ECONOMIC GEOLOGY

AND THE

BULLETIN OF THE SOCIETY OF

ECONOMIC GEOLOGISTS

VOL. 57 MARCH--APRIL, 1962 No. 2

TRANSPORTATION AND PRECIPITATION OF URANIUM AND

VANADIUM AT LOW TEMPERATURES, WITH SPECIAL
REFERENCE TO SANDSTONE-TYPE URANIUM
DEPOSITS t

P. B. HOSTETLER AND R. M. GARRELS

CONTENTS
PAGE

Abstract .............................................................. 137

Introduction .......................................................... 138
U-O-t-I20 system ..................................................... 139
U-O-t-I20-CO system ................................................ 142
Open system ....................................................... 143
Closedsystem ...................................................... 144
V-O-t-IO system ..................................................... 147
U-V-K-O-t-IO-CO system ........................................... 149
Discussion,with referenceto the ores of the ColoradoPlateau .............. 552
Correlationwith natural ground waters .............................. 152
Environmentof ore deposition....................................... 155
Formation of coffinite and vanadium silicates .......................... 158
Stunmary ............................................................. 160
Acknowledgments..................................................... 161
Appendix ............................................................. 162
References ............................................................ 166

ABSTRACT

Uranium and vanadium in sandstone-typedeposits of the western

United States apparently have been transportedto their present environ-
ment from external sources by low-temperature aqueous solutions. In
this paper an attempt is made to interpret the characteristicsof aqueous
solutions capable of transporting significant quantities of uranium and
vanadiumthrough continentalsedimentaryrocks,and the changesin these
Publication authorized by the Director, U.S. Geological Survey.
137
138 P. B. HOSTETLER .4ND R. M. G.4RRELS

characteristicsthat might result in precipitation of uraninite and other ore

minerals in concentrationsof ore grade. On the basis of present knowl-
edge, the transportationenvironmentis shown'to be that of weakly alka-
line, moderatelyreducing ground water, with an average or larger than
averageconcentrationof dissolvedcarbonatespecies(H2CO + HCOf +
COs--). Precipitationis inducedby reductiou,probablyby carbonaceous
material or hydrogensulfide,or both. Uranium is transportedmainly in
the form of the highly stable uranyl dicarbonate and tricarbonate com-
plexes. Precipitation results from reduction of hexavalent aqueous
uranium speciesto form uraninite, reduction of tetravalent vanadium to
form montroseite,and fixation of uranyl ions by combinationwith potas-
sium ions and quinquivalentvanadium to form the mineral carnotite.

INTRODUCTION

STUDIES
overthe pastseveralyearshaveshownthat the complexmineralogy
of the sandstone-typeuranium ores is the result of the oxidation of an
originallysimplemineralsuite (18, 37). Apparently,the mineralsof this
primary suite were precipitatedsimultaneouslyfrom aqueoussolutionlong
after depositionof the enclosingsediments(27, 30, 34, 35). In any given
depositthe primary minerals are few, the commonones being uraninite,
UO2+x,coffinite,UO2(SiO4)l_x(OH)4x, montroseite,VO(OH), vanadium
silicates,pyrite, chalcopyrite,
bornite,chalcocite,
calcite,quartz,and gypsum.
It is noteworthythat in these minerals the elementsare in the lowest
valence states characteristic of natural occurrences. Indeed, it has been
demonstratedthat the conditionsunder which the primary mineralscan co-
existat equilibriumin an aqueousmediumat low temperaturesand pressures
are near the most strongly reducing environmentsattainable (18). This
relationhas led to the postulationthat the presenceof ore is the result of
precipitationlargely inducedby reductionprocesses; a view bolsteredby the
widespreadoccurrenceof woodymaterialsreplacedby ore minerals,and by
the importantore occurrences reasonablyascribedto the reducingeffectsof
hydrogensulfide(26).
Coleman (11), in a summary of criteria suitablefor the estimationof
the environmentof precipitationof the ore minerals,arrived at a temperature
of the order of 100 C, and a pressureof the order of a few hundredatmos-
pheres. This temperatureestimate has been substantiatedby the recent
work of Bregerand Chandler(7), who investigated
the temperaturedepend-
ence of extraction of humic constituents from coalfried wood in mineralized
horizons. Thus deductionsbasedon low temperatureand pressurechemistry
are relevant to the actual processof ore formation.
Whatever the sourceof uranium and vanadium,all workers are in agree-
ment that in numerous instancesthese elements had to be carried long
distances(a mile or more) through the pores of the host sedimentsbefore
depositionoccurred (18, 27, 36). These sedimentsare characteristically
lenticularcontinentalsandstones
and mudstones. Quartz, mica, and feldspar
are commonminerals of these rocks and there is a variety of clay minerals,
includingchlorite, montmorillonite,mixed layer mica-montmorillonite, and
kaolinite.
 TRANSPORTATION OF URANIUM AND VANADIUM 139

The problemsof ore-elementtransportationand depositioncan be stated

in the followingquestions:
1. In what aqueousenvironments,within the range of natural conditions,
can all the ore-elementsbe transportedsimultaneouslyat some reasonable
concentration ?
2. If such environmentsexist, are they compatiblewith conditionsex-
pectedduringtransportationof watersthroughthe poresof the host rocks?
3. What changesin environment,inducedby localvariationsin the nature
of the host rocks,can be expectedto lead to precipitationof the ore mineral
suite ?

Our attemptsto answer these questionsmust be looked upon as a pre-

liminary study. In our present state of knowledgewe can hope only to
delineatesets of conditionsthat are favorablefor the transportationand
depositionof ore elements,and then to test these conditionsagainstactual
occurrences.We emphasizethe degreeof solubilityof uranium and vana-
dium in the presenceof other ions commonin subsurfacewaters in order to
elucidatepossiblemodesof ore-elementtransport. This simplifiesthe treat-
ment of stabilityrelationshipsamongthe insolubleore minerals,which are
of widespreadoccurrence, by eliminatingthe needfor consideration
of soluble,
locallydistributedminerals. Our approachis to beginwith a discussion of
uraniumsolubilityin pure water, and then to assessthe effectsof addition
of other constituents.
U--O2--H20 SYSTEM

Relationsamonguranium,oxygen,and water have been summarizedby

Garrels (19) and by Deltombe,de Zoubov,and Pourba{x (13). The only
solid phases expected to occur stably in the presenceof water and
variable concentrationsof oxygen are uraninite, UO,+x, and schoepite,
UO2(OI-I),'I-I,O, which spontaneouslydehydrateswhen exposedto the
atmosphere(10). Natural uraninitesapparentlymaintain the cubic UO,
structurefor x valuesup to 0.6 (8). UaOs, easilypreparedin the laboratory,
has not been found in nature (16) and is probablymetastable. Thermo-
chemicaldata are available only for stoichiometricUO,. Figure 1 shows
schematicallythe compositionrelations among UO,, UO,(OI-I),-I-I20,
O,, and I-I.O, and showsthat the compositionof uraninite in the presence
of hydrogenand water is UO., but that for conditionssufficientlyoxidizing
to permit the coexistenceof UO,(OI-I) ,-I-I20, its compositionis UO,.6.
Stabilityrangesof UO2(OI-I)2'I-I20 and stoichiometric UO2 are shown
as functionsof oxidationpotential (Eh) and hydrogen-ionactivity (pI-I) in
Figure 2. However, a gooddeal more informationthan is shownin Figure 2
can be gleanedfrom Eh-pH diagramsthat portray thermochemical considera-
tion of the solubilityof uraniniteand schoepite. This involvesthe calculation
of the activitiesof all the ions in the system,in equilibriumwith whichever
solidphaseis stableunderspecifiedEh and pI-I conditions. For the details
of suchcalculationsthe reader is referred to the papers of Garrels (19) and
Deltombeet al. (13). Deltombeet al. and Garrels consideredonly three
140 P. B. HOSTETLER AND R. M. GARRELS

(Schoe

///
//// //
['102 Uroninite 02

Fro. 1. Composition diagram showingprobablerelationsamonguraninite,

schoepite,
water,oxygen,andhydrogenat 25C and 1 atmosphere total pressure.
Percentagesare molecular. Hydrates of compositionlying betweenUO3 and
UO2(OH)_o'H_oOare not considered.

ions,UO2% U + and U(OH) +a. Uraniumis presentin the hexavalentstate

in UO2+andin thetetravalent statein U +andU(OH) +a. Figure3 shows
the relationshipof theseionsto schoepiteand uraniniteand to the additional
hexavalent uraniumspecies, HUOc andUO2(OH) +. Thermodynamic data
for HUOc and UO(OH) + werederivedfrom GayerandLeider (24) and
Bierrum et al. (4) respectively.
It shouldbe pointedout that for Figure3 and for ensuingdiagramsit is
not, in general,possible
to reachany selectedEh-pH valuewithinthe frame-
work of the simple systemunder discussion. The addition of one or more
components is necessary to reachselectedEh-pH values. The assumption
is that the additionalcomponents will not disturbthe diagrammed equilibrium
relationships, and in simple,dilute systemssuchis usuallythe case. We
return to a Inor(critical discussionof this 1)fintin the sectionon correlation
with natural ground waters.
In Figure 3 unbrokenlines drawn betweenthe stablesolid phasesand
the fields of the ions indicate that total ionic activity of dissolveduranium
 TRANSPORTATION OF URANIUM AND VANADIUM 141

