7/25/2019 Parry 2011 concreciones de hierro

1/16

Composition, nucleation, and growth of iron oxide concretions

W.T. Parry

Department of Geology and Geophysics, University of Utah, 115 South 1460 East, Room 383, Frederick A. Sutton Building, Salt Lake City, Utah 84112-0101, United States

a b s t r a c ta r t i c l e i n f o

Article history:

Received 4 June 2010

Received in revised form 20 October 2010

Accepted 26 October 2010

Available online 3 November 2010

Editor: B. Jones

Keywords:

Concretions

Goethite

Lepidocrocite

Sandstone diagenesis

Iron oxide concretions are formed from post depositional, paleogroundwater chemical interaction with iron

minerals in porous sedimentary rocks. The concretions record a history of iron mobilization and precipitation

caused by changes in pH, oxidation conditions, and activity of bacteria. Transport limited growth rates may be

used to estimate the duration of

uid

ow events. The Jurassic Navajo Sandstone, an important hydrocarbonreservoir and aquifer on the Colorado Plateau, USA, is an ideal stratum to study concretions because it is

widely distributed, well exposed and is the host for a variety of iron oxide concretions.

Many of the concretions are nearly spherical and some consist of a rind of goethite that nearly completely lls

the sandstone porosity and surroundsa central sandstone core.The interior and exterior host-rock sandstones

are similar in detrital minerals, but kaolinite and interstratied illitesmectite are less abundant in the

interior.Lepidocrociteis present as sand-grain rims in theexteriorsandstone, butnot present in theinteriorof

the concretions.

Widespread sandstone bleaching resulted from dissolution of early diagenetic hematite grain coatings by

chemically reducing water that gained access to the sandstone through fault conduits. The iron was

transported in solution and precipitated as iron oxide concretions by oxidation and increasing pH. Iron

diffusion and advection growth time models place limits on minimum duration of the diagenetic, uidow

events that formed the concretions. Concretion rinds 2 mm thick and 25 mm in radius would take place in

2000 years from transport by diffusion and advection and in 3600 years if transport was by diffusion only.

Solid concretions 10 mm in radius would grow in 3800 years by diffusion or 2800 years with diffusion and

advection.

Goethite (-FeO (OH)) and lepidocrocite (-FeO (OH)) nucleated on K-feldspar grains, on illite coatings onsand grains, and on pore-lling illite, but not on clean quartz grains. Model results show that regions of

detrital K-feldspar in the sandstone that consume H+ more rapidly than diffusion to the reaction site

determine concretion size, and spacing is related to diffusion and advection rates of supply of reactants Fe 2+,

O2, and H+.

2010 Elsevier B.V. All rights reserved.

1. Introduction

Diagenetic redistribution and recrystallization of iron minerals in

sandstone aquifers and hydrocarbon reservoirs are records of

paleouid ow paths, timing, and uid composition (Beitler et al.,

2005; Eichhubl et al., 2004). Redistribution and recrystallization leads

to prominent color changes and mobilized iron was precipitated as

concretions. Theoretical solubility and stability calculation together

with the chemical and mineralogical composition of concretions

constrains solution characteristics, precipitation mechanisms, and the

size of spherical concretions. Theoretical calculations of growth times

can be used to estimate duration of uid ow events. Study of

concretions has implications for developing a genetic model of

formation, diagenesis of iron minerals, and hydrocarbon migration.

The objectives of this study of concretions are to test three

hypotheses. First, the iron oxide concretions initiate or nucleate on a

single seed. This hypothesis will be tested by petrographic analysis to

identify nucleation sites and a conceptual model that accounts for

rinds of iron oxide cemented sandstone that surrounds uncemented

sandstone. Second, the concretions form quickly and record brief

diagenetic events. This hypothesis will be tested by numerical

calculations of concretion growth rates using estimates of tempera-

ture, diffusion coefcient, hydraulic conductivity, hydraulic gradient,

and solution composition. Third, concretions grow by diffusion, and

reactant transport limits the growth rate, not the rate of precipitation

or recrystallization of iron oxide. Calculation of mass balance of iron in

the sandstone, examination of the iron distribution in host sandstone,

and comparison of transport limited growth with precipitation and

recrystallization rates of iron oxide will test this hypothesis.

Testing these hypotheses requires selection of sandstone strata

that host concretions, and determination of the chemical, mineralog-

ical, and petrographic characteristics of concretions and surrounding

Sedimentary Geology 233 (2011) 5368

Tel.: +1 801 277 4124.

E-mail address:wparry@comcast.net.

0037-0738/$ see front matter 2010 Elsevier B.V. All rights reserved.

doi:10.1016/j.sedgeo.2010.10.009

Contents lists available at ScienceDirect

Sedimentary Geology

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s e d g e o

http://dx.doi.org/10.1016/j.sedgeo.2010.10.009http://dx.doi.org/10.1016/j.sedgeo.2010.10.009http://dx.doi.org/10.1016/j.sedgeo.2010.10.009mailto:wparry@comcast.nethttp://dx.doi.org/10.1016/j.sedgeo.2010.10.009http://www.sciencedirect.com/science/journal/00370738http://www.sciencedirect.com/science/journal/00370738http://dx.doi.org/10.1016/j.sedgeo.2010.10.009mailto:wparry@comcast.nethttp://dx.doi.org/10.1016/j.sedgeo.2010.10.009

7/25/2019 Parry 2011 concreciones de hierro

2/16

host rock. Datable materials are notably absent so formation times

must be deduced from numerical modeling. Theoretical solubility and

stability calculations identify precipitation mechanisms.

1.1. The Navajo Sandstone

The Jurassic Navajo Sandstone abundantly displays the effects of

redistribution and recrystallization of iron minerals and is host for a

complex variety of iron and manganese precipitates including manydiverse iron oxide concretions. The Navajo Sandstone is widely

distributed and well exposed on the Colorado Plateau of Utah (Fig. 1)

and is an ideal stratum for a study of concretions. Methods, calculation,

interpretations, and conclusions developed for the Navajo Sandstone

are applicable to iron oxide concretions in other sandstone strata.

The Navajo Sandstone and stratigraphically equivalent Nugget and

Aztec Sandstone, the largest eolian dune deposit in North America,

cover an area greater than 3.5105 km2 (Blakey, 1994; Blakey et al.,

1988). The Navajo, an important aquifer and hydrocarbon reservoir,

has produced 288 millionbarrels of oil and 5.1 trillion cubic feet of gas

(Chidsey and Morgan, 2005). The Navajo Sandstone is a well-sorted,

ne-grained quartz arenite. Color varies from moderate reddish

brown and moderate orange pink altered to white or pale orange with

later superimposed diffuse and concretionary iron oxides and

carbonate (Beitler et al., 2005).

Regional studies of Navajo Sandstones and stratigraphically

equivalent Nugget Sandstone describe early diagenetic hematite,

clay, and calcite and dolomite cements. Later burial diagenesis

includes compaction, quartz and feldspar overgrowths, and pore-

lling illite and kaolinite. (Beitler et al., 2005; Bergosh et al., 1982;

Bowen, 2005; Chan et al., 2000; Jordan, 1965; Lindquist, 1983, 1988;

Net, 2003; Parry et al., 2007, 2009; Tillman, 1989).

The Navajo Sandstone, normally red from early diagenesis

(Walker, 1967, 1975, 1979), exhibits widespread chemical reduction

fronts formed by migrating reducing aqueous uids (Beitler et al.,

2003, 2005; Bowen et al., 2007; Chan et al., 2000; Garden et al., 2001).

Chemical reduction produces a color change from moderate reddish

brown and moderate orange pink to white or very pale orange along

uid ow pathways (Beitler et al., 2005; Eichhubl et al., 2004). The

color is a function of iron oxide abundance, grain size, andmineralogy.

Iron mobilized by chemical reduction is precipitated at the oxidation-reduction front as iron mineral cement in the sandstone. Cemented

sandstone occurs as spheroidal forms from mm to cm scale, rinds,

buttons, disks, and large cylindrical pipes or columns some up to 10 m

high. Strata bound layers of iron oxide cemented sandstone and

conglomerate extend for10s of meters(Chan et al., 2000, 2004, 2005).

