in solution in
pores or sorbed on pore walls. Owing to the low solubility of
both lead carbonate and sulfate/sulfide species, and ability to
adsorb onto colloids, clays, and organic, lead is likely to be
present in low quantities in most ground-water systems.

There are three sources of lead-210 in soils: (1) upward
migration of radon-222, (2) radon-222 trapped in soil minerals,
and (3) atmospheric fallout (Moore and Poet, 1976) from decay of
radon in the open atmosphere. This fallout was estimated by Moore
et al. (1973) to average 0.8 dpm/cm2 yr in the western U.S. Lead-
210 was found to decrease in concentration with depth in soils.
The high lead-210 content near the soil surface is partly due to
upward migration of radon-222 and partly due to fallout.


1-13









Percolating ground water transports some of the fallout-generated
lead downward.

Lead-210 essentially controls the fate of its granddaughter,
polonium-210. Cowart et al. (1987) compared lead-210 and
polonium-210 in aquifers in central Florida. They found that
polonium-210 activities greatly exceed lead-210 in water and that
lead-210 averages about 3% of the polonium-210 activity. Thus,
most of the lead-210 is present as a sorbed or precipitated phase
in or on the aquifer matrix, not as an aqueous species.


Polonium Chemistry and Migration

Polonium is a radioactive, polyvalent metal with a number of
possible valence states. In natural conditions only the +4 and +2
states are likely, with the +4 state being the most common. The
half life of polonium-210, the longest-lived isotope of polonium,
is 138 days. Decay of polonium-210 is by alpha emission, and the
daughter is stable lead-206. "Crystalline" and colloidal phases
can be precipitated in the laboratory, but intense radioactive
decay rapidly destroys crystal structure and heats the solid
(Bagnall, 1957), which increases solubility. No work has been
done on the phases present in natural systems.

In reducing environments, PoS has been shown to precipitate
at elevated temperatures with a solubility constant of 10'29. PoS
precipitates in the presence of H2S in dilute acids. It is
insoluble in HCl, but can be oxidized.

In oxidizing environments, PoO2 and Po(OH)4 are the most
likely solid phases (Bagnall, 1952; Sedlet et al., 1964).
Polonium dioxide is structurally like the uranyl radical. It is
relatively insoluble in water, ammonia, or alkali carbonate
solutions. Polonium dioxide is soluble in HCl and other hydrogen
halides, and the result is volatile PoC12- and PoCl2-. Volatility
of polonium chloride and other halides, is a constraint in
analysis (Bernabee and Sill, 1982; Oural et al., 1986). Polonium
dioxide is also soluble in nitric acid, with the formation of
several nitrate complexes. Polonium hydroxide forms in aqueous,
alkaline environments. The hydroxide is slightly soluble in
dilute sulfuric acid according to the reaction

Po(OH)4 + 2H2S04 P04+ + 2S042 + 4H20.
(I.1)
The solubility product constant for the dissociation of polonium
hydroxide in water at 25*C


1 I-14









Po(OH)4 Po4 + 40H-
(1.2)

is 10'37 (Ziv and Efros, 1959), so equilibrium activities in pure
water would be in the picoCurie range. Polonium hydroxide is also
slightly soluble in phosphoric acid. Tetravalent polonium
complexes with a number of organic acids, including acetic acid,
formates, oxalates, and tartarates. Stability of the complexes
increases with acidity of the compound. Therefore, in ground-
water systems there is the potential for formation of trace
quantities of polonium complexes with sulfate, phosphate,
nitrate, chloride, fluoride, and organic.

An important property of polonium is its ability to form
"radiocolloids" (Sedlet et al., 1964). Radiocolloids are
anomalous particles that form in solutions of radioactive
isotopes at concentrations well below equilibrium with solid
phases. Thus, the radiocolloid behaves like a solid and is
filterable, although the isotope is present in insufficient
quantities for chemical nucleation. In ground-water systems
radiocolloids may be mechanically filtered as water passes
through pore throats in the aquifer or it may be filtered after
sample collection. The behavior of polonium as a radiocolloid is
controversial. There is not any doubt that radiocolloids form,
but reports of the conditions of colloid formation are
inconsistent. Acidity of the solution seems to control colloid
formation. Sedlet et al. (1964) summarized the literature on
polonium radiocolloids. They reported that, in neutral, or nearly
neutral, solutions (pH 6-8), polonium is soluble and only 0-20%
is filterable on 0.045im membranes. At pH's near 1 and 12,
polonium is nearly completely filtered. In contrast, Haissinsky
(1964) reported that polonium acts as a radiocolloid at pH's
above 3. Thus, polonium solubility is pH sensitive, but the
specific behavior is debatable.

Hansen and Watters (1971) characterized polonium-210
solubilities in soils, including 8 samples from 5 Florida soils.
They used a polonium dioxide tracer and found that polonium is
fixed in soils, especially in fine-grained soils. Distribution
coefficients (K,) were calculated from the ratio of polonium
fixed on the soil sample to that remaining in solution.
Distribution coefficients for Florida soils are as follows:


1-15









Soil Horizon K_
Adamsville A, 26 2
Blanton A, 35 3
Lakeland A, 25 2
Leon A, 17 1
A, 15 0.6
B 55 17
C 77 29
Ruskin A, 17 1

These distribution coefficients are low in comparison with sandy
loams from elsewhere, where Kd's of 500 to 7000 were reported.
However, they do indicate that polonium is effectively
immobilized in Florida soils. The highest level of fixation is in
the B zone of the Leon sand, which is rich in ferric hydroxides
and organic. The lowest levels of fixation are in fine quartz
sands of the Al zones of the Leon and Ruskin soils. Hansen and
Watters (1971) also determined regression equations to predict
the distribution coefficients as a function of soil pH and silt
content. The data base for the regression equation includes soil
samples from all horizons and a diversity of locations throughout
the U.S. The equations they obtained are

In Kd = 3.2 + (0.0460.007) (% silt)
(1.3)

for A horizons only, and

In Kd = -1.3 + (0.0340.007) (% silt) + (0.880.20) (pH)
(1.4)

for all soils. The multiple correlation coefficient (R) for
equation 1.3 is 0.84 (15 df), and R is 0.74 (48 df) for equation
1.4. An equation similar to 1.4, but using percent clay rather
than percent silt, has an R of 0.65, so sorption on clay is
somewhat less important than the presence of the silt fraction,
which often includes the oxyhydroxides. Since both pH and silt
are deemed important in predicting the distribution coefficient,
and the correlation (slope) with pH is positive, it appears that
sorption on clays is less important to fixation of polonium than
is precipitation of radiocolloids and/or co-precipitation of
polonium hydroxide with iron and/or aluminum oxyhydroxides.

Factors that mobilize polonium in ground water are unknown.
Possible causes of mobilization include (1) decay of soluble
lead-210 to produce polonium-210 at levels below equilibrium
concentrations, (2) "dissolution" of radiocolloids with change in
pH, (3) mobilization as an organic complex, (4) mobilization as
an inorganic chemical complex, (5) changes in solubility with
reduction or oxidation, (6) ion exchange reactions (desorption)


1-16








between water and aquifer matrix, and (7) bacterial metabolism.
Decay of lead-210 to polonium-210 is by beta emission, so recoil
is not a significant factor in introduction of polonium into
ground water.

At the concentrations typical of ground-water systems, lead
salts and sorbed lead represent trivial masses in aquifer
systems. Decay of lead-210 to polonium-210, therefore, could
result in small amounts of a chemically unstable, polonium
compound that could go into solution if the water is
undersaturated with respect to ithe compound. Also changes in pH,
or other chemical parameters, can lead to changes in solubility
of the compound. For example, polonium hydroxide can be used to
illustrate these conditions. Assuming that the dissolution
reaction of polonium hydroxide is

Po(OH)4 Po4 + 40H-,
(1.5)
and the K relationship at 25C is

K = 10-37 = ap. a.
(1.6)

where a is the activity of the species and the activity of
particulate polonium hydroxide is assumed to be 1, then the
equilibrium polonium activity can be calculated for any OH-
activity. Note that the assumption that the activity of
particulate polonium hydroxide is 1 causes underestimation of the
particle solubility. Activities should exceed 1 because of (1)
the small size and high surface-free-energy of colloid-sized
precipitates, and (2) the disruption of order in the interior of
the precipitates during radioactive decay.

Examination of equation 1.6 indicates that the solubility of
polonium hydroxide-is quite sensitive to hydroxyl activity, and,
by extension, pH. Low hydroxyl activities are associated with low
pH's, and, therefore, high solubility of the hydroxide.
Conversely, high hydroxyl activities encourage polonium hydroxide
precipitation. Assuming that there is excess polonium hydroxide
available so that equilibrium can be obtained, and assuming a
typical range of hydroxyl activities of 106-103 at 250C,
dissolved polonium radioactivities at chemical equilibrium range
from 10r6 to 10+6 pCi/l. The existing reservoir of polonium
hydroxide in most aquifer systems is, however, insufficient to
allow radioactivity to rise to 10*6 pCi/l, but Burnett et al.
(1987) have documented polonium activities in excess of 106
pCi/l. Clearly, polonium-210 is likely to go into solution in
acid waters, while it is precipitated in basic solutions.
Polonium activities in water are, however, ultimately limited by


1-17








the concentration and decay constant of the parent lead-210 in
the system.

Inorganic and organic completing have not been adequately
addressed in the literature. No literature on natural, inorganic
completing can be found. Laboratory studies (see -above) clearly
indicate that halide, nitrate, and other complexes are possible.
Harada et al., 1989 have found a correlation of polonium activity
with sulfide oxidation and a relationship with ability to recover
polonium in the laboratory. They speculate that completing and
microbial activity may play roles in mobilization of polonium,
with highest polonium mobilities in reducing, sulfide-rich
waters. Oural et al. (1987, 1989) adjusted the pH of organic-rich
ground water to separate humic and fulvic fractions. They
estimated that as much as 89% of the recoverable polonium was
associated with the fulvic fraction. Their studies suggest that
organic completing may also be important in polonium
mobilization.

There is little direct evidence that ion exchange or
reduction/oxidation reactions play a role in polonium
mobilization. Ion exchange is well documented for major elements
in ground-water systems, and the process can be expected to
occur, although the low chemical concentrations of polonium in
ground-water systems reduces the probability of exchange.
Upchurch (1987) suggested that exchange may play a role in
polonium mobilization in coastal aquifers; however, his data were
circumstantial. The predominant valence state of polonium over
the range of conditions prevalent in most aquifers is +4
(Bagnall, 1957). Existing literature does not indicate
significant reduction or oxidation of polonium in such systems.
Upchurch et al. (1987) suggested that variations in water
saturation in the Surficial aquifer in central Florida may lead
to reduction/oxidation reactions and variation in polonium
mobility. Harada et al. (1989) found that polonium solubility, as
indicated by the ability to filter on an 0.2 gm filter, is
highest in-reducing, sulfide-rich waters, and that particulate
polonium increases as sulfide is oxidized, perhaps by sulfur
bacteria.

Polonium-210 has been shown to be present in large
quantities in Surficial aquifer waters in Florida (Oural et al.,
1987; Chin et al., 1987; Upchurch et al., 1987; Burnett et al.,
1987; Cowart et al., 1987; Harada et al., 1989). Chin et al-.
(1987) found that polonium in the Central Florida Phosphate
District can be well above the state and federal Maximum
Contamination Limit (MCL) of 15 pCi/l gross-alpha radiation, with
a maximum of 500 pCi/l in filtered water and 2,500 pCi/l in
unfiltered water. Polonium chemistry and migration modes in the
Central Florida Phosphate District have been characterized by
Upchurch et al. (1987). They suggested that polonium-210 mobility


1-18








is controlled by complexation with soluble organic and
maintenance of a saturated zone with net chemical reduction of
the polonium to its divalent state. Upchurch (1987) compared
uranium-daughter occurrences within aquifers in central Florida
and found that high levels of polonium-210 are limited to sand
aquifers that are characterized by organic and variable redox
potentials. In Lee County, Florida, high polonium activities were
limited to the Surficial aquifer and the upper Hawthorn Group
sandstone aquifer. Burnett et al. (1987) and Cowart et al. (1987)
reported on a private well in southeast Hillsborough County,
Florida. They found polonium-210 in excess of 2500 pCi/l in
unfiltered samples and 500 pCi/l in filtered. These values
represent the highest polonium activities in ground water known.