speciesin equilibriumwith either solidphaseis 10-6. In otherwords,in the

presenceof uraninite or schoepite(whichever is stable), total ionic activity
is fixed at 10-6 for the Eh and pH conditionsshownby the solidlines. Along
theselinesthe solidphaseis no longerstableif total ionicactivityis lessthan
10-6; all uranium is in solution. In the fields shownfor the solid phases
total ionicactivity in equilibriumwith that solidphaseis lessthan 10-6, while
in the areasdiagrammedfor the ionstotal ionic activitymust rise to a value
greater than 10-6 beforeeither uraniniteor schoepitecan be precipitatedas a
stable phase (Fig. 2 showsthe stable solid phase). A dashedline repre-
sentinga total ionicactivityof 10- in equilibriumwith the stablesolidphase
is includedin Figure 3 to give the reader an idea of the activity gradient
involved. A family of contours,parallelto the 10-6 and 10- contours,could
be drawn to representother valuesof total ionic activity in equilibriumwith
the solidphase. There are discreteEh-pH conditionsfor which most of the
uraniumin solutionis tied up in oneion; therefore,linesdrawn betweenions
show conditions for which the activities of the two ions are identical.
The foregoingdiscussion
is basedon the conventions
of Eh-pH diagram-

o , ,

oe

uo2"
$choeplte
4 Schoep,te
HUO
4-

O.Z
U02
(OH}2
'H20

Uranlnlte

-oz Uranlnlte
U02

H20

-o 6

-08

-o! z 4 ' e
un,qs
,o -i c
)H un,ls
8 0 12

Fro. 2. (Left) Equilibrium diagram of the U-O,.H.,O systemat 25C and

1 atm. Stability rangesof solid phasesare plottedas a functionof hydrogen-ion
concentrationand oxidation potential.
Fro. 3 (Right) Aqueousequilibrimn diagram of the U-O?H._.O systemat
25C and 1 atto. Fields of solid phasesand dissolvedspeciesare plottedas a
function of hydrogen-ion concentration and oxidation potential. Boundaries of
solid phasesat total activity of uranium-bearingions: 10- in unbrokenlines, and
at total activity: 10- in dashedlines.
142 P. B. HOSTETLER AND R. M. G.4RRELS

ming developedby Pourbaix (31). It must be clearly recognizedthat the

solubilityareas, labeledwith the ion that is their chief contributor,do not
representstoichiometricsolubilities,but rather are areas in which the solu-
bility is at least the value shown for thermodynamicequilibrium between
solid phasesand ionic activities. True solubilityexceedsthat calculatedby
a factor that dependson total concentrationof all ions (ionic strength) and the
degreeof completeness of our knowledgeof the ionswhichmustbe considered.
In this systemnot all aqueousspecieshave been considered,for a number
of polyuranatecomplexeshave been reported (4). Thermodynamicdata
or equilibriumconstantsreportedfor thesespeciesare in poor internal agree-
ment. Nevertheless,it is clear that the polyuranatecomplexesare not suffi-
cientlystableto alter significantlythe relationshipsshownin Figure 3; their
total effectis to increaseslightlythe solubilityof schoepiteover that portrayed
in Figure 3.
Figure 3 showsthat in the U-O2-H20 systemuranium is solubleonly
in acid or distinctlyalkalinesolutionsand that the chief contributingions in
acid solutionare the UO2+"and UO(OH) + species,with small fields of the
U +4 and U(OH) +a ions also present. HUOc is the main contributorto
uranium solubilityin alkaline solutions.
No species are shownbelowthe water breakdownline, that is, for Eh-pH
values at which the partial pressureof hydrogenexceedsone atmosphere.
The assumptionis that any reducingagent sufficientlypowerfulto reduce
water and producehydrogenat pressuresof one atmosphereor greater
(such as the U *a ion) reactswith an excessof water to producea more
highly oxidizedspecies.
The field shownin Figure 2 as "UO shouldbe, as shownin Figure 1,
a field of a solid changingin composition from UO,at the water-hydrogen
boundaryto UO2.0at the uraninite-schoepite
boundary. In the absenceof
thermochemicaldata for UO2+x we have assumedthat only UO2 is present.
This placestheuraninite-schoepite boundary at its lowestpossible
positionon
the Eh scale;the boundarybetweenUO.6 and schoepite may be as muchas
a tenthof a volt higher,but parallelto the boundaryshown. Our procedure
givesuraniniteits minimumstability,and somewhatoveremphasizes the
strength
of thereducingagents
necessary
for its precipitation.
At anyrate,we canconclude
thatnearlypurewateris a poortransporting
agentfor uranium,exceptat pH valuesof about5 or less. As will be seen
later,suchpH valuesare incompatible
with the geologicoccurrences of the
ores. ,

U--02--I-120--C02 SYSTEM

Essentially
purewateris notto be expected
in subsurface
watersthatare
in intimate contactwith rock minerals. Examination of analysesof ground
watersrevealsthat dissolvedCO._,,in oneform or another,is an importantand
ubiquitous
constituent.The rangeof CO., contentis large,but valuesof
100-300ppmare common, whereasvaluesof lessthan 10 and morethan
10,000 ppm are uncommon.
Carbondioxidemarkedlyincreases
the solubilityof the soliduranium
 TRANSPORTATION OF URANIUM AND VANADIUM 143

phasespreviouslydiscussed. Bullwinkel (9) has studied its effects theo-

reticallyand experimentally,and concludesthat the solid phasesinvolvedin
the U-Os-I-IO-CO2 system are uraninite, schoepite,and rutherfordine,
UOsCO3. The dissolvedspecies,in additionto thosealreadyconsidered,are
[UOs(CO3)2'2HsO] -s and [UO(CO3)3] -4. Thesecomplexes are hereafter
referred to as UDC (uranyl dicarbonatecomplex) and UTC (uranyl tri-
carbonatecomplex) respectively. We have computedthe free-energyof
formationvalues of rutherfordine,UDC, and UTC using the data of Bull-
winkel (9) and the recent /xFf of schoepitegiven by Deltombeet al. (13).
The equilibrium constantfor the dissociationof UTC to UDC plus car-
bonate,reportedby Blake and others (5), is in good agreementwith that
reportedby Bullwinkel (9).
The U-Os-I-IsO-COs systemmay be consideredfrom two standpoints,
either open to an externally controlledPco2,or closed,with a fixed amount
of COs in the system. An Eh-pH plot with Pcosequalto the partial pressure
of CO2 in the atmosphereis usefulfor portrayingsurfaceweatheringcondi-
tions. However, atmosphericCOs has limited accessto percolatingground
waters,so it is more realisticto considersuchwatersas constitutinga closed
systemin which a fixed addition,or periodicadditions,of CO proportionate
among various CO-bearing speciesaccordingto an Eh-pH characterized
environment. In such a closedsystemPco is variable, a function of pH.
The total of the activitiesof all COs-bearingspeciesis constantfor any addi-
tion of COa reachingequilibrium. This constantis referredto as :COa
Details of the calculationsrequiredin constructingthe diagramspresented
here are given in the Appendix.
Open System.--Relationsamong uranium-bearingsolid and dissolved
speciesas functionsof Eh, pH, and a Pcosvalue of 10-3'4atm. (the partial
pressureof COs in the atmosphere)are shownin Figure 4. The general
Eh-pH-Pcosdiagramis shownin Figure 5 (Fig. 4 is a cross-section of Fig.
5 at Pcos= 10-3'4atm.) As in Figure 3, unbrokencontoursbetweenions
and solidphasesrepresenta total activityof dissolveduraniumspeciesof 10-,
while dashedcontoursrepresenta total activityof 10-4. For clarity, 10-4 total
ionicactivitycontoursare omittedin Figure 5. In Figure 4 and thereafter,
formulasdesignatingfields for solid phasesare underscored.
ComparingFigures 4 and 5 with Figure 3 showsthe marked effect of
COs, evenat low partial pressures.WhereasFigure 3 is composed largely
of areasindicatingfieldsof solidphases,Figures4 and 5 are dominatedby
areasof dissolvedspecies. Uranyl carbonatecomplexingcausesa wide field
of solubilityto exist under oxidizingconditionsat essentiallyall pH values.
The fieldof schoepite, UOs(OH),,.' HaO, as an insolublecompound is almost
entirelyeliminated,and althoughrutherfordine,UOaCO3,doesappearas a
new phase,it occupiesa minor part of the diagrams. For a giventempera-
ture, the schoepite-rutherfordine
transitiondependsonly on activity of water
and Pcoa. Assumingthat aaao-- 1, Garrels (20) has shownthat the equi-
librium Pco.,betweenthesetwo phasesis very closeto the partial pressure
of COs in the atmosphere.
144 P. B. HOSTETLER .4ND R. M. G.4RRELS

[,} unls

Fit3. 4. Aqueousequilibriumdiagramof the U-O-I-I_00-C02systemat 25C

and 1 atm., and with Pco2 constant at 10-s-4 atm. Boundaries of solid phases
are drawnfor total ionicactivityof 10- (unbrokenlines) and 10-4 (dashedlines).
Formulas designating fields for solid phasesare underlined.