Smaller concretions are more closely spaced than larger concretions

(Chan et al., 2004). Iron oxide also lines northeast striking, vertical

joints. Rind concretions up to 12 cm in diameter typically exhibit a

spheroidal rim b1 mm up to 1 cm thick of iron oxide lling the

sandstone porosity surrounding interior uncemented sandstone. Solid

concretions cemented from center to rim are seldom larger than

1.5 cm in diameter. Numerical simulations of advection combined

with diffusion produces periodic self-organized nucleation centers

through Liesegang-type double-diffusion of iron from interaction of

reduced formation water and oxygen from shallow, fresh waterChan

et al. (2007). Unique spherical concretions in the Navajo Sandstone

are similar in appearance to hematite concretions at Meridiani

Planum on Mars (Chan et al., 2004; Orm et al., 2004).Sefton-Nash

and Catling (2008) give a detailed analysis of the time scale for

formation of the Martian concretions.

In laboratory bench tests, precipitation of hydrated iron hydro-

xides forms rinds aroundan initial sphericalsource of iron (Chan et al.,

2007). Chemical gradients between the inside and outside of

precipitated spheres cause diffusion of Fe towards the outer perimeter

of the sphere forming a rind. The rind then grows inward due to

diffusion within the sphere.

Isotopic composition of Fe in the concretions is consistent with

chemical reduction of early diagenetic Fe3+mediated by bacteria and

later precipitated by complete oxidation of aqueous Fe2+ (Busignyand Dauphas, 2007; Chan et al., 2006). However organic compounds

could not be detected by gas chromatographymass spectrometry

analysis (Souza-Egipsy et al., 2006).

2. Stratigraphy

The Navajo Sandstone is overlain by 1.8 to 4.2 km of marine,

uvial, eolian, and lacustrine sediments (Fig. 2) (Doelling et al., 2000;

Hintze and Kowallis, 2009). The CarmelFormation, a sabkhasequence

of sandstone, siltstone, mudstone, limestone, anhydrite, and gypsum,

above the Navajo Sandstone forms a seal. Theuvial Chinle Formation

and the uvial to marine Moenkopi Formation that lie beneath the

Wingate Formation are ne-grained aquitards. The aquitards above

and below the Navajo Sandstone inhibit uid access except alongfaults or surface exposures.

3. Structure

The study area is located in the Colorado Plateau province, a

roughly circular area of about 384,000 km2 with many high plateaus

and isolated mountains. Elevations range from 900 to 4200 m and

average 1600 m. Monoclinal folds, anticlines, domes, basins and faults

deform the generally at lying Navajo Sandstone (Hunt, 1956). The

structural framework of the Colorado Plateau and the study area is

dominated by Laramide deformation (Dickinson and Snyder, 1978;

Dickinson et al., 1988) that occurred from uppermost Cretaceous to

Middle Eocene and later Basin and Range normal faults (Davis, 1999;

Hunt, 1956). Laramide compression formed the Kaibab, Circle Cliffs,

Navajo Sandstone Outcrop

Utah

S

C

CVB

M

KHF SF

P

L

U

AH

114 113 112 111 110 109

42

41

40

39

38

37

GSForm

erS

evie

rHighla

nd

s

200 Miles

Fig. 1. Mapof Utah's portion of theColoradoPlateaushowing areas of Navajo Sandstone

outcrop, major normal faults, and structural uplifts. A=Abajo Mountains, B=Boulder

Mountain, C=Circle Cliffs Uplift, CV=Caineville, H=Henry Mountains, HF=Hurri-

cane fault K= Kaibab Uplift, L =LaSal Mountains M =Monument Uplift, P =Paunsau-

gunt fault; S =San Rafael Uplift; SF= Sevier Fault, GS= Sample location in the Grand

Staircase Escalante National Monument, U=Uncompahgre Uplift.

54 W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

3/16

Escalante, Monument, and San Rafael monoclines (Davis, 1999; Hunt,

1956). The Escalante monocline, with a steeply dipping west limb and

gentle east limb, lies to the west of the study area and the Circle Cliffs

Uplift, with a steeply dipping east limb, lies to the east ( Fig. 1). The

EscalanteRiver and itstributarieshave cutdeep canyons into the west

limb of the Circle Cliffs Uplift. The study area is located on the gentle

west limb of the Circle Cliffs Uplift (Fig. 1).

Basin and Range extensional faulting began about 15 Ma and

continuestoday (Davis, 1999). Three major Basin andRangefaults are,

from west to east, the Hurricane fault, the Sevier fault and the

Paunsaugunt fault (Fig. 1). The Paunsaugunt fault is the easternmost

of the Basin and Range faults and is nearest the study area, 55 km to

the east (Fig. 1). Boulder Mountain, a portion of the Aquarius Plateau

protected from erosion by a volcanic cap lies north of the study area.

Uplift of theColorado Plateauregionbeganin themiddleEocene and

lasted until the Late Miocene elevating and tilting the area and

subjecting it to erosion, which continues today (Hunt, 1956). Mean

uplift of the Colorado Plateau near the study area since the Late

Cretaceous is estimated at near 2000 m (Pederson et al.,2002; Sahagian

et al., 2002). The present physiography of mesas, plateaus and deeply

incised canyons is the result of nearly continuous erosion throughout

Cenozoic time that has removed thousands of meters of sedimentary

rock and developed the present groundwater ow system.

4. Chemical, mineralogical, and petrographic methods

The location of the study is an area of abundant concretions that

reach 5 cm in diameter located in the Spencer Flat area of the Grand

Tr

iass

ic

Jurass

ic

Age Thickness (m)

Claron Fm

Pine Hollow, Grand CastleCanaan Peak Fms

Kaiparowits Fm

Wahweap Fm

Tropic Shale

Dakota and Cedar Mtn Fms

Morrison Fm

Entrada Ss.

Carmel-Page Fms

Navajo Sandstone

Kayenta Fm

Wingate Sandstone

Chinle Fm

Moenkopi Fm

Kaibab Limestone

yraitreT

suoecater

C

Perm

ian

427

0-396

610-914

305-457

274-549

152-229

1-313

0-290

0-305

55-317

396-457

46-107

30-107

130-283

134-351

Eolian sandstone

Eolian sandstone

Eolian sandstone

Eolian sandstone

Fluvial, eolian, and coastal plain sandstone,siltstone, mudstone, limestone, gypsum

Fluvial conglomerate, sandstone,claystone, mudstone, bentonite

Fluvial conglomerate, sandstone, siltstone,mudstone. Lacustrine limestone

Lacustrine limestone and dolomite

White Rim Sandstone

Marine mudstone and shale

Fluvial and lacustrine mudstone, siltstone,sandstone, and conglomerate

Lacustrine sandstone and siltstone

Fluvial conglomerate, sandstone, andmudstone

Marine dolomite

Fluvial sandstone, siltstone

Marine limestone, sandstonesiltstone, gypsum, and eolian sandstone

Fig. 2.Stratigraphy of the Grand Staircase-Escalante National Monument study area. Strata and thicknesses fromDoelling et al. (2000).

55W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

4/16

Staircase Escalante National Monument (Fig. 1). The sample site is in

the Jurassic Navajo Sandstone on the gently dipping west limb of the

Circle Cliffs Uplift. The Navajo Sandstone in the study area contains

numerous iron oxide concretions some of which weather out and

form an extensive lag deposit (Fig. 3A). The concretions consist of a

rind of iron oxide cemented sandstone surrounding interior, unce-

mented sandstone (Fig. 3B). Two large samples approximately

502020 cm each that included a rind concretion and surrounding

host rock (Fig. 3C, E) were subdivided in the laboratory to examinecharacteristics of the concretion interior, concretion rind, and host

rock. A series of sub-samples were collected systematically 2 to 4 cm

apart from the concretion center to 28 cm from the center in sample

GS1a (Fig. 3D) and concretion center to 13 cm from the center in

sample GS1b (Fig. 3C).

Mineralogy and textures were determined by petrographic

examination of thin sections, petrographic point counting, and X-ray

diffraction. Thin sections were stained for potassium feldspar ( Bailey

and Stevens, 1960). Thin sections were examined with a standard

petrographic microscope. Modal mineralogy was determined by a

combination of petrographic point counting (1000 points per thin

section, 300m step) and least squares t of mineralogy, mineral

chemistry, and whole-rock chemistry using computer software (Parry

et al., 1980). The computer program incorporates whole-rock

chemistry and petrographic point counting to produce mineral

abundance estimates that are consistent with both petrography and

chemistry.