Iron Chemistry and Migration

While understanding iron behavior in the aquifer systems and
recharge wells is not a major goal of this study, much data is
available and some observations may be helpful. Iron and sulfur
are associated with bacterial activity in ground waters, so color
associated with ferric hydroxide precipitates, iron colloids, and
bacterial activity cause related problems. For example, Gordon
Palm and Associates (1983), in a water quality survey of the
Central Florida Phosphate District, found that 16% of the
shallow-well samples from mine areas violated the water-quality
standards for color and 20% violated standards for iron. In deep
wells, 13% violated the color standard and 10% violated iron
standards. Recharge wells are particularly susceptible to
problems. The well screens plug with bacterial mats and ferric
hydroxides, and color and iron standards are frequently violated.

Iron has two valence states, +2 and +3, and is highly
susceptible to reduction/oxidation reactions. Hem (1970) has
summarized the stability relationships of iron in sulfur-rich
systems. In general, the sources of iron in ground waters include
(1) oxidation of pyrite (FeS2), (2) oxidation of organic
compounds, and (3) dissolution of iron oxide and silicate
minerals.

In general, in acid, reducing waters FeS2ald is the stable
iron phase. In acid, oxidizing waters ionic Fe2" and Fe3* should
go into solution. In basic, reducing waters, pyrite (Fe2S2) and
siderite (Fe2*CO3) are stable solids. And, in basic, oxidizing
waters, amorphous Fe(OH)3 should precipitate and then mature to
goethite (FeO(OH)). In the ground-water systems under
investigation, reducing environments tend to mobilize iron as the
ferrous (2+) ion. In oxidizing waters, ferric hydroxide
precipitates. The color problem is a result of this


1-19








precipitation.


Iron is regularly reported in water-quality data from the
phosphate district. However, no published studies exist that deal
with the problem. LaMoreaux and Associates (1978) conducted a
short study of iron in recharge wells. They found a crude
relationship between reduction/oxidation potential and iron
precipitation.


Sulfur Chemistry and Migration

Rightmire et al. (1974) examined the origin of sulfates in
the Floridan aquifer on the basis of sulfur-isotope composiiton.
They found that sources of sulfate include (1) recharge of
sulfate-rich maritime rainfall, (2) dissolution of
intraformational gypsum at depth, and (3) mixing with ocean-like
saline water. The maritime rainfall component, of course, passes
through the Surficial aquifer and constitute a major source of
sulfate in that aquifer as well. Isotopic fractionation during
sulfate reduction leads to a relative increase in ;S in the
data. Thus, sulfate reduction in the Floridan was documented. Rye
et al. (1981) studied the isotopic composition of sulfide in the
Floridan. They found that sulfide in the Floridan results from
slow, in situ microbial sulfate reduction. Sulfide concentrations
increase down flow paths and with increased residence times. The
source of the sulfate in long flow path waters was attributed to
dissolution of gypsum in the underlying confining beds. Waters in
the study area were found to be low in dissolved sulfide and
sulfate as a result of short flow paths and residence times.
Harada et al. (1989) studied the relationship of sulfide
oxidation in the Surficial aquifer (?) to polonium mobilization.
They suggested that polonium is mobile in waters in which sulfide
is present. Upon oxidation of the sulfide to sulfate the polonium
forms filterable colloids.


REGIONAL HYDROGEOLOGY

The Central Florida Phosphate District can be characterized
by three hydrostratigraphic horizons, using the terminology
developed by the Southeastern Geological Society, Ad Hoc
Committee on Florida Hydrostratigraphic Unit Definition (1986).
These are: (1) the Surficial aquifer system, (2) a complex of
locally-confined aquifers and semipermeable layers of the Miocene
Hawthorn Group, which is called the Intermediate aquifer system
and confining unit, and (3) the Floridan aquifer system
(Hutchinson, 1978).

Lithologic summaries of these horizons are based on core


1-20








data obtained as part of the monitor-well program of this study.
Core logs are presented in Appendix A.


Surficial Aquifer System

The Surficial aquifer consists of clean quartz sand to
clayey sand of eolian and marine origins. The age of these
deposits varies from Plio-Pleistocene to Recent. Organic and
ferruginous B horizons and hard pans are locally present. Near
the base of the aquifer, phosphate grains may be reworked from
the underlying Bone Valley Member of the Peace River Formation
(Hawthorn Group). The thickness of the Surficial aquifer is
highly variable, and ranges from a few centimeters to over 75 m
(Stewart, 1966) in Polk County. In the study area, the thickness
of the surficial deposits that overlie the confining unit ranges
from 7 to 15 m (Wolansky et al., 1979). The Surficial aquifer
constitutes the principal portion of the overburden removed
during mining. Cathcart (1963, 1966) mapped overburden
thicknesses in the study area, and found that overburden thickens
near fractures and other depressions in underlying units.

Cathcart (1963, 1966) also characterized the thickness and
extent of the "leached zone". The "leached zone" is a
discontinuous weathering profile developed on top of the
underlying phosphatic material of the Hawthorn Group. It is
enriched in aluminum phosphate minerals, such as wavellite
(A13(P04)2(OH)3.5H20) and crandallite (CaA1 (P4) (OH ), .H20), and
in uranium and uranium daughters. The "leached zone" has variable
thickness, with thickness tending to increase over depressions in
the underlying material. Some linear and circular thickening
patterns appear to be related to development of alluvial dolines
(sediment-choked sinkholes) and fractures. Where present, the
"leached zone" forms the base of the Surficial aquifer, and it
has the potential of introducing high levels of radioactivity,
fluorides, and other unwanted constituents to ground waters.

The Surficial aquifer is unconfined, and receives recharge
directly from precipitation and from surface-water bodies. It is
seldom utilized as a source of potable water, and the primary use
is small, low-yield domestic wells.

The sand, which comprises most of the Surficial aquifer, is
not cohesive and tends to slump when water saturated. Because of
the impact of slumping on mine cut stability and because of
excess water that the Surficial aquifer transmits to the mine
cut, it is desirable to dewater the Surficial aquifer prior to
mining. This is one of the primary motivations for use of
recharge wells to dewater the aquifer in lands scheduled to be
mined by the phosphate industry. Recharge wells are designed to


1-21









extract water from porous and permeable sandy zones in the
Surficial aquifer by means of gravity flow through slotted'casing
(Figure 1-2). The slots are located above the underlying
confining unit and/or "leached zone" to limit potential
contamination of receiving waters in the Floridan aquifer.

In general, Surficial aquifer waters are acid, lower in
dissolved ions, and lower in radium-226 than the other
hydrostratigraphic horizons. The water is sometimes colored with
organic, and has the odor of sulfides, which indicates reducing
conditions (Upchurch et al., 1979a,b). Radon-222 migrates upward
from the phosphatic strata below and enriches the Surficial
aquifer in both radon and radon daughters, such as polonium-210.


Intermediate Aquifer System and Confining Unit

The Intermediate aquifer system and confining unit includes
the Peace River Formation and portions of the underlying Arcadia
Formation (Hawthorn Group, Miocene to Pliocene). The entire
section is phosphatic to varying degrees. The Peace River
Formation is predominantly plastic in texture, and is composed of
sand and gravel beds, argillaceous sands, silts, and clays. Silt
and clay beds include siliciclastic particles and silt-sized
dolomite. The Peace River Formation includes the Bone Valley
Member (previously the Bone Valley formation) and phosphate rich
zones which represent the "ore" or "matrix" mined by the
phosphate industry. The Arcadia Formation also contains extensive
plastic horizons, but it is characterized by dolostone and
limestone. The Tampa Member (old Tampa formation) occurs near the
base of the Arcadia. These carbonate units have developed varying
degrees of cavernous porosity. The Tampa Member has a clay unit
near its base that separates the unit from the underlying
Floridan aquifer. The "leached zone" is usually included with
this horizon. Thickness of the confining unit ranges from 30 to
120 m (Buono et al., 1979).

The coarse plastic units and cavernous carbonate units serve
as local aquifers. Artesian conditions exist in many of these
horizons. The individual production zones in the plastic part of
the system are discontinuous and limited in extent. These
horizons have been called the "uppermost artesian aquifer"
(Stewart, 1966). The cavernous carbonates may be more extensive.
The Tampa Member serves as a local aquifer throughout the
phosphate district. It has been called the "upper Floridan
aquifer", "Intermediate aquifer system", "secondary artesian
aquifer", or the "shallow artesian aquifer" (Southeastern
Geological Society, Ad Hoc Committee on Florida
Hydrostratigraphic Unit Definition, 1986). The intermediate-
aquifer portion of the Tampa Member is separated from the
underlying Floridan aquifer system by a thick, regionally


1-22








extensive clay.


The "uppermost artesian aquifer" is seldom used as a source
of water because of its limited extent. The "secondary artesian
aquifer" is widely used as a source of industrial and potable
water. Head distribution is generally below the water table and
above the potentiometric surface of the underlying Floridan
aquifer. Depth to the "secondary artesian aquifer" is variable,
with depths averaging 45 to 60 m.

Recharge wells are cased through the plastic section of the
Intermediate aquifer and confining system (primarily the Peace
River Formation) to avoid problems of caving, turbidity, and
contamination of the underlying aquifers. Waters in the
"uppermost artesian aquifer" are typically rich in sulfides,
iron, fluoride, phosphate, and, possibly radium and polonium
(Upchurch et al., 1979a,b; Upchurch, 1987). The wells are
commonly cased into solid dolostone or limestone of the Arcadia,
and they are open hole well into the underlying Floridan aquifer.
Cavernous zones in the "secondary artesian aquifer" of the
Arcadia, therefore, receive discharge from the Surficial aquifer,
and in many of the wells reported upon in this report the
"secondary artesian aquifer" may be taking a significant portion
of this discharge. Waters of the "secondary artesian aquifer"
have a typical carbonate-aquifer chemistry, with calcium,
magnesium, bicarbonate, and sulfate/sulfide. Fluoride may be
slightly elevated and radium-226 may be present at activities
near the 5 pCi/l MCL.


Floridan Aquifer System

The Floridan aquifer system is the principal aquifer in
Florida, and it is one of the most productive aquifers in the
world. It includes limestones, with minor dolostone and plastic
strata, of Miocene to Eocene age. Formations involved include the
lowermost Arcadia Formation, Suwannee Limestone, Ocala Group,
Avon Park Limestone, and Oldsmar Limestone. The gypsiferous Cedar
Keys Limestone forms the lower confining unit at the base of the
aquifer.

Depth to the top of the Floridan aquifer varies from -15
to -75 m MSL in the study area (Buono and Rutledge, 1979). Depth
to the base of the aquifer ranges from -365 to -425 m MSL
(Wolansky, Barr, and Spechler, 1979). The Floridan is primarily a
fractures and karstic aquifer, with flow through cavernous
porosity, fractures, and, to a lesser extent, intergranular and
moldic porosity. The aquifer is confined and artesian in the
study area, and water rises in wells in excess of 30 m above the
top of the aquifer in many areas. The potentiometric surface is
below that of the Surficial aquifer, so localized recharge to the


1-23








Floridan by means of leakance through the confining beds and by
direct connection in active fractures and/or karst conduits
occurs. Recharge wells take advantage of this head difference to
drive downward discharge of the Surficial aquifer.

Water quality in the Floridan is variable. The water is
dominated by calcium, magnesium, and bicarbonate (Back and
Hanshaw, 1970), with significant amounts of sulfate. The pH
averages 7.6 in the study area. Portions of the upper aquifer
that are highly fractured and permeable have relatively low ionic
strengths, sulfate, and radium. Upper Floridan aquifer waters in
low permeability blocks and deep aquifer waters are elevated in
sulfates, radium, and ionic strength (Upchurch et al., 1979a,b).


1-24



























































Figure I-1. Uranium-238 decay series.


1-25























12" PVC Casing n




Water Table --.
S 22'
Slotted PVC

__ 42'




75'
Potentiometric r .-.
Surface






Bottom of Casing [> 120'


TD [>


Sediment Fill [


SAND
O u.
Undifferentiated IL
I PlIo-Plelstocene Sand Q
PHOSPHATIC SAND
. PHOSPHATIC SAND


(MIo-Pllocene)


- -


DOLOSTONE I DOLOSILT


PHOSPHATIC SAND
& CLAY



CAVERN 175' 179'


LIMESTONE


Hawthorn Formation
(Miocene)


Tampa Formation
Z r"
(Miocene) < L
Q L

O O
-.J <
u.