NotethatinFigure
4 a'fieldforrutherfordine
isonlydefined
fora total
activityof uraniumbearingions of 10-4; there is no field for this mineral
when total ionic activity dropsto 10-. Thus, in a weatheringenvironment
the concentrationof uranium in solution must be around 10-5 moles/liter
or greaterbeforerutherfordinecan be precipitatedunderthe mostfavorable
pH conditions(around pH 6.0).
In acid, oxidizing solutionsuranium is transportedas UO2*2 and
UO,OH) +, in neutralsolutionsas UDC, and in alkalinesolutionsas UTC.
Not onlydoesadditionof CO2tendto eliminateschoepite,
it causes
the field
of uraninite to diminish. As Pco increasesin the systemstronger and
strongerreducingagentsare requiredto removeuraniumfrom solutionas
uraninite.
ClosedSystem.--Thecomplexing
effectof CO on uraniumis alsovigor-
ousfor thoseconditionswhere the total amountof CO (1CO.,) in the system
maybe regardedas fixed. In Figure6 ICOois takenas 10-a (that is, the
sum of the activitiesof all ions that containCO.,, includingUDC and UTC,
is 10-:). This corresponds
to a modest44 ppm COgin solution. Uranium,
as UDC and UTC, is solublein a wide range of environments. The small
field of rutherfordine(again definedonly for total uraniumion activityof
10-4) indicatesthe restrictedconditions under whichthis mineralforms.
 TRANSPORTATION OF URANIUM AND VANADIUM 145

Schoepite is solubleunderall but very alkalineconditions

whenuraniumion
activityis 10-4. It has no field of stabilitywhenaqueousuraniumdrops
to 10-6. The expansion of the UTC field, at the expenseof the UDC field,
under mildly reducingconditionsis interesting. Recallingthat the total
amountof carbonateis fixed,it is obviousthat the activityof carbonate
and
bicarbonate
mustincreaseas the uranylcarbonatecomplexes beginto break

Io'11o

'o2

'1.O j .0-8
o
i

Io-i I

i ioo

FG. 5. Aqueous equilibriumdiagramof the U-OH._,O-CO._,

systemat 25C
and 1 atto. Pco.,is the third coordinate. Boundariesbetweensolid phasesand
dissolvedspeciesare drawnfor total ionicactivityof 10-6.
146 P. B. HOSTETLER ,4ND R. M. GARRELS

downunderreducingconditions. The increased activityof COa-2 andHCOa-

is reflectedby an increasein the UTC/UDC ratio, as can be seenfrom the
following reaction'

UO=(COa),..2H,.O- d- COa-2 = UO2(COa)a-4 d- 2HaO

The uranyl carbonatecomplexesare stableenoughto tie up considerable

concentrations
of uranium under slightly reducingconditions.
!CO2 is raised to 10-2 in Figure 7, thus enlargingthe dominantareas
of UDC and UTC. Increasing!CO. extendsthe area of UTC relative to

10-4

U02++
'
uo
o

units

Fro. 6. Aqueousequilibriumdiagram of the U-O-HO-CO systemat 25C

and 1 atm., and with !CO = 10-a. Boundariesof solid phasesare drawn for total
ionic activity of 10- (solid lines) and 10-4 (dashedlines).

UDC because UTC containsmorecarbonate. Conversely, the field of UTC

disappears entirely when !CO. falls to 10-". The upper boundaryof the
uraninitefieldin equilibriumwith the uranylcarbonates
is noticeablylowerin
Figure 7 than in Figure 6.
The trends discussedfor Figure 7 are continued for Figure where
:/CO2 is 10-.
Equilibrium between carbonatespecies,water, and graphite is ignored
in Figures7 and8 because it is generallyirreversible. The effectsof organic
matter, hydrocarbons,and other reducingagentson the solubilityof uranium
is consideredin a subsequentsection. It is curious to note that in spite of
 TRANSPORTATION OF URANIUM AND VANADIUM 147

the increasein COg, the field of rutherfordineremainsthe samesize, although

it doesmoveto slightlymore acidicconditions. The reasonfor this behavior
is that the activityof UO2+-oin equilibriumvith rutherfordinedecreases
vith
increasing:CO-oat exactlythe samerate that the activity of UDC in equi-
librium vith rutherfordine increases. Actually, the field of rutherfordine
increasesvery slightlyat the expenseof UO-o.
All the relationsdiscussedfor the closedsystemare pictoriallysummed
in Figure 9 with :CO-othe third coordinate. All boundariesbetveensolid

I
i

O6

04

O2
**
U02 "
'i
.- '- I
I ['u2iu

I i
- II
04

02
I

II
UOz{CO)
2 2H20-2 UOz{C03)
-4

0 xxj 0

0 xx -4 I
-0.2
I0- I -02 iO
*s XO_
4
I0- IO-

0
io-6
h20 -,

' -O8 pH unl?s

Fro. 7. (Left) Aqueousequilibrium diagram of the U-O-H.O-CO= system

at 25 C and 1 atto., and with :CO== 10--.
Fro. 8. (Right) Aqueousequilibriumdiagram of the U-O-H_oO-CO system
at 25 C and 1 atm., and with CO_-- 104.

phasesand ionsin solutionare shovnfor a total activityof uranium-bearing

ions of 10 -6 .
In short,then,the mostnotevorthyeffectof addingCO-oto the U-O-o-YI-oO
systemis to increaseenormouslythe solubilityof uranium.

V--O2--H-OO SYSTEM

The vanadium-vater systemhasbeenstudiedby Garrels(17), Deltombe,

de Zoubov,and Pourbaix (12), and Evans and Garrels (14). The V-O-O-
H-oOdiagramcomputedby Evans and Garrels (14) is shovnin Figure 10.
As the solid phase-ionicactivity contoursrepresenta total ionic activity of
148 P. B. HOoeTETLER _/hVD R. M. G.dRRELe

ofll
1's
6

Il'1

0.6

0.4
th

(vOL-r) 0

o.

0.4.

/.o

Io-

Fro. 9. Aqueousequilibrium diagram of the U-O-H20-CO systemat 25C

and 1 atm. CO is the third coordinate. Boundaries of solid phases at ionic
activity = 10- only.

10-2for dissolvedvanadiumspecies,the solubilityof vanadiumis quite marked

underany type of oxidizingconditions. The fieldsof the solids,V20, V204,
and V23, decreasewith decreasingionic activity. A comparisonof ionic
speciesin the diagramspresentedby Evansand Garrels (14) and Deltombe
et aI. (12) reveals differencesin interpretationof the hydration and poly-
 TRANSPORTATION OF URANIUM AND VANADIUM 149

merizationstatesof the ions,but thesedifferencesare of very little importance

in this study.
U--V--K--Og--It 20--CO2 SYSTEM

The additionof vanadiumand potassiumto the U-O2-H20-CO2 system

can now be discussed.No new ionic complexesare introduced. Carnotite,
K2(UO2)2V2Os' 1-3HO2, is the only new and well-definedsolidphasethat is
insoluble,althougha number of solublesolid phasesare known. Mention
will be made below of the rather widespread,but poorly defined,mineral,
rauvite. No measuredvalueof AF/of carnotiteis available,however,we were
ableto computea valuefrom experimentalsolubilitydeterminations conducted
by Marvin and Magin (Marvin, R. F., and Magin, G. B., written communi-
cation). Data resultingfrom theseexperimentsare not sufficientlyaccurate
to permita rigorousfree-energycalculation,nor is sucha calculation
essential
for the purposeof this paper. Instead,the lowestpossiblefree-energyvalue

FIG. 10. (Left) Aqueous equilibrium diagram of the V-O-HO system at

25 C and 1 atm.; after Evans and Garrels (14). Boundariesbetweensolid phases
and dissolvedspeciesare drawn for total ionic activity of 10-2. The formulas
designatingfields of solid phasesare underlined.
Fro. 11. (Right) Aqueousequilibrium diagram of the V-O_o-H-,Osystemat
25C and I atto. Ionic nomenclatureis from Deltombeet al. (12). Boundaries
betweensolid phasesand ions in solutionare drawn for total ionic activity of 10-a
(unbroken lines). Dashed lines representionic activity contoursof the H_VO,-
ion.
150 P. B. HOSTETLER .4ND R. M. G.4RRELS

was chosenconsistentwith Marvin and Magin's determinations. Thus, the

field of carnotiteshownin ensuingdiagramsis the largestpossible.
Briefly, calculations
were made in the followingmanner. In pure water
(devoidof CO2) at a pH of 7.0 carnotitehasa solubilityof 3 x 10-* moles/
liter at 25 C. Assumingthat 2:
K,.(UO,.)=V=Os.3H,.O+ 4H + = 2K + + 2UO=+2 + 2H=VO4- q- 3H=O. (1)
The concentration of eachof the ionicspeciesis 2 X 3 X 10-7: 10-6'22moles/
liter. As the solubilityof carnotitein pure water is very low, concentration
can be equatedwith activity. The activitygradientof LIO2+2in equilibrium
with schoepiteis two ordersof magnitudechangeper unit changein pH.
At a pH of 7, and in the absenceof CO2, the activityof LIO2+2in equilibrium
with schoepiteis 10-s'4. At this pH, however,most of the solubleuranium
is present as UO2(OH) +. The activity of this ion in equilibrium with
schoepiteshouldbe 10-5"5. Marvin and Magin's data, though,indicatethat
in pure water, carnotite slowly hydrolyzesto a colloidalhydrated uranium
oxide,whichsuggests eitherschoepite is somewhat morestable,or UO (OH) +
lessstable,than previouslyconsidered. The latter seemsmore probable.
The equilibriumconstantfor the reactiongiven by (1) is:
(K+)2 (UO2+2)2 (H 2VO4-)2
Ka= (H+)
4 (2)
Substituting ionic activities and assumingthat if ao2+rises above 10-8'1
carnotite, rather than schoepite, will be stable (to give a minimum
AFi ..... tite);
Ka = 10-'7. (3)
Since
zXF.... tion= RTIn K = -- 1.364 log K (4)
zXF o.... tio --- 18.6 Kcal. (5)
The free-energiesof all productsand reactants are known except for carno-
rite.
_ AF products
zXF.... tion-- -- AF .... rants (6)
zXFz..... tt =- 1,294(+0 to +5) Kcal. (7)
A roughcheckon this valuewasprovidedfrom datagivenby Barton (3).
Barton assumesV40 4 is the vanadiumion causingprecipitationof carnotite.
Barton'sexperimentsconsisted,in part, of a determinationof the time re-
quiredto precipitate
camoritefrom a concentrated
solutioncontaining
KzSO,,
2 Calculations on this and the following pages concerning camotire stability are made on
the basis of the HVO,- ion of Deltombe et al. (12) rather than the V,Ox-* ion of Evans and
Garrels (14) becausethis choice enormously simplifies such calculations. While this procedure
may be criticized for seeming to perpetuate a less probable interpretation of vanadium chem-
istry, careful comparisonof the two interpretations shows that for neutral, alkaline or reducing
conditions it makes no difference which interpretation is used in deriving camotire stability.
The schemeof Deltombe et al. (12) does give carnotite a slightly greater range of stability
in acid environments than does that of Evans and Garrels (14).
 TRANSPORTATION OF URANIUM AND VANADIUM 151