Whole-rock samples were analyzed for major elements by

inductively coupled plasma atomic emission spectroscopy. Analytical

accuracy and laboratory procedures conform to international stan-

dards ISO 9001:2000 and ISO 17025:2005. Analyses are accurate to

the signicantgures reported (Table 1).

Clay minerals were extracted from core samples by mild grinding

by hand and 2 micron particle size was separated by centrifugation.

Oriented smears on glass slides were X-rayed with a Rigaku 2000 V X-

ray spectrometer using a copper anode X-ray tube. Scanning speed

was 2 per minute. X-ray scans were done following air-drying and

vapor glycolation at 60 C for 12 h.

5. Chemical, mineralogical, and petrographic results

5.1. Chemical composition

Chemical analyses of sampling points (Fig. 3E) reveal that the

interior sample contains slightly less Al2O3and K2O than the exterior

samples (Table 1,Fig. 4). Samples from GS1a contain 0.03 to 0.08% C,

but no carbonate minerals were detected in petrographic or X-ray

diffraction examination (Table 1). Only the interior uncemented

sandstone and one exterior host sandstone sample contained

detectable S. Phosphorous was detected in all exterior samples at

concentrations of 0.01 to 0.07% (Table 1). Fe2O3 varies from 1.47

inside the concretion to 28.3 wt.% in the concretion rind to 1.70 wt.%

outside the concretion (Table 1, Fig. 4). No iron-depleted zone is

present outside theconcretion within28 cm of theconcretion or more

than half the distance to the next concretion.

5.2. Petrography and mineralogy

Sample GS1a averages 89% rounded quartz grains and 5.7% K-

feldspar grains (Table 1). Quartz grainsreach300500 m in diameter

and in some cases quartz overgrowths are present. The rind

concretions consist of a thin rind of goethite-cemented sandstone

that is 19.6 vol.% goethite (Table 1) surrounding an uncemented

Fig. 3. Photographsof thesample locality showingthe location of detailed sampling.A. Lag deposit of ~5 cm diameter concretions thathave weathered fromthe Navajo Sandstone. B.

Examples of concretion cross-sections of various sizes. C. Sample GS1b at the location of collection. D. Sample GS1a at the collection location. Detailed sampling points are shown. E.

Sample GS1a in the laboratory following collection of detailed samples from points shown.

56 W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

5/16

center. The rind porosity is almost completely lled with precipitated

goethite (Fig. 6A, B). The higher magnication photo ofFig. 6B shows

that little porosity remains open. Samples average 1.4% FeO (OH).

Goethite in the concretion rim occurs in three textures: a

microgranular texture with crystal size from less than 1 m up to

5 m that forms a coating on quartz grains; an equigranular texture

with grain size up to 15 m occurs in the centers of formerpore space;

and a radial-brous texture with sweeping extinction 20 to 40 m in

size that forms on quartz and feldspar grains. Porosity of the goethite

rim is estimated to be 1.8%.

Clay minerals consist of illite and kaolinite (Fig. 5A, B). Illite in thin

section occursas a porelling andas rims on quartz grains. Illite rims are

absent from the quartz grains in the rind or masked by semi opaque

goethite. Illite rims on quartz grains are present adjacent to kaolinite

lled porosity. Scattered red hematite grains less than 0.01 m arepresent within illite coating. Claymineral X-ray diffraction traces reveal

two ordered interstratications of illite with smectite. The rst (RN3, 3

or more illite layers together before a smectite layer is encountered)

contains 710% interstratied smectite and is present in both GS1a and

GS1b. The second (R1) contains 70% illite and 30% smectite and is

present in theexterior of theGS1b concretionsalong with theRN3 illite

(Fig. 5A, B). The concretion rind contains slightly more kaolinite than

interior and exterior sandstones and no illite (Table 1,Fig. 4). Samples

average 2.1% illite and 1.4% kaolinite (Table 1). Some K-feldspar grains

are altered to kaolinite and illite.

Modal mineralogy of sequential samples is remarkably uniform.

However, the sample from the concretion interior contains less

kaolinite and slightly less illite than the exterior samples and more K-

feldspar (Table 1,Fig. 4). Exterior host sandstone samples commonly

show a light tan iron oxide coating on sand grains. X-ray diffraction

analysis of these samples identied lepidocrocite (-FeO (OH))

(Fig. 5A, B). Illite grain coatings are present inside the tan iron oxide

coating. Also the concretion interior sandstone sample is free of

lepidocrocite that is common in exterior samples.

Iron oxide precipitate clearly nucleates on K-feldspar in the host

sandstone (Fig. 6E, F). In some cases the iron oxide precipitate

completely surrounds K-feldspar grains (Fig. 6F).

6. Numerical growth time models

Chemicaland mineralogical study did not detect dateablematerials in

the concretions. The rate of growth of the spherical concretions can be

calculated by numerical growth models that constrain the minimum

duration of diagenetic uid events involved in formation. Goethitelling

the porosity in the concretions forms from solution by a combination of

diffusion and advective transport of reactants through the porous

sandstone, concurrent precipitation reactions, and recrystallization of

the precipitate to goethite. Growth time calculations require estimates of

the temperature of transport and precipitation, identity, stability,

solubility, and precipitation mechanism of the iron oxide forming the

concretionary cemented sandstone, hydrologic conductivity of thesandstone, hydraulic gradient driving advection, advective velocity, and

concentration of iron in solution in the groundwater.

6.1. Temperature

Diffusion and chemical reactions are temperature dependent

processes. Few constraints on temperature of formation of the

concretions are available. Goethite in 10's of micrometer particles is

stable relative to hematite up to 60 C (Navrotsky et al., 2008). Burial

histories for the region suggest depth of burial of the Navajo

Sandstone did not exceed 1.8 to 4.2 km (Fig. 2) limiting the maximum

temperature to 60to 100 C (Gardenet al., 2001; Huntoon et al., 1999;

Nuccio and Condon, 2000). Hematite is the main iron oxide

precipitate in the Navajo Sandstone in the Covenant oil eld where

0

0.5

1

1.5

2

2.5

3

3.5

0

K2O

Al2O3

Fe2O3

Kaolinite

Distance from Concretion Center (cm)

We

ightor

Vo

lume

%

28.3 %

rin

d

30252015105

Fig. 4. Plot of chemical constituents and minerals as a function of distance from a

concretion center.

Table 1

Chemical composition and modal mineralogy of GS1a samples.

Sample 0 rind* 1 2 3 4 5 6 7 8 9 10

SiO2 94.7 63.5 94.6 94.2 94.0 93.8 92.8 94.4 94.3 94.5 94.2 94.0

Al2O3 1.96 1.93 2.11 2.14 2.32 2.12 2.11 2.14 2.20 2.25 2.15 2.33

Fe2O3 1.47 28.3 1.23 1.55 1.51 1.46 1.04 1.44 1.17 1.30 1.40 1.70

CaO 0.07 0.06 0.04 0.05 0.05 0.04 0.09 0.03 0.03 0.03 0.03 0.03

MgO 0.07 0.08 0.07 0.07 0.07 0.07 0.06 0.07 0.07 0.08 0.07 0.08

Na2O 0.12 0.16 0.07 0.06 0.07 0.05 0.06 0.05 0.05 0.05 0.07 0.05

K2O 1.14 0.83 1.20 1.22 1.32 1.22 1.20 1.22 1.24 1.26 1.22 1.30C 0.05 0.03 0.04 0.03 0.03 0.08 0.08 0.03 0.03 0.03 0.05

S 0.02 0.01 0.01

P2O5 b0.01 0.02 0.02 0.01 0.01 0.04 0.01 0.02 0.01 0.07 0.02

Modal Mineralogy (Volume %)

Quartz grns 75.5 69.4 74.5 71.7 76.0 75.5 74.9 74.9 73.3 73.5 74.1 71.5

Overgrowth 0.1 0.3 0.1 0.5 0.5 0.5 0.5 0 0.9 0.5 0.3

Kfeldspar 5.1 6.6 4.6 5.3 6.2 5.2 5.4 5.2 5.3 6.1 5.7 6.4

Goethite 0.7 19.1 0.6 0.7 0.7 0.7 0.5 0.7 0.6 0.6 0.7 0.8

Illite rims 1.7 2.9 1.7 1.7 1.4 1.4 1.6 2.0 1.6 1.0 1.3

Pore illite 0.4 0.03 0.6 0.4 0.6 0.3 0.6 0.3 0.3 0.5 0.4

Kaolinite 0.1 3.2 0.4 0.5 0.9 1.2 1.4 1.0 1.0 1.2 1.2 0.8

Porosity 16.4 1.8 16.8 19.3 14.1 14.9 15.6 15.6 17.6 15.9 16.4 18.5

*rind thickness of 0.7 mm prevented large enough sample for chemical analysis. This analysis is a thicker rind from Bowen (2005).