Figure 1-2. Diagram showing the design of recharge well KR-98B
and the hydrostratigraphy of a typical recharge well.


1-26


,
I
1
1
I








II. METHODS


SAMPLE COLLECTION

Well Selection

Two sets of recharge-well data were utilized as part of this
study. A set of 30 existing wells was selected for detailed
chemical analysis, and a larger set of 270 historical well
records was used for time-series analysis. The set of 270 wells
is based on radiation records (gross alpha and radium-226
activities) submitted to the Southwest Florida Water Management
District by the phosphate companies from19-76 through 1986. This
data set includes virtually all recharge wells in existence
during that 10-year interval. The 270-well data set is discussed
in Chapter V.

The number of recharge wells available for the chemical
study was limited by three factors. First, there were only about
50 wells remaining in the district, and many of them are
clustered, so areal coverage was reduced. Second, wells scheduled
for abandonment were excluded in order to maximize the time
available for sampling. Third, a few companies declined us access
to wells. Given these constraints, 31 recharge wells were
selected for analysis. Locations of these wells are listed in
Table II-1.


Monitor Wells

In addition to sampling of the recharge wells, monitor wells
were installed adjacent to several recharge wells. Four monitor
wells were installed for us at recharge well KR-98B in January,
1984 by International Minerals and Chemical Corp. These wells,
which consist of 4 in. (0.10 m) PVC, have been described by Oural
et al. (1986). Deep and shallow monitor wells were installed both
up-gradient and down-gradient from the recharge well. Regional
gradient is from northeast to southwest. The shallow wells, which
are designated KR-98B-US and -DS (Up-Shallow, Down-Shallow), were
cased to a depth of 0.6 m (20 ft.), and had 0.6 m of slotted PVC
screen. The screen was set at the same horizon as the screen in
recharge well KR-98B. The deep wells were completed into the
upper portion of the Floridan aquifer with total depths of 63 m
(207 ft.) for the south, down-gradient well (KR-98B-DD), and 63.7
m (209 ft.) for the north, up-gradient well (KR-98B-UD). In both
deep wells, casing was set and cemented to 36.6 m (120 ft.),
which is within the Arcadia Formation dolostones. The open-hole
portion of the deep monitor wells coincides with the portion of
the recharge well that was receiving the falling water. In fact,
geophysical logs (Oural et al., 1986) indicate that a cavernous
zone at 53-54.6 m (175-179 ft.) below land surface is the primary


II-1









receiving zone and that the recharge well was "silted up" to a
depth of 63.7 m. Shallow monitor wells were intended to be within
the "cone of depression" of the recharge well and were within 8 m
of the recharge well. Deep wells were designed to be within the
"cone of impression" of the recharge well and were within 30 m of
it. Geophysical logs of recharge well and monitor wells were
provided by the Florida Department of Environmental Regulation,
Southwest Florida Water Management District, and International
Minerals and Chemical Corp. The recharge well and all monitor
wells were plugged and abandoned in the summer of 1988 just prior
to mining. Appendix A gives lithologic logs of KR-98B.

The Florida Department of Environmental Regulation installed
monitor wells on recharge wells KR-83 and KR-127 (Table II-1) in
cooperation with this study. These wells were installed in early
1985. Each recharge well had a Surficial aquifer and a Floridan
aquifer monitor, each located to be within the influence of the
recharge well. Notation used to indicate the monitor wells is -S
for Surficial aquifer monitor and -D for Deep aquifer monitor.
All monitor wells are within 30 m of the recharge wells. Shallow
monitor wells are 0.05 m (2 in.) in diameter. KR-83-S has 6 m (20
ft.) of black-iron casing and a 1.5 m (5 ft.) stainless screen.
KR-127-S has 8.5 m (28 ft.) of black iron casing and 1.5 m (5
ft.) of wound-wire stainless steel screen. Pea gravel pack was
used opposite the screen and the casing was cemented to the land
surface. Deep monitor wells are 0.10 m (4 in.) in diameter. KR-
83-D was drilled to a depth of 80.8 m (265 ft.) and was cased to
39 m (128 ft.). KR-127-D was drilled to 82.8 m (271.75 ft.) and
cased to 52.7 m (173 ft.). The wells are "open hole" below
casing, which is cemented to the land surface. Casing is "Tri-
Lock" PVC, with the deepest 1.5 m (5 ft.) consisting of stainless
steel casing in order to allow seating of a packer when using
bladder pumps for sampling. The deep monitor well at KR-127 can
be shown to be in direct communication with the recharge well as
cavitated gas is forcibly discharged from the monitor well. These
well complexes were removed and the wells were abandoned prior to
mining in 1989.

All wells were developed by air lift and allowed to
stabilize for at least 4 weeks prior to sampling.


Sampling Protocols

Recharge wells were sampled at several horizons. Oural et
al. (1986) showed that there is a minor change in water chemistry
from falling water, to standing water, to bottom of the well.
These changes are a result of mixing of recharging water with
host water and, possibly, leakage from the Intermediate aquifer
into the recharge-well bore. After examining water chemistry and
difficulties in obtaining valid samples from falling water, it


11-2 s








was determined that a sample of standing water, just below the
potentiometric surface or peak of the "cone of impression", and a
sample from the bottom of the well would suffice to characterize
"end-member" compositions in the recharge wells.

Because of having to pass through the cascading water, it is
difficult to operate bladder pumps in the recharge wells, so
stainless steel bailers with PVC check valves were designed. The
bailers were constructed with 0.04 and 0.10 m (1.5 and 4.0 in.)
diameters. The bailers were designed with a seated, conical, PVC
valve in either end of the stainless-steel tube. Tension on the
rope from which the bailer was suspended was used to keep the
valve seated during lowering and raising-of-the bailer. The check
valves were opened-by releasing tension when the bailer was at
the desired depth and closed for retrieval of the sample, thus
preventing contamination. The top of each bailer was designed to
shed water and debris from the well casing. In spite of the
closed bailer, considerable difficulty was encountered in
micropore filtering the water samples owing to extreme turbidity.
The turbidity was a result of ferric hydroxide and microbial
filaments scraped off the casing as the bailer was lowered by
hand. It was discovered that use of a tripod to center the bailer
over the recharge well and a gasoline-powered "cat-head" to
smoothly lower and raise the bailer eliminated the problem. Since
the recharge wells are constantly flowing, there was no need to
purge the wells before sampling.

Monitor wells and a number of recharge wells were sampled by
means of Well Wizard bladder pumps powered by a portable,
gasoline-powered air compressor/controller. The Surficial aquifer
monitor wells were sampled with 0.05 m (1 7/8 in.) pumps, and the
Floridan aquifer monitors were purged with a 0.08 m (3 in.)
diameter pump and sampled with a 0.05 m (1 7/8 in.). pump. All
pumps have Teflon bladders, PVC shells, and stainless fittings,
and are fitted with an inflatable packer 0.15 m above the top of
the pump. The Surficial aquifer monitor wells were "packed off"
just above the top of the screened interval, and 5 well volumes
purged before sampling. The Floridan well monitors were first
purged by means of a 0.08 m air-lift pump, then "packed off"
against the casing just above open hole. Then the bladder pump
was allowed to pump for a minimum of 15 minutes prior to sample
collection.

The monitor wells were sampled at the same time as the
recharge wells in order to determine Surficial and Floridan
aquifer water compositions before and after being disturbed by
the free-fall down the recharge well.

All samples were archived and "chain-of-custody" procedures
maintained to insure sample validity. Forms were used to track
sample analysis and record results at all stages of the study.


II-3









ANALYTICAL METHODS


Table II-2 lists the parameters measured as part of the
chemical study. All measurements were made using standard methods
unless otherwise indicated.


Field Methods

In the field, notation was made of well designation and
condition, weather, and other factors that might affect sample
results. Depth to the pump or bailer was determined from the
amount of line in the well. Water temperature, specific
conductivity, pH, Eh, dissolved oxygen, and alkalinity were
determined at the time of sample collection.

Temperature and conductivity were determined with a
calibrated YSI S-C-T meter and temperature was checked with a
laboratory thermometer. When the bladder pumps were used, Eh and
pH were measured in a closed reservoir with constant
replenishment. When bailers were used, these measurements were
made as quickly as possible from an undisturbed sample. Eh was
measured with a platinum electrode calibrated in the field to
Zobel's standard solutions. The pH was measured with a
combination glass electrode calibrated to NBS pH=4 and 7
standards in the field. Dissolved oxygen was measured with a YSI
DO meter, however, the results are inconsistent and were not used
in further interpretation of the data. Alkalinity was determined
by titration at the time of sample collection.


Laboratory Methods

Table II-2 lists the methods used for most analytical
procedures. Metals were analyzed from samples filtered at the
time of collection on a 0.045 im micropore filter and then
acidified according to specified protocol (see references in
Table 11-2).

All chemical analytical procedures included running of
standards every fifth analysis. Radioisotope analyses used
internal standards. All standards can be traced to U.S.
Environmental Protection Agency (USEPA)-approved procedures
and/or standard solutions. USEPA check samples were included in
each procedure. Replicate samples were analyzed on the average
every four samples. Analyses were redone if analytical errors
exceeded those indicated in the published method, or if the USEPA
check sample concentration disagreed with the measured
concentration allowing for published confidence limits.


11-4







Radioisotope analytical methods are described in detail in
Oural et al. (1986). Only modifications to these methods are
discussed below. There were no modifications to the uranium-238
and uranium-234 procedures.

Gross-alpha radiation method is the procedure specified in
American Society for Testing and Materials (1981). The method
used is D 1943-66(1977). Allowed sample storage time is 1 year.
The gross-alpha method is considered to be a screening procedure
at best, and results presented in this report indicate that it is
highly unreliable in comparison with isotopic analysis. For
example, Oural et al. (1988) have shown that, if polonium-210
represents a major proportion of the alpha-emitting isotopes in
the water, gross-alpha activity can underestimate actual activity
because of volatilization of polonium during evaporation on the
planchet. In addition, the short-lived radon daughters can
significantly influence gross-alpha analytical results (Oural et
al. 1987, 1989). Presumably, radon is lost through agitation and
degassing at the time of sample collection, or during evaporation
on the planchet. The alpha-emitting daughters are not entirely
lost and, until they decay to a level supported by radium in the
water, a time-dependent alpha activity that has little
correspondence with actual water chemistry results. Consequently,
the length of time required for evaporation of the water on the
planchet can affect gross-alpha activity. Low total dissolved
solids (TDS) waters are evaporated over a long period of time in
order to accumulated sufficient alpha activity to optimize
counting statistics. The radon daughters deposited early in the
evaporation have time to decay to lead-210, those deposited later
in the procedure do not. In contrast, the volume of high TDS
waters that can be deposited on the planchet is limited because
of self-absorption problems of the residue. Therefore, short-
lived radon daughters may not have time to decay away prior to
counting and the resultant analysis is high. This leads to the
often-described correlation between TDS and gross-alpha activity.
Finally, the holding time of the sample and/or planchet affects
activity measurement. Samples collected, prepared and run
immediately can have high alpha activities because of the
presence of radon progeny. Samples held for a few weeks, or
prepared and then stored for a few days have relatively low alpha
activities because of loss of radon daughters. These problems are
discussed and modeled in Oural et al. (1987, 1989).

It would be most desirable if gross-alpha samples could be
analyzed immediately so that maximum exposure potential could be
evaluated. Unfortunately, laboratories cannot operate on such a
short "turn-around" time, so the only alternative is to avoid
providing inconsistent analyses by allowing the radon daughters
in each sample to grow-out before analysis. This was done for the
gross-alpha analyses reported herein.


II-5








Lead-210 was analyzed by first precipitating metals from the
sample with ammonium hydroxide. Lead and polonium co-precipitate
with ferric hydroxide in the sample, and.the precipitate can be
stored indefinitely without fear of losses from sorption on
sample container walls. To analyze for lead-210, which is a beta
emitter and difficult to determine, the precipitate was stored
for 6 months to allow polonium-210 that is supported by lead-210
to grow in. Analysis for polonium-210 in sample splits soon after
precipitation (t=0) and after the "grow-in" period (t) allows
determination of lead-210 by the analytical expression

A2 (2 1X) + Ao2e-l2 (1 12)
12 (e-l-e-^)
(II.1)

where A01 = activity of lead-210 at time t = 0, A02 = activity of
polonium-210 at time t = 0, A2 = activity of polonium-210 at time
t, and 11 and 12 = decay constants for lead-210 and polonium-210,
respectively.