Na4(UO2)(COa)a and NaVOa, at various pH values. We obtainedthe

approximatepH value of sucha solutionjust saturatedwith carnotiteby
extrapolatingprecipitationtime to infinity. Concentrations
were modified
by estimatedactivitycoefficients.As the extrapolation
was onlyan approxi-
mation,we deriveda free-energyvalue for carnotitewhich lies in the range
between - 1288 and - 1293 Kcal.
The questionof carnotitestabilityrelative to that of other solid phases
in the systemis resolvedby a comparison of the activityof the commonion,
eitherUO2+2or HVO-, in equilibriumwith eachsolidphaseat a specified
Eh and pH. The activity of the commonion is not only a functionof Eh
and pH, but also dependson the activitiesof the other two ions involvedin
the precipitationof carnotite. In this study carnotitestabilityis shownon
the frameworkof the U-O2-H20-:CO2 systemand potassiumand vanadium
ai'eaddedto this systemin sucha way that K + = 10-a and total activity of all
vanadium-bearing ions (11V) = 10-a. In groundwaters,an activity of 10-s
for 142
+ is quite reasonable. The :V value given is high, but it may have
been achievedlocally during the time of ore depositionon the Colorado
Plateauand hascertainlybeenachievedin subsequent leachingand oxidation.
:CO is variedbetween10-a and 10-h Contoursrepresenting the activityof
HzVO- in equilibrium
with the variousoxidesof the V-O2-H20 systemare
show in Figure 11. These contoursare drawn on the basisthat the sum of
the activities of all vanadium ions is 10 -a.
Assume,for purposes
of illustration,a pH of 4.0, an Eh above4-0.38volts,
and a :CO value of 10% so that either carnotiteor rutherfordine,UO2CO8,
(Fig. 6) is the stablesolid. The activity of K + = 10-a and the activity of
H2VO-= 10-- (Fig. 11). From equations(2) and (3):

( (
= (8)
UO9.+9'= (10-1ti')
= 10-8-4tj. (9)

Comparingthis activity of UO2+ with that of UO * in equilibriumwith

rutherfordineunder the aboveconditions,from Figure 6 (ao2*2= 104'*),
indicatesthat carnotiteis the stablephase. Based on similar calculations
(see Appendix), Figure 12 was drawn for :CO2: 10-. Solid phases
are numbered,with carnotiteoccupyingthe lion's share of the diagram.
"Montroseite"(Fig. 12) representssynthetic,crystallineVO, for which
thermodynamic data are available. A free-energyvalue for true montroseite
(VOOH or VOa'H20) would enlargethe diagrammedfield of "montrose-
ite" slightly. Areas indicatinga certaindegreeof solubility,shownstippled
on Figures 12, 13, and 14, are basedon total activitiesof 10-6 for uranium-
bearingions (:U = 10-), 10-a for vanadium-bearing ions, and 10- for K +,
indicatingthat 40 ppm of potassium,51 ppm of vanadium,and 24 ppm of
uranium are carried in solution. The weight percentratio of vanadiumto
uranium in solutionis 2/1, a value approximatingthe generalratio of these
elementsin many parts of the ColoradoPlateau.
152 P. B. HOSTETLER .4ND R. M. G.4RREL$

Note that in Figure 12 the only solubilityindicatedis underquite acid

conditions. ICO2 is increasedto 10--in Figure 13 wherea further area of
solubilityis shownto occurfor a mildlyreducing,neutralto alkalineenviron-
ment. Increased solubility under these conditionsis due to the increased
activityof the uranylcarbonate
complexes in equilibriumwith carnotiteand
uraninite. Vanadiumin this environment is transported
in the tetravalent
state(Fig. 10). A hypotheticalsolutioncarryingthe amounts of potassium,
uranium,and vanadiumgivenabovewouldthusprecipitatecarnotiteif sub-
jectedto oxidizingconditions. Uraninite and oneof the tetravalentvanadium
minerals,h//ggiteor doloresite,wouldform if the solutionwereacidified. Para-
montroseiteis presumablymetastable and formsby oxidationof montroseite
(14). Uraniniteand montroseite
wouldbe precipitated
if the solutionhad
accessto a distinctlyreducingenvironmentcontainingeither carbonaceous
material or H2S.
A ICO2 valueof 10-2 is reasonable
for naturalgroundwaters. For ex-
ample,a seriesof spring and well water samplestaken acrossthe southern
part of the PowderRiver Basinin Wyoming(W. N. Sharp,writtencommu-
nication)showtotal alkalivalues,as HCOa-, varyingfrom 175 to 320 ppm.
Thesevaluesapproach the ICO2-- 10-2 valueof 440 ppm. The pH of these
watersis around8.0, varying from 7.6 to 8.4. Shouldthesewatersbe sub-
jectedto slightlyreducingconditions,
they would be capableof carrying
considerable vanadium and uranium in solution.
The carnotite"curtain" continuesto rise, and the UO2 field to shrink in
Figure 14, where ICO2 = 10-L The ionic strength of solutionscontaining
this much CO2 precludesthe equalityof activity and molarity, hencethe
amountsof uranium, potassiumand vanadiumactually carried in solution
are somewhatgreater than given above.

DISCUSSION WITH REFERENCE TO THE ORES OF THE COLORADO PLATEAU

The portionsof Figures 12, 13 and 14 which indicatethe Eh-pH condi-

tions necessaryfor aqueoustransport of uranium and vanadium should be
criticallyexaminedfrom two standpoints:
1) correlationwith natural ground waters.
2) the environmentof ore precipitationfrom suchwaters.
Correlationwith Natural Ground I/Faters.--Consideringthe distanceof
migration of ore solutionson the Plateau, and the absenceof intensivewall-
rockalterationthere,it is very unlikelythat the ore solutionswere sufficiently
acid (pH 3 or less) to satisfythe acid requirementsfor aqueoussolubility
shownin Figures 12, 13, and 14.
Solutionscharacterizedby thoseareasof Figures 13 and 14 which show
solubility as a function of UDC and UTC have already been mentioned
as having characteristicsin commonwith some natural ground waters (neu-
tral to alkaline,mildly reducing,moderatelylarge amountsof CO2). None
of these characteristicsis at variance with the commonlyassumedaqueous
environmentof ore depositionon the Plateau. The effects of other ionic
TRANSPORTATION OF URANIUM AND VANADIUM 153

(SlIOA) q
154 P. B. HOSTETLER AND R. M. G/1RRELS

speciescommonlyfound in subsurfacewatersand the associated problemof

electrostaticallybalancinga large excessof HCOs- and COs-- ions remain
to be discussed.
Significantadditionsof new ionic speciesinto solutionsof the U-V-K-
O2-H20-CO2 system requires evaluation of three possible phenomena.
First, suchintroductionmay causea reactionwith existingionic speciesto
precipitatenew insolublesolid phases(we are not concernedhere with the
solubleones). Secondly,new ionic complexes,characteristicof definiteEh
and pH conditions,may be formed,thus causingan increasein solubility.
Lastly, evenif the first two effectsdo not occur,any additionof new species
increasesthe ionic strength of the solution. This effect is not shown on our
diagrams,inasmuchas they are drawn for thermodynamicsolubility (activi-
ties). Increasingthe ionic strengthcausessolubility,in moles/liter, to
exceed calculatedthermodynamicsolubility. Most subsurfacewaters are
sufficientlydilute (concentrationof major specieslessthan 0.1 molal) so that
true and calculatedsolubilitiesare approximatelyequal.
Specifically,subsurfacewaters containingnotable uranium, vanadium,
potassiumand CO2 will certainlyalsocontainsignificantamountsof at least
severalof the following: sulfate,chloride,sodium,calciumand magnesium.
No insolublephasesare formedas a resultof additionof SO4-2 and CI-, but
theseionswill form complexes with the uranyland uranousions. The uranyl
sulfatecomplexes,
UO2(SO4)2-2 and UO2(SO)s-, are stableonly for pH
values less than 3.5, while the uranoussulfate complexes,U(SO)a -2 and
U (SO)c , formunderevenmoreacidconditions(29), hencethe importance
of suchcomplexes in mostnaturalwatersis negligible. From the work of
Ahrland and his associates (1, 2) it is clearthat uranyl and uranouschloride
complexes have stabilityrangesvery similarto their sulfateanalogues.
The cations,Na+, Mg +2,and Ca+2,must be appealedto for electrostatic
balancein natural solutions. Ca+2can be quantitativelyconsideredwith the
followinglimitations. For calcite:
CaCO = Ca+2+ CO-' (10)
From the thermochemical
data of Latimer (28):
K= (Ca+2)(CO-2) = 10-8'5 (11)
If :CO2= 10-2 or 10-x, maximumac,valuesin solutioncanbe calculated
with the aid of aeo-determinedfor the constructionof Figures 7 and 8
(seeAppendix). If :CO2-- 10-, calciumionscanelectrostatically balance
all HCO- andCO8- ionsin solutionif the pH is no higherthanabout7.5;
if :CO2= 10-x, the highestpH is only about6.0. Similarly,dolomiteor
magnesite wouldcontroltheamountof magnesium ionsthatcouldbecarried
in solution.
Althoughnosolubility
dataareavailable
fortyuyamunite,
Ca(UO.)
5H20, we suppose
that it is oneor two ordersof magnitude
moresoluble
than carnotite. If suchis the case,Ca+ activitiesof 10-2 to 10- would suffice
 TRANSPORTATION OF URANIUM AND VANADIUM 155