57W.T. Parry / Sedimentary Geology 233 (2011) 5368

http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B4%80

7/25/2019 Parry 2011 concreciones de hierro

6/16

the temperature is 87 C (Parry et al., 2009). Mobilization of iron is

associated with hydrocarbon migration that took place in Late

Cretaceous to Early Tertiary and precipitation of iron oxide concre-

tions must followiron reduction andmobilization. Temperaturein the

Navajo Sandstone cooled to 50 C, a temperature at which goethite

could precipitate, during exhumation in the late Tertiary (Huntoon

et al., 1999). The stability diagram and diffusion models have been

calculated using a temperature of 50 C.

5 30

GS1a0

GS1a1

GS1a2

GS1a3

GS1a4

GS1a5

250 Counts

Intens

ity

2 (CuK)

GS1a6

GS1a7

GS1a8

GS1a9

GS1a10

Illite

95/Smec

tite

Illite

95/Smec

tite

Kao

linite

Kao

linite

Lep

idocroc

ite

K-fe

ldspar

Quartz

Illite

95/Smec

tite

0

250

500

750

1000

1250

Intens

ity

(Coun

ts)

Illite

95/Smec

tite

Illite

70/Smec

tite

Illite

70/Smec

tite

Illite

95/Smec

tite

Kao

linite

Kao

linite

Lep

idocrocite

K-fe

ldspar

Quartz

Illite9

5/Smec

tite

Quartz

GS1b0

GS1b1

GS1b2

GS1b3

GS1b4

GS1b5

GS1b6

25201510

5 30

2 (CuK)25201510

A

B

Fig. 5.X-ray diffraction patterns of 2clay extracts. A. X-ray diffraction patterns of -2 extracts of sample GS1a. B. X-ray diffraction patterns of 2extracts of sample GS1b.

58 W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

7/16

6.2. Solubility and precipitation mechanisms

A genetic model of iron oxide concretion formation should include

stability, solubility, and precipitation mechanism. Goethite is stable

with respect to magnetite at higher oxidation and lower pH (Fig. 7).

Pyrite is stable at lower oxidation where sulfur is reduced as well as

iron. Goethite is in equilibrium with aqueous Fe2+

species at higher

pH values as log fO2 decreases. Lepidocrocite is metastable with

respect to goethite (dashed line onFig. 7).

Solubilities of goethite and lepidocrocite were calculated using a

moderately saline water composition (ionic strength 0.18) from the

Navajo Sandstoneaquifer in southeasternUtah (Spangler et al., 1996).

Activity coefcients for aqueous species were calculated with

Geochemists Workbench (Bethke, 1998). Stability constants, free

Fig. 6.Photomicrographs of thin sections. A. Goethite cemented rim on sample GS1a. B. Goethite cement in concretion rim. C. Goethite nucleating on K-feldspar (yellow) in sample

GS1a-1 outside the rim. D. Goethite nucleating on K-feldspar in sample GS1a-4 outside the rim. E. Goethite nucleating on K-feldspar in sample GS1a-4 outside the rim. F. Goethite

protective layer surrounding K-feldspar in sample GS1a-1 outside the rim.

59W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

8/16

energies, activity coefcients and equilibrium constants used in the

solubility calculations are given inTable 2.

Goethite has a solubility minimum that increases in pH from 7 to 9

as log fO2 decreases from 20 to 50 (Fig. 8). The solubility of

lepidocrocite is more than 10 times greater than goethite at all pH

values (Fig. 8). The solubility minimum for lepidocrocite at log fO2=

20 is at pH=7 to 7.5. No change in solubility above log fO2= 20

is evident in the calculations as the solution is fully oxidized in the pH

range from 4 to 12. Precipitation of goethite can be initiated by an

increase in pH or an increase in fO2or a combination (Figs.7 and 8).

Iron species in solution calculated as a function of pH at log fO2=

20 illustrate Fe++ species areimportant at lowpH andlowervalues

of log fO2(Fig. 9). At pH =4 the dominant species are Fe++ (52%), Fe

(OH)++ (37%), then Fe (OH)2+ (10%), followed by Fe (OH) 3

o (1%), and

at pH above about 7.5 Fe (OH) 4- becomes the dominant species.

Aqueous complexes of Fe with Cl, SO4 and HCO3 have negligible

concentrations except at low pH values.

6.3. Hydraulic conductivity and hydraulic gradient

The product of hydraulic gradient and hydraulic conductivity

yieldsthe advectivewatervelocity neededin thegrowth time models.Paleo groundwater and hydrocarbon ow is assumed to be in the

direction of the paleosurface (Sanford, 1995). Topographically driven

groundwater ow affecting the study area in Late Cretaceous was

northeast toward the study area from the Sevier highlands approx-

imately 300 kmwestof the study area(Sanford, 1995) (Fig. 1). During

this time, the Navajo Sandstone was near sea level and the Sevier

2 3 4 5 6 7 8 9 10 11 1280

70

60

50

40

30

20

10

0

pH

log

fO2(

g)

Fe++

FeSO4+

Goethite

Magnetite

Pyrite

Pyrrhotite

50C

Metastabl

eLepidocrocite

Fig. 7. Log fO2-pHdiagram showing thestabilityof goethiterelativeto pyrite, magnetite,

siderite, and solution. A dashed line shows metastable lepidocrocite. The diagram wasconstructed using Geochemists Workbench (Bethke, 1998). The diagram was

constructed using log aFe= 6 (about 0.05 mg/l), log aSO4= 3 (about 100 mg/l).

Table 2

Parameters used in model calculations.

Parameter Reaction or species Value (50 C) Source

Stability Constants

Log1 Fe3++H2O=Fe(OH)

2++H+ 1.57 1

Log2 Fe3++2H2O=Fe(OH)2

++2H+ 5.73 1

Log3 Fe3++H2O=Fe(OH)3+3H

+ 12.95 1

Log4 Fe3++H2O=Fe(OH)4

-+4H+ 20.26 1

Activity Coefcients

Fe3+ 0.1268 2

Fe2+ 0.3344 2

Fe(OH)2+ 0.3006 2

Fe(OH)3 1.000 2

Fe(OH)2+

, Fe(OH)4-

, Fe(OH)3-

, 0.7308 2 Fe(OH)+ 0.7334 2

Fe(OH)2 1.000 2

Gibbs Free Energy

Gof Fe2+

21,231 cal 3

Gof Fe3+

2,431 3

Gof O2(g) 1,233 3

Gof H+ 0 3

Gof FeO(OH) Goethite 117,384 4

Gof Fe)(OH) Lepidocrocite 115,098 4

Log Equilibrium Constants

Goet hite FeO (OH)+2H+=Fe2++1/4O2+3/2H2O 6.873 5

Lepidocrocite FeO(OH)+ 2H+=Fe2++1/4O2+3/2H2O 5.327 5

Log Keq Fe3++1/2H2O=Fe

2++1/4O2+H+

6.395 5

1. Liu and Millero (1999); 2. Geochemist's Workbench; 3. SUPCRT92 Johnson et al.

(1992); 4. Majzlan et al. (2003a,b); 5 Calculated.

0.001

0.01

0.1

1

10

100

1000

10000

100000

1000000

4

pH

So

lubility

(nanograms

Fe

/kg

)Goethite

logfO2=

-50

logfO2=

-40

logfO2=-20

log fO2=-20

Lepidocrocite

121086

Fig. 8. Calculated solubility of goethite and lepidocrocite. Both minerals have a

solubility minimum that shifts with shifting fO2. Parameters used in the calculations are

given inTable 2.