The method utilized for polonium-210 is described in Oural
et al. (1986). The method was modified slightly by addition of a
oxidant to break up organic complexes that were thought to
interfere with recovery of polonium. These complexes were
described in Oural et al. (1987, 1989). To destroy these
complexes 1-1.5 ml of 30% H202 was added to each liter of sample
upon return to the laboratory. After allowing the sample to react
for 3 days, the polonium was coprecipitated with ferric hydroxide
according to the published method. Addition of hydrogen peroxide
significantly enhanced recovery of the polonium.

Improved accuracy and counting statistics were obtained by
adopting the radium-recovery technique of Moore and Reid (1973).
This method involves preparation of manganese oxide-dioxide
impregnated acrylic fibers. These fibers quantitatively extract
radium from water, and can be utilized for analysis of radium in
ground water (Michel et al., 1982). Eight to ten liters of water
were passed through a cuvette of manganese-treated fibers to
extract the radium. Analysis of fibers from a second cuvette in
line with the first indicates complete removal of the radium on
the first cuvette. The fibers were then removed, ashed to reduce
volume and improve counting geometry, and sealed in a tin "snuff
can" for analysis by gamma spectroscopy (Michel et al., 1982).


Isotherms

Isotherms are laboratory models that allow approximation of
the ability of small amounts of dissolved chemicals to sorb onto


II-6








a solid substrate. Briefly, 4-5 solutions of the chemical of
interest are prepared in varying concentrations. These are placed
in contact with known masses of substrate (rock or soil in this
case) and agitated for a fixed length of time. The acidity of the
mixtures may or may not be buffered, depending upon the goals of
the experiment. After agitation, the supernatant is analyzed for
the chemical. The difference between original concentration and
final concentration in each solution is the amount adsorbed. The
"rate" of sorption as a function of starting concentration is
used to characterize the sorptive capacity of the material and
approximate a maximum sorption capacity.

Two isotherms are commonly used (Ellis and Knezek, 1972).
These are the Langmuir and Freundlich isotherms. The Freundlich
isotherm is given by the equation

1
x = k.Cn,,
m
(11.2)


where x the quantity of ions adsorbed per unit weight of
m
-adsorber, C = the equilibrium concentration of the adsorbate
after adsorption has occurred, k and n = constants. The equation
can be rearranged as

log () = log C + log k
\m n
(11.3)

and a graph of log (M) versus C yields a straight line with a
slope of 1/n and an intercept of k, if the Freundlich isotherm is
followed. A limitation of the Freundlich isotherm is that it does
not predict an adsorption maximum.

The Langmuir isotherm equation is

x K-M-b
m 1 + K-M'
(11.4)

where x = meq of ion adsorbed per 100 g of adsorber, M =
m
activity of ion in question in moles/liter, b = the absorption
maximum of ion M in meq/100 g, and K is a constant derived from
the ratio of adsorption to desorption rates. Equation 11.4 can be
rewritten as


II-7








M 1 M

\mj
(11.5)

which can be evaluated by linear regression techniques. The
adsorption maximum (b) is the reciprocal of the slope of the
regression, and the "adsorption rate" (K) is the reciprocal of
the intercept divided by b. K is not a time-based rate, and
cannot be used to characterize reaction kinetics.

Polonium sorption has been evaluated by Hansen and Watters
(1971), and it was found that, in typical Florida soils, 17 to 77
times as much polonium is sorbed or precipitated as remains in
the solution. Since strong sorption is indicated and the results
of this study suggest those chemical processes that can cause
desorption, it was concluded that sorption isotherms for polonium
need not be determined.

A major question remains as to the mobility of the polonium
predecessor, lead-210. Isotherms were obtained for lead using (1)
slightly argillaceous and ferruginous quartz sand, which is
typical of the Surficial aquifer; (2) typical, smectite- and
palygorskite-rich, Peace River Formation clay; and (3) calcitic,
Suwannee Limestone from cores at KR-98B and KR-127D. Stock
solutions of 1, 5, 10, and 20 mg/1 lead were prepared from PbC12
in deionized water. Room temperature was maintained at 26C. The
samples were agitated on a shaker table for 24 hours. The pH of
the solutions before and after contact with the adsorbant and the
pH of a blank were monitored. Table II-3 gives the conditions of
the experiment. Since starting and final pH's are similar to
actual field pH's, and buffering of the pH by reaction with
carbonate is a predicted reaction as recharge-well water
interacts with the Floridan aquifer, no attempt was made to
control pH's by addition of buffers. The alkalinity obtained in
the experiment using limestone is sufficient to cause
precipitation of PbCO3 (cerussite), rather than induce
adsorption.


CHEMICAL MODELS

Equilibrium Models

The interactions of the chemical species in ground water are
extremely complex and use of simple concentration data often
obscures relationships of chemical species in the system. The
solutions investigated contain sufficient dissolved ionic species
to cause ion pairing. For example, not all calcium is present as
Ca+2, some of the calcium is present as chemical complexes with


II-8








sulfate, carbonate, bicarbonate, fluoride, and other anionic
groups. Thus, when determining the relationship of calcium to
other chemical constituents, it is necessary to identify that
portion of the total calcium which is actually free calcium ion
and available to participate in chemical reactions. The
concentration of a free ion, as opposed to total concentration,
is known as the activity (a), where activity is related to the
concentration by

a = y m,
(11.6)

where r = the activity coefficient, and m =-molality of the
constituent. Molality is the total, analytical concentration, and
7 is a proportionality constant that denotes the fraction of the
total, analytical concentration that is present in chemical
complexes.

The amount of completing, or r, of an constituent is a
function of the total charge concentration in the solution. This
is calculated as the ionic strength (I) by the equation (Garrels
and Christ, 1965)

I=
2
(11.7)

where mi is the molality of charged species i, and zi is the
charge of species i. The r can then be calculated by a number of
different equations. The equation used in this study is the
Davies equation, which is best for ionic strengths less than 0.5.
The equation is

log Y i = -AziR,
(11.8)

where yi = the activity coefficient of species-i, A is a
constant, zi is the charge on the ion i, and R is determined
from the ionic strength by

R = T 0.31.
1 + /T
(11.9)

For Ca2 ,Mg2I,Na,K',C1-,S04~,C02,, and HCO3, the extended Debye-
Huckel equations (Truesdell and Jones, 1974; Drever, 1988) are
used. The extended Debye-Huckel equation is


II-9










log i = z + bl,
1 + Bax +l
(II.10)

where A and B are constants that depend upon pressure and
temperature, and a ~ is the hydrated ionic radius of the species
in question.

To solve the simultaneous equations necessary to calculate
the activities of important dissolved constituents and complexes
(ion pairs), a computer program, WATEQF, by Plummer et al. (1976)
was used. This program calculates activity coefficients and
activities of 193 aqueous species and minerals. It also
calculates solubility products and degree of saturation of the
water with respect to the mineral phases.


Statistical Models

Once the complex ions were identified, chemical models were
constructed to explain behavior of the radioisotopes of interest.
These models are based on (1) first principles of chemistry, such
as basic chemical reactions of major species, and (2) a
statistical search for chemical relationships by use of
correlation analysis. The correlation analysis search was
necessary because the radioisotopes are present in such minute
concentrations that the presence of even trivial activities of
complex ions may cause unpredicted chemical complexes and
mobilization of the isotopes of interest.

Two modes of correlation analysis were undertaken in the
search procedure, which involved use of ABSTAT, a commercially-
available statistical package (Anderson-Bell, 1987). Pearson
product-moment correlation and Spearman rank correlation
coefficients were calculated for the monitor-well data and for
the recharge-well data.

Pearson product-moment correlation coefficients (r's)
require that the data be normally distributed, and they can be
rigorously tested for significance (Sokal and Rohlf, 1969).
Missing data are omitted from the analysis in a pairwise
deletion, so Pearson product-moment correlations allow
maintenance of large sample sizes and, therefore, better
confidence estimates. Levels of significance of r's were
determined by two criteria: (1) reference to a table of
significance (Rohlf and Sokal, 1981) for critical levels of r,
given an a level of 0.05 (P = 95%) and appropriate degrees of
freedom, and (2) calculation of r to determine the amount of
covariance accounted for by the correlation. Only correlations


II-10









that account for at least 50% of the covariance (r2 = 0.50, r =
0.707) were accepted as representing a significant relationship.

Spearman rank correlation coefficients (p) are based on
principles of nonparametric statistics and do not require
normalcy of the data. However, the significance of the
correlation coefficient is not easily tested (Sokal and Rohlf,
1969). Also, since the method is a ranking system, missing data
are omitted listwise, which eliminates all variables for any
sample with missing data. The result is a greatly reduced sample
size, and ability to make inference, in comparison to product-
moment correlation. Quantiles of the Spearman test statistic were
used to test for significance (Conover, 1971). An a level of 0.05
(P = 95%) was used to test for significance.


SEQUENTIAL-DATA ANALYSIS

Task 3 of this study involves determination of the temporal
and spatial distributions of gross-alpha radiation in recharge
well water. An unique data set exists for this determination. The
Southwest Florida Water Management District (SWFWMD) has
accumulated analyses submitted by the companies for many years.
All totaled, the data set assembled for the time-series analysis
included sequential data from 270 recharge wells installed in the
Central Florida Phosphate District. A list of these data sources
is presented in Section V of this report. Most (48.9%) of the
wells were installed in the Keysville Quadrangle. Homeland
(13.7%), Bradley Junction (11.5%) and Duette NE (11.5%)
quadrangles account for most of the remainder of the available
data. Data from the recharge wells range from only 1 gross-alpha
radiation analysis to 84 months of data. The average well has 9.4
monthly analyses.


Time-Series Analysis

While most of the records are not suitable for time-series
analysis, some are excellent for this purpose. Data sets were
selected for time-series analysis according to the following
criteria:

1. Only well data from companies shown by Oural et al.
(1986) to have good sampling and analytical procedures were
used, and

2. Only data sets that have continuous monthly records of at
least 24 months were used.

Ten data sets met these criteria. Table II-4 lists the wells used
for time-series analysis. Data sets which have single


II-11









observations that are anomalously high were avoided because it is
likely that the anomalous activities represent interference from
short-lived radon progeny and the observations are not comparable
to data from other months (Oural et al., 1987, 1989). Attempts at
time-series analyses of these "spiky" data sets failed because
the peaks are artifacts of sample collection and analysis, and
they are randomly distributed throughout the time of data
collection.

There were no monitor wells with water-level data from the
Surficial aquifer that were suitable for correlation
with the gross-alpha data provided by the. water management
district. There is, however, a well-maintained precipitation
gauge at the International Minerals and Chemical Corp. New Wales
Chemical Plant located at the northwest corner of the Kingsford
Mine, Keysville Quadrangle. Data from this gauge correspond in
time and interval with the gross-alpha data, so comparisons with
precipitation are possible. By extension, variations in the
precipitation data should correspond in a general way with
variations in water levels in the Surficial aquifer. These data
were used to relate the gross-alpha radiation data to hydrologic
conditions in the Surficial aquifer. All of the data used in this
paper are from wells within approximately 8 km of the
precipitation gauge. Much of the drainage in the mines is
internal, so water levels in the Surficial aquifer respond
directly to precipitation.

Smoothing, autocorrelation and cross-correlation techniques
(Davis, 1973) were used to identify temporal patterns in the
data. Correlations with precipitation were used to determine the
behavior of the gross-alpha radiation with water-table
fluctuations. The smoothing algorithm used in this study is

i.caic. = 01.Yi-2 + 0.2Yl + 0.4Y, + 0.2Yi + O. Yi+2,
(II.11)

where Yi,cac,. is the smoothed value of Y at time i, and
i-2z ... ,Yi2 are the observed values of Y at times i-2,
i-1,...,i+2.


Spatial-Analysis Methods

It is well known that radon gas moves along fracture traces
(Oural et al., 1986; Banwell and Parizek, 1988) and, since
polonium-210 is a radon daughter, an attempt was made to
correlate the locations of wells that chronically exceeded the 15
pCi/l gross-alpha radioactivity MCL with photolinear features,
which are presumed to represent fracture traces (Parizek, 1976).
Since the majority of the data are from wells on the Keysville


II-12








Quadrangle, and since there are two independent fracture-trace
studies (Upchurch and Kaufmann, 1979b; Culbreth, 1983) and an
excellent geological study (Cathcart, 1963) of that quadrangle,
it was selected for spatial analysis. No judgments as to the
presence or absence of photolinears were made as part of this
study. Cathcart (1963) did not locate photolinear features, but
his maps of overburden thickness and thickness of the "leached
zone" show linear trends that coincide with many of the features
identified by other workers as photolinears, and greatly assist
interpretation of the processes operative in the area. No
selection was made on the basis of company or sampling and
analytical procedure, but the data are:from companies found to
have adequate procedures (Oural et al., 1986). Only wells with
over 10 months of record were used in the spatial analysis.