to give a stabilityfield for tyuyamunitesimilarto that of carnotitein Figures

12-14.
Rauvite, xUO'yV20'zH20, is fairly widespreadon the Colorado
Plateau. Apparently y is somewhatgreater than x (14). Two analyses
reportedby Frondel (16) show about 2.7 percentCaO each,but it is not
known whethercalciumis an essentialconstituentof rauvite. Low pH and
moderatelylarge concentrations of pentavalentvanadiumshouldtheoretically
favor the formationof rauviteratherthan carnotiteor tyuyamunite. Indeed,
Barton (3) hasfoundthat rauvite,ratherthan carnotite,is precipitatedfrom
acid solutionsof pH 1.5-2.0 and 0.01 molar in uranium and vanadium. The
association of rauvitewith navahoite,V.O,.3H.O, at MonumentValley (38)
suggests that whereaqueouspentavalentvanadiumis sufficientlyconcentrated
in an acid environment for the formation of navahoite, then rauvite is stable
relative to camoriteor tyuyamunite. As the formationof rauvite appears
to be restrictedto acid, oxidizingenvironments it is unnecessary to consider
furtherthe stabilityof thismineralwithinthe frameworkof a mildly reducing,
neutral-alkalinegroundwater.
Na* is the most likely choiceto balancea large excessof anions. The
additionof Na* to the systemsdiscussed aboveproducesno insolublemin-
erals, henceNa* can balancethe solutionselectrostatically over the entire
Eh-pH rangeof interest.
The net effect of the addition of those ionic speciescommonin natural
groundwatersto the neutral-alkalinesolutionsof the type indicatedin Fig-
ures 13 and 14 is that the boundariesbetween solid phasesand solution
remainvalid as drawn (no significantcomplexingwith SOc ' or C1-for these
pH conditions)and that sufficientquantitiesof the cations,Ca*', Mg+, and
especiallyNa*, can be carried in solutionto electrostatically balanceall the
anionspresent. Accordingly,part of the area of solubilityshownin Figure
13 is particularlycompatible with subsurface watersbecause of its pH values
and reasonable CO.content(440 ppm). The narrow Eh rangeof the solu-
bilitiesshowndefinitelysuggests that ore depositionis characterized by certain
redox potentialsand indicatesthe desirabilityof evaluatingactivity gradients
for this area of solubility. This evaluationis shownin Figure 15 where the
incrementof decreasingtotal ionic activity is an order of magnitudeand
CO. = 10-2. Shadedlinesin Figure 15 thusrepresentsolidphase-solution
contoursfor a total activity of uranium-bearingions equalto 10-*, and total
activitiesof 10-4 for K + and vanadium-bearing ions. Five ppm vanadium,
4 ppm potassiumand 2.4 ppm uraniumcan be carriedin solutionunder condi-
tionsindicatedby the shadedlines. The conditions for solubilityindicatedon
Figure 13 are shownhere with dashedlines. For an Eh-pH characterized
environment
varyingfrom the shadedto the dashedlines,vanadiumsolubility
may increasefrom 5 to 50 ppm, potassiumfrom 4 to 40 ppm, and uranium
from 2.4 to 24 ppm.
Environmentof Ore Deposition.--In Figure 15 ionicactivitycontoursfor
ions in equilibriumwith carnotiteare more widely spacedthan contours
for ionsin equilibriumwith uraniniteand montroseite,
indicatingthat the
156 P. B. HOSTETLER AND R. M. GIRRELS

stabilityof carnotiteis muchmoresensitive

to decreasing
amountsof uranium
and vanadiumin solutionthan is the stability of uraninite or montroseire.
Indeed,if the activityof potassium,
uraniumand vanadiumionsshouldeach
dropto 10-6'acamoritewouldcompletely
disappear
fromFigures13 and 15,
whereasuraninitehas a sizeablefield of stabilityunder suchconditions(Fig.
7). Suchvalues,reasonable for uraniumand vanadium,are distinctlylow
for potassium,
but evenif K+ is raisedto 10-a, the fieldof camoritestability
is very small.
Considera potentialore solutionmigratingslowlythroughpore spaces
in the host rock under thousandsof feet of sediments. Such a solution would
almostcertainlybe devoidof significant
oxygen. Redoxpotentialwouldbe

c $
O.I

Paramontroseit/,.
'a.. - '</,.
Carnotite
-
=
II
I
-O.1 Rulherrdine
'<//,. = II
-0.2
H20 h..... [
,- -0,3

-0.4

-0.5
and
-'==='Montrose
'1 ;
-0.6

S!
-0.
0 2 4 6 7 8 I0 12 C
pH units

FIG. 15. Aqueousequilibriumdiagramof the U-V-K-O-H._,O-CO2 systemat

25C and 1 atm., and with CO2 fixed at 10-g. Boundariesbetweensolid phases
and solutionare fixed for :U--10 -5, V = 10-4, and :K----10-4 (shadedlines),
whereas dashedlines representsolid phase-solutionboundariesfor :V--10 -a,
K: 10-a and U: 10-4.

poorlybuffered. Sinceearnotiteis stableovera widerangeof Eh andpH

conditionsits formation would be most critically controlledby the amounts
of uranium, vanadiumand COg in solution.
Migrationthroughpockets of carbonaceousmaterialor areasof entrapped
HgS would,of course,makethe solutionEh more reducing.Two likely
redoxpotentials are shownin Figure 15 by linesc-c' and s-s'. Line c-c'
representsa calculatedequilibriumpotentialbetween graphiteand aqueous
carbonatespecieswhich is analogous to an equilibriumpotentialbetween
carbonaceous materialand carbonatespecies.Line c-c' was calculatedfor
CO2: 10% but the changein potentialis only 15 my. whenCO2 is
alteredby a factorof 10. Line s-s' showsthe redoxpotentialof equalcon-
 TRANSPORTATION OF URANIUM AND VANADIUM 157

centrations of SO4-2 andsulfideionsascalculated by GarrelsandNaeser(22).

Migrating solutionsat 25 C containingSO4- and encountering gaseousH2S
wouldreacha potentialvery closeto this line independently of the amountof
sulfurspecies in eachphasebecause of the very steepactivitygradientinvolved
for eachaqueoussulfur speciesaway from line s-s'.
BetweenpH 6 and pH 10 bothpotentialsare very closeto the bottomof
the area designatingsolubilityin Figure 15. At 25C and 1 atmosphere
pressurethe reducingenvironmentindicatedby thesepotentialsis sufficient
to precipitateuraninite and montroseitcif solution concentrationis about
0.5 ppmor higherfor uraniumand 2 ppm or higherfor vanadium. Natural
uraniniteprecipitatedin suchan environment,however,is probablycloser
to UO. than to UO2 indicatingthat the saturationconcentration of uranium
is somewhatlessthan 0.5 ppm because UO. is more stableand thereforeless
solublethan UO. An environmentof. rising temperatureshouldnot signifi-
cantly reducethe effectiveness of HS and woody material as precipitation
agentsfor uraninite and montroseitcfrom natural solutions. Rafalsky (32)
heateduranyl sulfatesolutionswith aluminumsulfidein closedsilicaampules
and foundthat at 150 C the concentration of solubleuraniumin the presence
of uraniniteis 1.6 ppm. At 125 C Miller (29) notedabout1.2 ppm uranium
in solutionafter precipitation of uraniniteby HS. It thusseemslikely that
the Plateauuraninitewasdeposited from solutionsthat containeda minimum
of about0.5 ppmuranium.
It is noteworthythat the uranyl carbonatecomplexes becomeincreasingly
unstableabove 120 C. Rafalsky's data (32) indicate that uranium solu-
bility in carbonatesolutionsdiminishesby a factor of about 100x between
100 and 200 C and that the decreasein solubilityis slight between100 and
120C but becomesmuch greater above 120 C. A similar temperature
range (120-215 C) wasusedin Miller's experiments(29) and the resulting
decreasein uranium solubilitywas found to be one order of magnitude.
Neither authoroffersany data for uraniumsolubilityin carbonatesolutions
between25 and 100 C, but Rafalsky'swork between100 and 120 C suggests
that solubilityis either fairly constantin the 25-100 C intervalor may even
rise to a maximumsomewhatunder 100 C. IncreasedI-I2S pressure(up to
9,000lbs/sqinch) alsodecreases uraniumsolubility,according to Miller (29),
but its total effect is small.
The foregoing considerationslead to a consistentpattern of aqueous
transportand depositionof ore elementson the ColoradoPlateau. The
ubiquitousspatialand temporalrelationshipof uranium and vanadiumore
mineralspointsto a commonorigin and transportationmediumfor these
elements. Althoughthis commonorigin cannotbe pinpointed,it seemslikely
that in late Cretaceousor early Tertiary time, perhapseven earlier, a heavy
load of overlyingsedimentsforceduranium-and vanadium-bearing solutions
out of sedimentsrich in volcanic ash, such as the Chinle Formation of Late
Triassicage and the Brushy Basin Member of the Morrison Formationof
Late Jurassicage (36). These solutions,comparableto the type discussed
in the previoustwo sectionsof this study,migratedthroughpore spacesand
158 P. B. HOSTETLER IND R. M. GIRRELS