0

10

20

30

40

50

60

70

80

90

100

4

pH

Proport

iono

fSpec

ies

Fe(OH)4-

Fe(OH)2+

Fe(OH)3

Fe++

Fe(OH)++

121086

Fig. 9. Fespeciationas a function of pHat logfO2= 20. At elevated pH thecomplex Fe

(OH)4- increases solubilityfrom theminimumvalue. Solubilityincreases substantiallyas

pH is lowered below the solubility minimum.

60 W.T. Parry / Sedimentary Geology 233 (2011) 5368

http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B7%80

7/25/2019 Parry 2011 concreciones de hierro

9/16

highlands were as high as 3.2 km (DeCelles and Coogan, 2006)

resulting in a topographic gradient of 0.01. From Middle Eocene to

Late Miocene, groundwater ow was westward from the crest of the

Circle Cliffs Uplift. From late Miocene to the present, ow was

generally southwestward from the Boulder Mountain portion of the

Aquarius Plateau (Sanford, 1995).

The hydraulic gradient from the Circle Cliffs Uplift can be

estimated by restoring the strata above the exposed Permian bed in

the core of the uplift. The present crest of the uplift is 197 m above thestudy area. Restoring the strata from Permian to the Jurassic Navajo

Sandstone (Fig. 2) indicates that the Navajo Sandstone in the restored

crest of the uplift would have been 1044 m above and 34 km

northeast of the sample location for a hydraulic gradient of 0.03.

The Boulder Mountain portion of the Aquarius Plateau lies 42.5 km

north and 1543 m above the sample location. Abundant lakes and

springs from 3358 m elevation down to 1879 m suggest the present

hydraulic gradient is near the topographic gradient of 0.036.

Numerous studies of the important Navajo Sandstone regional

aquifer include measurements of hydraulic conductivity. The mea-

surements nearest the study area at Caineville (Fig. 1) show that the

Navajo Sandstone is hydraulically isotropic with an average hydraulic

conductivity of 0.15 m/d (54.8 m/yr) (Hood and Danielson, 1979). The

product of hydraulic conductivity of 54.8 m/yr and the hydraulic

gradient of 0.036 yields the ow velocity of 2 m/yr.

6.4. Groundwater composition

The iron concentration in groundwater must be estimated to

calculate growth times. Iron concentrations ranges from 1.8 to 5 mg/

l in the region 100 km northeast of the study area (Hood and

Danielson, 1981) and 1.5 to 19 mg/l in the Paradox Basin area 200 km

eastof the study area(Spangler et al., 1996). Waterproducedfrom the

Covenant oil eld Navajo Sandstone reservoir contains 14 mg/

l dissolved Fe (Chidsey et al., 2007). The following calculations use a

value of 5 mg/l, the maximum observed nearest the study area by

Hood and Danielson (1981).

6.5. Transport controlled concretion growth time

Numerical growthtime modelscan be used to deducethe lengthof

time required for formation of concretions and the duration of the

uid ow events that form them. Concretions to be modeled are solid

concretions up to 1.5 cm in diameter and rind concretions with a thin

rind of precipitated matter surrounding a spheroidal core containing

littleof theprecipitate. Mass of Fe used in calculationsof growthtimes

for solid concretions assume that concretion mineral matter cements

the sandstone from center to edge, but no assumption of a nucleus is

made. To calculate the time for growth of the rind, the timefor growth

of a spherical concretion of the radius of the interior is rst calculated,

and then the time for growth of the radius of the exterior of the rind is

calculated. The difference between the two growth times is the time

required to grow the rind. The growth time for concretions wascalculated using a temperature of 50 C, 5 mg/l dissolved Fe, diffusion

only as a minimum no ow condition, and a ow velocity of 2 m/yr

using the following 5 equations.

Growth time models described below include formation of solid

concretions and rind concentrations formed by diffusion and by a

combination of diffusion and advection of reactants. Thegrowth times

not only indicate the length of time required for growth of the

concretions, but also place limits on the minimum duration of the

uid ow events that led to the formation of the concretionary iron

oxide cements.

Eq. (1) was developed for Fe supplied to the concretion by

diffusion only (Berner, 1968, Lasaga, 1998) assuming the volume

fraction of precipitated iron oxide in the concretion, Fp, equals the

porosity. Terms and values used in the calculations are dened in

Table 3. Derivation of the equation assumes a linear concentration

gradient of less than 0.05 mole/l over some large distance.

t= R

2

2vD ccR

1

Dissolving original red hematite grain coating near the concretion

could have formed a concretion and an iron-depleted zone. The

equation for this is (Berner, 1980):

t=

R2 Fp

Fd1

2vD cLcR 2

As the volume fraction of dissolving material, Fd, becomes smaller

the time necessary for concretion growth becomes larger. For the

values of Fp=0.15 and Fd=0.002 in the Navajo Sandstone host rock,

the concretion growth times estimated from equation 2 are greater

than equation 1 by a factor of 74. However, chemical and

mineralogical analysis of host rock near a concretion shows no

depleted zone (Table 1,Fig. 4).

Eq. (3) accounts for the supply of iron by both diffusion and

owing groundwater (Berner, 1968, 1980).

t= R D0:715U

1 + RUD 0:715

+ D0:715U

1:715Uv c

cR

3

Table 3

Variables in calculation of concretion growth times in Eqs. (1)(5)(note all variables

converted to meters, years, moles).

R=radius of concretion

D= diffusion coefcient. Here we follow Berner (1980) and Drever (1997) to

correct for tortuosity. The effect of tortuosity on the diffusion coefcient is

approximated by the effect on electrical conductivity which is measured by

Archie's Law and the formation factor F. The diffusion coefcient for a porous

sediment, Dsed=D solution/F, where F is the formation factor. The formation

factor F =2. Further, the effective diffusion coefcient is Dsed/ so

Deffective=Dsolution. Signicant Fe is present in solution only as Fe2+ species.

Precipitation is initiated by either an increase in fO2 or pH. The diffusion

coefcient for Fe2+ at 50 C in aqueous solution is 1.5105 cm2/sec (Oelkers

and Helgeson, 1988). Correcting this for tortuosity and converting to

appropriate units of m2/year produces Deffective=710-3m2/year.

Fc=Volume fraction of the host rock occupied by concretions. Concretions 5 cm in

diameter are spaced approximately 40 cm apart. Therefore, each concretion

occupies 65.45 cm3 in a volume of rock 4188.8 cm3 or a volume fraction of

0.0156.

Fd= Volume fraction of the host rock occupied by dissolving source material.

Average hematite concentration in original red Navajo Sandstone is 0.5

weight %. For a sandstone that is 89% quartz, 5.7% K-feldspar, 2.1% illite, 1.4%

kaolinite and 15% porosity, the volume fraction hematite source material is

0.002. The average sandstone in sample GS1a is 1.3% Fe2O3, however much of

this is introduced and is present as lepidocrocite.

Fp= Volume fraction of the concretion occupied by precipitated material.

Precipitated goethite occupies all of the porosity. Fp then=0.15 (see below

for porosity)=Volume fraction of the host rock occupied by pore space (porosity). Porosity of

undeformed Navajo Sandstone ranges from 10 to 30% ( Cooley et al., 1969;

Cordova, 1978;Hood and Patterson, 1984;Freethey and Cordy, 1991;Shipton

et al., 2002). Average porosity of dune facies Navajo Sandstone in the Covenant

oileld, central Utah is 12% (Chidsey et al., 2007). We have chosen a nominal

15% porosity for these calculations.

=Molar volume of precipitate (goethite)=2.0810 5 m3/mole

ccR=Concentration of Fe2+ far away from a concretion - concentration at the

concretion. Concentration of Fe2+ in oil eld reservoir produced water is

12 mg/l at the Covenant oil eld (Chidsey et al., 2007), and concentration

in the model calculations of Parry et al. (2004)where dissolved CH4 is

reacted with Navajo Sandstone reaches a maximum of 14 mg/kg. We

choose a conservative 5 mg/kg for these calculations =.0895 moles/m3.

cL=concentration at the outer edge of the depleted zone.

U=Pore uid velocity. The average hydraulic conductivity of 58.4 m/yr and

hydraulicgradient of0.03 wereused to calculate a darcian velocity of 2 m/yr as

discussed in the text.

61W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

10/16

The model for simple diffusion (Berner, 1968) was modied to

account fora steady state concentration gradient that is not linear. The

volume fraction of precipitated material within the concretion, Fp,

and the volume fraction of dissolving material, Fd, within the host

rock were also introduced (Wilkinson, 1991; Wilkinson and Dampier,

1990).

t=

r2 1Fd

Fp !

13

0@ 1A2vD cLcR

4

Pore-uidow and nonlinear steady state concentration gradient

were introduced in Eq.(5)(Wilkinson and Dampier, 1990).

t=R D0:715U

1 + RUD 0:715

+ D0:715U

h i 1 FdFp

13

1:715Uv ccR

5

The results of the calculations for each equation are shown in

Figs. 10 and 11. Eq.(1) and (4)produce nearly identical results for

diffusion-only growth, and Eq.(3) and (5)produce nearly identical

results for concretion growth by diffusion and advection. For example,

calculations suggest that a solid concretion with a radius of 25 mm

grows by diffusion in about 24,000 years (Fig. 10). A 25 mm radiusconcretion would form in 15,000 years by diffusion and advection.

Growth time is 3,600 years for a 2 mm thick rind with a radius of

25 mm formed by diffusion (Fig. 11A). The two models involvingboth

diffusion and uid ow predict the 2 mm thick rind formed in slightly

less than 2,000 years. Concretions 25 mm in radius form rinds 1 mm

thick in 1,000 years to 5,000 years for a 5 mm thick rind when

advection accompanies diffusion (Fig. 11B).

6.6. Oxidation rate of Fe2+

The calculated growth times assume transport controls the rate of

growth of the concretion. If the rate of growth is instead controlled by

chemical reaction, then these growth times are minimum values. The

chemical reactions that formed the concretions are oxidation rate ofFe2+ and the rate of conversion of precipitated ferric hydroxide to

goethite. Above pH= 4 oxidation rate is related to the overall reaction

(Langmuir, 1997)

Fe2 +

+:2502 + 2:5H2O= FeOH3 + 2H

The empirical rate law for the reaction is (Langmuir, 1997):

d Fe2 +

dt

= kFe

2 +

H 2 PO2

The rate of oxidation of ferrous ion is greater at high pH and PO2.

The rate constant, k+, is 1.21011 mol2/(bar day) at 20 C and theactivation energy is estimated to be 23 kcal/mol (Langmuir, 1997).

The rate constant at 50 C is therefore estimated to be 4.711010

mol2/(bar day). If pH and PO2 are specied the rate law becomes

pseudo rst order. With pH=6 the pseudo rst order rate constant,

k', is 94/day for PO2=0.2 and 0.00471 for PO2=1105 yielding

half-lives of .007 days and 147 days respectively. Bacterial mediation

of the oxidation reaction increases the rate by a factor of one million

(Langmuir, 1997).

Theprecipitate, Fe (OH)3, recrystallizes to hematite or goethite at a

rate that is dependent on pH, temperature, and solution composition.

At 50 C and pH=6.4, the half life for recrystallization is 8.3 days and

recrystallization is 99% complete in 55 days calculated with the rate

law given in Shaw et al. (2005) and therate constant ofFischer (1971)

reported inShaw et al. (2005).Concretion growth was transport controlled because of the rapid

oxidation of Fe2+with PO2above 1 105, the rapid conversion of Fe

(OH)3 to goethite, and the lack of a depleted zone surrounding

concretions.

7. Conceptual reaction-nucleation model

Concretion growth models previously proposed incorporate

diffusion and advective transport of reactants, but do not consider

chemical reaction with minerals in the host rock, nor do the growth

models explain the spacing between concretions or the presence of

the rinds. The purpose of this section is to explain Liesegang-Ostwald

self-organization and the role of pore solution reacting with K-

feldspar in concretion nucleation and growth.

0

5000

10000

15000

20000

25000

30000

35000

40000

0

Concretion Radius (mm)

Grow

thTime

(years

)

Equa

tion1

Equa

tion3an

d5

Equa

tion4

30252015105

Fig. 10. Growth time for FeO(OH) concretions calculated for diffusion and diffusion

coupled with pore water ow.

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

1 2 3 4 5

Rind Thickness (mm)

Go

wthTime

(years

)

Equation 1Equation 4

Equations 3 and 5

0

1000

2000

3000

4000

5000

5

Concretion Radius (mm)

Grow

thTime

(years

)

Rind Thickness (mm)

5

4

3

2

1

A

B

25201510

Fig. 11.A. Growth times for rinds of various thickness and 5 cm diameter calculated for

diffusion and diffusion coupled with pore water ow. B. Growth time for FeO(OH)

concretion rinds of various thicknesses and diameters calculated for diffusion coupled

with pore water ow.

62 W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

11/16

The reactants for precipitation of goethite, Fe2+ and O2, cannot be

carried within the same solution; two separate solutions are

necessary. Precipitation can be initiated by addition of oxygen or

consumption of H+ by reactions with K-feldspar. Iron concretions and

bands form by an Ostwald-Liesegang cycle (Ortoleva, 1994) and were

interpreted for the Navajo Sandstone by Chan et al. (2000, 2007). In

this cycle, reducing solutions owed upward along faults into the

Navajo Sandstone, reduced the iron, and dissolved the hematite. The

reducing water now charged with Fe

2+

owed through the NavajoSandstone, reacted with K-feldspar, and encountered oxygenated

groundwater (Fig. 12A). Upstream diffusion of O2 together with

downstream diffusion of Fe2+ and consumption of H+ by reaction

with K-feldspar elevated the reaction product until it reached a level

of supersaturation that promoted nucleation and precipitation of iron

oxide (Fig. 12B-D). The reaction product then fell to its equilibrium

value (Fig. 12E). In this model reactants Fe and O2were supplied by

diffusion, H+ was consumed by the K-feldspar reaction and the

concretion grew. The Fe concentration at the concretion precipitation

site was very low at goethite solubility (Fig. 8) so Fe diffused to the

concretion from all directions causing nearly spherical growth. When

the advection front of reducing, Fe2+-charged water proceeded

downstream, diffusion of O2 was no longer effective in maintaining

saturation, so a new supersaturation-nucleation level was reached

downstream and growth of a new concretion began (Fig. 12F-I).

Therefore, spacing and size of the concretions depends on reactant

Fe2+, O2, and H+ supply rates.

For K-feldspar reaction to raise the pH and cause precipitation of

goethite, H+must be consumed more rapidly than it can be supplied

by diffusion. The following conceptual model incorporates reaction

with host-rock K-feldspar and evaluates reaction rates compared with

H+ diffusion rates to the site of reaction. The rate at which H+ is

consumed for the volume of K-feldspar contained in the interior of

rind concretions of various sizes was calculated using the following

rate model (Bethke, 1998). Variables are shown inTable 4.

Rate= Ask

anH 1Q= K inmole = cm

2= s

The rate constant at an elevated temperature is calculated usingthe Arrhenius equation

kt =Ae

E=RT

Numerous eld studies have demonstrated that the natural

feldspar dissolution rates are 2 to 5 orders of magnitude slower

than laboratory rates under similar conditions of pH and temperature

(e. g. Blum and Stillings, 1995; Velbel, 1993; White and Brantley,

2003; Zhu, 2005). K-feldspar grains from the Navajo Sandstone at

some localities are coated with clay minerals that drastically retard

feldspar reaction (Zhu, 2005). However, the small rock volumes in a

concretion interior considered here and the lack of clay coatings on

the K-feldspar in host sandstone suggest reaction rates may not

approach the slowest rates observed in eld studies. Calculationspresented below are based on a reaction rate that is 3 orders of

magnitude slower than the laboratory rate.

K-feldspar reacts with H+ that diffuses from solution in surround-

ing porous rock. The concretions are regularly spaced with spacing

that varies with the concretion size (Chan et al., 2004). The rate at

which H+ would diffuse into the volume of reacting K-feldspar was

calculated from the diffusion distance of the half-spacing between the

concretions, and the difference in H+ concentration from ambient

solution pH and H+ at the solubility minimum. The H+ diffusionux

Fe2+

Fe2+

Fe2+

Fe2+

Fe2+

Fe2+

Fe2+

Fe2+

Fe2+

O2(aq)

O2(aq)

O2(aq)

O2(aq)

O2(aq)

O2(aq)

O2(aq)

O2(aq)

O2(aq)

Nucleation Threshold

Equilibrium Saturation1.0

0.5

0

1.5

1.0

0.5

0

1.5

1.0

0.5

0

1.5

1.0

0.5

0

1.5

1.0

0.5

0

1.5

1.0

0.5

0

1.5

1.0

0.5

0

1.5

1.0

0.5

0

1.5

1.0

0.5

0

1.5

A

B

C

D

E

F

G

H

I

Goe

thiteSa

tura

tion

Leve

l

Goethite saturation level

Moving front

Distance

Fig. 12. Conceptual model of the Ostwald-Liesegang cycle that is thought to govern

concretion spacing. Ironbearingwaterowingto theright andoxygenfrom theoxygen

rich water on the right diffuses upstream in A. The saturation index rises in B and C and

reaches the nucleation threshold in D causing precipitation of goethite and lowering Fe

content to very low values. Fe diffuses to the concretion and the concretion grows. The

saturation index falls to the equilibrium value in E. As the front moves downstream,

oxygen diffusion no longer reaches the concretion in E so that a new saturation index

rises in F and G, reaches the nucleation threshold in H and a new concretion forms.

63W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

12/16

is calculated using the equation for spherical diffusion developed by

Lasaga (1998):

Flux= Drbrs

cbcs

rbrs

Variables in this equation are inTable 4,and the main components

of the physical-chemical conceptual model are illustrated inFig. 13.

The results of these calculations (Fig. 14) illustrate that the rate of

diffusion of H+ to the concretion and the volume of sandstone

determines the rate of H+ consumption by reacting K-feldspar. The

pH at precipitation of ferric oxide is assumed to be at the solubility

minimum (Fig. 8). When the rate of consumption exceeds the rate of

supply, the pH will rise and cause precipitation of ferric oxide. For the

lowest H+ concentration difference (cb-cs=41013 mol/cm3)

considered, the rate of consumption of H+ by reacting K-feldspar at

pH=6 or 7 exceeds the rate of supply of H+ by diffusion at a

concretion radiusof 14 mm (Fig. 14). If K-feldspar reacts at pH=5 the

radius is 20 mm. For an intermediate H+ concentration difference of

121013 mol/cm3, the concretion size when K-feldspar consump-

tion of H+ exceeds supply is 21 mm radius for K-feldspar reacting at

pH 6 to 7 or 12 mm in radius if the reaction takes place at pH=5. For

the greatest H+ concentration difference of 231013 mol/cm3, a

volume of sandstone 29 mm in radius consumes H+ faster than

supplied by diffusion if the reaction takes place at pH=6 or 7 and

18 mm if the reaction takes place at pH=5. A single K-feldspar grain

can raise the pH and nucleate goethite only if the pH is very close to

the saturation value. For a solution pH level far from goethite

saturation, a larger mass of K-feldspar is needed to consume H+

faster than supplied by diffusion, so that pH will rise and causeprecipitation. With this model, concretion size is related to the pH of

solution reacting with K-feldspar and the pH gradient surrounding the

concretion. Concretion spacing is determined in the Liesegang-

Ostwald cycle by the upstream diffusion of O2and ow velocity.

8. Mass balance calculations

Mass balance estimates presented below evaluate the possibility

that the iron in concretions is derived from diffusion of reactants in

static pore uid, from redistribution of iron by diffusion in

immediately surrounding rock, or from pore uid ow. The spacing

between spherical concretions is related to the concretion diameter.

The average spacing of 5 cm diameter concretions is 40 cm (Chan et

al., 2004). The volume of a 40 cm diameter volume of porous rocksurrounding each concretion is 33,000 cm3. The porosity of the

sandstone is approximately 15% so the pore volume is 5,000 cm3.

Chemical modeling of reduction of hematite in the original sandstone

by dissolved methane (Parry et al., 2004) suggests the maximum

dissolved iron in the pore space is near 15 mg/l. Static pore uid

surrounding a concretion could contain a maximum of 0.075 g of Fe. A

5 cm diameterconcretion with a rind 2 mm thick has a rind volume of

14 cm3 that is 20% goethite and contains 8 g of Fe, but 0.075 g of Fe is

available in the surrounding pore uid. Fe must therefore be supplied

by pore uidow (100 pore volumes) or redistribution of Fe mineral

in the surrounding rock. Navajo Sandstone with hematite grain

coatings from early diagenesis originally contained on average 0.53

wt. % Fe2O3(range 0.18 to 1.25%, Beitler et al., 2005) or 350 g of Fe,

ample to supply the Fe in the concretion rind. If the Fe in host rock

Table 4

Variables used in calculation of the reaction model.

As=K-feldspar surface area; 374 cm2/g (57.9 in2/g) calculated using the equation of

Brantley and Mellott (2000)for mean grain size of 0.125 mm (0.0049 in).

k+=thelaboratory rateconstant at 25 C is 1.16810 14 mol/cm2/s. The reaction

model assumes a dissolution rate 3 orders of magnitude slower than the

laboratory rate.

aH+=activity of hydrogen ion.

Q=reaction quotient.

K=equilibrium constant.

n (exponent on H+)=0.5 (Blum and Stillings, 1995)

A= pre-exponential factor= 1.34710-5.

E=activation energy=51.7 kJ/mol for acid solutions ( Blum and Stillings, 1995).

R= gas constant.

T=absolute temperature.

D= diffusion coefcient.

=porosity.

rb=outer radius of H+ depleted zone (half spacing between concretions).

rs=radius of concretion.

cb=concentration of H+ at the outer radius of the depleted zone.

cs=concentration of H+ at the concretion surface.

Fig. 13. Conceptual reaction-nucleation model. A. regional geological cross section

showing chemically reducinguids gainingaccess to theNavajo Sandstone along a fault

conduit, bleaching the sandstone, and precipitating concretions at the chemical

reaction front. B. Diagram of an array of concretions showing dimensions used in the

reaction-nucleation calculations.

-10.0

-9.5

-9.0

-8.5

-8.0

-7.5

5

Concretion Radius (mm)Log

(H+

Fluxor

Reac

tion

)(mo

les

/yr)

K-feldspa

rreaction

pH=6-7

K-feldspa

rreaction

pH=5

(cb-c

s) mol/cm3

4x10-13

12x10-13

23x10-13

10 252015

Fig. 14.Comparison of H+ diffusion with H+ consumption by reaction with K-feldspar

volumes of various diameters. As K-feldspar volume increases, lower pH is required for

H+ diffusion to the reacting K-feldspar. If K-feldspar consumes H+ faster than diffusion

can supply it, then the pH raises causing precipitation of hydrous iron oxide.

64 W.T. Parry / Sedimentary Geology 233 (2011) 5368

http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4http://localhost/var/www/apps/conversion/tmp/scratch_3/image%20of%20Fig.%E0%B1%B4

7/25/2019 Parry 2011 concreciones de hierro

13/16

near a concretion were dissolved, transport of the Fe to the concretion

would be related to distance with more Fe transported from shorter

distances producing an iron depletion halo in the host rock. However,

no Fe-depleted halo is apparent surrounding the concretions (Table 1,

Fig. 4). The host rock surrounding the concretion averages 1.39% Fe as

goethite and lepidocrocite nuclei on illite-coated quartz grains and K-

feldspar grains from advective addition of iron. The original iron was

not redistributed to form the concretion rind but iron was added to

the rock by advection. The source of the iron is chemical reduction ofhematite coatings on sand grains by a reducing, hydrocarbon-bearing

solution. The reduced iron is transported in solution to the site of

oxidation and concretion formation.

9. Discussion

Concretions represent an end stage in diagenesis of iron minerals

in red beds. The iron mineral diagenesis begins with an early stage of

reddening that forms as ferrous silicates oxidize and form hematite

coatings on detrital grains during deposition and early burial.

Hydrocarbons from underlying source beds ow into the sediment

along fault conduits and react with the hematite coatings to

chemically reduce and mobilize iron and bleach the sandstone

(Parry and Blamey, 2010). Chemical reduction and iron removal

changes the color from reddish brown to nearly white recording uid

ow pathways. The mobilized iron remains in solution during

transport until an oxidizing solution is encountered. The iron in

solution is then precipitated as a result of oxidation. Bacteria may

mediate both the chemical reduction and oxidation of iron.

Concretion growth rates calculated in numerical growth models

are sensitive to concentration of reactants in solution, temperature,

and uid ow rate, but do not consider reaction ofuids with host

rock minerals such as K-feldspar, identity of precipitate, nor timing of

concretion growth. Growth times using the iron concentration in the

modeling ofChan et al. (2007) that are 10 orders of magnitude less

than those used here produce growth times that exceed the Jurassic

age of the Navajo Sandstone. Doubling the iron content of ground-

water to 10 mg/l cuts growth time by half. Increasing temperature to

100 C doubles the iron diffusion coefcient and decreases growthtime by a factor of 2, but hematite would be more likely to precipitate

instead of goethite. Decreasing the ow velocity below the 2 m/yr

used in the modelsproducesgrowth times between the diffusion only

calculations and the diffusion with advection.

Petrographic observation of FeO(OH) nuclei forming on K-feldspar

and illite suggests that H+ consumption by reacting silicates as well as

oxidation causes the precipitation. Reacting K-feldspar could cause a

pH gradient to form if H+ consumption rate exceeds the rate of H+

supply by diffusion. Calculations suggest that the concretion rind is

related to ambient solution pH. Low solution pH reacting with regions

of K-feldspar results in growthof a rind with a large diameter, because

the larger volume of K-feldspar consumes H+ at a greater rate than

can be supplied by diffusion; pH then rises causing precipitation of

goethite. The increased pH caused by K-feldspar reaction of smallervolumes with higher pH solutions causes precipitation of goethite

with smaller volumes of K-feldspar and thus smaller concretions.

The goethite rim forms a protective layer of reduced porosity and

permeability that inhibits chemical reactions with K-feldspar in the

concretion interior withexteriorsolutions. The porosity of the goethite

rim is 1.8% (Table 1) reducing permeability by 5 orders of magnitude

(Chidsey et al., 2007). Thus, solutionsthat precipitated lepidocrocitein

host sandstone were inhibited from reaching the concretion interior.

Claymineralogy differs in interiorversus exteriorsamples. The interior

is protected from reactants in solution that form kaolinite from

alteration of K-feldspar so the interior contains one-third to one-half

the kaolinite as the exterior samples (Table 1,Fig. 4). Formation of 1%

kaolinitein theexteriorhost rock would require approximately 1 m. y.

from rates of K-feldspar reaction shown inFig. 14.

Lepidocrocite present in host sandstone outside concretion rinds is

10 times more soluble than goethite (Fig. 8) so some mechanismmust

prevent precipitation of lower solubility goethite. Precipitation of

lepidocrocite over goethite is favored by phosphate, high oxidation

rate, and pH above 7, and inhibited by dissolved Si and dissolved CO2(Carlson and Schwertmann, 1990; Cumplido et al., 2000; Schwertman

and Thalmann, 1976). The host sandstone contains up to 0.07 wt. %

P2O5 (Table 1). Phosphate concentrations in groundwater from the

Navajo Sandstone are seldom above 0.03 mg/l, but do reach amaximum of 0.16 mg/l (Hood and Danielson, 1979, 1981; Hood and

Patterson, 1984). The ratio P/Fe varies from b0.003 to 0.016 at the

assumed Fe concentration used in the modeling. The rate of oxidation

of Fe2+ increases with an increase in P/Fe ratio up to 0.2 (Cumplido et

al., 2000). High oxidationrate is therefore suggestedas the most likely

cause of metastable lepidocrocite precipitation rather than more

stable goethite.

9.1. Timing of concretion formation

Formation of goethite concretions involves mobilizing iron from

hematite rims on detrital grains of quartz and feldspar that formed

during early diagenesis (Fig. 15). Then, the iron precipitates as

concretions when the iron-bearing solutions encountered oxidizing

conditions. Concentration of Fe3+ in equilibrium with iron oxides is

negligibly small so the likely mechanism of mobilization is chemical

reduction of the iron to Fe2+ by organic carbon. Other reductants

considered include hydrogen sulde that would immobilize the iron

as pyrite (Fig. 7), methane, or organic acids (Garden et al., 2001). The

timing of the mobilization step was related to hydrocarbon migration,

forexample,the iron content of produced water from theCovenantoil

eld reservoir in the Navajo Sandstone is 14 mg/kg (Chidsey et al.,

2007) near the maximum that can be dissolved by chemical reduction

at near-neutral pH.

Hydrocarbon migrated from Mississippian source beds during

Cretaceous thrust faulting into the Lower Permian White Rim

Sandstone reservoir during Middle Tertiary Laramide deformation

(Huntoon et al., 1999) (Fig. 15). Oil migrated into the Navajo

Sandstonein the Covenant oileld during Cretaceous to Early Tertiary(Chidsey et al., 2007). In the Moab area, hydrocarbon inux and iron

reduction occurred at 55-60 Ma (Garden et al., 2001).

The original porosity of the Navajo Sandstone is unknown, but

could have been as much as 49% (Atkins and McBride, 1992). The

average present intergranular volume (sum of intergranular porosity

and cements) is 20.8% (Table 1), the intergranular volume predicted

for 2 to 4 km of burial (Lander and Walderhaug, 1999). The iron oxide

cementation is expected to preserve the intergranular volume at the

time of cementation and must have taken place after 2 to 4 km of

burial compaction. The Navajo Sandstone is overlain by 1.9 to 4.2 km

of Jurassic to Eocene strata (Fig. 2). An inux of oxidizing meteoric

water would have been caused by local topographically driven ow

gradients that accompanied uplift and exhumation of the Colorado

Plateau. Uplift of the Colorado began in middle Eocene and continuedto late Miocene subjecting the area to exhumation and development

of the Colorado River system that continues today.

9.2. Bacterial mediation of iron reduction and oxidation

Formation of lepidocrocite and goethite by microbial Fe2+

reduction does not differ from abiotic Fe2+ oxidation with O2; but

solution geochemistry is responsible for the mineralogy of most Fe3+

precipitates (Larese-Casanova et al., 2010). Iron oxides in two typical

iron oxide concretions from Utah resemble known biosignatures, but

organic compounds could not be detected with gas chromatography-

mass spectrometry analysis (Souza-Egipsy et al., 2006). Host rock

samples for the present study average 0.04 wt. % C, but no carbonate

minerals could be detected. The isotopic composition of Fe in the

65W.T. Parry / Sedimentary Geology 233 (2011) 5368

7/25/2019 Parry 2011 concreciones de hierro

14/16

concretions provides the only evidence of bacterialinvolvement in the

iron diagenesis system. The Fe-oxide concretions show a relatively

large range in 56Fe from +0.9 to 2.0 (Chan et al., 2006). The

negative values result not from simple, closed-system remobilization

of 0 Fe-oxide, but instead formed in an open system. The

concretions probably formed by complete oxidation of aqueous

Fe2+, and the negative 56Fe must have resulted from negative values

of Fe2+ in solution that resulted from mobilization of Fe by chemical

reduction mediated by Fe3+ reducing bacteria (Chan et al., 2006).

Mediation of iron oxidation reactions by bacteria has no effect on the

concretion growth models, because the growth is limited by transport

of iron not by the rate of oxidation.

10. Conclusions

A physical, chemical, diffusion, and advection transport model has

been applied to mobilization of iron and deposition of hydrous iron

oxideconcretions. Iron oxide concretions form in porous sandstone bychemical reduction and solution of iron minerals, transport of Fe2+ in

solution, and precipitation of hydrous ferric oxide by oxidation.

Concretions form quickly in 1000's of years by diffusion and

advection, and K-feldspar consumption of H+ initiates nucleation.

The Jurassic Navajo Sandstone, host for the iron oxide concretions

in this study, averages (volume % one standard deviation) 74 2 %

quartz, 5.60.6 % K-feldspar, 2.10.2 % illite, 1.10.8 % kaolinite

and 0.7 0.1 % FeO(OH) as goethite or lepidocrocite. Concretions

consist of an outer rind of goethite almost completely lling the

porosity (19.6 volume%) in sandstone. A smaller amount of goethite is

also disseminated within the interior of the concretions. Lepidocrocite

is present in the exterior sandstone as light tan rims on sand grains.

Concretions form by diffusion and advection of iron that was

previously dissolved and mobilized during chemical reduction.Calculated solubility of goethite and lepidocrocite are extremely low

and signicant iron mobility requires reduction to Fe2+ species. A

calculated mineral stability diagram shows that precipitation would

result from an increase in pH a