To clarify any relationship between fracture-trace location
and wells that show high gross-alpha radioactivity levels, the
shortest distance from each well with at least 10 months of
record to the nearest known photolinear feature was measured.
This distance, a "nearest-neighbor" statistic, was used to
investigate the correlation of proximity to photolinear features
with radiation distribution.


II-13




































Table II-I. Recharge uells sampled for chemical and radiological analysis in this study. n.d. = not yet determined.

i---------- i --------- USGS --Conpany --Conpany County Latitude Longrtude Section, Conents
Sample Pernit Uell ID Quadrangle Well ID rounship, &
Number Nunber Range Loc.
RO4-- -ak---ity-- --AR- -- --- -nato----- ------ ----- 3----3----2-1-----------------
RG-Ol.- 20o0029 UR-4 .nyakka City NM U.R. Grace RK-Ol i Manatee 27.28.7 82.11.15 3s-34-21
URG-02- 200029 MRG-5 tyakka City NU U.R. Grace RK-02 Manatee 27.28.55 82.11.01 26-34-21
APRU-2- 200332 RMX-14 Keysville lnAX/MOBIL PRW-2 Hillsborough 27.46.26 82.04.02 24-31-22
APRU-6- 200332 RnX-18 HMnRX/OBIL PRU-6 27.43.41 82.04.33
RPRU-9 200332 AHX-22 Duette NE RMRX/MOBIL PRU-9 Hillsborough 27.43.56 82.03.20 01-32-22
APRU-10- 200332 RMHt-23 Duette NE RMAfR/MOBIL PRU-10 Hillsborough 27.43.56 82.04.46 02-32-22
TRRW-17- 200332 RnH-31 Duette NE HRMX/nOBIL TRU-17 Hillsborough 27.43.09 82.04.54 02-32-22
RTRU-28- 200332 RnX-42 Keysville flfRMX/OBIL TRU-28 Hillsborouqh 27.45.15 82.04.52 26-31-22
RrRU-29- 200332 RtM-43 Duette HE RMRX/nOBIL rRW-29 Hillsborough 27.42.32 82.04.26 11-32-22
RrRU-30- 200332 RMH-44 Duette NE RMRX/HOBIL rRU-30 Hillsborough 27.42.36 82.04.21 11-32-22
KR-22- 203053 IMC-61 Keysville Inc KR-22 Polk 27.49.03 82.00.56 04-31-23
KR-80- 203053 IHC-65 Keysville Inc KR-80 Polk 27.48.45 82.01.28 05-31-23
KR-83- 203053 ItC-68 Keysville InC KR-83 Polk 27.48.39 82.01.53 05-31-23 Monitor wells
KR-86- 203053 IMC-71 Keysville Inc KR-86 Polk 27.48.36 82.02.06 05-31-23
KR-87R- 203053 ItC-72 Keysville IMC KR-87R Polk 27.48.45 82.02.13 05-31-23
KR-95- 203053 IMC-76 Keysville Inc KR-95 Polk 27.48.52 82.02.50 06-31-23
KR-98B- 203053 IftC-80 Keysville InC KR-98B Polk 27.49.10 82.00.55 04-31-23 Monitor wells
KR-117- 203053 IMC-97 Keysville IMC KR-117 Polk 27.48.48 82.01.07 04-31-23
KR-118- 203053 IlC-98 Keysville IMC KR-11B Polk 27.48.33 82.01.30 05-31-23
KR-120- 203053 IMC-100 Keysville IMC KR-120 Polk 27.48.31 82.01.21 04-31-23
KR-122- 203053 IIC-67 Keysville Inc KR-122 Polk 27.48.50 82.02.25 06-31-23
KR-127- 203053 IMC-104 Keysville IlC KR-127 Polk 27.48.52 82.02.39 06-31-23 Monitor uells
KR-135- 203053 ItMC-112 Keysville Inc KR-135 Polk 27.48.33 82.00.35 04-31-23
MR-2- Inc NR-2 Polk
NR-3- Inc NR-3 Polk
NR-4- Inc NR-4 Polk
MR-5- INC NR-5 Polk
NR-6- Inc NR-6 Polk
NR-7- InC NR-7 Polk
NR-8- InC NR-9 Polk
NR-9- InC .NR-9 Polk
--------- ----------------------------------------------------------------------------


II-14









Table 11-2. Variables measured and methods used in the chemical
study. APHA = Rand et al., 1976; USEPA = Environmental Monitoring
and Support Laboratory, 1979; USGSa = Wershaw et al., 1983; USGSb
= Fishman and Friedman, 1985; ASTM = Amer. Soc. Testing and
Materials, 1981.


Variable Units Method Used


Field Measurements
Temperature
Eh redoxx pot.)
pH
Dissolv. Oxygen
Alkalinity


Laboratory Measurements
Bicarbonate
Chloride
Fluoride
Phosphate (sol.)
Silica
Sulfate

Aluminum
Calcium
Iron (tot.)
Lead
Magnesium
Manganese
Potassium
Sodium
Strontium

Tot. Dissol. Solids
Tot. Org. Carbon
Tot. Volatile Solids

Lead-210
Polonium-210
Radium-226
Uranium-238
Uranium-234
Gross Alpha Radiation


deg. Celsius
millivolts
units
mg/l
mg/l


mg/1
mg/1
mg/l
mg/1
mg/1
mg/1

Ag/l
mg/l
Ag/1
Sg/1
mg/1
Ig/1
mg/l
mg/1
Ag/1

mg/l
mg/1
mg/l

pCi/l
pCi/l
pCi/l
pCi/l
pCi/l
pCi/l


APHA (USEPA 170.1)
Orion Research, 1978
USEPA (150.1)
USEPA (360.1)
USEPA (310.1)


Calculated from alkal.
USGSb
USEPA (340.2)
APHA
USEPA (370.1)
USEPA (375.1)


USEPA
USEPA
USEPA
USEPA
USEPA
USEPA
USEPA
USEPA
USGSb


(202.2)
(215.1)
(236.2)
(239.2)
(242.1)
(243.2)
(258.1)
(273.1)


USGSb
USGSa,b
USGSb


See text
See text
See text
Oural et al. (1986)
Oural et al. (1986)
ASTM (D 1943-66)


II-15









Table 11.3 Conditions of the lead isotherm determinations.


Sample pH


Starting solutions

20 mg/l PbC12 solution 4.98
10 mg/l PbCI2 solution 5.13
5 mg/l PbCI2 solution 5.28
1 mg/l PbC12 solution 5.70
Deionized water blank 5.70

Sample SW-6, sl. arQillaceous and ferrugineous quartz sand from
KR-98B after equilibration

20 mg/l PbC12 solution 5.76
10 mg/l PbCl2 solution 5.78
5 mg/l PbC12 solution 5.78
1 mg/l PbClI solution 5.68

Sample of Suwannee Limestone from 267 ft. BLS at KR-127-D after
equilibration

20 mg/l PbCI2 solution 8.92
10 mg/l PbC12 solution 8.98
5 mg/l PbCI2 solution 8.87
1 mg/l PbCl2 solution 8.70

Sample of Peace River Formation clay from 259 ft. BLS at KR-127-D
after equilibration

20 mg/l PbCz1 solution 5.20
10 mg/l PbCl2 solution 4.91
.5 mg/l PbCI2 solution 4.83
1 mg/l PbCI2 solution 4.74


II-16








Table 11-4. Sources of historical gross-alpha radiation data
utilized for time-series analysis. Data are from records
collected by the Southwest Florida Water Management District. See
Table V-l for more details.


Well Record duration Months of continuous
record


a. IMC-68 44 months (1978 1985) 35 months
b. IMC-85 40 months (1978 1985) 32 months
c. IMC-97 42 months (1978 1986)- 35 months
d. IMC-104 42 months (1978 1986) 37 months
e. AMX-14 37 months (1977 1986) 31 months
f. WRG-4 84 months (1978 1985) 85 months
g. WRG-5 83 months (1978 1985) 84 months
h. MOB-43 25 months (1981 1984) 24 months
i. MOB-67 32 months (1982 1985) 34 months
j. MOB-69 31 months (1982 1985) 33 months


II-17








III. RESULTS CHEMICAL DATA


DATA EVALUATION

Two sets of data were collected as part of this study. Data
from the monitor wells represent undisturbed samples that reflect
the ambient chemistry of the aquifer system. These data are
important for understanding the chemical environments in which
radionuclides are mobilized or fixed, and, therefore, monitor-
well data are the main focus of the interpretation portion of
this report. The recharge-well data reflect water that is in the
well environment for a short time, and, therefore, in
disequilibrium. These data are best used for determination of the
extent of any problem and for determination of any chemical
changes caused by recharge.


Monitor-Well Data

Introduction The chemical data derived from monitor wells are
given in Appendix B. These data allow characterization of the
Surficial and upper Floridan aquifers outside of the recharge
well environment. A detailed chemical analysis of a sample from
HK-1, the well with the highest polonium-210 activity yet
reported in the state (Cowart et al., 1987), is included for
comparison in Appendix B. Table III-1 summarizes chemical
conditions in the two aquifers. The aquifers differ significantly
in bulk composition. The major-element chemistry of the two
aquifers represents the lithologies of the respective aquifer
frameworks.


Water Type Bulk water chemistry is controlled by interactions
with rock materials. The bulk chemistry differs greatly from a
siliciclastic aquifer, such as the Surficial aquifer, where
chemical reactions are minimized owing to the low reactivities of
quartz and other minerals, to a carbonate aquifer, such as the
Floridan aquifer, where highly reactive minerals are present.


Surficial aquifer. Water compositions from the Surficial
aquifer monitor wells range from sodium-sulfate to mixed waters
with sub-equal proportions of calcium, magnesium, and sodium and
a linear mixture of sulfate and bicarbonate (Figure III-1).
Average total dissolved solids (TDS) is 496 mg/l. The Eh-pH
conditions average -96.6 my and 5.3 pH units, with ranges in
values of -266.7 my to +17.6 my and 4.3 to 6.4 pH. Total organic
carbon (TOC) averages 8.3 mg/1, with a range of 6.0 to 13.1 mg/l.


The Surficial aquifer is a quartz sand aquifer with minor


III-1









carbonate mineral content. Particulate organic are widely
distributed through the aquifer, and are concentrated in organic-
rich hardpans. Siliciclastic clays are present and may form
argillaceous beds, especially near the base of the aquifer. In
the absence of carbonates to buffer the system, the aquifer water
tends to be acid and the predominant anions are acid radicals,
such as sulfate (Table III-1, Appendix B).


Floridan Aquifer. The upper Floridan aquifer is a karstic
limestone that contains calcite and minor dolomite. The upper
portions of the aquifer include minor phosphate, siliciclastic
clay, and quartz.

Floridan aquifer water is a calcium-magnesium-bicarbonate
water (Figure III-1) that averages 289 mg/l TDS. Calcium is
slightly predominant(approximately 45% of the cation
equivalents), magnesium represents about 40% of the equivalents,
and sodium constitutes about 15%. Bicarbonate is the predominant
anionic constituent, and represents 90% of the anionic
equivalents. Chloride and sulfate are minor constituents. Eh
averages -139 mv, and ranges from -235.7 to 64 my. The pH
averages 7.3 and ranges from 7.1 to 7.5. Total organic carbon
averages 4.4 mg/l, with a range of 3.4 to 4.9 mg/l.

The high pH and low range in pH values in the Floridan
aquifer are a result of buffering by carbonate, where acidity is
consumed according to reactions, such as dissolution of calcite

CaCO3 + H' Ca2+ + HCO ,
(III.la)

and dolomite

CaMg(C03)2 + 2H+ -Ca2' + Mg2 + 2HCO .
(III.lb)

In reactions of this sort, acidity is consumed and the
predominant reaction products include cations, such as calcium
and magnesium, and bicarbonate.