permeablechannelsin the ShinarumpMember of the Chinle Formation or

the Salt Wash Memberof the MorrisonFormation,wherethey undoubtedly
encounteredreducingenvironmentswith concomitantore precipitation(see
Jensen (26), p. 611-615, for a more completediscussionof this point).
Theseenvironments may havebeenpocketsof carbonaceous materialand/or
I-t2S, or even H2 which may have been locallyproducedby the reactionof
I-tS with Fe2Oa. Gaseoushydrogenwill precipitateuraniniteand montroseite
from very dilute solutions.
By the end of Cretaceous time the thicknessof the stratigraphiccolumn
atop the Salt Wash Member in the Uravan districtwas nearly the sameas
that abovethe ShinarumpMember in northeasternArizona, a thicknessof
about 6,000 feet (L. C. Craig, oral communication). Normal geothermal
gradientsacrossthis thicknessduring the time of ore-solutionmigration
wouldmeantemperatures around75 C. This is well belowthe temperature
(120 C) at which the uranyl carbonatecomplexesbegin to break down.
Of course,a deeperdepth of burial, or higher geothermalgradients,might
haveproducedtemperatures in excessof 120 C in the vicinityof the migrating
solutions,in which caseuraniniteshouldprecipitateas a result of decomposi-
tion of UDC and UTC. Becausethis control of ore formation, unlike those
given earlier, is selectivetowardsuranium,there is no reasonfor uraninite
so formed to be found in associationwith trivalent vanadium minerals, sulfides
or carbonaceous material. Lack of such ore on the Plateau supportsthe
hypothesisthat temperaturesin the environmentof ore depositionwere not
above 100 to 120 C.
Formationof Colnite and VanadiumSilicates.--Ore depositionhas many
ramificationsthat are not discussed
here, suchas comparisonof precipitation
conditionsfor locallyabundantcopper,molybdenumand seleniumminerals
of the Plateauwith thoseprecipitationconditionsgiven here for uranium and
vanadium minerals. Such a comparisonwould make an excellent future
study (many of the pertinentthermochemical data are known for 25 C).
Some remarksabout the stabilityof coffinite,U(SiO4)x_x(OI-t)4x,and the
vanadiumsilicatesare appropriatebecauseof their widespreadoccurrenceas
primary mineralson the Plateau. No thermochemical data are availablefor
these minerals.
Coffiniteand uraninite are the only known minerals from the Colorado
Plateau in which the dominant uranium is in the +4 oxidation state, and
their intimate associationindicatesmutual low solubilityunder reducingcon-
ditions(21). The uraninite-coffiniteequilibriummaybe written:
U(SiO4)_x(OH)4: UOo.+ (1 - x)SiO.o+ 2xH.oO. (13)
At a given temperatureand total pressure,equilibriumbetweenthe solid
phasesis a functionof the activity of water and aqueoussilica. In high-
temperaturehydrothermal veinsand in pegmatites reaction(13) goesto the
right. The environmentof ore depositionon the Plateaumust have been
suchthat reaction(13) went eitherway, depending on smallaa2oand asio
fluctuations.Descriptionsof corrodedquartz grainsin impregnatedsand-
 TRANSPORTATION OF URANIUM AND VANADIUM 159

stoneand nearbysecondary quartz overgrowths(33, 39) indicatea liquid

phasethat hoveredaroundquartz saturationvalues.
As colfiniteis rapidly alteredunder oxidizingconditions(21), it seems
probablethat in the presenceof excessquartz coffiniteoccupiesmuch the
samepositionas uraninitein the Eh-pH diagramspreviouslyshown.
The vanadiumsilicatesof the Plateauore depositsare the vanadiummica
(roscoelite, K(A1,V) 2(A1,Sia)O0(OH,F) 2), and vanadium-bearingclays
(chlorites and mixed layer montmorillonite-micasand montmorillonite-
chlorites). Severalworkers(15, 23, 39), emphasizing the pronounced chemi-
cal and mineralogical similaritybetweenmineralizedand nearbybarrenclays,
have concludedthat most of vanadiumclays originatedby replacementof
octahedrally coordinated aluminumby vanadiumin preexistingclay minerals.
Roscoeliteand some of the vanadiumclays probablyprecipitateddirectly
from solution(15).
The presenceof montroseiteat those localitieson the Plateau where
roscoeliteis abundant(6) suggeststhat, at low temperaturesand moderate
concentrations of K +, A1 and +3 vanadium,roscoelitehas a field of stability
similarto montroseite on Figures12-15. Apparently,a decreasing V/U ratio
is accompanied by a relativedecreaseof roscoeliteand increaseof montroseite
in primary vanadiumore (39). Abundant clay mineralsin both types of
ore (high roscoeliteor high montroseite)indicatethat solubleK, A1, and Si
do not changenoticeablyas the V/U ratio fluctuates and that the formation
of roscoeliteis probablymore sensitiveto decreasingconcentrations of +3
vanadium than is montroseite.
Most of the vanadiumin the mineralizedclays is tetravalent,even in
oxidizedore (15, 23). This observationhas led to the suggestion(Garrels
et al., (23)) that the vanadiumwas transportedin the tetravalentstate in
the ore solutions. Reasoningbasedon availablethermochemical data (Figs.
13-15) supportsthe notion that vanadiumwas tetravalentduring transport.
However, if this reasoningis correct, an equally fundamentalquestionre-
mains;namely,why shouldonly thoseclay mineralsin ore depositscontain
+4 vanadium? Barren clay mineralsa few inchesfrom mineralizedclays
containonly one tenth as much vanadiumas the mineralizedclays (15).
This selectivesubstitutionof octahedralA1 by +4 vanadiumis especially
puzzling as the ore solutionsprobablymigrated considerabledistances(27,
36). We concludethis paper by speculatinga bit on this question.
First, +3 vanadiummay replaceoctahedrallycoordinatedaluminummore
readily than does+4 vanadium. If this is true then most of the vanadium
was originally emplacedin the clay minerals in the trivalent state. The Eh
valuesof the sulfide-sulfateand carbon-carbonate redox potentialswhich we
believecontrolprimarymineralization(lines s-s' and c-c' of Fig. 15) require
most,but not all, of the aqueousvanadiumat thesepotentialsto be trivalent.
The association of vanadiumclay mineralswith montroseite,as in the Uravan
district (33, 39) and with roscoeliteand montroseite,as at Placervilleand
Rifle (6), alsoimpliestrivalentaqueousvanadiumat the site of clay minerali-
zation. The intimate associationof roscoelitewith the vanadiumclays is
160 P. B. HOSTETLER .lIND R. M. G.4RRELS

particularlysuggestive of original -3-3vanadiumin the clays,for Hathaway

(25) has shownthat completelytransitionalmixed layeringoccursbetween
pure roscoeliteand randomlyinterstratifiedroscoelite-montmorillonite clays.
From her study of the presentoxidationstatesof vanadiumin selected,
mixed-layerroscoelite-montmorillonite claysin partially oxidizedore, Foster
(15) concludesthat the vanadiumwas originally emplacedin the +4 state.
Foster (15) arguesthat assumingthat all vanadiumwas originally trivalent
leadsto abnormallylargenegativechargesfor the sumof the tetrahedraland
octahedral layersof the clayminerals. Fosterassumes, though,that noneof
the vanadiumis lost in the theoreticaloxidation of vanadium in theseclays to
the -t-4 state. The increasein stabilityof aqueousionic speciesbearing -t-4
vanadium comparedto those containing+3 vanadium (cf. Fig. 2 in Evans
and Garrels (14)) suggeststhat oxidationof vanadiumfrom the +3 to +4
statein clayswouldbe accompanied by a modestlossof vanadiumto solution.
If the claysdescribed by Fosterwereto loseonlyten percentof their vanadium
during theoreticaloxidationthe total negativechargeof the tetrahedraland
octahedrallayerswould be reasonablefor mixed layer roscoelite-montmoril-
lonites (about-0.90).
Clearly, analysesare neededof the oxidation state of vanadium in
vanadiumclayscloselyassociated with unoxidizedassemblages.
A secondhypothesis, put forward by Roachand Thompson(33) to ex-
plain the presenceof vanadiumclay in otherwiseunmineralizedsandstone
at the Peanut Mine, states that these clays originated from tetravalent
vanadiummadeavailableduring oxidationof primary vanadiumminerals.
A third, but lessplausible,methodfor selectivelyreplacingA1+aby V +4
within the clay mineralsof the ore depositappealsto relativelyimpermeable
boundaries,such as festoonsurfaces,to detain migrating ore solutionsfor a
sufficientlength of time for replacementto occur. Replacementby tetra-
valent vanadiumpresumablyoccurredaway from local "hot spots"of car-
bonaceousmaterial. The releasedaluminum might eventually react with
aqueouspotassium,silica, and trivalent vanadium (available nearer car-
bonaceousmaterial) to form roscoelite.

SUMMARY

1. Eh-pH diagramsat 25 C were constructed with the aid of published

thermodynamicdata for the U-O2-H20 and U-O-HO-CO2 systems.
Uraninite,UO+x,and schoepite, UO2(OH)2-H20, are the only solidphases
expectedto be stablein the presence of waterand variableconcentrationsof
oxygenin the systemU-O-H20. The additionof CO,oto this system
causesformationof stableuranyl carbonatecomplexesand a significantin-
creasein the solubilityof uraniniteand schoepite.
2. Followinga rough calculationof the free-energyof carnotite,Eh-pH
diagramsof the U-V-K-O=-YIO-CO., systemweredrawnfor variousvalues
of total carbonate(:CO2). Thesediagramsshowstabilityfieldsfor carno-
tite andotheruraniumandvanadiumsolidphases,and indicatethe conditions
 TRANSPORTATION OF URANIUM AND VANADIUM 161

necessaryfor vanadiumand uraniumsolubility. One set of conditionswas

foundto indicatesolubilityin a mildly reducing,neutralto alkalineenviron-
ment containingfairly abundantCO2, and thus is comparablewith some
natural ground waters.
3. Analysisof solubilitycontrolsindicatesthat increasingCO2 increases
solubility. Oxidizingconditionscausecarnotiteto precipitatefrom solution
while more reducingconditionscausemontroseireand uraniniteto form.
4. The possibleeffectsof other ionicspeciescommonlyfoundin natural
ground waters were investigatedwithin the Eh-pH framework of aqueous
transport in the U-V-K-O-HO-CO system. No significantchangesin
suchtransportwere found. Sodiumand, to a lesserdegree,calciumand
magnesiumcan balanceelectrostatically the concentrations of HCOa-, COa-,
and otheranionsin solution. Activity gradientsshowingvaryingtransporta-
tion conditionswere presented.
5. In the presence of excessquartzthe fieldof coilfinite
stabilityis probably
very similar to that of uraninite. With favorableconditions(moderatecon-
centration of its constituentelements) the field of formation of roscoelite
shouldapproximatethat of montroseite.
6. Carnotitestabilityis very sensitiveto changesin ionic activity rather
than changesin Eh and pH (except for distinctly reducingconditions);
its precipitationis criticallycontrolledby the amountsof uranium,vanadium
and CO2 in solution. The reducing environmentprovided by ore-solution
migrationintocarbonaceous
and/or HS richareaswasshownto be sufficient
to precipitate uraninite and montroseitefrom dilute solutionsat 25 C.
Temperaturesabove 120C might be sufficientto precipitateuraninitefrom
the postulatedore-bearingsolutionsbecauseof the breakdownof the uranyl
carbonatecomplexes. Sulfides,carbonaceous materialand trivalentvanadium
mineralswould not be expectedto occurwith uraninite formed in this way.
The lack of suchore is interpretedas supportingevidencefor a depositional
temperatureof 120 C or less.

ACKNOWLEDGMENTS

We are gratefulto our colleagues

of the U.S. Geological
Survey,Alice D.
Weeks, Paul B. Barton, Richard P. Fischer,and especiallyCharlesL. Christ,
for stimulatingcriticism and helpful suggestionsconcerningmuch of the
material of this study. TheodoreBotinelly,Lawrence C. Craig, John C.
Hathaway,and J. J. Hemley of the U.S. GeologicalSurvey,and Clifford
Frondel of Harvard University, also contributedmany ideasduring discus-
sionsof the problemspresentedhere,and to them we extendour thanks.
U.S. GEOLOGICAL
SURVEY,
DENVER,COLORADO
AND

LABORATORY
OF MINING GEOI.OGY
HARVARDUNIVERSITY,
CAMBRIDGE,MASSACHUSETTS
162 P. B. HOSTETLER AND R. M. GARRELS

APPENDIX

This Appendix was prepared for those readerswho desire the details of
construction of the Eh-pH diagrams here presented. An introduction to
the methodsof Eh-pH diagramming,basedon the useof free-energydata to
calculateequilibrium constants,is given by Pourbaix (31). All free-energy
values usedin our calculationshave been taken from Latimer (28) except as
noted below. The values for uraninite and schoepiteare from Deltombe
et al. (13), while those for rutherfordine and the uranyl carbonate complexes
are from Bullwinkel's data (9) correctedfor the recent XFof schoepitegiven
by Deltombe et al. A value for HUO4- was calculatedfrom the published
equilibrium constantsof Gayer and Leider (24). Bjerrum et al. (4) tabu-
late six fairly consistent constants for the dissociationof UO2(OH) + at
25 C. We averagedtheseand so obtained a aF of UO2(OH) + at 25 C.
The value for camotire was derived in the text. Free-energyvalues for the
vanadium oxides are taken from Evans and Garrels (14), while that of
HVO4- is fromDeltombeet al. (12). Allfree-energy
valuespertinentto
the constructionof the diagramspresentedin this paperare listedin Table 1.
TABLE 1

FREE--ENERGY VALUES FROM THE ELEMENTS, AT 25 C AND 1

ATMOSPHERE TOTAL PRESSURE

Aqueous species AF.t' (Kcal) Solid phases AF.t' (Kcal)

HO -- 56.69 UO. -- 246.6

H + 0.0 UO2 (OH) 2' H20 -- 398.8
OH- -- 37.6 UO2CO3 -- 385.0
HCOf -- 149.00 -- 271.0
HCO- -- 140.31 V204 -- 381.0
CO-- --126.22 -- 344.0
-- 67.46 K.(UO 2)2V2Os3 H 20 -- 1,294.0
HVO4- --248.8
--236.4
U+4 --138.4
U (OH)* --193.5
UO2 (OH) * --287.4
HUO4- --317.6
UO (CO) 2' 2 H 20-- --630.0
U02 (COs) - --648.0

U--O2--H20--Pc02 DIAGRAMS

The followingequationsare pertinent:

CO(g) + HO(1) = H.CO3
H.CO = H + + HCOa- (2)
ttCO- = H + + CO-- (3)

UO(CO02-2H20-- = UO,,.
++ + 2CO3-- + 2H,O (4)

UO2(CO8)34-- = UO2'+"+'
+ 3COi (5)
 TRANSPORTATION OF URANIUM AND VANADIUM 163

The thermodynamic equilibrium constants for reactions (1)-(5) are as

follows:
HCOs
-- 10--1'47 ' (6)
Pco

(HCOF) (H +)
10-6'38 (7)
HC0s

(C0s--) (H +)
----10--10'1 (8)
HC0a-

(U0,.++) (C0a--)
= 10-2'4 (9)
UO (COs) 2H,.O--

- 10-24'. (10)
UO2 (COs)s4-

For a fixed Pco, HCOs is calculated from (6). For a given pH value
COs-- can be calculated from equations (7) and (8). Next, all ions con-
taining uranium are consideredto be in equilibrium with schoepite.
UO,.(OH).HO d- 2H + -- UO++ d- 3HO (11)
or,
UO++
K- (H+)
. ,.(12)
Hence, UO++ can be calculatedfrom stipulatedpH valuesby (12). Lastly,
activities for the uranyl carbonatecomplexescan be derived from equations
(9) and (10) with the aid of COs-- and UO++ values calculatedabove.
If Pco2is high enough, rutherfordine (UOCOs) is the stable hexavalent
uranium solid phaserather than schoepite.
UO(OH).HO d- CO() = UOCOs d- 2HO (13)
1

K- Pco2' (14)
When Pco2is 10-s.s the two solid phasesrutherfordineand schoepiteare in
equilibrium. For Pco2values higher than this, UO ++ must be determined
from COs-- values (above) as follows'

UOCOs = UO++ d- COs-- (15)

(UO++) (COs--) = 10-6'4. (16)
For reducingconditionsuraninite becomesthe stablesolidphase.
UO d- 3HO = UO(OH).HO d- 2H + d- 2e (17)

UOa d- COs-- = UOaCOsd- 2e. (18)

164 P. B. HOSTETLER AND R. M. G.4RRELS

Employing the Nernst equation for equations (17) and (18):

0.06
Eh= Eq-- log(Itq (19)

O.O6 1

Eh= E+ --log(COa--)' (20)

E for each reaction is evaluated from free-energydata. Depending on the
Po= value under consideration,either (19) or (20) is chosen. If Pois less
than 10-a-8the Eh-pH boundary between uraninite and schoepiteis found by
substituting pH values in (19). For Po2greater than 10-a-8 the ruther-
fordine-uraninite boundary is establishedby substituting the COa-- values
derived from equations (7) and (8) in equation (20). In each case the
solid-solid boundary has a linear slope of 60 mv/pH unit. In the UO2
field then:
UO2 = UO2++ q- 2e (21)
0.06
Eh = Eq- log(UO2++). (22)
UO2++ is calculated from (22) for arbitrary Eh values (UO2++ is not pH
dependentin the UO2 field as expressedin (21)). UCD and UTC activities
are then calculated as before; these ion activities are now both Eh and pH
dependent.
U--02-I20-- C02 DIAGRAMS

Equation (1) is no longer applicable since Po2is variable. Equations

(2)-(5) and, hence, (7)-(10) are used, but note that now, after selecting
arbitrary pH values, COa-- cannot be solved from equations (7) and (8)
because there are three unknowns. Since 21CO2 is fixed, the following
expressionis used.
21CO2 = H2COa + HCOa- + COa-- + 2UO2(CO3)2-2H20--
q- 3UO2(CO0 4-. (23)

Rewrite equations (7)-(10) so that ionic activity soughtis a function of

UO++2 and COa--:
(COa--) (H +)
HCOa-
= 10_o
m (24)
(HCO-) (H+) (CO3--) (H+)2
H2CO3
= 10_6.a8= 10_16.6
9 (25)
(U02 ++) (CO--)2
UO2 (COa)2' 2H 20-- = 10 --20.4
(26)

uo2(coa)
4-= (uo,.++)
(coo-)
10-24.2
3 (27)
 TRANSPORTATION OF URANIUM AND VANADIUM 165

If UO2(OH)2-H=O is the stablesolid phase,UO2++ is fixed for arbitrary

pH values. Equations(24)-(27) are substitutedin equation (23) which is
solved for COs'-- using the desiredZ CO. value. Equation (23) is best
solvedby approximationas it is a third-orderequation. Remainingion
activities are then determined using (24)- (27).
Equation (16) is used to determine the UO2CO3- UO2(OH)2.H20
boundary.
10-.4
UO2++
- COa-- (16)
Substitute this expressionfor UO2++ in equations(26) and (27). As above,
substitute (24)-(27) in equation (23) and solve for CO3--. Finally, use
this CO3-- value to determine UO.++ in equilibrium with rutherfordine from
(16). If this value for UO=++ is lessthan UO=++ in equilibrium with schoe-
pite for a givenpH then UO2CO3is the stablesolid. The boundarybetween
the two phases (UO=+ values in equilibrium with each are identical) is
dependenton both CO=and pH, but not on Eh.
The rutherfordine-uraninite and uraninite-schoepitebounding equations
have been given ((17)-(20)). The uraninite-schoepiteboundary is not
dependenton CO=. Use the CO3-- valuesobtainedfrom (23) and (16) for
establishingthe uraninite-rutherfordineboundary.
Ionic activities in equilibrium with UO= are cumbersometo determine
when :z CO2 is fixed. UO2++ values for (26) and (27) are derived from (22)
for arbitrary Eh values. Substituting (26) and (27) in (23) indicatesthat
COa-- is both pH and Eh dependent. Activities for each ion must be cal-
culated on a point-by-point basisin the field of stability of uraninite.
U--V--K--O2--H20--CO2 DIAGRAMS

Carnotite stability must be consideredin relation to both uranium and

vanadium solid phases.

K2(UO02V2Os-3H20 q- 4H + = 2K + q- 2UO2++
q- 2H=VO4- q- 3H20 (28)
(K +)'(U 02++)=(H 2VO
= 10-a'L (29)
(H+) 4

For any Eh-pH value, the activity of H=VO4- is taken from Figure 11 (see
text). K + = 10-3, and UO2++ values are taken from the results of calcula-
tions for the U-O=-H20-:Z CO= diagrams. When these values, for the
particular Eh and pH in question,are inserted in the left-hand side of equa-
tion (29), carnotite is the stable solid if the value of the left-hand side of the
equation exceeds10-3'7. Carnotite is not stable relative to the other ura-
nium and vanadium solidsif the left-hand side of (29) is lessthan 10-3'L
Equation (29) is thus usedas a criterion for determiningat specifiedEh and
pH conditions, whether caronitite is or is not stable relative to the combina-
tion of the stable uranium oxide (or carbonate) plus the stable vanadium
166 P. B. HOSTETLER ,'tND R. M. G,'tRRELS

oxide plus potassiumhydroxide. The solid-solidboundariesfor Figures

12-14 are drawnfor thoseEh-pH conditionswherethe left-handsideof (29)
is equal to 10-a'7.
KeepingZ CO2constant,equations(23)--(27)remainapplicable:
23CO. = H2CO3 q- HCO3- q- CO
q- 2UO2(COa).-2H20-- q- 3UO2(COa)a
4- (23)

HCOa-= (COa--)
(H+)
10_0.
(24)

= (COa--)
H2CO (H+)2
10_,0.69 (25)
(UO2++) (COa--)2
UO2(COa)2.2H20--
= 10_.0.4 (26)
(UO2++)(CO--)
UO2(CO)
4-= 10_24.
2 (27)
Rearrangingequation (29):

UO2++
=[ (K+)2
(10-a')(H+)4]'
(H2VO4-)21 (30)
UO2++ valuesfromequation(30) are substitutedin (26) and (27), and ion
activities in equilibriumwith carnotite are determinedin the manner
described above.
REFERENCES

1. Ahrland,S., 1951,The complex

chemistry
of the uranylion. VI. The complexity
of uranyl
chloride,bromide,and nitrate: Acta Chem. Scand.,v. 5, p. 1271-1282.
2. --., andLarssen,R., 1954,The complexity
of uranium(IV) chloride,bromide,
andthio-
cyanate: Acta Chem. Stand., v. 8, p. 137-150.
3. Barton,P. B., Jr., 1958, Synthesis
and propertiesof earnotiteand its alkali analogues:
Am. Mineralogist, v. 43, p. 799-817.
4. Bjerrum, J., Sehwarzenbaeh, G., and Silldn, L. G., 1958, Stabilityconstants,
Part II:
InorganicLigands:Spec.Pub. No. 7, The ChemicalSociety,London.
5. Blake,C. A., Coleman,
C. F., Brown,K. B., Hill, D. G., Lowtie,R. S., and Schmitt,J. M.,
1956, Studiesin the carbonate-uranium system:Jour. Am. Chem. Soc., v. 78, p.
5978-5983.
6. Botinelly,T., and Fischer,R. P., 1959,Mineralogyand geologyof the Rifle and Garfield
Mines, Garfield County, Colorado:U.S. Geol. Survey Prof. Paper 320, Pt. 19, p.
213-218.
7. Breger,I. A., and Chandler,J. C., 1960,Extractabilityof humieacid from eoalifiedlogs
as a guideto temperatures in ColoradoPlateausediments:Ecoa. Gv.o.,v. 55, p
1039-1047.
8. Brooker,E. J., and Nuffield,E. W., 1952, Studiesof radioactivecompounds:
IV, piteh-
blende from Lake Athabaska, Canada: Am. Mineralogist, v. 37, p. 363-385.
9. Bullwinkel,E. P., 1954,The chemistryof uraniumin carbonatesolutions:TopicalReport
RMO-2614, The Merrill Co., San Francisco.
10. Christ, C. L., and Clark, J. R., 1960, Crystal chemicalstudiesof someuranyl oxide
hydrates: Am. Mineralogist,v. 45, p. 1026-1061.
11. Coleman,R. G., 1957, Mineralogiealevidenceon the temperatureof formationof the
Colorado Plateau uranium deposits: Ecoa. Gzo., v. 52, p. 1-4.
12. Deltombe,E., de Zoubov,N., and Pourbaix,M., 1956, Comportement
dleetroehimique
du
vanadium:RapportTechniqueNo. 29, CentreBelged'lgtudede la Corrosion.
 TRANSPORTATION OF URANIUM AND VANADIUM 167

13. --, 1956, Comportementglectrochimique

de l'uranium: Rapport TechniqueNo. 31, Centre
Belge d'Ptude de la Corrosion.
14. Evans, H. T., Jr., and Garrels, R. M., 1958, Thermodynamicequilibria of vanadium in
aqueoussystemsas appliedto the interpretationof the ColoradoPlateau ore deposits:
Geochim. et Cosmochim. Acta, v. 15, p. 131-149.
15. Foster, M.D., 1959, Chemicalstudy of the mineralizedclays: U.S. Geol. Survey Prof.
Paper 320, Pt. 10, p. 121-132.
16. Frondel, C., 1958, Systematicmineralogy of uranium and thorium: U.S. Geol. Survey
Bull. 1064.
17. Garrels, R. M., 1953, Some thermodynamicrelations among the vanadium oxides and their
relation to the oxidation state of the vanadium ores of the Colorado Plateaus: Am.
Mineralogist, v. 38, p. 1251-1265.
18. Garrels, R. M., 1955, Geochemistry of oxidation of the uranium deposits of the Colorado
Plateau: preprint 250, Nuclear Engineering and Science Congress, published by Am.
Inst. Chem. Engineers, New York.
19.--, 1955, Some thermodynamicrelations among the uranium oxides and their relation
to the oxidation states of the uranium ores of the Colorado Plateaus: Am. Mineralogist,
v. 40, p. 1004-1021.
20.--, 1957, Some free-energy values from geologic relations: Am. Mineralogist, v. 42,
p. 780-791.
21. ., and Christ, C. L., 1954, Behavior of Colorado Plateau uranium mineraIs during
oxidation: U.S. Geol. Survey Prof. Paper 320, Pt. 6, p. 81-89.
22. --, and Naeser, C. R., 1958, Equilibrium distribution of dissolvedsulfur speciesin water
at 25 C and 1 attn. total pressure: Geochim. et Cosmochim. Acta, v. 15, p. 113-130.
23.--., Larsen, E. S., 3rd, Poreruer, A.M., and Coleman, R. G., 1959, Detailed chemical
and mineralogical relations in two vanadium-uranium ores: U.S. Geol. Survey Prof.
Paper 320, Pt. 15, p. 165-184.
24. Gayer, K. H., and Leider, H., 1955, The solubility of uranium trioxide, UOa-HO, in solu-
tions of sodium hydroxide and perchloric acid at 25C: Jour. Am. Chem. Sot., v. 77,
p. 1448-1450.
25. Hathaway, J. C., 1959, Mixed-layered structures in vanadium clays: U.S. Geol. Survey
Prof. Paper 320, Pt. 11, p. 133-138.
26. Jensen, M. L., 1958, Sulfur isotopesand the origin of the sandstone-typeuranium deposits:
EcoN. Gzox., v. 53, p. 598-616.
27. Kerr, P. F., 1958, Criteria of hydrothermal eraplacementin Plateau uranium strata: Proe.
of the Second U. N. Int. Conf. on the Peaceful Uses of Atomic Energy, v. 2, p. 330-334.
28. Latimer, W. M., 1952, Oxidation potentials: 2nd edit., Prentice-Hall Inc., New York.
29. Miller, L. J., 1958, The chemical environment of pitchblende: Ecor. Gzox., v. 53, p.
521-545. '
30. Miller, S. D., and Kulp, J. L., 1958, Isotopic study of some Colorado Plateau ores:
Ecor. Gzox., v. 53, p. 937-948.
31. Pourbaix, M., 1949, Thermodynamics of dilute aqueous solutions: Edward Arnold and Co.,
London.
32. Rafalsky, R. P., 1958, The experimental investigation of the conditions of uranium
transport and deposition by hydrothermal solutions: Proc. of the Second U. N. Int.
Conf. on the Peaceful Uses of Atomic Energy, v. 2, p. 432-444.
33. Roach, C. H., and Thompson, M. E., 1959, Sedimentary structures and localization and
oxidation of ore at the Peanut Mine, Montrose County, Colorado: U.S. Geol. Survey
Prof. Paper 320, Pt. 17, p. 197-202.
34. Stieff, L. R., Stern, T. W., and Milkey, R. G., 1953, A preliminary determination of the
age of some uranium ores of the Colorado Plateau by the lead-uranium method: U.S.
'Geol. Survey Circ. 271.
35. Stieff, L. R., and Stern, T. W., 1959, Isotopic study of some Colorado Plateau ores:
EcoN. Gzox., v. 54, p. 752.
36. Waters, A. C., and Granger, H. C., 1953, Volcanic dbris in uraniferous sandstonesand its
possiblebearing on the origin and precipitation of uranium: U.S. Geol. Survey Circ. 224.
37. Weeks, A.D., 1955, Mineralogy and oxidation of the Colorado Plateau uranium ores:
U.S. Geol. Survey Prof. Paper 300, p. 187-193.
38. --., and Thompson, M. E., 1954, Identification and occurrenceof uranium and vanadium
minerals from the Colorado Plateaus: U.S. Geol. Survey Bull. 1009-B, p. 13-62.
39.. , Coleman, R. G., and Thompson, M. E., 1959, Geochemistry and mineralogy of the
Colorado Plateau uranium ores; summary of the ore mineralogy: U.S. Geol. Survey
Prof. Paper 320, Pt. 5, p. 65-79.