Sulfur Species The distributions of sulfur species, especially
sulfate (S04-) and sulfide (S2-) species, is critical to mobility
of polonium, and perhaps other isotopes.


Surficial aquifer. The high sulfate content of Surficial
aquifer monitor-well water appears to be a result of three


III-2







processes: (1) oxidation of pyrite in clay-rich zones of the
aquifer, (2) destruction of sulfide- and sulfate-bearing
organic, and (3) influx of waters from a nearby waste gypsum
disposal area (gypstack) and cooling pond. Hutchinson (1978)
included analyses of seven Surficial aquifer water samples in the
Alafia RIver basin. He found two calcium-sulfate water samples
from wells that are distant from the phosphochemical plants.
Therefore, high sulfates are not necessarily a product of
fertilizer plants.

Comparison of the Eh-pH conditions of the Surficial aquifer
(Figure III-2) indicates that, at 25*C, the water falls at or
below the boundary between SO e and -H2 aous (Garrels and
Christ, 1965), with over half of the samples well within the
HzSaqeous stability field. Equilibrium kinetics between the two
sulfur species are dominated by bacteria, and metastable species
can exist if bacterial action is inhibited. Connell and Patrick
(1968) investigated the stability of sulfate and sulfide in
waterlogged soils, such as the Surficial aquifer. They found that
microbial sulfate reduction is inhibited and sulfide formation is
minimized at redox potentials above -150 my and at pH's outside
the range of 6.5 to 8.5. Since optimal conditions for microbial
sulfate reduction may not-be present (low pH values and Eh values
that tend to be above -150 my), sulfate, not sulfide, may
predominate as a metastable species (Figure III-2).


Floridan aquifer. Sulfate in the Floridan aquifer is formed
by oxidation of pyrite or vertical movement of sulfur-rich water.
Downward leakance of Surficial aquifer water introduces sulfate
and up-coning of waters that have been in contact with gypsum
(CaS04.2H20) and/or anhydrite (CaSO4) near the base of the
aquifer can be a source, albeit an unlikely one in this data set.
Eh-pH relationships (Figure III-2) indicate that sulfate and
sulfide are stable in the Floridan in those portions of the
aquifer sampled. The low absolute concentrations of sulfate (X =
8.3 mg/l, range = 0.0 to 50.0 mg/l) in the Floridan indicate that
sulfate reduction, sorption/precipitation reactions, and dilution
reduce sulfate in comparison with the Surficial aquifer. The Eh-
pH range in the Floridan is suitable for microbial sulfate
reduction according to the criteria of Connell and Patrick (1968).

Iron Species As Figure III-3 indicates, iron is also sensitive
to Eh-pH conditions. Ferric-hydroxide precipitates are well known
scavengers of metals in aqueous systems. Therefore, it is
critical to understand how iron speciation affects radionuclide
mobility.


III-3









Surficial aquifer. Total iron concentrations in the
Surficial aquifer average 5.4 mg/l, and range from 1.5 to 11.0
mg/l. Comparison with the Eh-pH stability fields for iron species
2+
(Figure III-3) indicates that Feaqeous is the stable iron species.
Clearly, iron is soluble and mobile in the Surficial aquifer.


Floridan aquifer. Iron is present in very low quantities in
the Floridan aquifer, as compared to the Surficial aquifer. Iron
concentrations average 12.6 /g/l, and range from 0.0 to 43.5
gg/l. Eh-pH stability field data (Figure III-3) suggest that
Feaueous is the stable species, so one can only conclude that iron
concentrations are reduced en route to the aquifer by
precipitation of ferrous sulfide or ferric oxyhydroxides, by
sorption on aquifer matrix minerals, or by dilution.


Organics A very important component of the chemical make-up of
Surficial aquifer water is its organic content. Many naturally
occurring, soluble organic are strong chelating agents, so
metals and radionuclides may be mobilized in the presence of the
organic.


Surficial aquifer. Total organic carbon averages 8.3 mg/l.
and concentrations as high as 13.1 mg/l were reported. These are
very high organic carbon values with respect to natural organic
in aquifer systems. Meybeck (1981, 1982) summarized organic
carbon in rivers. He estimated that the average dissolved organic
carbon content is 5.8 mg/l. Ground waters are generally
considered to contain less organic carbon that do surface waters
(Hem, 1985). Hem's (1985) citation of an aquifer with
exceptionally high organic carbon is the study by Thorstenson et
al. (1979) in which TOC's as high as 30 mg/l were reported.


Floridan aquifer. Total organic carbon is low in the
Floridan aquifer, as compared to the Surficial aquifer. However,
an average of 4.4 mg/l is still high compared to other aquifer
systems and the deep Floridan. The TOC present in the Floridan at
the monitor wells is probably a result of the influence of the
recharge wells. Since there is a significant increase in pH and
alkalinity from the Surficial aquifer to the Floridan, humic
acids and other pH sensitive organic groups should be flocculated
and form advantageous substrates for bacteria in the vicinity of
the recharge wells.


Radionuclides In comparison with the Floridan aquifer, the
Surficial aquifer is typically high in polonium-210 and gross-


III-4








alpha radioactivity. Other radioisotopes exhibit smaller
differences in activity between the two aquifers.


Surficial aquifer. There is little difference in uranium-238
and -235 activities in the Surficial aquifer as compared to the
Floridan aquifer. However, the activity ratio (234U/23U) averages
2.8, with a range of 1.3 to 4.5. This is significantly greater
than the ratio in the Floridan and suggests that there is a
significant reservoir of fixed uranium available to the aquifer.
While the Eh and pH of the system limit uranium solubility, alpha
recoil of 234U in the surficial reservoir causes elevated activity
ratios. This reservoir is the "leached zone".

Radium-226 averages 1.6 pCi/l, with a range of 0.2 to 3.4.
As indicated earlier, radium can form soluble complexes with
sulfate, which may enhance radium mobility. The activities
detected in the monitor wells suggest that either radium is
limited in availability, or that mobilization is minor because
the radium is fixed as a precipitate or is sorbed onto soil
particles.

Lead-210 and polonium-210 represent an interesting pair in
the Surficial aquifer. Lead-210 is the parent of polonium-210,
and the two should be in approximate radioactive equilibrium in
the ground water, given similar chemical properties. In fact the
average lead-210 activity is 1 pCi/l (range 0.0 4.3)-and
polonium-210 is 18.8 pCi/l (range 0.2 54.4). Thus, an average
activity ratio of 210Po/21 Pb is 18.8. That is, polonium-210 is
almost 19 times as mobile as lead-210 in the Surficial aquifer.
In the Surficial aquifer, chemical conditions favor precipitation
and/or sorption of lead. In contrast, the acid, organic-rich
Surficial aquifer water enhances mobilization of polonium.


Floridan aquifer. Near the recharge wells, radionuclides are
less abundant in the Floridan aquifer than in the Surficial
aquifer (Table III-1). Gross-alpha activities are low, which
indicates low activities of radium and polonium.

Uranium isotope activities are similar to those in the
Surficial aquifer, but the activity ratio is nearer to unity.
This indicates that there is less uranium-234 injected into the
aquifer waters by alpha recoil, which is probably an artifact of
a small reservoir of uranium-238.

Radium is low in activity. Although other analyses (Upchurch
et al., 1979a, 1979b) of Floridan aquifer waters in the same
general area have shown moderately high radium activities, the
data presented herein suggest that the recharge wells studied are
not contributing radium to the Floridan. The high radium waters


III-5









are found in deep, relatively stagnant portions of the aquifer.

Lead-210 is low owing to precipitation of lead carbonates
and polonium is equally low, although the mechanism of fixation
is not well known. Polonium hydroxides and radiocolloids are pH
sensitive and appear to precipitate in neutral to alkaline
waters.


Recharge Wells

Introduction Chemical data from the recharge wells are given in
Appendix B and summarized in Table III-2. Comparison with Table
III-1 indicates that recharge well water is chemically similar to
water in the Surficial aquifer, which discharges into the wells.
Significant differences in Eh of the water, however, affect the
behavior of iron and sulfur.


Water Type The bulk composition of typical recharge-well water
is statistically identical with Surficial aquifer water. The
water is a sodium-sulfate to mixed water. The average pH of the
water from wells associated with the monitor wells is 5.1 (the
Surficial aquifer water pH is 5.3). Average Eh, however, is 23.7
my, with a range of -158.1 to 97.9 my. This represents a
significant increase in oxidation potential within the well as
compared to both the Surficial aquifer (mean Eh = -96.6) and
Floridan aquifer (mean Eh = -139.1 my). Examination of the data
in Appendix B also shows that most of the recharge-well samples
had positive Eh values.

Clearly, the dynamics of water (1) entering the well
environment, (2) free falling to the top of the cone of
impression, and then (3) flowing down the well to an exit horizon
represents potentially rapid changes in chemical environment. As
the water enters the recharge well environment it is mildly
reducing in nature, as it flows through the well it is turbulent
and becomes oxidizing, and then, as it enters the Floridan, it
becomes reducing again.


Sulfur Species The Eh-pH relationships are such that most of
the samples fall within the sulfate stability field (Figure III-
2). Clearly, there is a strong tendency to oxidize sulfur within
the recharge-well environment. Well screens and casings are
usually covered by a mat of pale gray, filamentous bacteria and
other organisms. These biomats include elemental sulfur, and
indicate that sulfate reduction is possible within the protection
of the biomat environment.


S III-6








Iron Species Iron concentrations in the recharge wells reflect
the Surficial aquifer waters. However, the average iron
concentration is down from 5.4 mg/1 to 1.6 mg/l. This partly
reflects precipitation of ferric hydroxides in the oxidizing
environment, and partly a larger sample set than the monitor-
well set. The hydroxides remain admixed with the biomat and/or as
free particles that are removed during microfiltration. Eh-pH
e2*
(Figure III-3) conditions indicate that ferrous iron (Fe eous)
remains the stable iron species in standing water. Cascading
water and, possibly, the biomat are the primary loci of ferric
oxyhydroxide fixation.


Organic Carbon Organic carbon is present in the recharge wells
in two forms.

Particulate organic are present in the water. These
represent the filamentous microbial biomats that thrive on the
well screens opposite the Surficial aquifer and on the casing in
the splash zone. In sampling, the biomat is disturbed and
particulate content of the water increases. Undoubtedly, biomat
fragments and unattached microbes are continually entering the
water; however, it appears that most of this material is an
artifact of sampling. When a tripod was used to center the
sampling vessel over the well and minimize scraping along casing
walls or when a bladder pump was used, the amount of filterable,
filamentous material declined to near zero. From this, it can be
concluded that microbe contributions to the recharge-well water
are limited.

Dissolved organic carbon averaged 6.4 mg/l, which represents
an unusual amount of organic carbon in an aquifer. Note, however,
that the average TOC in recharge wells is less than in the
Surficial aquifer (Table III-1).


Radionuclides Radioactivity in the recharge wells also reflects
radioactivity in the Surficial aquifer. Average activity ratios
calculated by dividing the mean activity in the aquifer of
interest by the mean activity in the recharge wells are given in
Table III-3. Use of means to calculate the ratios in Table III-3
is somewhat misleading because the sample sizes differ and the
recharge-well population includes a number of wells that have
little problem with radioactivity, while the monitor well
population includes a significant representation from KR-98B,
which does have a radioactivity problem. Comparison of the data,
however, illustrates an important relationship. That is, for all
of the uranium-238 daughters, radioactivity in the Surficial
aquifer exceeds or equals radioactivity in the recharge wells,
while radioactivity in the Floridan aquifer is less than


III-7









radioactivity in the recharge wells.


Comparison of the recharge well data (Appendix B) with state
and federal water quality standards indicates that 8 out of 62
samples violated the 15 pCi/l MCL for gross-alpha radiation.
These 8 samples were taken from 5 different wells. Clearly, most
of the wells that have been chronic violators of the MCL have
been closed and abandoned. Four of the well samples in this study
violated the 5 pCi/l MCL for radium.


CHANGES IN WATER CHEMISTRY IN THE RECHARGE WELL SYSTEM

Reliability of Monitor-Well Data

There is good chemical evidence to indicate that the
Surficial aquifer monitor wells are located within the "cones of
depression" of the recharge wells. While there are vertical
heterogeneities in the Surficial aquifer because of hardpans and
minor clay lenses, lateral flow is through homogeneous media.
Therefore, one can assume that flow is essentially radial to the
wells. A direct relationship between the recharge wells and
monitor wells in the karstic Floridan aquifer is more difficult
to establish. Most of the flow in the limestone/dolostone aquifer
is through fractures or karst conduits, and, while one can argue
that the monitor wells are within the "cone of impression", or
mound, developed by recharge, individual monitor wells may not be
in the same conduit as the recharge wells they serve. In one
case, KR-127, there is little doubt because the deep monitor well
discharges gases entrained in the recharge-well water during free
fall. As will be illustrated below, there is chemical continuity
in the data, which also suggests that the deep monitor wells
provide samples that reflect the chemical changes that recharge-
well water undergoes upon entering, and mixing with, native
Floridan aquifer water.


Chemical Changes During Recharge

Two groups of chemical changes occur as water passes from
the Surficial aquifer to the recharge wells and then to the
Floridan aquifer. Bulk chemistry changes as the recharge-well
water equilibrates with the carbonate minerals of the Floridan,
and Eh and pH change at each stage of the path.


Eh-pH Changes Figures III-2 and III-3 illustrate the changes in
Eh and pH along the flow path.

As water passes from the Surficial aquifer into the recharge
well, Eh increases, but pH does not change appreciably. There is


III-8








insufficient time for pH to change, and buffering by sulfates and
organic retard significant changes in acidity. Eh changes
because of the aeration induced during free fall in the recharge
wells. The net effect of the transition from Surficial aquifer to
recharge well is minimal. The only important change is an
increase in the stability of sulfate as opposed to hydrogen
sulfide (Figure III-2).

Examination of the data in Appendix B and the data of Oural
et al. (1986) indicate that there are minor chemical changes from
the top of standing water in the recharge well downward. These
are a result of mixing with upwardly advected Floridan water or,
possibly, Hawthorn Group waters; re-establishment of reducing
conditions with depth; and possible sorption on clays and
particulate organic.

Once the water enters the Floridan aquifer, Eh quickly
returns to a more strongly reducing environment (Figures III-2,
III-3), and the pH increases by 1 to 2 units. The increase in pH
is a result of equilibration with carbonate minerals in the
aquifer according to chemical reactions III.la and b. The return
to more reducing conditions accompanied by an increase in pH has
little affect on sulfur or iron species. Radionuclides, however,
are greatly reduced by precipitation, dilution, and sorption (see
Chapter IV).


Reaction Paths Figure III-1 shows the bulk chemical reaction
paths taken by samples from recharge wells. In general,
Surficial-aquifer and recharge-well waters evolve from being
enriched in sodium and sulfate to a calcium-bicarbonate water.
Most of the chemical change is accomplished by equilibration with
the calcite and dolomite and mixing with native water in the
Floridan. It is this equilibration that accounts for the increase
in pH of the waters in the Floridan. In effect, reaction equation
III.1 goes to the right, with consumption of acidity and
carbonate minerals, and release of bicarbonate and dissolved
calcium, and magnesium if dolomite is involved.


Saturation State of Water

The reaction pathways can be compared by examining the
saturation state of the water with respect to certain critical
minerals. Saturation state is defined as

SI log( IAP,
Ksp
(111.2)


III-9









where SI = the saturation index, IAP = the ion activity product,
and Kp = solubility product constant at environmental
temperature. If the SI is negative, the water is undersaturated
with respect to the mineral in question. If it is equal to 0.0
0.2, the water is saturated; and if it is positive,
oversaturation exists. Note that these calculations, which were
done with program WATEQF, do not account for reaction kinetics.
Therefore, undersaturation predicts that the mineral will
dissolve in the water, and oversaturation predicts precipitation.
Neither may necessarily occur.

Table III-4 compares the average SI values at each step of
the flow path. The minerals chosen are those likely to play a
significant role in changes in water chemistry. Calcite (CaCO3)
and dolomite (CaMg(CO,)2) are the dominant minerals in limestone
and dolostone, respectively. Ferric hydroxide (Fe(OH)3),
manganese oxide (MnO2), aluminum hydroxide (A.(OH)3) are examples
of amorphous oxyhydroxides that exist in area soils, serve as
strong sorption sites for metal ions, and may plug well screens.
Boehmite and diaspore are dimorphs with the composition AlO(OH).
They, too, form in area soils.

The trend illustrated in Figure III-1 is also indicated by
the saturation states of water with respect to calcite and
dolomite. Surficial aquifer and recharge-well waters are strongly
undersaturated with respect to the two minerals. However,
Floridan aquifer waters are very close to saturation, indicating
that the recharge-well water reacts with aquifer materials, which
are dissolved until equilibrium is obtained.

The oxyhydroxides, boehmite, and diaspore saturation indices
show predictable trends. Water is undersaturated with respect to
all three phases. However, ferric hydroxide shows an increase in
saturation state, which reflects the oxidation present in the
recharge-well environment. Aluminum compounds are known-to form
in area soils as a result of lateritic weathering (Altschuler et
al., 1956). The saturation-index data indicate that the
Surficial-aquifer water is at or near saturation with respect to
diaspore. The water is less saturated in the recharge-well
environment, and no aluminum was detected in the Floridan aquifer
water. The absence of detectable aluminum in the Floridan is
probably a result of dilution.


CHEMICAL SPECIATION

Introduction

Correlation analysis of chemical analytical data indicates


III-10









that analytical data (i.e., total concentrations) are not
necessarily appropriate for determining the chemical behavior of
the system. Correlations are often weak, statistically
insignificant, and spurious. Part of the reason that these
relationships are obscure is that the radionuclides of interest
are present in such small quantities that, if they were not
radioactive, they would not be detectable. These isotopes may be
strongly affected by chemical complexes that are also present in
small quantities. In order to more accurately identify chemical
reactions that govern radionuclide mobility, activities of
specific species, including ion-pair complexes, were calculated
by means of WATEQF. WATEQF calculates chemical speciation based
on ionic strength, Eh, pH, temperature, and analytical species
present. For example, the samples were analyzed for 14 major and
minor constituents (Appendix B). Sulfate was the only sulfur
species determined. Calculation of species, however, determined
the activity of sulfate (as opposed to analytical concentrations)
and identified 8 sulfate-metal ion pairs. Thus, not all
analytically determined sulfate is available to function as ionic
sulfate in chemical reactions, and only that fraction of the
sulfate that can function as simple, ionic sulfate is of
importance. By isolating the activities of the various species,
much stronger correlations can be identified, and less risk is
involved in interpreting the data.

Only those species related to mobilization of the
radionuclides are discussed in this report. Pearson product-
moment and Spearman rank correlation analysis, and first
principles of chemistry (Chapter I), indicate which chemical
systems affect the radioisotopes of interest. Only chemical
relationships that are statistically significant at a = 0.05 and
r2 > 0.50 are discussed. This section discusses the gross
chemical systems. Chapter IV discusses the specific chemistry of
the radionuclides of interest.


Sulfate Chemistry

Both sulfate and sulfide, as indicated previously, are
thermodynamically favored species at the Eh and pH conditions
present in the aquifer systems. Microbial processes and chemical
kinetics determine the rates of conversion from one species to
the other (Connell and Patrick, 1968; Rye et al., 1981), so
metastable sulfur species may be present in the water samples.
Harada et al. (1989) have shown that both species may be
important in determining polonium mobility. Unfortunately we did
not analyze for sulfide, so only the role of sulfate can be
documented in this study.


Relationship to Ionic Strength Examination of the monitor-well


III-11









data indicates that the activity (as opposed to concentration) of
sulfate is closely related to ionic strength (Figure III-4).

Sulfate is one of the dominant anions in the Surficial
aquifer, and its activity in the water is a major control on
total dissolved solids and ionic strength. The increase in
sulfate activity that is documented in these data reflects the
high sulfate wells, particularly KR-98B, at the Kingsford mine.


Relationship to Eh-pH As Figure III-5 indicates there is a
negative relationship of sulfate activity with pH. That is, more
acid solutions contain higher sulfate activities. In part, this
is because oxidation of sulfides produces acid according to the
reaction

R-S2 + 3.502 + H20 R2+ + 2SO2~ + 2H+,
(III.3a)

where R-S2 is a metal-sulfide compound, such as pyrite,
containing reduced sulfur, and R2+ is a metal that is
simultaneously involved in the reaction

R2' + 0.2502 + 2.5H20 R3 (OH)3 + 2H.H
(III.3b)

If the reaction involves pyrite, R-S2 is FeS2 and R is ferrous
iron, which is oxidized to ferric iron by reaction III.3b. Thus,
production of sulfate and metal oxidation produce hydrogen ion,
and reduce the pH. Note that Floridan aquifer monitor well data
indicate low sulfate activities and high pH values. Water in the
Floridan aquifer is buffered by dissolution of calcium carbonate
(calcite), which yields the high pH values. Surficial aquifer
water is not buffered in this way.

Sulfate should be related to Eh, because equations III.3a
and b are dependent upon the availability of oxygen, or some
other electron receptor. Examination of Figure III-2 indicates
that the monitor well and recharge well data fall on the boundary
between the sulfate and sulfide stability fields. Hydrogen
sulfide odor was detected in many wells, and one can conclude
that sulfide oxidation is an active process in the system. Figure
III-6 shows the relationship of sulfate activity with Eh. Note
that, with two exceptions, sulfate does not appear to be
important below Eh values of -150 my. This is in agreement with
the findings of Connell and Patrick (1968), who concluded that
sulfate reduction in saturated soils is inhibited at Eh's greater
than -150 mv. Therefore, one can conclude that those samples with
Eh's less than -150 my are most likely subject to net sulfate


III-12








reduction. Only one Surficial aquifer monitor-well sample (11% of
sample set), three Floridan aquifer monitor-well samples (33%),
and two recharge-well samples (14%) had Eh values of -150 my or
less.

Therefore, sulfate exists at Eh's in which microbial sulfate
reduction is inhibited, including Eh's from -150 to 0 mv. This
constraint leads to a chemical system in which either sulfuric
acid is produced and maintained by sulfide oxidation (eq. III.3)
or sulfate is directly introduced to the system by marine
aerosols or chemical plant effluents. As sulfates increase, pH
decreases and ionic strength increases (compare Figures III-5,
III-7). Note in Figure III-7 that the samples from the Floridan
aquifer do not participate in this relationship because of
buffering with carbonate minerals.

Strontium and sulfate combine to form SrSO4 (celestite) in a
manner that is chemically similar to RaSO4. Strontium is not
clearly correlated with sulfate. Figure III-8 shows that there is
little correlation between strontium and sulfate for the entire
data set. The data may indicate several, separate patterns with
positive slopes. The pattern of four high-sulfate activities
represents data from the monitor wells at KR-98B, which have the
highest sulfate values measured in monitor wells. The fact that
the data show no coherent pattern suggests that strontium
activities are too low for equilibrium reactions to occur.
Celestite does not precipitate, and the correlation with sulfate
reflects net increase in dissolved salts during equilibration
with aquifer matrix. KR-98B is located in a trough that contains
significant organic as is near a cooling pond associated with a
phosphochemical plant, either of which may constitute the source
of sulfate for that well complex.

Iron Chemistry

Iron speciation and solubility are controlled by Eh and pH
conditions (Figure III-3). Recharge-well environments approach
conditions favorable for ferric hydroxide precipitation.
Otherwise, the data indicate that ferrous ion (Fe2*) is stable.

Figure III-9 illustrates the relationship of total iron
concentration with pH. Note that the solubility of iron increases
with acidity. Thus, in the Surficial aquifer the more acid
environments are most likely to contain high iron and to be
stripped of particulate, oxyhydroxide compounds, such as ferric
hydroxide, by dissolution.

The relationship of iron concentration with Eh is less clear
(Figure III-10). In the Surficial aquifer, a positive correlation


III-13









exists between the two variables (see regression line on Figure
III-10). However, examination of the data indicates that the high
iron concentrations are at Eh's greater than -150 mv. This is the
threshold predicted by Connell and Patrick (1968) for sulfate
reduction, and any samples that show Eh's less than -150 my are
likely to also have low iron owing to precipitation of ferrous
sulfides. Iron availability is restricted in the Floridan aquifer
because of the high pH's, which encourage precipitation of
minerals, such as ferric oxyhydroxdes, carbonates, and
phosphates. Thus, no pattern exists in the Floridan-aquifer
derived samples.


Organic Chemistry

The humic substances that occur in ground water are affected
by pH. Total organic carbon (TOC) was used to estimate the
organic fraction. TOC measurements are incapable of
distinguishing between humic substances, however. Because many
humic substances are known ligands, it is worthwhile to determine
the environmental conditions that favor formation and/or
preservation of humic substances.

Figure III-11 illustrates the relationship between TOC and
pH. Note that the regression line suggests that a weak
correlation exists between the two variables within the Surficial
aquifer and recharge wells. The most acid samples are from KR-98B
and HK-1, where peaty sediments and phosphochemical plant
effluents are known to occur nearby. The most basic samples are
from the Floridan aquifer, where organic are reduced through
flocculation and microbial activity.

The relation with Eh (Figure III-12),shows that TOC's exist
at all Eh's. Note that TOC's are highest in the Surficial aquifer
and that a very weak negative correlation (not significant at eF
0.05) with Eh may exist in the Surficial aquifer samples. Note
also that the majority of high TOC samples occur at Eh's near or
above -150 mv. This may reflect sulfate reduction processes
where, below -150 mv, microbial metabolism, which requires a
source of carbon, may consume TOC.

It has been suggested in this study that the high sulfate
concentrations in the Surficial aquifer may be a result of
release of sulfate from humic substances or by microbial
oxidation of sulfide. Figure III-13 shows that there is a
negative correlation between TOC and sulfate in the Surficial
aquifer. This supports both arguments, which are related, that
destruction of organic compounds is paralleled by release of
sulfate. No pattern can be discerned in the Floridan aquifer
data.


III-14









Iron is a major metal commonly bound by natural organic. It
has been shown to serve as a receptor for phosphate sorption on
organic (Young and Comstock, 1986). Therefore, soluble organic
may serve as vehicles for iron and phosphorus mobilization in
ground water. Figure III-14 illustrates the relationship of iron
to TOC. In the Surficial aquifer there is a noisy negative
correlation between the two variables. Since the water samples
were micropore filtered before metal analysis, large organic
molecules and floccules that contain iron may have been removed
from the sample. Thus, one could argue that high TOC waters may
contain chelated iron that was removed by filtration and not
detected. Furthermore, the pattern is consistent with other data
in that iron can be released during oxidation of organic. Thus,
loss of organic, and the chelating capacity they represent,
leads to an increase in soluble iron.


III-15










Table III-1. Means (X) and standard deviations (a) of raw data
from monitor wells. See Chapter II for analytical procedures.

Surficial Floridan
aquifer aquifer
Variable X a X a


Temp. (C)
Cond. (jmhos
cm" )
pH
Eh (mv)
D.O. (mg/1)

Alk. (mg/1)
Cl (mg/1)
F (mg/1)
HCO3 (mg/1)
P04 (mg/1)
SO4 (mg/1)

SiO2 (mg/1)
TOC (mg/1)
TDS (mg/1)
VDS (mg/1)


Al (gg/1)
Ca (mg/1)
Fe (4g/1)
K (mg/1)
Mg (mg/1)
Mn (gg/1)
Na (mg/1)
Pb (4g/1)
Sr (Mg/1)


Pb-210 (pCi/1)
Po-210 (pCi/1)
Ra-226 (pCi/1)
U-234 (pCi/1)
U-238 (pCi/1)
A.R.
Gross Alpha
(pCi/1)


24.2
731.2

5.3
-96.6
1.2

68.7
26.7
0.1
83.7
0.3
243.2

4.0
8.3
496.0
90.6


71.5
26.3
5,414.0
0.4
14.2
10.7
87.4
5.7
36.8


0.7
264.7

0.8
83.9
0.3

69.9
14.8
0.1
85.2
0.2
177.0

1.7
2.1
188.6
83.9


81.7
23.8
3,218.5
0.1
7.7
4.8
71.4
2.3
30.2


1.0
18.8
1.6
0.3
0.1
2.8
38.3


1.7
21.2
1.1
0.1
0.0
1.1
40.5


24.4
388.3


7.3
-139.1
1.5

183.6
8.9
0.7
223.9
0.1
8.3

36.6
4.4
288.9
79.1


0.0
35.2
12.6
1.2
17.6
0.7
14.6
5.2
45.9

0.2
0.8
1.0
0.1
0.1
1.8
5.8


1.2
109.4

0.2
64.1
0.3

13.0
1.6
0.1
15.8
0.1
15.0

6.4
0.5
96.8
51.7

0.0
3.8
16.5
1.1
4.3
0.2
1.4
1.8
23.4

0.2
0.9
1.0
0.2
0.1
1.0
2.6


III-16









Table III-2. Means (X) and standard deviations (a) of raw data
from recharge wells. See Chapter II for analytical procedures.


Variable X o

Temp. (*C) 22.8 5.2
Cond. (Amhos/cm) 193.4 168.8
pH 5.1 1.3
Eh (mv) 23.7 94.9
D.O. (mg/i) 3.4 1.9

Alk. (mg/1) 39.4 42.9
Cl (mg/1) 10.4 7.9
F (mg/1) 0.5 0.4
HCO3 (mg/1) 48.0 52.3
P04 (mg/1) 1.9 1.7
SO4 (mg/1) 45.2 67.7

SiO2 (mg/1) 5.9 3.2
TOC (mg/1) 6.4 3.7
TDS (mg/1) 174.7 128.1
VDS (mg/1) 62.9 49..8

Al (jg/1) 412.8 1283.3
Ca (mg/1) 13.6 8.4
Fe (Ag/l) 1565.9 1814.3
K (mg/1) 0.6 0.5
Mg (mg/1) 6.7 5.0
Mn (jg/l) 12.1 15.3
Na (mg/1) 19.6 31.7
Pb (Ag/l) 4.3 4.0
Sr (ig/1) 11.2 17.1

Pb-210 (pCi/1) 0.3 0.7
Po-210 (pCi/1) 3.9 5.1
Ra-226 (pCi/1) 1.2 2.0
U-234 (pCi/1) 0.2 0.2
U-238 (pCi/1) 0.1 0.2
A.R. 1.9 1.4
Gross Alpha (pCi/1) 8.4 7.9










III-17









Table III-3. Activity ratios of aquifer water as compared to
recharge well water. Ratios are calculated from mean values.

Surficial aquifer Floridan aquifer
Isotope Recharge wells Recharge wells


Uranium-238 0.64 0.64
Uranium-234 1.19 0.52
Radium-226 1.31 0.83
Lead-210 2.85 0.50
Polonium-210 4.81 0.20

Gross-alpha 4.56 0.69
activity


III-18









Table III-4. Average log-saturation indices of aquifer and
recharge-well water with respect to selected mineral species.


Surfic. Aquifer Recharge Wells Floridan Aquifer

Species X a X a X a

Calcite -2.23 1.85 -3.42 0.91 -0.26 0.16
Dolomite -4.40 3.58 -6.71 1.80 -0.81 0.96
Fe(OH)3 -8.05 10.40 -7.86 2.36 -4.60 1.55
Mn(OH)3 -24.65 2.27 -24.14 2.42 -19.67 1.82
Al(OH)3 -2.16 1.85 -4.00 1.62 nd nd
Boehmite -1.04 1.12 -2.21 1.62 nd nd
Diaspore -0.02 0.82 -0.63 1.60 nd nd

nd = no aluminum detected



































III-19

















CATIuOI x Or TOMTA. M/L M1&MIS

SURFICIAL
AQUIFER


4


cimuu o0r =?TAJL A AIO NM
SI uImi


RECHARGE
WELLS


CAtmons 2 or TO"AL a/1. AXoIM
9 SAMim


FLORIDAN
AQUIFER
Figure III-1. Piper diagram showing major-element compositions of
Surficial-aquifer, recharge well, and Floridan aquifer samples.


III-.?0


C.


"d











+0.4


+0.2


Eh (v) 0.0


-0.2


-0.4




S]
o ]


2.0 4.0 6.0 8.0
pH

Surficial Aquifer Wells
Recharge Wells
Floridan Aquifer Wells


Figure III-2. Eh-pH diagram showing stability fields of major
sulfur species.


III-21











+0.4


+0.2


Eh (v) 0.0


-0.2


-0.4


2.0 4.0 6.0 8.0
pH


+-

S


Surficial Aquifer Wells
Recharge Wells
Floridan Aquifer Wells


Figure III-3. Eh-pH diagram showing stability fields of major
iron species.


III-22









S04 activity (mg/1)
600



500-



400



300-



200



100


S4- +-I
0---- -- M----,,,----,----

0 0.005 0.01 0.015 0.02
Ionic Strength

Surficial aq. + Floridan aq.
SRecharge wells

Figure III-4. Scatter diagram showing the relationship of
calculated sulfate activity to ionic strength. Line is a linear
regression on Surficial-aquifer samples.


III-23









S04 activity (mg/1)
600




500-




400




300 -




200-




100



O +

4 5 6 7


- Surficial aq.


+ Floridan aq.


SRecharge wells


Figure III-5. Scatter diagram showing the relationship of
calculated sulfate activity to pH. Line is a linear regression on
Surficial aquifer data.


III-24















S04 activity (mg/I)


-150 my threshold


+
I L. I _4- IL-


-250


-200


-150






- 50


-50


M Surficial aq.


+ Floridan aq.


* Recharge wells


Figure III-6. Scatter diagram showing relationship of sulfate
activity to redox potential (Eh). Note that, with the exception
of two samples, all high sulfate samples occur at Eh's in excess
of -150 mv.


III-25


.00


.00


00 1


300


200


0
-300


-100
Eh (mv)


-


I +t II....


^
















++


0,005 0.01


Ionic Strength


Surficial aq.
SRecharge wells


+ Floridan aq.
O HK-1


Figure III-7. Scatter diagram showing relationship of pH to ionic
strength. Regression line is on Surficial aquifer samples.


III-26


*f


0.015


0. 0i


.










804 activity (mg/1)


++


I I I I i l I l l


I I I I I I I'


10
Strontium


S Surficial aq.

* Recharge wells


100
concentration (ug/I)


+ Floridan aq,

0 HK-1


Figure III-8. Scatter diagram showing the relationship of
strontium to log sulfate concentration. Note that there is no
correlation within the data set.


III-27


1000









100









10


I I II I


10C


I I I I I









Total Iron Concentration (mg/I)


2


~~~~II I I I I Iii't**I*


4 4.5


5 5.5


6 6.5


7 75


pH


- Surficial aquifer


+ Floridan aquifer


*)- Recharge wells

Figure III-9. Scatter diagram showing the relationship of total
iron concentration to pH. Line is a linear regression on the
Surficial aquifer data.


III-28


\\


*~i~







Total Iron Concentration (mg/1)
12




10 -














4
+ 4 +

















0
-300 -200 -100 0 100
Eh (mv)

Surficial aq, + Floridan aq. Recharge welil

Figure III-10. Scatter diagram showing relationship of total iron
concentration to redox potential (Eh). Note that, with the
exception of one sample, all high iron concentrations occur at
Eh's in excess of -150 mv. Regression line is for Surficial
aquifer sample data. Compare with Figure 111-6.


III-29










14




12




10




8




6




4




2
4


Total Organic Carbon (mg/1)


5 6 7


-*- Surficial aq.
SRecharge wells


+ Floridan aq.
O HK-1


Figure III-11. Scatter diagram showing the relationship of total
organic carbon (TOC) to pH. Weak correlation in the Surficial
aquifer data is suggested by the linear regression.


III-30









Total Organic Carbon (mg/1)


14



12



10


-300 -250 -200 -150 -100 -50


0 50


Eh (mv)

3


Surficial aq.
* Recharge wells


+ Floridan aq.
S. HK-1


Figure III-12. Scatter diagram showing the relationship of total
organic carbon (TOC) to redox potential (Eh). Regression line is
based on Surficial aquifer data.


III-31


+/U


8


6h


4h


10C


-









S04 activity (mg/I)
600




500




400




300 -




200
\


100-




2 4 6 8 10 12
2 4 6 8 10 12


Total Organic Carbon (mg/l)


- Surficial aq.
SRecharge wells


+ Floridan aq.
O HK-1


Figure III-13. Scatter diagram showing relationship of total
organic carbon (TOC) to sulfate concentration. Regression line is
on Surficial aquifer data.


III-32









Total Iron Concentration (mg/l)


2 4 6 8 10 12


Total Organic Carbon (mg/I)


Surficial aq.
SRecharge wells


+ Floridan aq.
] -HK-1


Figure III-14. Scatter diagram showing relationship of total
organic carbon (TOC) to total iron concentration. Regression line
suggests a weak correlation in the Surficial aquifer.


III-33


12




10




8




6



4




2




0




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs