University of Southern California Dissertations and Theses

Self-organized chemical precipitates: laboratory and field studies

(USC Thesis Other)

Self-organized chemical precipitates: laboratory and field studies

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

SELF-ORGANIZED CHEMICAL PRECIPITATES:
LABORATORY AND FIELD STUDIES

by

Laura Marie J. Barge

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GEOLOGICAL SCIENCES)

December 2009

Copyright 2009 Laura Marie J. Barge

ii
Acknowledgments

I would like to thank all the people who have inspired and helped me throughout
my graduate study. I especially thank my thesis committee, Ken Nealson, John Petruska,
Doug Hammond, Dave Bottjer, and Frank Corsetti, for their encouragement and valuable
suggestions. This work has benefited from collaboration with Marjorie Chan and Sally
Potter at the University of Utah, and I would like to thank them for their assistance with
field work, and for helpful comments and discussion. I would also like to thank Tim
Parker for sponsoring me as a research assistant at the Jet Propulsion Laboratory during
the beginning of my graduate study.
My thesis work was financially supported by a number of institutions. In
particular I would like to acknowledge the generous support from the NASA Harriett G.
Jenkins Pre-doctoral Fellowship Program and the NASA Astrobiology Institute / Lewis
and Clark Field Research Grant. I have had many valuable travel and networking
opportunities over the years thanks to the support of the NASA Astrobiology Institute,
the USC Earth Sciences department, the American Astronomical Society, the USC Dean
Joan M. Schaefer scholarship fund, the USC Women in Science and Engineering
program, the Delaware Space Grant / NASA Academy, and others.
I also want to thank the numerous friends, fellow grad students, and conference /
travel buddies who have provided entertainment and a support network for me these past
five and a half years. Finally I would like to thank my family and Luke for their
encouragement and emotional support throughout my graduate student career.

iii
Table of Contents

Acknowledgments

ii

List of Tables

v

List of Figures

vi

Abstract

ix

Chapter 1. Introduction

1
1.1. Periodic Patterns in Natural Systems 2
1.2. Experimental Studies of Periodic Banding 4
1.3. Theories and Quantitative Investigations of Liesegang
Phenomena
8
1.4. Objectives and Significance of the Dissertation 17
Chapter 1 References 21

Chapter 2. Periodic Precipitation Patterns in Diffusion Gradients: Chemical
Factors Involved in Their Formation

26
Summary 26
2.1. Introduction 27
2.2. Methods 30
2.2.1. Gel Preparation and Purification 32
2.2.2. Precipitation Reactions
2.2.3. Additional Experiments
33
34
2.3. Results 35
2.3.1. Silver Chromate / Dichromate Precipitation 35
2.3.2. Iron Precipitation Experiments 36
2.3.3. Experiments With Precipitation of Two or More Iron
Compounds
38
2.3.4. Changes in Pattern Morphology With Time 39
2.4. Discussion 46
2.4.1. Formation of Banded Precipitates 46
2.4.2. Effect of Initial Chemical Conditions on Precipitate
Morphology
50
2.5. Conclusion 53
Chapter 2 References 55

iv
Chapter 3. Precipitation Patterns Formed by Self-Organizing Processes in
Porous Media: Comparisons of Laboratory and Field Examples
58

Summary 58
3.1. Introduction 58
3.2. Experimental 62
3.2.1. Glass Bead Experiments 63
3.2.2. Gel Experiments 67
3.2.3. Gel / Glass Bead Mixtures 69
3.3. Results 72
3.3.1. Precipitation in Glass Beads 72
3.3.2. Precipitation in Gels 72
3.3.3. Precipitation in Glass Bead / Gel Mixtures 73
3.4. Field Studies: Self-Organized Precipitates from the Navajo
Sandstone, Southern Utah, USA
81

3.5. Discussion 92
3.5.1. Comparison to Field Examples 96
3.6. Conclusions 97
Chapter 3 References 99

Chapter 4. Effects of Amino Acids on Morphological Transitions in Inorganic
Precipitates

103
Summary 103
4.1. Introduction 103
4.2. Experimental 105
4.2.1. Gel Preparation 105
4.2.2. Analysis of Precipitates 107
4.3. Results 110
4.3.1. Gel Diffusion Experiments 110
4.3.2. Band Measurements - Spacing and Width Laws 119
4.3.3. Crystal Structure 124
4.4. Discussion 131
4.5. Conclusions 135
Chapter 4 References 137

Chapter 5. Conclusions

140

References

142

v
List of Tables

Table 2.1: Silver chromate precipitation experiments. 44

Table 2.2: Iron compound precipitation. 45

Table 3.1: Porosity and formation factor measured for compacted glass beads of
various sizes.
66

Table 3.2: Silver chromate precipitates in gels, glass beads, and glass beads +
gel.
71

Table 4.1: pH measurements. 108

Table 4.2: Dependence of the spacing coefficient on biochemical concentration. 121

Table 4.3: Dependence of the width law exponent α on biochemical
concentration.
123

vi
List of Figures

Figure 1.1: Sample experimental setup for Liesegang band formation in gels. 6

Figure 1.2: Examples of Liesegang ring formation in gels. 7

Figure 1.3: The formation of Liesegang bands in two interfaced solutions, from
Smith (1984).
12

Figure 1.4: Numerical simulations of pattern formation via competitive particle
growth from Feeney et al. (1983).
15

Figure 1.5: Two-dimensional simulation of a double-centered silver/chromate
Liesegang system, using an extended CPG model, from Krug and Brandtstädter
(1999).
16

Figure 2.1: Effects of gel medium on precipitation of silver chromate and
dichromate.
40

Figure 2.2: Silver chromate precipitate in Petri dishes. 41

Figure 2.3: Comparison of iron phosphate precipitates in agarose and silica gel.

42

Figure 2.4: Effects of pH changes on iron compound precipitation in silica gel
and agarose.
43

Figure 3.1: Silver chromate precipitates in glass bead matrix of various bead
size saturated with potassium chromate solution.
75

Figure 3.2: Gel controls. 76

Figure 3.3: Silver chromate precipitates in silica gel and glass beads of varying
size.
77

Figure 3.4: Silver chromate precipitates in agarose gel and glass beads of
varying size.
78

Figure 3.5: Silver chromate precipitates in agar gel and glass beads of varying
size.
79

Figure 3.6: Silver chromate precipitates in gelatin gel and glass beads of
varying size.
80

vii
Figure 3.7: Utah/ northern Arizona map and prominent mineralization
geometries.
86

Figure 3.8: Mini-concretions in the Navajo Sandstone, UT. 87

Figure 3.9: "Freckles" in situ. 88

Figure 3.10: Rind concretions from the Navajo Sandstone. 89

Figure 3.11: Liesegang banding in the Navajo Sandstone. 90

Figure 3.12: Mini-concretions in situ, associated with Liesegang bands. 91

Figure 4.1: Amino acids used in this study. 109

Figure 4.2: Silver chromate precipitation in agarose containing glycine. 112

Figure 4.3: Silver chromate precipitation patterns in agarose gel containing
Alanine.
113

Figure 4.4: Silver chromate precipitation patterns in agarose gel containing
Leucine.
114

Figure 4.5: Silver chromate precipitation in agarose gel containing N-acetyl
glycine.
115

Figure 4.6: Silver chromate precipitation patterns in agarose gel containing N-
acetyl alanine.
116

Figure 4.7: Silver chromate precipitates in agarose gel containing N-acetyl
leucine.
117

Figure 4.8: Density plots of silver chromate precipitates. 118

Figure 4.9: Graph of spacing constants as a function of concentration for all
biochemicals.
122

Figure 4.10: Silver chromate crystals in purified agarose (no biochemical
added).
126

Figure 4.11: Silver chromate precipitate in agarose containing alanine. 127

Figure 4.12: Silver chromate precipitates in agarose containing N-acetyl
glycine.
128

viii
Figure 4.13: Silver chromate precipitates in agarose containing N-acetyl
alanine.
129

Figure 4.14: Silver chromate precipitate in agarose containing N-acetyl leucine.

130

ix

Abstract

Self-organized patterns can be formed in diffusion experiments where two
interdiffusing electrolytes react to form an insoluble precipitate in a medium (such as
gelatin) that permits diffusive motion of ions, but prevents product particles from moving
from their site of formation. Various pattern morphologies such as periodic bands,
dendritic crystals, and continuous precipitates can be formed in many types of diffusion
media. We hypothesized that interfering ions that are not part of the dominant
precipitation reaction could affect the formation of self-organized patterns in a
reproducible fashion, and that the same precipitation reaction occurring in different
diffusion media would produce different pattern morphologies (depending on the
physical and chemical properties of that medium).
To initially characterize the patterns formed by various diffusion/reaction
systems, we performed experiments with ferrous carbonate, phosphate, and hydroxide
precipitation in agarose gel. Ferrous compounds were precipitated under reducing
conditions, forming self-organized patterns such as periodic bands, and were
subsequently oxidized as atmospheric oxygen diffused into the gel. Mineral replacement
occurred upon introduction of a reactant that formed a more insoluble compound than the
one already present, and after replacement the banding pattern remained unaltered,
allowing identification of the original precipitate.
We also characterized the effects of the diffusion medium and interfering ions on
self-organized precipitation in silver nitrate/potassium chromate and silver

x
nitrate/potassium dichromate reaction systems. We observed that precipitate morphology
was characteristic of the reactants that were initially present and the type of gel medium
in which they precipitated. It was found that soluble impurities in agarose gel were
responsible for the slight banding produced in silver chromate precipitation experiments.
We then tested the effects of organic compounds (simple amino acids and their N-
acetylated derivatives) on morphologies of silver chromate precipitates in purified
agarose gel. High concentrations of pure amino acids caused periodic banding of
crystals, and N-acetyl amino acids were more effective than pure amino acids, because of
their negative charge. We hypothesized that the length and orientation of the neutral side
chain of the amino acids, as well as charge, affects the degree of binding to the crystal
surface, and hence the ability to induce banded patterns.
To better apply these studies to self-organized patterns that form in natural
systems, where the diffusion medium may be inhomogeneous (such as sand or
sandstone), we studied laboratory and field examples of self-organized mineral
precipitates in porous media. Silver chromate precipitation experiments in tubes of glass
beads and glass bead/gel mixtures produced structures such as finger fluid fronts,
periodic banding, and millimeter-size spherules. The spheroidal precipitates produced in
our experiments nucleated via self-organizing processes throughout the glass bead
medium, and bear morphological resemblance to iron oxide concretions formed via self-
organizing processes in the Navajo Sandstone, UT, that preserve records of fluid flow
events in a porous and permeable sandstone.

1

Chapter 1. Introduction

Periodic patterns are widespread in geological and biological systems, and are
found in structures such as rhythmic banding in agates, concentric growth of bacterial
colonies, and pattern formation in embryonic development. Such patterns can form
spontaneously in a specific range of physical and chemical conditions, and the presence
of ordered patterns in geology and biology do not necessarily require an external template
(such as, banding in rock being produced by periodically-deposited sedimentary layers,
or periodic patterning in biology being specified in an organism's DNA). Periodic
precipitation in chemical systems was first investigated by R. E. Liesegang in 1896, when
he discovered concentric rings of silver dichromate precipitating in a glass plate filled
with gelatin (Henisch, 2005; Liesegang, 1896). "Liesegang patterns" have since been
observed in a wide variety of chemical systems (Henisch, 2005, and references therein),
and their formation has been generally attributed to supersaturation-nucleation-depletion
cycles combined with competitive particle growth (Feeney et al., 1983; Henisch, 2005;
Ortoleva, 1982; Ostwald, 1897; Smith, 1984). Periodic patterning of this type is not
limited to concentric rings or bands: self-organizing chemical processes can give rise to
morphologies such as bands of crystals, speckled patterns, spirals, or even spheroidal
concretionary precipitates, depending on the initial conditions and properties of the
diffusion matrix (Chan et al., 2007; Henisch, 2005; Ortoleva, 1984, 1994). The overall
objective of this dissertation is to shed light on the chemical processes that lead to

2
periodic precipitates in diffusion-controlled environments such as gel and porous media,
which has implications for understanding the self-organized patterning often observed in
geology and biology.

1.1. Periodic Patterns in Natural Systems

Self-organization in diffusion gradients is of particular interest in biology since
patterned morphologies in biological development are determined by gradients of
chemical substances known as morphogens, which determine gene expression as a
function of concentration (Lander et al., 2002; Turing, 1952). Turing (1952) proposed
that an initially homogenous gradient may begin to develop patterns due to random
instabilities, and that this may be the mechanism responsible for certain organized
morphologies that are seen in biological systems. Diffusion may be a main mechanism
responsible for formation of morphogen gradients in embryonic development (Crick,
1970; Gregor et al., 2005; Lander et al., 2002), and so it is possible that periodic patterns
of cell development could arise from two cross-diffusing morphogens, in the absence of
an external template.
Bacterial colonies are also known to display self-organized growth in the form of
concentric rings and/or dendritic patterns. Bacillus subtilis colonies commonly exhibit
various patterns, ranging from branching/dendritic growth to concentric rings (Fujikawa,
1992; Mimura et al., 2000, and references therein). Proteus mirabilis colonies grown in
agar display concentric banded morphology similar to Liesegang banding (Rauprich et

3
al., 1996). Feeney et al. (1956) also observed Liesegang-like rings of bacterial growth
around regions of concentrated chelating agents or metal ions. Various bacterial strains
have been observed to create dendritic branching patterns during growth (Kozlovsky et
al., 1999), and these patterns have been modeled by reaction-diffusion equations
(Golding et al., 1998; Kozlovsky et al., 1999; Mimura et al., 2000). Liesegang rings
have even been found in human tissues (e.g., Tuur et al., 1987; Raso et al., 1998, and
references therein).
Periodic pattern formation is also of importance to studies of geological
phenomena such as concretions, banding, and crystal growth that commonly occur in
sedimentary rock. It was once thought that seemingly ordered structures in the rock
record must have either been a product of biological processes (and hence, could be used
as a biosignature), or must have been a product of external processes (e.g., periodic
sedimentation). Now it is known that organized patterns can form through purely
chemical processes, and the use of morphology alone as a biosignature is discouraged
(Garcia-Ruiz et al., 2002). Various geological phenomena that were once described as
layering from periodic sedimentation and/or recrystallization are now thought to be
caused by diffusion-controlled self-organization processes—for example, "zebra rock"
(Loughnan and Roberts, 1990) and rhythmic banding in agate (Heaney and Davis, 1995).
Reaction-diffusion processes that lead to the formation of Liesegang bands can also result
in the formation of distributed spherical precipitates when operating in heterogeneous
media such as sand or sediment (Chan et al., 2007). Spheroidal or oblate concretions
distributed in a self-organized fashion are found in many geological environments: a few

4
examples include chert nodules with banded interiors (McBride et al., 1999), iron oxide
concretions of various forms found in sandstone (Chan et al., 2000, 2005), and the
hematite concretions that are found at the Mars Exploration Rover Opportunity landing
site at Meridiani Planum, Mars (Squyres et al., 2004).

1.2. Experimental Studies of Periodic Banding

Chemical precipitation in diffusion-controlled environments has been the subject
of many experimental investigations over the past century. One basic experimental setup
is the gel diffusion experiment, in which a gel is made to contain a concentration of one
reactive ion (the "inner electrolyte"), and another reactive ion (the "outer electrolyte") is
applied to the top of the gel in concentrated form so that it diffuses into the gel (Fig. 1.1).
The structure of gels prevents motion of fluid via convection or flow, but allows diffusion
of ions and small molecules between pores. The particles of precipitate formed are too
large to move between pores so sedimentation (as would occur if the same reaction were
performed in a fluid only) is prevented. Alternately, the experiment can be performed
with two reagents diffusing into the gel from opposite sides. In either case, precipitates
are able to develop self-organized patterns that may exhibit periodicity, depending on the
experimental parameters. This type of experiment can be performed in any system that
allows for diffusive movement of ions/molecules only; gels are an obvious scenario but
similar experiments have also been performed in sand, glass beads, mixtures of glass

5
beads and gel, paper saturated with fluid, jelly, and even in a gas (Henisch, 2005, and
references therein; Manley and Stern, 1955; Stern, 1954; Stern and Shniad, 1958).
Gel diffusion experiments have largely been pursued as a method for obtaining
large crystals of compounds that precipitate immediately in amorphous form when the
reaction occurs in a liquid (Henisch, 2005); however, many other interesting
morphologies may be produced depending on the specific reactants used, reactant
concentrations, and properties of the gel. Large crystals are best grown in gels such as
agarose that suppress nucleation of new crystals (Lagzi and Ueyama, 2009; Toramaru et
al., 2003), preventing a high number density of particles from forming and allowing
growth on existing crystals to be the dominant precipitation process. In gels such as
gelatin, which permit a higher rate of heterogeneous nucleation, a very high number
density of crystals may form (Lagzi and Ueyama, 2009), and this can result in a
continuous precipitate or in periodic bands. Liesegang rings (also called Liesegang bands
when the experiment is done in a tube of gel, since disks of precipitate form) can be
formed in many different salt systems, and the bands themselves can range from almost
amorphous composition to obviously macrocrystalline (Fig. 1.2).

6

FIGURE 1.1: Sample experimental setup for Liesegang band formation in gels. Reactant A is
typically distributed in dilute solution throughout the gel, and reactant B is introduced in concentrated form
to diffuse into the gel. As B diffuses into the gel it reacts with A to produce an insoluble precipitate AB
that can take various morphologies depending on experimental conditions. Left: Diffusion in a test tube
that produces a planar reaction front, and periodic precipitates form rhythmic bands. Right: Reactant B is
applied in a concentrated drop on top of a petrie dish containing gel, so that diffusion proceeds in a radial
fashion, and periodic precipitates form concentric rings ("Liesegang rings"). A and B can be any two ions
that react to produce an insoluble precipitate; for example 2Ag
+
+ Cr
2
O
7
-2
⇔ Ag
2
Cr
2
O
7
(s).

7

FIGURE 1.2: Examples of Liesegang ring formation in gels. A) Silver chromate precipitating in
laboratory gelatin. B) Silver chromate precipitating in commercial Knox gelatin. C) Silver dichromate
precipitating in laboratory gelatin. D) Silver dichromate precipitating in commercial Knox gelatin. The
patterns shown above took approximately one week to form.

8
1.3. Theories and Quantitative Investigations of Liesegang Phenomena

Studies of Liesegang bands formed in the laboratory have determined several
general mathematical relationships that govern the appearance of the bands.
First is the time law (Morse and Pierce, 1903) that states that the distance x
n
of the
nth band from the gel interface is proportional to the square root of the time of its
formation t
n
:
!
x
n
~ t
n
.
The time law arises as a consequence of the diffusive nature of the reaction front.
Second is the spacing law (Jablczynski, 1923) which states that the ratio of the
distances of consecutive bands from the gel interface approaches a constant (1 + p)
known as the spacing coefficient:
!
x
n +1
x
n
"(1+ p).
The spacing constant depends on the initial concentrations of both reacting ions (C
A0
and
C
B0
) as determined by Matalon and Packter (1954):
!
p =F(C
B0
) +
G(C
B0
)
C
A0
.
Normally the spacing between bands increases with x
n
; however, the opposite situation
has been observed experimentally and is known as "revert" banding (Flicker and Ross,
1974).
Finally, the width law states that the width of each band w
n
increases with x
n
.
Initially, this was thought to be a roughly linear relation. However, observations of

9
models that produce a large number of bands show that w
n
and x
n
are more accurately
related by a power law
w
n
∼ x
n
α

where the width exponent α depends on the initial concentrations C
A0
and C
B0
(Chopard
et al., 1994a, 1994b; Droz et al., 1999).
There have been many subsequent investigations to further refine these laws
(Chopard et al., 1999; George et al., 2003; George and Varghese, 2002; Lagzi and Izsák,
2004); however, the exact description of a Liesegang-banded system according to these
laws is problematic, since errors in determining band width and time of formation can be
large. Definition of the "position" or "width" of a band is somewhat ambiguous, since
bands can be composed of large crystals that are well-separated from one another, or of
amorphous particles that do not give clear edge definition of the band (Henisch, 2005,
and references therein). More complicated morphologies are often seen in Liesegang
banding studies that are not easily described using these laws, such as formation of
secondary patterns (Krug and Brandtstädter, 1999), revert band systems (Flicker and
Ross, 1974), spirals or discontinuous bands (Henisch, 2005; Krug and Brandtstädter,
1999), multiple superimposed band systems from simultaneous precipitation of two or
more compounds (Henisch, 2005), band doublets and triplets (Msharrafieh and Sultan,
2005), and even propagating patterns that move down the tube (Nasreddine and Sultan,
1999; Shrief et al., 2004).
The basic mechanisms of periodic band formation are relatively well understood.
The development of Liesegang bands is due to reaction-diffusion processes and can

10
therefore be described by coupled diffusion and reaction differential equations. However,
the exact mechanisms governing precipitation in these systems are uncertain, and many
attempts have been made to model the Liesegang phenomenon, with varying degrees of
success.
Typically in Liesegang banding experiments, the inner electrolyte is at much
lower concentration than the outer electrolyte diffusing in, creating an imposed
concentration gradient. The supersaturation theory of band formation developed by
Ostwald (Ostwald, 1897) suggests that the nucleation of a precipitate occurs when the
concentration product reaches some critical "nucleation threshold," and when
precipitation depletes the immediate area of the less abundant reactant, a zone of no
precipitation will occur. Alternately, more recent sol coagulation models suggest that
reactants A and B first produce a species C which also diffuses, and when C reaches a
concentration threshold, nucleation occurs (Chopard et al., 1994a, 1994b). Reactant B
(the outer electrolyte) would then continue to diffuse through the gel, and when the
reaction front moves far enough from the first precipitate that the concentration product
increases, precipitation will occur again when the nucleation threshold is reached. This
leads to periodic deposition of insoluble products, with band positions that can
theoretically be determined if the concentrations and diffusivities of both reactants are
known (Fig. 1.3). Models of this type produce banding that conforms to the spacing and
time laws as described above (Dee, 1986; Le Van and Ross, 1987; Prager, 1956), but do
not fit the width law (Chopard et al., 1994a) that is observed in experimental systems.
Supersaturation theory was later modeled using a cellular-automata model and assuming

11
the existence of an intermediate compound which also diffuses, and nucleates into a
precipitate at some threshold value (Chopard et al., 1994a). This produced Liesegang
bands that obeyed the time, spacing, and width laws.

12

FIGURE 1.3: The formation of Liesegang bands in two interfaced solutions, from Smith (1984).
a(x,t) and b(x,t) represent the concentration profiles of the outer and inner electrolyte, respectively. (This is
opposite the notation from Figure 1 - here A is the outer electrolyte.) H(x,t) is the concentration product. In
(a) the solubility product H(x,t) is just below the nucleation threshold H
c
. In (b) precipitation has occurred
at x = x
n
, catalyzing further precipitation and terminating when b(x,t) falls to zero. The concentration of B
must remain at zero at the location of the band at x
n
, because any increase in b would catalyze further
precipitation (since the concentration of A is much greater than B). In (c), B diffuses toward the band and
continues to precipitate, while A diffuses to the right (thus increasing the concentration of A). The
concentration product H(x,t) builds up until a band at x
n+1
forms in the same way.

13

A complicating factor, however, is that a concentration gradient is not necessary
to produce Liesegang bands (Flicker and Ross, 1974; Ortoleva, 1982). This was
experimentally verified by Flicker and Ross who produced Liesegang bands from an
initially homogenous lead iodide sol in gel (Flicker and Ross, 1974). The competitive
particle growth (CPG) model (Feeney et al., 1983; Ortoleva 1982) suggests that periodic
bands (and other self-organized patterns) can form because of particle growth dynamics
as the system ages. This effect arises because large particles/crystals are more
thermodynamically stable than smaller ones, so small particles can dissolve and larger
particles grow at their expense (Lifshitz and Slyozov, 1961), leading to pattern formation
from initial fluctuations in particle size distribution (a phenomenon also known as
"Ostwald ripening"). This model is in accordance with experimental data; it is commonly
observed in gel systems that band appearance continues to mature with time (Henisch,
2005), sometimes leading to dissolution of bands or formation of new ones (Feeney et al.,
1983; Ortoleva, 1984), and particle density decreases with time while average particle
size increases (Panjarian and Sultan, 2001). CPG models have successfully reproduced
many precipitation pattern types that have been seen in the laboratory, including "revert"
banding, secondary structure (Fig. 1.4), band splitting, spirals, and speckled patterns (Fig.
1.5) (Feeney et al., 1983; Flicker and Ross, 1974; Krug and Brandtstädter, 1999;
Ortoleva, 1984). In recent modeling studies the formation of Liesegang bands is treated
as a moving boundary problem, rather than keeping the interface of reactants stationary
(George et al., 2003; George and Varghese, 2005). Various other experimental factors
have been observed to have an effect on banding—for example, electric field, gravity,

14
light, dissociation of one reactant—and some of these have been successfully explained
quantitatively, as well (Chopard et al., 1999; García-Ruiz et al., 1996; Lagzi and Izsák,
2004; , Swami and Kant, 1967).

15

FIGURE 1.4: Numerical simulations of pattern formation via competitive particle growth from
Feeney et al. (1983). A) Evolution of a slightly random particle-size distribution into a macroscopic
pattern. B) Evolution of an initial band into secondary banding due to a gradient of particle size across the
band. C) Simulation of an experiment where a gel containing a coarser sol (left) is interfaced with a gel
containing a finer sol (right). Banding is induced in both gels.

16

FIGURE 1.5: Two-dimensional simulation of a double-centered silver/chromate Liesegang system,
using an extended CPG model, from Krug and Brandtstädter (1999). Silver ions diffuse from two
identical drops. When the diffusion fields of the two drops intersect, precipitation only occurs at the points
of intersection, producing a speckled pattern.

17

1.4. Objectives and Significance of the Dissertation

Computer models of Liesegang phenomena have successfully reproduced many
self-organized patterns that are observed in the laboratory; yet ideally, a complete
understanding of self-organized pattern formation would enable us to reconstruct the
conditions under which a pattern formed, based on its morphology. Modeling laboratory
scenarios involving idealized systems and simple reactions may not be sufficient since
self-organized patterns observed in geology and biology may involve other complicating
factors (such as diffusion through an irregular porous material, oxidation and replacement
of minerals in geology, or influence of organic compounds in biology). The goal of this
dissertation is to examine some of the specific chemical factors that may affect
morphologies of periodic patterns in natural diffusion-controlled systems, utilizing simple
gel/sand experiments that allow for study of individual factors. Specific objectives
include:

• Morphological characterization of mineral precipitates in gels. Precipitates
formed by silver chromate and silver dichromate (the original reactants used in
Liesegang's experiments) as well as ferrous iron compounds were produced in
silica, agarose, and gelatin gels. Varying gel characteristics and reactants
produced very different morphologies that were indicative of the initial conditions
in the system.

18
• Creating self-organized spheroidal precipitates in columns of glass beads (as a
laboratory analog to precipitation in quartz-based sandstone). Silver chromate
precipitation experiments were performed using glass beads or glass bead/gel
mixtures as the diffusion medium, to simulate self-organized concretion
precipitation in sandstone.
• Field work: Self-organized precipitates in the Navajo Sandstone, Utah. This
objective included field studies of periodic banding and various types of
concretions in the Jurassic Navajo Sandstone formation in Utah, to compare
natural self-organized precipitates with those produced in the laboratory.
• Investigating the ability of biomolecules to induce pattern formation in inorganic
precipitates in gels. In natural geological or biological systems other ions besides
the primary reactants are present; this objective aimed to determine the effect of
"interfering" ions on periodic pattern formation. In particular we focused on the
ability of amino acids to induce periodic pattern formation in silver chromate.

This thesis is divided into three chapters, each consisting of a separate manuscript.
Chapter 2 is a qualitative study of mineral precipitation morphologies in gel diffusion
experiments. A variety of reaction systems were tested, including the following: silver
nitrate/potassium chromate, silver nitrate/potassium dichromate, ferrous sulfate/sodium
hydroxide, ferrous sulfate/potassium phosphate, and ferrous sulfate/sodium carbonate.
Precipitation occurred in silica, agarose and gelatin gels, and the effect of the gel type on
precipitation morphology was examined. It was observed that the physical and chemical

19
properties of the gel had an obvious effect on precipitation structure; for example, the
same chemical reaction of silver nitrate + potassium dichromate → silver dichromate
produced dendritic crystals when performed in agarose gel, and periodic bands when
performed in gelatin gel. Iron minerals precipitated in periodic and continuous patterns,
and the effects of oxidation were observed to have an effect on morphology as well.
These results suggest that self-organized precipitates in geological and/or biological
environments may be greatly affected by physical and chemical properties of the
diffusion medium, and also that a precipitate morphology may be diagnostic of the initial
reaction that produced it, even if chemical replacement later occurs.
Chapter 3 expands on the observation that diffusion medium has a strong effect on the
morphology of precipitate, and extends this study of self-organized precipitation to a
natural analog. This chapter consists first of a laboratory component, in which silver
chromate precipitation experiments are performed in glass beads and gel media, to
approximate precipitation of concretions and periodic bands in a sandstone. Various
precipitation morphologies including millimeter-size spherules, rhythmic bands, fluid
fronts, and crystals were observed, depending on characteristics such as glass bead size
and the type of gel saturating the pores. The second component of this chapter is a field
study of self-organized iron mineral precipitates that formed in a natural environment.
Field studies of various locations in the Jurassic Navajo Sandstone formation in Utah
revealed iron precipitates in many forms that morphologically resemble the silver
chromate precipitates formed in our laboratory experiments. The mechanisms for self-
organized spherule and band formation in sand-like media are discussed.

20
Chapter 4 completes this work by studying the effects of biological components on
periodic precipitation in gels. We performed a study of the effects of biochemicals
(amino acids and their N-acetylated derivatives) on the precipitation of silver chromate in
purified agarose. It was found that crystal growth can be suppressed by soluble organic
compounds present in the gel, and that higher concentrations of organic compounds cause
precipitation to become more banded. We observed that N-acetyl derivatives of amino
acids are most effective at inducing banding. This observation that biochemicals can
induce banding of inorganic precipitates has implications for the formation of periodic
patterns in diffusion gradients in biology.

21
Chapter 1 References
Chan, M. A., Parry, W. T., & Bowman, J. R. (2000). Diagenetic Hematite and
Manganese Oxides and Fault-Related Fluid Flow in Jurassic Sandstones,
Southeastern Utah. AAPG Bulletin, 84(9), 1281-1310.

Chan, M. A., Beitler, B., Parry, W. T., Ormö, J., & Komatsu, G. (2005). Red rock and red
planet diagenesis: Comparisons of Earth and Mars concretions. GSA Today,
15(8), 4-10.

Chan, M. A., Ormö, J., Park, A. J., Stich, M., Souza-Egipsy, V., & Komatsu, G. (2007).
Models of iron oxide concretion formation: field, numerical and laboratory
comparisons. Geofluids 7, 1-14.

Chopard, B., Luthi, P., & Droz, M. (1994a). Reaction-Diffusion Cellular Automata
Model for the Formation of Liesegang Patterns. Physical Review Letters, 72(9),
1384-1387.

Chopard, B., Luthi, P., & Droz, M. (1994b). Microscopic Approach to the Formation of
Liesegang Patterns. Journal of Statistical Physics, 76(1/2), 661-677.

Chopard, B., Droz, M., Magnin, J., Rácz, Z., & Zrinyi, M. (1999). Liesegang Patterns:
Effect of Dissociation of the Invading Electrolyte. The Journal of Physical
Chemistry, 103, 1432-1436.

Crick, F. (1970). Diffusion in Embryogenesis. Nature, 225, 420-422.

Dee, G. T. (1986). The Patterns Produced by Precipitation at a Moving Reaction Front.
Physica 23D, 340-344.

Droz, M., Magnin, J., & Zrinyi, M. (1999). Liesegang patterns: Studies on the width law.
Journal of Chemical Physics, 110(19), 9618-9622.

Feeney, R. E., Petersen, I. M., & Sahinkaya, H. (1956). “Liesegang-like” rings of growth
and inhibition of bacteria in agar caused by metal ions and chelating agents.
Journal of Bacteriology 73, 279-283.

Feeney, R. E., Schmidt, S. L., Strickholm, P., Chadam, J., & Ortoleva, P. (1983). Periodic
precipitation and coarsening waves: Applications of the competitive particle
growth model. Journal of Chemical Physics, 78(3), 1293-1311.

Flicker, M., & Ross, J. (1974). Mechanism of chemical instability for periodic
precipitation phenomena. Journal of Chemical Physics, 60(9), 3458-3465.

22
Fujikawa, H. (1992). Periodic growth of Bacillus subtilis colonies on agar plates. Physica
A, 189, 15-21.

García-Ruiz, J. M., Carnerup, A., Christy, A. G., Welham, N. J., & Hyde, S. T. (2002).
Morphology: An Ambiguous Indicator of Biogenecity. Astrobiology, 2, 353-369.

García-Ruiz, J. M., Rondón, D., García-Romero, A., & Otálora, F. (1996). Role of
Gravity in the Formation of Liesegang Patterns. The Journal of Physical
Chemistry, 100, 8854-8860.

George, J. & Varghese, G. (2002). Formation of periodic precipitation patterns: a moving
boundary problem. Chemical Physics Letters 362, 8-12.

George, J., Paul, I., Varughese, P. A., & Varghese, G. (2003). Rhythmic pattern
formations in gels and Matalon-Packter law: A fresh perspective. Pramana,
60(6), 1259.

George, J. & Varghese, G. (2005). Intermediate colloidal formation and the varying
width of periodic precipitation bands in reaction-diffusion systems. Journal of
Colloid and Interface Science, 282, 397-402.

Golding, I., Kozlovsky, Y., Cohen, I., & Ben-Jacob, E. (1998). Studies of bacterial
branching growth using reaction-diffusion models for colonial development.
Physica A, 260, 510-554.

Gregor, T., Bialek, W., de Ruyter van Steveninck, R. R., Tank, D. W., & Weischaus, E.
F. (2005). Diffusion and scaling during early embryonic pattern formation.
Proceedings of the National Academy of Sciences, 102(51), , 18403-18407.

Heaney, P. J. & Davis, A. M. (1995). Observation and Origin of Self-Organized Textures
in Agates. Science, 269, 1562-1565.

Henisch, H. K. (2005). Crystals in Gels and Liesegang Rings. Cambridge: Cambridge
University Press.

Jablczynski, K. (1923). Bulletin de la Société Chimique de France, 33, 1592.

Kozlovsky, Y., Cohen, I., Golding, I., & Ben-Jacob, E. (1999). Lubricating bacteria
model for branching growth of bacterial colonies. Physical Review E, 59(6),
7025-7035.

Krug, H. J. & Brandtstädter, H. (1999). Morphological characteristics of Liesegang Rings
and Their Simulations. The Journal of Physical Chemistry A, 103, 7811-7820.

23
Lagzi, I. & Izsák, F. (2004). Stabilization and destabilization effects of the electric field
on stochastic precipitate pattern formation. Chemical Physics, 303, 151-155.

Lagzi, I. & Ueyama, D. (2009). Pattern transition between periodic Liesegang pattern
and crystal growth regime in reaction-diffusion systems. Chemical Physics
Letters, 468, 188-192.

Lander, A. D., Nie, Q., & Wan, F. Y. M. (2002). Do Morphogen Gradients Arise by
Diffusion? Developmental Cell, 2, 785-796.

LeVan, M. E. & Ross, J. (1987). Measurements and a Hypothesis on Periodic
Precipitation Processes. The Journal of Physical Chemistry, 91, 6300-6308.

Liesegang, R. E. (1896). Naturwiss. Wochenschr., 11, 353.

Lifshitz, I. M. & Slyozov, V. V. (1961). The Kinetics of Precipitation From
Supersaturated Solid Solutions. Journal of Physics and Chemistry of Solids,
19(1/2), 35-50.

Loughnan, F. C. & Roberts, F. I. (1990). Composition and origin of the 'zebra rock' from
the East Kimberly region of Western Australia. Australian Journal of Earth
Sciences, 37, 201-205.

Manley, D. R. & Stern, K. H. (1955). Liesegang rings in inhomogeneous media:
powdered glass. Journal of Colloid Science, 10, 409-412.

Matalon, R. & Packter, A. (1954). The Liesegang Phenomenon: I. Sol Protection and
Diffusion. Journal of Colloid Science, 10(1), 46-62.

McBride, E. F., Abdel-Wahab, A., & El-Younsy, A. R. M. (1999). Origin of spheroidal
chert nodules, Drunka Formation (Lower Eocene), Egypt. Sedimentology, 46,
733-755.

Mimura, M., Sakaguchi, H., & Matsushita, M. (2000). Reaction-diffusion modeling of
bacterial colony patterns. Physica A, 282, 283-303.

Morse, H. W. & Pierce, G. W. (1903). Diffusion and Supersaturation in Gelatine.
Proceedings of the American Academy of Arts and Sciences, 38(22), 625-648.

Msharrafieh, M. & Sultan, R. (2005). Patterns with High Rhythmicity Levels in
Multicomponent Liesegang Systems. ChemPhysChem, 6, 2647-2653.

24
Nasreddine, V. & Sultan, R. (1999). Propagating Fronts and Chaotic Dynamics in
Co(OH)
2
Liesegang Systems. The Journal of Physical Chemistry A, 103, 2934-
2940.

Ortoleva, P. (1982). Solute Reaction Mediated Precipitate Patterns in Cross Gradient Free
Systems. Zeitschrift für Physik B: Condensed Matter, 49, 149-156.

Ortoleva, P. (1984). From Nonlinear Waves to Spiral and Speckle Patterns:
Nonequilibrium Phenomena in Geological and Biological Systems. Physica 12D,
305-320.

Ortoleva, P. (1994). Geochemical Self-Organization. Oxford Monographs on Geology
and Geophysics No. 23. Oxford: Oxford University Press.

Ostwald, W. (1897). Zeitschrift für Physikalische Chemie – Leipzig, 23, 356.

Panjarian, S., & Sultan, R. (2001). Crystal selection and Liesegang banding in dynamic
precipitate systems. Collection of the Czechoslovak Chemical Communications,
66, 514–541.

Prager, S. (1956). Periodic Precipitation. Journal of Chemical Physics, 25(2), 279-283.

Raso, D. S., Greene, W. B., Finley, J. L., & Silverman, J. F. (1998). Morphology and
Pathogenesis of Liesegang Rings in Cyst Aspirates: Report of Two Cases With
Ancillary Studies. Diagnostic Cytopathology, 19(2), 116-119.

Rauprich, O., Matsushita, M., Weijer, C. J., Siegert, F., Esipov, S. E., & Shapiro, J. A.
(1996). Periodic Phenomena in Proteus mirabilis Swarm Colony Development.
Journal of Bacteriology, 178(2), 6525-6538.

Shreif, Z., Mandalian, L., Abi-Haydar, A., & Sultan, R. (2004). Taming ring morphology
in 2D Co(OH)
2
Liesegang patterns. Physical Chemistry Chemical Physics, 6,
3461-3466.

Smith, D. A. (1984). On Ostwald's supersaturation theory of rhythmic precipitation
(Liesegang's rings). Journal of Chemical Physics, 81(7), 3102-3115.

Squyres, S. W., Grotzinger, J. P., Arvidson, R. E., Bell III, J. F., Calvin, W., Christensen,
P. R., Clark, B. C., Crisp, J. A., Farrand, W. H., Herkenhoff, K. E., Johnson, J. R.,
Klingelhofer, G., Knoll, A. H., McLennan, S. M., McSween Jr., H. Y., Morris, R.
V., Rice Jr., J. W., Rieder, R., & Soderblom, L. A. (2004). In Situ Evidence for
an Ancient Aqueous Environment at Meridiani Planum, Mars. Science, 306,
1709-1714.

25
Stern, K. H. (1954). Liesegang rings in inhomogeneous media: partially coagulated
gelatin. Journal of Colloid Science, 9(4), 329-337.

Stern, K. H., & Shniad, H. (1958). Diffusion in inhomogeneous media: capillary
diffusion in beds of glass beads. Journal of Colloid Science, 13, 24-31.

Swami, S. N., & Kant, K. (1967). Effect of Light on the Formation of Liesegang Rings of
Copper Chromate in Agar Agar Gel. Colloid and Polymer Science, 215(1), 60-
61.

Toramaru, A., Harada, T., & Okamura, T. (2003). Experimental Pattern Transitions in a
Liesegang system. Physica D, 183, 133-140.

Turing, A. M. (1952). The Chemical Basis of Morphogenesis. Philosophical
Transactions of the Royal Society of London. Series B, Biological Sciences,
237(641), 37-72.

Tuur, S. M., Nelson, A. M., Gibson, A. M., Neafie, R. C., Johnson, F. B., Mostofi, F. K.,
& Connor, D. H. (1987). Liesegang rings in tissue: How to distinguish Liesegang
rings from the giant kidney worm, Dioctophyma renale. The American Journal of
Surgical Pathology, 11(8), 598-605.

26
Chapter 2. Periodic Precipitation Patterns in Diffusion Gradients:
Chemical Factors Involved in Their Formation

Summary

We present results of studies of the morphologies of precipitates formed in silica
gel, agarose, and gelatin. When ionic impurities were present in agarose gels, silver
chromate precipitates were banded, most likely due to the interference of these ions with
crystal growth. In each case, the morphology of the precipitates formed (bands,
periodically banded crystals, dendritic crystals) was characteristic of the reactants that
were initially present, as well as of the properties of the gel medium in which they were
precipitated. Ferrous carbonate, phosphate, and hydroxide precipitates that were
precipitated under reducing conditions were subsequently oxidized as atmospheric
oxygen diffused into the gel. Mineral replacement occurred upon introduction of a
reactant that formed a more insoluble compound than the one already present, and after
replacement the banding pattern remained unaltered, allowing identification of the
original precipitate. These results provide a basis for beginning to understand how
physical and chemical factors can lead to the formation of complex, self-organized
chemical precipitates.

27
2.1. Introduction

In this study we investigate some of the chemical factors that control the
morphology of inorganic precipitates formed in diffusion gradients, in order to better
understand how we might use morphology of a natural mineral structure to assess the
characteristics of the environment in which it formed and matured. In geological and
biological systems, diffusion gradients can lead to highly-organized structures such as
rhythmic banding, dendritic crystals, or oscillatory zoning of crystals, which do not
necessarily reflect underlying templates. This phenomenon of emergence of ordered
patterns without an underlying template is referred to as self-organization (Ortoleva,
1994). Highly-ordered structures that arise from self-organizing processes are found as
complex living forms in biology (Lander, 2007) but also can be found in simpler non-
living forms in geology. A few geological examples include rhythmic bands of hematite
in dolomite beds (Cochran and Elmore, 1987), bands of chalcedony and quartz in agate
(Heaney and Davis, 1995), and chert nodules with rhythmically banded interiors
(McBride et al., 1999). In some cases self-organized precipitation in geological
environments also results in spheroidal structures, one notable example of which are the
hematite spherules found at Meridiani Planum, Mars (Squyres et al., 2004), which are
morphologically similar to solid millimeter-size concretions found in terrestrial
environments (e.g. Chan et al., 2004, 2005).
The existence of such ordered macroscopic structures in nature was once thought
to be either a byproduct of biological processes (such as banded or branching growth

28
structures that can be produced by bacterial colonies (Feeney et al., 1956; Fujikawa,
1992; Golding et al., 1998; Kozlovsky et al., 1999; Mimura et al., 2000; Rauprich et al.,
1996)), or an underlying geological template (such as layers in rock corresponding to
periodic sedimentation); however, it is now known from experimental studies that a
variety of self-organized structures can form through inorganic chemical processes in
diffusion-controlled environments such as gels (Henisch, 2005; Ortoleva, 1994). Studies
of inorganic precipitation in gels have produced periodic precipitates in many different
salt systems, in varying morphologies depending on initial conditions. Rhythmic banding
in precipitates (known as Liesegang banding (Liesegang, 1896)) has been attributed to
supersaturation-nucleation-depletion processes (Ortoleva, 1994). A few examples
include oscillatory zoning of silver chromate crystals in agarose (Henisch, 2005),
rhythmic bands of cobalt/nickel/magnesium oxide (Msharrafieh and Sultan, 2005), and
lead iodide precipitates that vary from inhomogeneous "blobs" to rhythmic bands to
bands of dendritic crystals (Muller and Ross, 2003; Ripszam et al., 2005; Toramaru et al.,
2003). Volford et al. (2007) even reported dynamical patterns such as traveling waves in
an aluminum hydroxide precipitation system in agarose. Stone and Goldstein (2004) also
reported layered bands of iron oxide/hydroxide in agarose gel forming in an oxidation
gradient.
The chemical nature of the gel medium has long been known to have an effect on
the precipitation morphology; this is seen in silver chromate precipitation experiments
where crystals or bands of crystals are formed in agarose but much sharper rhythmic
bands of fine particles are formed in gelatin (Henisch, 2005; Liesegang, 1896). Physical

29
effects are important as well — for example, Toramaru et al. (2003) found that increased
gel concentration caused lead iodide to precipitate in bands while at low gel
concentrations only dendritic crystals were formed. Inhomogeneities in the gel medium
can affect precipitation morphology, causing instabilities within the banded patterns to
form, which is one proposed mechanism behind "zebra rock" (Krug et al., 1996). In a
porous medium such as sandstone or sediment the inhomogeneous porosity and
permeability (compared to a gel) could lead minerals to precipitate in other morphologies
that are commonly seen in geological environments, such as finger-like fluid fronts or
spheroidal concretions (Ortoleva, 1994).
Previous work on precipitation in gels and porous media has shown that
morphology of self-organized structures is greatly affected by chemical and physical
conditions, but there is still much to learn about the specifics of how certain parameters
affect precipitation patterns. In geological environments the composition of a mineral
precipitate may give information about the environment in which it was deposited, but
the chemical makeup of a structure could easily be replaced or altered, in which case
interpretations based only on morphology may prove useful. In this work we aim to
understand the factors that govern the pattern morphology of self-organized chemical
precipitates in gels, in order to develop a methodology for determining past chemical and
physical factors based on morphological analysis. We focus on precipitation in gels
under various chemical conditions to examine the effects of impurities, gel type, reactant
type and oxidation on pattern morphology in diffusion-controlled systems.

30
2.2. Methods
Our approach involves diffusion of reactive ions through different types of gels.
Gels are used because they prevent bulk movement of fluid while allowing diffusion (by
random walk) of small molecules and ions, and prevent product particles greater than a
certain size from moving from their sites of formation. In this work we used a variety of
gels (agarose, gelatin, and silica hydrogel) and different combinations of reactants (silver
+ chromate, silver + dichromate, ferrous iron + hydroxide, ferrous iron + phosphate, and
ferrous iron + carbonate).
We chose these three gels for the silver chromate precipitation experiments
because they represent very different chemical environments. Gelatin is a complex
organic matrix of collagen polypeptides and associated mucopolysaccharides
(glycosamino glycans) obtained from animal connective tissues, containing many organic
and inorganic ions. Agarose is a simpler organic matrix obtained from seaweed
consisting mainly of methylated cellulose, with a much lower ionic content and therefore
lower conductivity suitable for use in gel electrophoresis. Silica gel is a purely inorganic
matrix composed of polymerized acidified silicate that contains some silicate ions in
solution, and also anions of the acid used to create the gel (e.g. acetic acid). However,
unlike neutral agarose, the matrix of silica gel has an inherent negative charge, and the
pH of the silica gel after gelling increases with age (Henisch, 2005).
Using different types of gel varies the amount of impurities in the fluid
environment, and varying the combinations of reactants allows identification of
characteristic morphologies for different reactions. The chromate and dichromate anions

31
were chosen in this study because they are colored and structurally similar to phosphate
and pyrophosphate, the dominant anions in biology, which are colorless. The silver
cation (Ag
+
) was chosen because it is a monovalent cation that forms insoluble
precipitates with not only chromate and dichromate but also phosphate and
pyrophosphate. Experiments with iron compound precipitation were performed to test
the morphological characteristics of ferrous iron reacting with various geochemically
plausible ions, like hydroxide, carbonate and phosphate. Iron precipitates can form in
neutral agarose gels, where reduced iron (Fe
2+
) can be maintained in the gel to precipitate
with other reactive ions diffusing in (and the first precipitate can react with atmospheric
oxygen as it diffuses in), which allows us to simulate the reactions that may occur in an
iron redox system.
Chemical precipitates were formed by introducing a solution of one reactant (A)
to a gel containing a dilute concentration of another reactant (B). (A) then diffuses
throughout the gel containing (B) and the reaction proceeds as A (sol.) + B (sol.)  AB
(insol.), yielding an insoluble AB compound that precipitates out of solution. These
experiments can be performed in test tubes so that diffusion of reactant A proceeds
vertically into the gel, or in petrie dishes so precipitates form in a radial fashion around a
center drop of solution containing reactant A. Precipitation occurred throughout the
entire gel in 4-5 days; however, oxidation of iron precipitates occurred over a longer
period (weeks to a month).

32
2.2.1. Gel Preparation and Purification

Gelatin gels (made with commercial Knox gelatin) were prepared at 3.33 weight
percent to contain a 5 mM concentration of potassium chromate or potassium dichromate.
Agarose gels (made with Molecular Biology Agarose, Shelton Scientific) were
prepared at 0.66 weight percent to contain 5 mM of a soluble salt of reactant B. When
agarose gels were prepared to contain iron sulfate, the mixture of agarose powder and
deionized water was allowed to boil uncovered for 5 minutes to allow oxygen to be
expelled from the gel, then ferrous sulfate crystals were added after the gel was removed
from the heat. For all other experiments (with gels containing potassium chromate,
potassium dichromate, or potassium phosphate) the agarose powder was boiled in a 5
mM solution of the desired salt.
In several experiments the agarose powder was purified before use, by extracting
with deionized water, then centrifuging for 5 minutes and pouring off the water
containing the gel impurities. This extraction was repeated five times to purify the
agarose before it was used in the procedures described above. One experiment was
performed in which 50 mM sodium acetate (0.205 g CH
3
COONa per 50 ml of gel) was
added to already purified agarose gels containing potassium chromate, to test whether
banding could be induced by introduction of a non-precipitating, negatively-charged ion
(CH
3
COO
-
).
Silica gels were prepared by mixing a solution of sodium silicate (1.25 ml of
reagent grade sodium silicate solution (containing ~10.6% Na
2
O and ~26.5% SiO
2
)

33
diluted to 8 ml with deionized H
2
O) with a solution of dilute acetic acid and soluble salt
of reactant B (360 µl of glacial acetic acid diluted with 7.6 ml of a 10.5 mM solution of
reactant B, so that after mixing with the silicate solution the entire gel will have a 5 mM
concentration of reactant B). The sodium silicate solution was poured into the acid
solution in a 20 ml tube and the tube was sealed and inverted slowly (not shaken) five
times to thoroughly mix the gel. All gels were allowed to set in test tubes at room
temperature before the concentrated solution containing reactant A was added. In all
cases the tubes were exposed to air so that atmospheric oxygen diffused into the gel.
After the gels hardened, 150 µl of a concentrated solution of a soluble salt of
reactant A (e.g. silver nitrate, potassium phosphate, sodium hydroxide) was added to the
top of the gel.

2.2.2. Precipitation Reactions

Silver chromate/dichromate precipitates:
When the gel was prepared to contain potassium chromate or potassium
dichromate, application of a silver nitrate solution produced precipitates of silver
chromate or dichromate according to the following reactions:
2Ag
+
+ CrO
4
-2
→ Ag
2
CrO
4
(insol.)
2Ag
+
+ Cr
2
O
7
-2
→ Ag
2
Cr
2
O
7
(insol.)

34
Iron compound precipitates:
When the gel was prepared to contain iron sulfate, solutions containing potassium
phosphate (saturated solution), sodium carbonate (saturated solution), or sodium
hydroxide (0.1 M) were applied, which produced precipitates according to the following
reactions:
Fe
+2
+ PO
4
-3
→ Fe
3
(PO
4
)
2
(insol.)
Fe
+2
+ CO
3
-2
→ FeCO
3
(insol.)
Fe
+2
+ 2OH
-
→ Fe(OH)
2
(insol.)
As atmospheric oxygen diffused into the gel, the initial (reduced) iron precipitates
oxidized and color changes were seen. The different combinations of gel type, reactants,
and expected precipitates for all experiments performed are listed in Tables 2.1 and 2.2.

2.2.3. Additional Experiments

In two iron precipitation experiments the solution (A) added to the top of the gel
contained two reactants (potassium phosphate and sodium hydroxide, and sodium
carbonate and sodium hydroxide).
In one iron phosphate precipitation experiment in agarose, the positions of the
electrolytes were reversed, so the agarose contained 5 mM of potassium phosphate and 1
g of iron sulfate crystals were added to the top of the gel.

35
Mineral replacement was tested by adding 1 ml of 0.1 M sodium hydroxide
solution to the top of a tube of gel in which precipitation of iron phosphate had already
occurred.

2.3. Results

2.3.1. Silver Chromate / Dichromate Precipitation

When a solution containing silver nitrate was applied to a gelatin gel containing
potassium chromate or dichromate (Figs. 2.1 and 2.2; Table 2.1), both silver chromate
and silver dichromate precipitated in bands, but the bands formed by silver dichromate
were sharper and had more well-defined edges than those formed by silver chromate
(Figs. 2.1A, 2.1D). In silica gel, both silver chromate and dichromate precipitated in
bands of small crystals (Figs. 2.1C and 2.1H), while in agarose, silver chromate formed
small crystals and silver dichromate formed dendritic crystals (Fig. 2.1B). In unpurified
agarose the silver chromate crystals exhibited some banding (Fig. 2.1E), and the banding
effect disappeared when the agarose was purified (Fig. 2.1F). When a purified agarose
gel was made to include sodium acetate in solution, silver chromate crystals once again
precipitated in bands (Fig. 2.1G). The dendritic crystals formed by silver dichromate did
not exhibit banding, regardless of whether the agarose was purified to remove any
interfering ionic components.

36
Similar results were observed for silver chromate and dichromate experiments in
petri dishes, where the precipitate extended radially out from the center in the form of
concentric rings (for silver chromate or dichromate in gelatin), dendritic crystals (for
silver dichromate in agarose), or small crystals (for silver chromate in agarose) (Fig. 2.2).

2.3.2. Iron Precipitation Experiments

Iron precipitation experiments were performed in agarose and silica gels.
Experiments started with reduced iron (Fe
2+
) in the gel (with one exception where
phosphate (PO
4
-3
) was in the gel), and the first precipitates were ferrous iron compounds
that later oxidized as oxygen diffused into the gel.
In the iron-plus-hydroxide experiments, a gray continuous precipitate was initially
observed, which eventually separated into several thick bands over a period of
approximately one month. The gray precipitate turned orange as oxygen diffused into the
gel (Fig. 2.4A).
In the iron-plus-phosphate experiments, Liesegang bands were observed. In the
agarose experiment where Fe
2+
was in the gel and phosphate was introduced, bands of
fine gray crystals (presumably Fe(II)-phosphate) formed. These bands then turned green
as oxygen diffused into the gel (a property of vivianite). This experiment was repeated
using purified agarose, and unlike the silver chromate experiment where banding
disappeared when impurities were removed from the agarose, no change in iron
phosphate precipitation was observed (Fig. 2.3A). An agarose experiment was performed

37
in which the electrolytes were reversed, so that phosphate was in the gel and ferrous
sulfate was introduced in concentrated form (Fig. 2.3C). In this case bands of fine gray
crystals formed, similar to the first iron phosphate precipitates, but as they oxidized, the
bands as well as the spaces between bands turned cloudy yellow, and after oxidation, the
banding pattern was mostly obscured. Iron phosphate precipitation was also observed in
silica gel (with Fe
+2
in the gel and phosphate in solution) as gray/white crystals that
precipitated in several thick periodic bands, turning green as the gel oxidized (Fig. 2.3D).
In silica gel the bands were thicker than in any of the agarose experiments: the band
width increased towards the bottom of the tube and eventually individual crystals were
visible and well separated from one another.
In the iron-plus-carbonate experiments, several types of carbonate solutions at
different pHs were used. In agarose experiments (containing Fe
2+
in the gel) where
carbonate was introduced with saturated NaHCO
3
solution at pH 7.94, a region of
continuous white precipitate followed by indistinct rhythmic bands was observed. After
oxidation, the region of continuous precipitate contained many fine layers of orange
compounds (presumably stratified iron oxides/hydroxides) and after several months when
oxidation was complete, the original rhythmic bands formed by precipitation were mostly
obscured (Fig. 2.4B). In agarose experiments where carbonate was introduced in a
solution of equal molarities NaHCO
3
/Na
2
CO
3
, a homogenous precipitate was observed
that formed many fine layers of iron hydroxide/oxide after oxidation, and no rhythmic
banding was observed (Fig. 2.4C). In silica gel experiments iron-plus-carbonate

38
produced only a slight continuous green precipitate that turned orange after oxidation (not
pictured).

2.3.3. Experiments With Precipitation of Two or More Iron Compounds

One experiment was performed in which sodium hydroxide solution was added to
a tube that already contained periodic bands of green (oxidized) iron phosphate, produced
in the iron-plus-phosphate agarose experiment. The second solution diffused through the
gel in approximately one week, and at the new reaction front green bands became orange,
presumably as Fe-hydroxide replaced Fe-phosphate (Fig. 2.3B). The banding pattern was
not altered by the chemical change.
Both iron-phosphate and iron-carbonate precipitation experiments in agarose were
also repeated, using higher pH phosphate and carbonate solutions. The result is that in
each experiment there were two precipitating anions, since at the higher pH hydroxide
was present in solution as well as phosphate or carbonate.
In the iron-phosphate/hydroxide experiment, first a homogenous white precipitate
was observed that was morphologically similar to the precipitate formed in the iron
hydroxide precipitation experiment. This continuous precipitate was followed by
periodic bands of white/grey crystals that were morphologically similar to those produced
in the iron phosphate precipitation experiments. After oxidation the continuous
precipitate turned orange and exhibited many fine layers of different colored compounds,
and the rhythmic bands below turned green (Fig. 2.4D).

39
A similar result was observed for the iron-carbonate/hydroxide experiment.
When a solution containing both carbonate and hydroxide ion was added to the top of the
agarose gel containing iron sulfate, first a continuous white precipitate was observed,
followed by rhythmic bands (similar to those observed in the pure iron plus carbonate
experiment). The continuous precipitate oxidized to form many thin layers of different
colored oxides, and after oxidation was complete, the rhythmic bands of white/grey
precipitate were obscured (Fig. 2.4E).

2.3.4. Changes in Pattern Morphology With Time

For the iron experiments the patterns formed by precipitation in gels continued to
change over time. In all of our experiments (including silver chromate and dichromate
precipitation), precipitation began immediately and patterns were fully formed within
three or four days. However, changes in structure and oxidation continued to occur over
periods of weeks to months. Not surprisingly, one common change in the iron
experiments was progressive oxidation, i.e., the color change that occurred as O
2
diffused
throughout the gel matrix. However the precipitation patterns themselves also underwent
a maturation process: for example, the amorphous precipitate formed by ferrous iron
reacting with hydroxide eventually separated into several thick bands (Fig. 2.4A), and the
thin layers of various oxides formed after oxidation of Fe-carbonate changed in thickness
and color and had clearer edge definition as time passed (Fig. 2.4B).

40

FIGURE 2.1: Effects of gel medium on precipitation of silver chromate and dichromate.
(A) Bands of silver dichromate in gelatin. (B) Silver dichromate crystals in agarose. (C) Silver dichromate
crystals in silica gel. (D) Liesegang bands of silver chromate in gelatin. (E) Silver chromate crystals in
agarose. (F) Silver chromate crystals in purified agarose. Periodic precipitation does not occur. (G) Silver
chromate crystals in purified agarose with sodium acetate added. (H) Silver chromate crystals in silica gel.

41

FIGURE 2.2: Silver chromate precipitate in Petri dishes.
(A) Silver dichromate dendritic crystals in agarose. (B) Silver dichromate Liesegang bands in gelatin.

42

FIGURE 2.3: Comparison of iron phosphate precipitates in agarose and silica gel.
(A) Iron and phosphate form bands in purified agarose, which turn green in the process of oxidizing. (B) A
solution of hydroxide replacing the phosphate in already-formed bands in agarose, causing a color change
from green to yellow/orange. (C) Iron and phosphate form bands in agarose. Fe
+2
in high concentration in
solution causes oxidation to occur in between the bands, obscuring the pattern. (D) Iron and phosphate
form bands and crystals in silica gel.

43

FIGURE 2.4: Effects of pH changes on iron compound precipitation in silica gel and agarose.
(A) Precipitation and oxidation of iron hydroxide in agarose. (B) Fe
+2
in agarose reacting with a solution of
saturated NaHCO
3
, at pH 7.94. Continuous white precipitate and indistinct rhythmic bands were observed,
and after oxidation several thick layers of various colored (hydr)oxides were formed, obscuring the original
rhythmic bands. (C) Fe
+2
in agarose reacting with a solution containing equal molarities of
NaHCO
3
/Na
2
CO
3
. Homogenous precipitate forms, that later oxidizes and forms fine layers of iron
(hydr)oxides. (D) Fe
+2
in agarose reacting with K
2
HPO
4
/ NaOH in solution. Precipitates similar to (A)
form first, followed by patterns similar to the phosphate experiments in Fig. 3. (E) Fe
+2
in agarose reacting
with a saturated solution of Na
2
CO
3
at high pH, so two reactants (carbonate and hydroxide) are present.
Patterns similar to (A) form first, followed by a slightly banded pattern that resembles that formed in the
carbonate experiment.

44

Precipitate Medium Reactant Concentration Pattern
B K
2
CrO
4
5 mM
Agarose
A AgNO
3
6 M
Bands composed of small
crystals
B K
2
CrO
4
5 mM
Purified agarose
A AgNO
3
6 M
Small crystals, no banding
B K
2
CrO
4
5 mM
Gelatin
A AgNO
3
6 M
Sharp bands composed of non-
crystalline particles
B K
2
CrO
4
5 mM
Silver chromate
Silica gel
A AgNO
3
6 M
Bands of crystals
B K
2
Cr
2
O
7
5 mM
Agarose
A AgNO
3
6 M
Continuous dendritic crystals
B K
2
Cr
2
O
7
5 mM
Gelatin
A AgNO
3
6 M
Rhythmic bands
B K
2
Cr
2
O
7
5 mM
Silver
dichromate
Silica gel
A AgNO
3
6 M
Bands of crystals

TABLE 2.1: Silver chromate precipitation experiments.

45
Precipitate Gel Reactants Concentration pH* Pattern Notes
B FeSO
4
5 mM 5.5
Agarose
A NaOH 0.1 M 13
Continuous
precipitate
Oxidation begins
immediately, turning
grey precipitate to
orange
B FeSO
4
5 mM **
Iron
hydroxide
Silica gel
A NaOH 0.1 M 13
Continuous
precipitate
Oxidation begins
immediately, turning
grey precipitate to
orange
B FeSO
4
5 mM 5.5
A K
2
HPO
4
saturated 10.10
Rhythmic
bands
Bands of white
precipitate form,
turning green after
oxidation
B FeSO
4
5 mM 5.5
A K
2
HPO
4
saturated 10.10
Rhythmic
bands
1 ml of 0.1 M NaOH
solution was added
after bands of Fe-
phosphate formed
B K
2
HPO
4
5 mM 9.20
Agarose

A FeSO
4
crystals n/a
Rhythmic
bands
Oxidation turns bands
yellow, and oxidation
obscures the pattern
B FeSO
4
5 mM 5.5
Purified
agarose A K
2
HPO
4
saturated 10.10
Rhythmic
bands
Precipitate is
indistinguishable from
that in unpurified
agarose
B FeSO
4
5 mM **
Iron
phosphate
Silica gel
A K
2
HPO
4
saturated 10.10
Bands of
crystals;
crystal size
increases
toward bottom
of tube
Bands of white crystals
form, turning green
after oxidation
B FeSO
4
5 mM 5.5
A NaHCO
3
saturated 7.94
Continuous
precipitate
followed by
some slight
banding
Oxidation begins
immediately, turning
grey precipitate to
orange; bands form but
are later obscured
B FeSO
4
5 mM 5.5
Agarose

A
Na
2
CO
3

+
NaHCO
3

saturated 9.66
Continuous
precipitate
Oxidation produces
many fine bands of
orange compounds
(likely Fe-
oxides/hydroxides).
B FeSO
4
5 mM **
Iron
carbonate
Silica gel
A NaHCO
3
saturated 7.94
Continuous
precipitate
Oxidation begins
immediately, turning
grey precipitate to
orange
B FeSO
4
5 mM 5.5
Iron
hydroxide
+ iron
phosphate
Agarose
A
K
2
HPO
4
+
NaOH
saturated 12.0
Continuous
precipitate
followed by
banding
The first (continuous)
precipitate turns
orange and the second
(banded) precipitate
turns green as
oxidation occurs.
B FeSO
4
5 mM 5.5 Iron
hydroxide
+ iron
carbonate
Agarose
A Na
2
CO
3
saturated 11.10
Continuous
precipitate
followed by
some banding
Oxidation begins
immediately, turning
grey precipitate to
orange
TABLE 2.2: Iron compound precipitation.
*pH of reactant B is the pH of the liquid that agarose/gel powder is added to, unless otherwise noted.
**pH of silica gels is constantly changing as they age (Henisch 2005).

46

2.4. Discussion

2.4.1. Formation of Banded Precipitates

Since the discovery of banded precipitation patterns in gels by Liesegang over one
hundred years ago (Liesegang, 1896), many theories have been proposed to explain the
underlying mechanisms of rhythmic precipitation (Henisch, 2005, and references
therein). In gel diffusion experiments, periodic band formation appears to be the result of
a moving reaction front as the outer electrolyte (A) diffuses through the gel and reacts
with (B) to form a precipitate when the concentration product exceeds some nucleation
threshold. In a diffusion-controlled system where reactant (A) is at significantly higher
concentration than (B), formation of the first precipitate acts as a sink to deplete the
immediate area of (B) and nucleation of new particles is prevented, leading to a zone of
no precipitate (see Fig. 1.3). At a certain distance from this first band, the concentration
of B increases enough that the nucleation threshold is exceeded and precipitation occurs
again (Ortoleva, 1994). Another mechanism affecting pattern formation is competitive
particle growth (also known as Ostwald ripening): since smaller particles are more
soluble than larger ones, small particles can dissolve and larger particles will grow at
their expense (Lifshitz and Slyozov, 1961; Muller and Ross, 2003; Ortoleva, 1984). This
leads to an increase in average particle size and a decrease in average particle density as
the gel system matures with time (Morse and Casey, 1988), and this competitive particle

47
growth can lead to morphological changes in the precipitate as long as the gel remains
aqueous. Such "maturation" phenomena are commonly observed in gel precipitation
experiments (e.g., Panjarian and Sultan, 2001) and can lead to the dissolution of bands
after they have formed and can cause large bands to split into several smaller ones.
Although the physical mechanisms of Liesegang band formation in gels have
been investigated many times (Henisch, 2005, and references therein; Ortoleva, 1994)
and the general process is fairly well understood, the morphological effect of interaction
of other ions with the primary reactants represents an area of the unknown to this point.
In these experiments we have observed that impure gel compositions as well as non-
precipitating ions present in the solution appear to encourage the formation of banded
precipitates. Silver chromate and silver dichromate produce morphologically distinct
precipitates; however, in gelatin both compounds precipitate in bands, and in agarose
both precipitate in crystal form. There are considerably more organic impurities present
in gelatin than agarose and this may be the cause of the banding pattern produced. This is
also seen when comparing the result of silver chromate precipitation in purified versus
unpurified agarose. When the agarose was purified, the banding normally produced in
agarose was suppressed and crystals only precipitated continuously (Fig. 2.1F), but the
addition of specific non-precipitating ions (sodium acetate) to purified agarose caused
banding to occur again (Fig. 2.1H). Acetate ions (CH
3
COO
-
) are present in our silica gels
as well, where silver chromate crystals also precipitate in a banded pattern (Fig. 2.1H).
One hypothesis that is consistent with these observations is that the negatively charged
acetate ions present in solution inhibit silver chromate crystal formation by binding to

48
positively charged silver ions, interfering with the addition of chromate. When low (mM)
concentrations of chromate anions are distributed throughout the gel in our experiments,
even a small amount of interfering ions could have a discernable effect and prevent large
crystal formation. By suppressing the growth of crystals by binding to crystal faces,
impurities may cause precipitation to occur in smaller crystals that aggregate to form
rhythmic bands. In this way any negatively charged impurities that are present in solution
could affect the final pattern morphology, even if they are not part of the dominant
precipitation reaction.
In the case of gelatin, a gel composed of collagen fragments, banding may be
produced by interaction of the primary reactants with negatively-charged carboxyl groups
at the end of peptides. Gelatin contains many more soluble ionic impurities than agarose,
and these could effectively suppress the growth of existing crystals. However, increased
gel concentration may also be a factor. In a lead iodide precipitation system in agarose
gel, Toramaru et al. (2003) observed that increasing agarose concentration caused
crystalline precipitates to become more banded. It is possible that this is because
increasing the agarose concentration also increased the level of soluble impurities. An
alternate explanation is that more concentrated gels provide more nucleation sites and
thus promote nucleation of new particles instead of the growth of existing ones (Lagzi
and Ueyama, 2009; Toramaru et al., 2003). In our experiments, gelatin provides more
nucleation sites than agarose (since the gel is more concentrated - 3.0% gelatin compared
to 0.5% agarose) and this could lower the supersaturation needed to initiate nucleation,

49
leading to precipitation in a high number density of small particles instead of growth of
large crystals.
In our iron phosphate experiments, precipitation occurred in rhythmic bands
whether or not the gel is initially free of non-precipitating ions. In the iron-phosphate
reaction system, there are ions besides the primary reactants present whether or not the
gel is purified, because of the pH of the phosphate solution. In the iron phosphate
experiment the phosphate solution was made with concentrated K
2
HPO
4
which produced
a pH of 9.53, and according to the equilibrium reaction between HPO
4
2-
and PO
4
3-
,
K1 K2 K3
H
3
PO
4
 H
2
PO
4
-
+ H
+
 HPO
4
2-
+ H
+
 PO
4
3-
+ H
+

pK
1
= 2.13 pK
2
= 7.21 pK
3
= 12.32

at this pH there should be about 200 times as much HPO
4
2-
as PO
4
3-
. The visible
precipitate is likely Fe
3
(PO
4
)
2
since FeHPO
4
is soluble, so it is possible that the excess
HPO
4
2-
ions in solution are interfering with precipitation of Fe-phosphate similar to how
agarose impurities interfere with silver chromate crystal growth. Rhythmic banding was
consistently produced in iron phosphate precipitation experiments, regardless of the
positions of the electrolytes or the type of gel used. A possible explanation for this may
be that HPO
4
2-
ions could bind to Fe
+2
and prevent larger crystals of Fe
3
(PO
4
)
2
from
forming, a phenomenon which would be independent of gel characteristics.

50
2.4.2. Effect of Initial Chemical Conditions on Precipitate Morphology

Other factors that affect the morphology of precipitation structures in our
experiments include which primary reactants are present in solution, chemical
replacement, and oxidation of iron compounds. It is clear that different reaction systems
(silver chromate versus silver dichromate, for example) lead to very different
morphological results, even when other properties such as gel type and ion concentrations
are held constant. One obvious example of this is the larger continuous dendritic crystals
formed by precipitation of silver dichromate in agarose, compared to the much smaller
crystals formed by precipitation of silver chromate. Likewise, the rhythmic bands
produced by silver chromate in gelatin have blurry edges and are generally less distinct
than the bands produced by silver dichromate in gelatin. Characteristic morphologies
corresponding to specific reactions are seen in our iron precipitation experiments as well;
for example the precipitation of iron phosphate in bands (which, incidentally, occurs no
matter which type of gel we use, or whether the iron or the phosphate is the ion present in
the gel) compared to the colloidal, continuous precipitate formed by iron hydroxide in
agarose. When two anions were introduced in the same solution, as in Figure 2.4D
(hydroxide and phosphate) and Figure 2.4E (hydroxide and carbonate), the characteristic
morphologies of each separate reaction formed in order of the solubility of the respective
compounds. For example, a saturated phosphate solution adjusted to pH 11.98 with
sodium hydroxide applied to an agarose gel containing Fe
+2
, produced first the expected
continuous iron-hydroxide/oxide precipitate, then a “dead zone” where no precipitate was

51
visible, then bands similar to those produced in the iron phosphate experiments (Fig.
2.4D). Even when multiple precipitation reactions occurred in the same tube, the
precipitation morphology of a given compound was reproducible.
One question of interest is whether changes in chemical composition will alter the
morphology of a precipitation pattern. In the experiment in which we replaced banded
iron phosphate with iron hydroxide/oxide (Fig. 2.3B), the color of the mineral changed
from green to yellow/orange, and when the second reaction front had completely passed
through, the bands were still preserved. In the experiment in which iron and hydroxide
were the only primary reactants, only a continuous precipitate formed, and although this
precipitate later separated into several bands (likely due to Ostwald ripening since band
formation occurred before oxidation), these were morphologically distinct from the
rhythmic bands formed by iron phosphate. This result suggests that the initial reactants
in a gel system determine the morphology of the precipitate, even if composition is later
altered.
In all our experiments the tubes were exposed to air, so that atmospheric oxygen
could diffuse into the gels. The initial precipitation reaction occurred throughout the gel
in a matter of days, but the oxidation of precipitates took place over several weeks. The
result is a progressive change in oxidation along the length of the tube and a sharp
boundary between minerals containing oxidized vs. reduced iron, as seen in Figure 2.4A
(showing precipitation of iron hydroxide). Oxidation of the chemical precipitates
resulted in an obvious color change from white/grey to orange in the iron hydroxide and
iron carbonate experiments, and from white/grey to green in the iron phosphate

52
experiments. Another interesting effect of the oxidation of iron precipitates was the
formation of thin layers of yellow and orange compounds, presumably iron
oxides/hydroxides, throughout a continuous precipitate where previously no banding had
been visible (Figs. 2.4B, 2.4C). Similar banding was observed by Stone and Goldstein
(2004) in experiments in which ammonia diffused into a gel containing iron sulfate.
They observed various bands of oxides/hydroxides forming as an oxidation gradient was
established. The layered bands in our iron hydroxide and carbonate precipitation
experiments also probably resulted from stratification of iron oxides/hydroxides as the
oxidation front traveled throughout the gel. These layered bands underwent a maturation
process, slightly changing thickness and color, until the gel had dried out.
The eventual oxidation of all iron precipitates also further distinguished the iron-
limited and iron-concentrated phosphate experiments from one another (Figs. 2.3A and
2.3C, respectively). We performed precipitation experiments first with iron in the gel and
phosphate in concentrated solution, and then reversed the electrolytes so that phosphate
was in the gel and iron was introduced in concentrated form (in this case, ferrous sulfate
crystals). The reactant in the gel was at 5 mM concentration while the second reactant
was at much higher concentration, so in the first experiment there was an abundance of
phosphate ion and in the second experiment there was an abundance of iron. Rhythmic
bands of iron phosphate formed regardless of the positions of the electrolytes, and in the
iron-limited experiment (Fig. 2.3A) the bands turned green after oxidation. However in
the iron-concentrated experiment (Fig. 2.3C), as atmospheric oxygen diffused into the
system, it oxidized not only the precipitate in the bands but also the gel in between the

53
bands because excess Fe
+2
ions were abundant in solution. This obscured the banded
pattern by the time oxidation was complete. The end morphological results for these two
experiments were therefore very different, even though the same chemical reaction was
taking place.

2.5. Conclusions

In this study we find that a variety of self-organized precipitation patterns are
produced in gels, by varying reactants, gel type, and amount of ionic impurities.
Ionic components in the gel had a powerful effect on the banding observed in the
formation of ionic precipitates. In the case of agarose, we have been able to remove the
soluble ionic components by simply extracting with water. It appears that in the case of
agarose, the soluble ionic impurities were responsible for the banding observed in silver
chromate experiments. We then presume that in the case of gelatin, the banding observed
resulted from the ionic components of the gel, some of which may be in solution and
some of which were part of the matrix itself, and the same may be true for silica gel.
We were able to replace an already-formed iron phosphate precipitate with an iron
hydroxide precipitate by introducing hydroxide as a concentrated solution atop the gel.
The pattern was not altered and was characteristic of the original reactant in the system.
This replacement process is analogous to fossilization. When oxidation of iron hydroxide
or carbonate precipitates occurred, the morphology of the precipitate changed

54
dramatically, producing layers of oxidized iron compounds in a previously
undifferentiated precipitate.
Our work described here marks the beginning of a more detailed investigation
into the factors (physical and chemical) that affect morphology of chemical precipitates
in gels. These few experiments have shown that an incredible amount of morphological
variety can be produced in simple gel diffusion/precipitation systems. Further study is
needed to classify such precipitation patterns into useful groups based on morphology, to
increase their predictive capability by determining which features of a precipitate can be
used to gain information about the chemical conditions in which it was deposited.
Specifically, our future studies in this area will include determination of which ionic
properties best induce banding, in systems containing biomolecules such as amino acids
or carboxylic acids.

55
Chapter 2 References
Chan, M. A., Beitler, B., Parry, W. T., Ormö, J., & Komatsu, G. (2004). A possible
terrestrial analogue for haematite concretions on Mars. Nature, 429, 731-734.

Chan, M. A., Beitler, B., Parry, W. T., Ormö, J., & Komatsu, G. (2005). Red rock and red
planet diagenesis: Comparisons of Earth and Mars concretions. GSA Today,
15(8), 4-10.

Cochran, K. A., & Elmore, R. D. (1987). Paleomagnetic dating of Liesegang bands.
Journal of Sedimentary Petrology, 57(4), 701-708.

Feeney, R. E., Petersen, I. M., & Sahinkaya, H. (1956). “Liesegang-like” rings of growth
and inhibition of bacteria in agar caused by metal ions and chelating agents.
Journal of Bacteriology, 73, 279-283.

Fujikawa, H. (1992). Periodic growth of Bacillus subtilis colonies on agar plates. Physica
A, 189, 15-21.

Golding, I., Kozlovsky, Y., Cohen, I., & Ben-Jacob, E. (1998). Studies of bacterial
branching growth using reaction-diffusion models for colonial development.
Physica A, 260, 510-554.

Heaney, P. J., & Davis, A. M. (1995). Observation and Origin of Self-Organized
Textures in Agates. Science, 269, 1562-1565.

Henisch, H. K. (2005). Crystals in Gels and Liesegang Rings. Cambridge: Cambridge
University Press.

Kozlovsky, Y., Cohen, I., Golding, I., & Ben-Jacob, E. (1999). Lubricating bacteria
model for branching growth of bacterial colonies. Physical Review E, 59(6),
7025-7035.

Krug, H. J., Brandtstädter, H., & Jacob, K. H. (1996). Morphological instabilities in
pattern formation by precipitation and crystallization processes. Geologische
Rundschau, 85, 19-28.

Lagzi, I., & Ueyama, D. (2009). Pattern transition between periodic Liesegang pattern
and crystal growth regime in reaction-diffusion systems. Chemical Physics
Letters, 468, 188-192.

Lander, A. D. (2007). Morpheus Unbound: Reimagining the Morphogen Gradient. Cell,
128, 245-255.

56
Liesegang, R. E. (1896). Naturwiss. Wochenschr. 11, 353.

Lifshitz, I. M., & Slyozov, V. V. (1961). The Kinetics of Precipitation From
Supersaturated Solid Solutions. Journal of Physics and Chemistry of Solids,
19(1/2), 35-50.

McBride, E. F., Abdel-Wahab, A., & El-Younsy, A. R. M. (1999). Origin of spheroidal
chert nodules, Drunka Formation (Lower Eocene), Egypt. Sedimentology, 46,
733-755.

Mimura, M., Sakaguchi, H., & Matsushita, M. (2000). Reaction-diffusion modeling of
bacterial colony patterns. Physica A, 282, 283-303.

Morse, J. W., & Casey, W. H. (1988). Ostwald Processes and Mineral Paragenesis in
Sediments. American Journal of Science, 288, 537-560.

Msharrafieh, M., & Sultan, R. (2005). Patterns with High Rhythmicity Levels in
Multicomponent Liesegang Systems. ChemPhysChem, 6, 2647-2653.

Muller, S. C., & Ross, J. (2003). Spatial Structure Formation in Precipitation Reactions.
Journal of Physical Chemistry A, 107, 7997-8008.

Ortoleva, P. (1984). The Self Organization of Liesegang Bands and Other Precipitate
Patterns. In: G. Nicolis and F. Baras (eds.), Chemical Instabilities: Applications
in Chemistry, Engineering, Geology, and Materials Science, 289-297.

Ortoleva, P. (1994). Geochemical Self-Organization. Oxford: Oxford University Press.

Panjarian, S., & Sultan, R. (2001). Crystal Selection and Liesegang Banding in Dynamic
Precipitate Systems. Collection of Czechoslovak Chemical Communications, 66,
541-554.

Rauprich, O., Matsushita, M., Weijer, C. J., Siegert, F., Esipov, S. E., & Shapiro, J. A.
(1996). Periodic Phenomena in Proteus mirabilis Swarm Colony Development.
Journal of Bacteriology, 178(2), 6525-6538.

Ripszam, M., Nago, A., Volford, A., Izsak, F., & Lagzi, I. (2005). The Liesegang eyes
phenomenon. Chemical Physics Letters, 414, 384-388.

57
Squyres, S. W., Grotzinger, J. P., Arvidson, R. E., Bell III, J. F., Calvin, W., Christensen,
P. R., Clark, B. C., Crisp, J. A., Farrand, W. H., Herkenhoff, K. E., Johnson, J. R.,
Klingelhöfer, G., Knoll, A. H., McLennan, S. M., McSween Jr., H. Y., Morris, R.
V., Rice Jr., J. W., Rieder, R., & Soderblom, L. A. (2004). In Situ Evidence for
an Ancient Aqueous Environment at Meridiani Planum, Mars. Science, 306,
1709-1714.

Stone, D. A., & Goldstein, R. E. (2004). Tubular precipitation and redox gradients on a
bubbling template. Proceedings of the National Academy of Sciences, 101(32),
11537-11541.

Toramaru, A., Harada, T., Okamura, T. (2003). Experimental Pattern Transitions in a
Liesegang system. Physica D, 183, 133-140.

Volford, A., Izsak, F., Ripszam, M., & Lagzi, I. (2007). Pattern Formation and Self-
Organization in a Simple Precipitation System. Langmuir, 23(3), 961-964.

58
Chapter 3. Precipitation Patterns Formed by Self-Organizing Processes
in Porous Media: Comparison of Laboratory and Field Examples
*

Summary

We compare laboratory and field examples of self-organized mineral precipitates
in porous media. Laboratory tests of silver chromate precipitation in glass beads and
glass bead/gel mixtures produce structures such as finger fluid fronts, periodic banding,
and millimeter-size spherules. These are morphologically similar to the varied forms of
iron oxide precipitates in the Jurassic Navajo Sandstone, UT, that preserve records of
fluid flow events in a porous and permeable sandstone. Experimental studies of periodic
precipitates in porous media can provide valuable insight for understanding the
diagenetic history of similar precipitates in natural environments.

3.1. Introduction

In this paper we examine precipitates that form via self-organizing processes in
porous and permeable media. Self-organizing patterns in diffusion-controlled chemical
systems have been studied for over 100 years, since the discovery by R. E. Liesegang in
1896 that the interaction between inter-diffusing silver and chromate ions in a gelatin gel
can produce periodically-spaced bands of precipitate. These periodic patterns, termed
"Liesegang bands", have since been produced using various combinations of reactants

*
Chapter 3 has been submitted to Geofluids with an authorship of: L. M. Barge, D.
Hammond, S. Potter, M. A. Chan, J. Petruska, and K. Nealson.

59
(Henisch, 2005, and references therein) and in many different types of gels, and the
pattern morphology can vary widely depending on initial experimental conditions. While
the spacing of bands may not be strictly "periodic" in the spatial sense, we will retain the
original description of Liesegang (1896).
The silver chromate reaction system initially utilized by Liesegang has since been
extensively studied and the effects of many parameters on silver chromate and
dichromate precipitation patterns are known. The main function of the gel in such
experiments is to maintain a diffusion-controlled system by eliminating convection, and
precipitates are held in place by the gel so that patterns can form. However, the physical
and chemical properties of the gel do have effects on the pattern formation; parameters
that have been observed to affect formation of periodic precipitates include type of gel
(Henisch, 2005; Lagzi and Ueyama, 2009), gel concentration (Toramaru et al., 2003), and
soluble impurities in the gel (Chapters 2 and 4). Precipitation of silver chromate in
agarose gel or silica gel typically produces crystals, and in gelatin produces periodic
bands (Chapter 2; Henisch, 2005). The transition between crystals and banding in gel
systems has to do with relative rates of heterogeneous and surface nucleation, which in
turn can depend on the chemical properties of the medium (e.g., gel concentration) (Lagzi
and Ueyama, 2009; Toramaru et al., 2003). Periodic precipitation of silver chromate has
also been observed in inhomogeneous media such as partially-coagulated gelatin (Stern,
1954) and in beds of glass beads (Stern and Shniad, 1958). The formation of Liesegang
patterns has been attributed to various mechanisms including supersaturation-nucleation-
depletion processes (in which the first precipitate depletes the immediate area of one

60
reactant, resulting in the next precipitate being deposited some distance from the first
(Ostwald, 1897; Smith, 1984)) and competitive particle growth (in which larger particles
are more insoluble and grow at the expense of smaller ones - a phenomenon known as
Ostwald ripening (Lifshitz and Slyozov, 1961; Muller and Ross, 2003; Ortoleva, 1982)).
Self-organizing processes that can produce periodic patterns in laboratory gels
also have applications for organized patterning in natural systems. There are many
examples of periodic patterning in geology, such as banding and crystal growth in agate
(Heaney and Davis, 1995), banding of precipitates in rock or sediment (e.g., Chan et al.,
2000; Cochran and Elmore, 1987; Marko et al., 2003), or periodically-distributed
spheroidal concretions that lack an obvious nucleus (Chan et al., 2007; Squyres et al.,
2004), which have been attributed to chemical self-organizing processes operating while
reactive fluids permeate a porous medium. Liesegang band formation in gel diffusion
experiments is well understood; however, in geological systems, self-organized
precipitation morphologies will also be affected by the inhomogeneous nature of
sand/sediment as well as chemical properties of the permeating fluid(s).
One particularly well-exposed example of how varying fluid flow conditions can
affect the morphology of self-organized chemical precipitates is the Navajo Sandstone
formation in the western USA. Previous studies of this region (Chan et al., 2000) indicate
that a reducing fluid (probably hydrocarbon) infiltrated the reservoir along blind thrust
faults and mobilized the iron that had been precipitated syndepositionally as hematite
grain coatings in the host rock. Iron oxide/hydroxide precipitation later occurred upon
contact with a more oxidizing body of water (Beitler et al., 2005). Many geometries of

61
precipitates occur in this region, including Liesegang bands and vast numbers of tiny
(millimeter-sized) spheroidal concretions that are likely nucleation centers (Chan et al.,
2007) (these have been suggested as an analog for the hematite concretions, informally
called "blueberries", that have been found on Mars (Chan et al., 2004; Squyres et al.,
2004)). Chan et al. (2007) studied the formation of self-organized precipitates in the
Navajo Sandstone, with numerical models of how bands and periodically-distributed
nucleation centers form in a heterogeneous sand medium via Liesegang-type processes,
as well as laboratory bench tests of iron hydroxide/oxide precipitation in agarose gel to
simulate how concretions might grow after nucleation.
Experimentally, however, the nucleation and growth of spherical concretions in
porous media requires further investigation, since the processes that operate to produce
rhythmic banding in gel diffusion experiments can lead to other morphologies when
operating in sand or sediment. The Chan et al. (2007) numerical model of concretion
nucleation predicts that relative rates of advection and diffusion determine the number
and spacing of self-organized concretion "nuclei", as well as the morphological transition
between spherules and bands. However, the pattern transition between periodically-
spaced nucleation centers (concretions) and Liesegang bands also depends on fluid
chemistry, and physical properties of the porous medium. The aim of this study is to
investigate how physical and chemical parameters affect the nucleation and growth of
self-organized precipitates in porous media. In particular, we examine the effect of rates
of heterogeneous and surface nucleation, and effects of grain size, on the morphology of
periodic precipitates. These laboratory results are compared to field samples of self-

62
organized precipitates from the Navajo Sandstone, where there are a variety of
geometries present from different chemical reaction fronts, and established literature.

3.2. Experimental

Our approach investigates the formation of precipitates in inhomogeneous media
that may develop into spheroidal "concretions" or Liesegang bands, depending on initial
conditions. We hypothesize that the grain size of the porous medium and the amount of
soluble impurities will be two important factors affecting the transition from spherules to
bands.
The Chan et al. (2007) approach demonstrated formation of periodically-spaced
nucleation centers in numerical models of Fe and oxygen cross-diffusion, and simulated
concretion growth in laboratory experiments that precipitated Fe minerals in agarose gels.
Ferrous or ferric iron has previously been used as a reactant in Liesegang-type
experiments (Chapter 2, Chan et al., 2007, Stone and Goldstein, 2004), but the eventual
oxidation of precipitates changes the pattern morphology due to stratification of various
oxides (Chapter 2, Stone and Goldstein, 2004). To avoid the complications of aqueous
iron chemistry and oxidation of precipitates, we chose to investigate precipitation in
porous media using the silver nitrate/potassium chromate system, which has been more
widely used in studies of periodic precipitation and so the effects of many parameters on
precipitation morphology are known. An additional advantage to using the silver
nitrate/potassium chromate system is that it produces a red precipitate which allows easy

63
observation in a column of glass beads (unlike many iron precipitates, which are colorless
or white). As is typical of Liesegang band experiments, we used quite high reactant
concentrations compared to what would be found in natural systems; however, this
allows us to test the effects of other parameters on precipitate morphology on the short
time scales feasible for laboratory experiments.
Pattern formation was observed in porous media where the following reaction
occurs to produce an insoluble precipitate:
2Ag
+
+ CrO
4
-2
→ Ag
2
CrO
4

Precipitate formation occurred in tubes of gel (gelatin, agar, agarose, or silica
gels), glass beads, and glass bead/gel mixtures. In all experiments the diffusion medium
contained dissolved potassium chromate, and after preparation silver nitrate solution of
fixed concentration (6 M) was placed onto the glass bead or gel interface. We tested the
effects of two factors - grain size and gel type - on the resulting morphology of
precipitate.

3.2.1. Glass Bead Experiments

Glass beads (Sigma-Aldrich, Biospec) were used to simulate a compacted sand or
sandstone environment in which reactants can diffuse from one pore to another, and
convection may occur when pore sizes are large enough. Four sizes of glass beads were
used (100 µm, 150-212 µm, 212-300 µm, and 450-600 µm). This grain size range of

64
100-600 µm is typical of quartz grain sizes in the Navajo Sandstone concretions (Busigny
and Dauphas, 2007).
Glass beads were poured into tubes containing 3-4 ml of 5 mM potassium
chromate solution. The tube was tapped intermittently to compact the glass beads as
much as possible. Once the glass beads reached the 6 ml mark the supernatant liquid was
removed, leaving a column of glass beads saturated with potassium chromate solution.
150 µl of 6 M silver nitrate solution was added to the top of the glass bead column and
allowed to diffuse or advect through to react with the chromate. Precipitates were
monitored after 5 days.
The porosity of the glass bead columns was measured by filling a centrifuge tube
to the 5 ml mark with deionized water, then adding beads to the 5 ml mark (tapping the
tube to compact as much as possible) and measuring the displaced water which is roughly
equal to the grain volume. Porosity was then calculated as: φ = (V(t) - V(g))/V(t) where
V(t) is the total volume (= 5 ml) and V(g) is the average grain volume (= average volume
of water displaced by the glass beads). Average porosities for the four glass bead sizes
were as follows: 1) 100 µm, φ = 40%; 2) 150-212 µm, φ = 39%; 3) 250-300 µm, φ =
37%; 4) 450-600 µm, φ = 37%.
The relative diffusivities through columns of different sized glass beads were
estimated with conductivity measurements. Conductivity was measured across 1) a U-
tube of 0.745 g/L KCl solution and 2) a U-tube filled to the same mark with glass beads
saturated with 0.745 g/L KCl solution. This ratio of the pore solution conductivity σ
p
to
the bulk conductivity σ
b
is the formation factor F, which is approximately equal to the

65
ratio of the diffusion coefficient of an ion in pore solution (D
p
) to the bulk diffusion
coefficient of that ion (D
b
):
!
F =
"
p
"
b
=
D
p
D
b

The average formation factors measured for the four glass bead sizes were: 1) 100 µm, F
= 2.9; 2) 150-212 µm, F = 2.0; 3) 250-300 µm, F = 2.5; 4) 450-600 µm, F = 2.8. This
indicates that in solution (and in a gel, which should have similar diffusion coefficient to
pure solution), silver and chromate ions diffuse between 2 and 3 times faster than in a
column of glass beads.

66

Glass bead size Porosity Formation factor
100 µm 0.40 2.9
150-212 µm 0.39 2.0
250-300 µm 0.37 2.5
450-600 µm 0.37 2.8

TABLE 3.1: Porosity and formation factor measured for compacted glass beads of various sizes.

67
3.2.2. Gel Experiments

Gels are commonly used in periodic precipitation experiments because their small
pore size prevents bulk movement of fluid while allowing diffusion of small molecules
and ions, and prevents product particles greater than a certain size from moving from
their sites of formation. We performed silver chromate precipitation experiments in four
types of gel (silica gel, agarose, agar, and gelatin) to test effects of gel type on
precipitation morphology. Silica gel is the most geochemically plausible gelatinous
material, since quartz and other silicates may have originated from a silica gel precursor
(Garcia-Ruiz, 1994). Silica gel is an inorganic matrix composed of polymerized acidified
silicate that contains some silicate ions in solution, and also anions of the acid used to
create the gel (e.g., acetic acid). Agarose, a simple organic matrix obtained from
seaweed consisting mainly of methylated cellulose, creates a charge-neutral diffusion-
controlled environment. Agarose gels contain a small amount of impurities but agarose is
typically purified to have a very low ionic content and therefore lower conductivity
suitable for use in gel electrophoresis. Gelatin and agar gels were used to test the effects
of precipitation in a diffusion medium that contains many organic and inorganic ionic
impurities. Gelatin in particular is known for suppressing surface nucleation and
promoting the formation of periodic bands (Henisch, 2005; Lagzi and Ueyama, 2009).
The chemical composition of the gel partly controls the type of nucleation that is favored
(heterogeneous vs. surface nucleation (Lagzi and Ueyama, 2009)), and these different gel

68
types were used to investigate effects of nucleation mechanism on the precipitate
morphology.
All gels were prepared to contain 2 mM potassium chromate (to obtain a similar
total amount of chromate in 6 ml of diffusion medium, since the glass bead experiments
had about 40% porosity).
Silica gels were prepared by combining two solutions: an acid solution of 360 µl
glacial acetic acid added to 7.6 ml of 4.2 mM potassium chromate solution, and a sodium
silicate solution prepared by diluting 1.25 ml of commercial water glass (containing
~10.6% Na
2
O and ~26.5% SiO
2
) to 8 ml with deionized H
2
O. The sodium silicate
solution was poured into the acid solution in a 20 ml tube. The tube was sealed and then
inverted slowly (not shaken) five times to thoroughly mix the gel. (The 4.2 mM
potassium chromate solution is used to produce a 2 mM concentration in 16 ml of silica
gel; however, the acid and silicate solutions must be kept separate during preparation.)
The liquid gel is then poured into 6 ml tubes, and allowed to gel for 24 hours.
Agarose, agar, and gelatin gels were prepared by mixing gel powder with 2 mM
potassium chromate solution and bringing to a boil. 0.5% agarose gels (Molecular
Biology Certified, Shelton Scientific), 0.5% agar gels, and 3.0% gelatin gels (commercial
Knox gelatin) were used. Once heated, gels were poured into 6 ml test tubes and allowed
to set for 24 hours. All gels were allowed to set at room temperature, except for gelatin
which was kept at 4°C.

69
In all experiments, 150 µl of 6 M silver nitrate solution was added to the top of
the gel column after it had set. A visible precipitate formed immediately at the gel
interface, and pattern formation was monitored after 5 days.

3.2.3. Gel / Glass Bead Mixtures

Experiments were performed in which columns of glass beads were saturated with
gel containing potassium chromate. This is similar to the pore fluid experiments (2.1), but
the inclusion of gel in the pores of the glass bead matrix maintains a diffusion-controlled
environment, even when larger glass beads are used. This allows self-organized patterns
such as concretions or bands to form in the glass bead columns where convection might
occur in the absence of gel.
Experiments were done with all combinations of the four glass bead sizes (100
µm, 150-212 µm, 212-300 µm, and 450-600 µm) and the four types of gel (silica gel,
agarose, agar, and gelatin) mentioned above.
Gels were prepared in the same manner as described in section 2.2, except all gels
were made to contain 5 mM of potassium chromate. Silica gels were prepared by
combining an acid solution (360 µl glacial acetic acid + 7.6 ml of 10.5 mM potassium
chromate solution) and a sodium silicate solution (prepared by diluting 1.25 ml of
commercial water gel to 8 ml with deionized H
2
O) in a 20 ml tube. The tube was sealed
and then inverted slowly (not shaken) five times to thoroughly mix the gel. This
produces 16 ml of silica gel with a final potassium chromate concentration of 5 mM;

70
however, the acid and silicate solutions must be kept separate during preparation. 0.5%
agarose, 0.5% agar and 3.0% gelatin gels were prepared by adding gel powder to 5 mM
potassium chromate solutions and bringing to a boil.
While gels were still liquid, 3-4 ml of gel was poured into a test tube. Glass beads
were then added to the gel, and the tubes were tapped to compact the beads as much as
possible. After glass beads had been added to the 6 ml mark, any supernatant liquid gel
was removed. The glass bead/gel mixtures were left to set for 24 hours (experiments
containing silica gel, agarose, or agar were allowed to set at room temperature, and
experiments with gelatin were allowed to set at 4°C).
After the gels had hardened, 150 µl of 6 M silver nitrate solution was added to the
top of the gel/glass bead column. Precipitation was observed immediately and was
monitored after 5 days.

71

Diffusion medium Gel type
Glass bead size
(µm)
Precipitate
100 Amorphous spherules
150-212 Finger fluid fronts
212-300 Fluid fronts
Glass beads + 5mM
potassium
chromate solution
n/a
450-600 Continuous precipitate
100 Many crystals and mm-sized spherules
150-212 Small crystals
212-300
Crystals and one spherule (d~2.5 mm)
several cm from the gel interface
Agarose gel (0.5%)
450-600
One large spherule (d~3.5 mm) near the
bottom of the tube
100 Many mm-sized spherules
150-212 Small crystals
212-300 One larger (~3 mm) spherule
Agar gel (0.5%)
450-600 Small crystals
100 Mm-sized spherules and some banding
150-212 Rhythmic bands
212-300 Some banding
Gelatin gel (3.0%)
450-600 One large band
100 Many mm-sized spherules
150-212 No visible precipitate
212-300 Several isolated mm-size spherules
Glass beads + gel
containing 5mM
potassium
chromate
Silica hydrogel
450-600 No visible precipitate
Agarose gel (0.5%) Small crystals
Agar gel (0.5%) Rhythmic bands and crystals
Gelatin gel (3.0%) Rhythmic bands
Gel controls (2mM
potassium
chromate)
Silica hydrogel
n/a
Large crystals

TABLE 3.2: Silver chromate precipitates in gels, glass beads, and glass beads + gel. All gels and pore
solutions that saturated the glass bead columns contained 5 mM potassium chromate solution. In the pure
gel experiments, 2 mM potassium chromate was used to account for the ~40% porosity of the glass bead
columns. In all experiments 150 µl of 6 M silver nitrate solution was added to the top of the glass bead/gel
matrix.

72
3.3. Results

3.3.1. Precipitation in Glass Beads

Precipitation began immediately when silver nitrate was added to the top of the
glass bead column, and was monitored for 5 days. In the largest bead size (425-600-µm),
precipitation occurred in a few seconds throughout the tube. This may be due to fluid
convection (as the silver nitrate solution is more dense than the potassium chromate
solution) or to silver chromate crystals falling through the large pores (Stern and Shniad,
1958) (Fig. 3.1D). In 250-300-µm and 150-212-µm beads, fluid convected in the upper
part of the tube and some finger fluid front shapes were formed (Figs. 3.1B, 3.1C). In
100-µm glass beads, fluid did not convect and small spheroidal precipitates (1-2 mm in
diameter) were observed after 5 days (Fig. 3.1A). These spheroidal precipitates were
examined under a microscope, and they were found to be a coating of precipitate on the
outside of a cluster of glass particles and in the enclosed pores. The glass beads were not
well cemented or otherwise altered within the precipitation region, and they easily
disaggregated.

3.3.2. Precipitation in Gels

In silica gel, silver chromate precipitated in large crystals. In agarose experiments
precipitation occurred in small, randomly oriented, needle-like crystals. In agar gel,

73
silver chromate precipitated in small crystals that aggregated to form periodic bands. In
gelatin, precipitation occurred in tiny non-crystalline particles that formed periodic bands
(Fig. 3.2).

3.3.3. Precipitation in Glass Bead / Gel Mixtures

When glass beads were mixed with silica gel, numerous spheroidal precipitates
ranging from 0.5-2 mm diameter were formed in the 100-µm glass bead size (Fig. 3.3A).
A few spheroidal precipitates (~2.5 mm diameter) were observed in the 250-300-µm
glass beads as well, near the bottom of the tube (Fig. 3.3C).
When glass beads were mixed with agarose, small spheroidal precipitates formed
in the 100-µm glass beads (Fig. 3.4A). Only crystals were observed in the 150-212-µm
beads (Fig. 3.4B). In the 212-300- and 450-600-µm beads, single larger spherules 3-4
mm in diameter formed closer to the bottom of the tube (Figs. 3.4C and 3.4D). In all
glass bead/agarose experiments small evenly distributed crystals formed as well that did
not exhibit any periodicity or spherical shape.
When the glass beads were mixed with agar gel, small spheroidal precipitates 1-3
mm in diameter formed in the 100 µm beads (Fig. 3.5A). Some small crystals formed in
the larger three bead sizes, and in the 250-300-µm beads a single spherule ~3 mm in
diameter formed about one centimeter from the top of the gel column (Fig. 3.5C).
When the glass beads were mixed with gelatin, spheroidal precipitates 1-2 mm in
diameter and several bands formed in the 100-µm grain size (Fig 3.6A). In the other

74
grain sizes spheroidal precipitates did not form, and instead various forms of periodic
bands were observed. Liesegang bands were especially distinct in the 150-212-µm bead
size (Fig 3.6B).
In the 100-µm glass bead/gel mixtures, we also occasionally observed the
nucleation of two or more concretions next to one another that eventually grew to become
fused, a phenomenon known as "twinning" which is commonly observed in the Navajo
Sandstone concretions (Chan et al., 2005a, b).

75

FIGURE 3.1: Silver chromate precipitates in glass bead matrix of various bead size saturated with
potassium chromate solution. A) 100-µm-beads, B) 150-212-µm beads, C) 250-300-µm beads, D) 450-
600-µm beads.

76

FIGURE 3.2: Gel controls. 150 µl of 6 M silver nitrate reacting with 2 mM potassium chromate
distributed throughout the gel. 2 mM chromate concentrations were used instead of 5 mM to account for
the ~40% porosity of the glass bead matrices. A) Silica gel, B) Agarose, C) Agar, D) Gelatin.

77

FIGURE 3.3: Silver chromate precipitates in silica gel and glass beads of varying size. Bead sizes are
A) 100 µm, B) 150-212 µm, C) 250-300 µm, D) 450-600 µm.

78

FIGURE 3.4: Silver chromate precipitates in agarose gel and glass beads of varying size. Bead sizes
are A) 100 µm, B) 150-212 µm, C) 250-300 µm, D) 450-600 µm.

79

FIGURE 3.5: Silver chromate precipitates in agar gel and glass beads of varying size. Bead sizes are
A) 100 µm, B) 150-212 µm, C) 250-300 µm, D) 450-600 µm.

80

FIGURE 3.6: Silver chromate precipitates in gelatin gel and glass beads of varying size. Bead sizes
are A) 100 µm, B) 150-212 µm, C) 250-300 µm, D) 450-600 µm

81
3.4. Field Studies: Self-Organized Precipitates from the Navajo
Sandstone, Southern Utah, USA

In many ways the Navajo Sandstone is an appropriate terrestrial analog to
compare to the experimental results, since the field example shows many forms of
banding and spheroidal precipitates, and the homogeneity of the glass bead medium is
similar to the well-sorted, quartz rich, wind-blown sandstone. The eolian Navajo
Sandstone, deposited during the early Jurassic (~190 ma) is a clean quartz arenite. It
represents the largest erg to ever exist on Earth, originally encompassing an area of over
350,000 km
2
. The modern Navajo Sandstone and its equivalents extend throughout Utah,
and into Wyoming, Arizona, Nevada and Colorado (Blakey et al., 1988; Blakey, 1994).
The Navajo Sandstone hosts iron oxide precipitates that span a wide variety of
morphologies, including abundant Liesegang bands and spheroidal concretions that lack
any obvious nucleus (Chan et al., 2004, 2005a), suggesting nucleation via self-organizing
processes. Field studies were conducted in the Grand Staircase Escalante National
Monument (southern Utah) and Vermillion Cliffs National Monument (northern Arizona)
(Fig. 3.7).
The Navajo Sandstone is a porous and permeable eolian sandstone. Previous
studies of the Nugget Sandstone (the equivalent of the Navajo SS in Wyoming) show that
the porosity ranges from a few percent to approximately 25%. Permeability is widely
variable as well, ranging over five orders of magnitude from hundredths of millidarcies to
Darcies. These measurements shift from the lower porosity/permeablility sabkha facies at

82
the bottom of the unit to a more porous and permeable eolian dune facies at the top
(Lindquist, 1988). In southern Utah and northern Arizona, the Navajo Sandstone
exhibits dramatic coloration due to regional bleaching and hydrous ferric oxide
mineralization that create distinct layers of red, yellow, orange and white (Beitler et al.,
2003, 2005, 2007; Nielsen and Chan, 2006; Parry et al., 2004; Seiler, 2008). The
bleaching and mineralization of the sandstone is due to iron cycling in this reservoir rock
(Chan et al., 2000, 2006). A model for iron cycling in the Navajo Sandstone has been
proposed by Chan et al. (2000, 2005a). This model has three phases: 1) Source – Fe
3+
was
deposited as hematite grain coatings sourced from the breakdown of ferromagnesian
minerals upon deposition of the erg. 2) Mobilization – Fe
3+
was reduced to Fe
2+
when a
reducing fluid (possibly hydrocarbon) infiltrated the reservoir. 3) Mineralization – Fe
3+

precipitated as this reducing fluid met and mixed with an oxidizing fluid. This iron
cycling has resulted in several diagenetic facies such as a primary red facies, a bleached
facies and two different diffuse color variations due to the remineralization of bleached
rock (Beitler et al., 2005). For the purpose of this paper, the diffuse color variations are
herein called secondary red facies. Several hydrous ferric oxide (HFO) mineralogies
occur, and are generally referred to as HFO phases that include ferrihydrite, goethite, and
hematite.
Throughout southern Utah and northern Arizona, the Navajo Sandstone contains
different mineralization geometries. Two of the most prominent are abundant spheroidal
HFO concretions, and large-scale, prominent HFO-lined northeast (NE)-striking joints
with associated asymmetrical mineralization. Both of these represent different

83
hydrologic regimes of diffusion and fluid flow (respectively) through the reservoir. The
inter-diffusion of Fe
2+
and oxidizing fluids in the porous sandstone has resulted in
periodically distributed nucleation centers and Liesegang bands.
Abundant "mini-concretions", HFO concretions approximately 1-2 mm in
diameter, are commonly distributed throughout the host rock (Fig. 3.8). These
concretions are well-cemented throughout the entire concretion and they are resistant to
weathering, so they weather out and collect in topographic lows. These mini-concretions
are evenly cemented and do not have an internal structure nor a visible nucleus (Chan et
al., 2000). They are ubiquitous throughout southern Utah and northern Arizona, but their
size and spatial distribution vary from location to location (Fig. 3.8, 3.12). We also
observed a variety of mini-concretions that was only weakly cemented (herein termed
"freckles") in the Grand Staircase/Escalante National Monument field area that do not
weather out of the sandstone. These are similar in appearance to mini-concretions, are
approximately 2-3 mm in diameter and are friable (Fig. 3.9).
"Rind" concretions, consisting of a well-cemented hydrous ferric oxide (HFO)
rind surrounding a friable interior depleted of cement, are abundant in the southern Utah,
Grand Staircase/Escalante National Monument (GSENM) field area (Fig. 3.10). Rind
concretions are between 2 mm and 10 cm in diameter and the rinds range from 1-5 mm in
thickness. The concretion interiors exhibit a variety of HFO mineralization patterns
ranging from Liesegang bands to HFO-colored spots in the center or 'target' shapes (Fig.
3.10). These internal structural variations appear to be a result of a late stage
mineralization event that occurred after concretion formation. Navajo Sandstone

84
concretions show a wide variety of morphologies; however rind concretions in the same
region tend to exhibit similar size, interior pattern, and spatial distribution, suggesting a
self-organized nucleation process.
The large-scale, prominent N.E. joints have preferentially cemented lines or bands
extending to the southeast. These lines are interpreted as flow lines (Eichhubl et al.,
2004) that formed when a fluid infiltrated the reservoir and flowed to the southeast. This
flow created asymmetrical mineralization along fluid flow paths in the sandstone.
Liesegang bands occur parallel to the lines and perpendicular to the joints, suggesting that
there is a diffusive component to the reactant transfer as reactants diffused parallel to the
direction of fluid flow.
In the GSENM field area, Liesegang bands also extend throughout the sandstone
and contour around in-situ rind concretions (Fig. 3.11C), possibly indicating some related
processes in their formation. Occasionally rind concretions exhibit Liesegang banded
interiors (Fig. 3.10), which we interpret to be a separate process from the formation of the
large-scale band patterns (probably occurring after concretion formation, perhaps as Fe
from the rind was re-dissolved into solution and precipitated in the concretion interior).
This might require repeated oscillations of the redox conditions in the formation.
Liesegang bands are particularly prevalent in the Vermillion Cliffs National
Monument where bands create dramatic coloration that cross-cuts bedding (Fig. 3.11).
These bands tend to occur at the edges of reaction fronts, often separating bleached from
unbleached sandstone. In many cases, the edges of these reaction fronts contain related
clusters of mini-concretions (Figs. 3.11, 3.12). The clusters of mini-concretions are

85
usually found behind the Liesegang-banded "boundary" of the reaction front, however
occasionally are also found behind individual bands (Fig. 3.5B). The distribution of
millimeter-size concretions related to Liesegang bands suggests diffusion-controlled
nucleation via self-organizing processes, though advection also plays an important role in
concretion growth in the Navajo Sandstone (Chan et al., 2000).

86

FIGURE 3.7: Utah/ northern Arizona map and prominent mineralization geometries. A) Grand
Staircase Escalante National Monument and Vermillion Cliffs National Monument (red stars) in Utah and
northern Arizona. B) Spheroidal HFO concretions weathered out and collected in a topographic low. C)
HFO-lined joint (dark colored face) with associated asymmetrical mineralization. Blue arrow points in the
direction of fluid flow. Figure provided by Sally Potter (University of Utah).

87

FIGURE 3.8: Mini-concretions in the Navajo Sandstone, UT. A) Mini-concretions weathered out. B)
Mini-concretions in situ.

88

FIGURE 3.9: "Freckles" in situ. Weakly cemented mini-concretions in the Navajo Sandstone.

89

FIGURE 3.10: Rind concretions from the Navajo Sandstone. Most have a cemented outer rind and
friable sandstone interior, although occasionally cementation reaches nearly to the center of the concretion.
Many have diffusion patterns, Liesegang banding, and "freckles" in their interiors.

90

FIGURE 3.11: Liesegang banding in the Navajo Sandstone. A) Liesegang bands associated with micro-
concretions (1-2 mm size). B) Liesegang bands. (Black dots are lichen.) White card is 9 cm long. C)
Liesegang bands contouring around an in situ rind concretion.

91

FIGURE 3.12: Mini-concretions in situ, associated with Liesegang bands.

92
3.5. Discussion

In this work, varying types of self-organized silver chromate precipitates were
produced in columns of compact glass beads saturated with either pore fluid or gel. In
100-µm glass bead / gel mixtures, randomly distributed spheroidal precipitates of silver
chromate ranging from ~0.5-3 mm in diameter formed, morphologically resembling the
millimeter-size spheroidal HFO concretions that occur in our field localities in the Navajo
Sandstone. Spherules occasionally formed in larger bead sizes as well, but these were
generally larger (~3-5 mm diameter) and at a greater distance from the gel/bead interface.
We also observed formation of silver chromate crystals and periodic bands, depending on
what type of gel was used as the pore-filling agent. Finger-like fluid fronts were
observed when the glass bead pores were filled with potassium chromate solution only.
All of these precipitation morphologies (spherules, bands, fluid fronts) are observed in
the iron precipitates in the Navajo Sandstone as well (Figs. 3.8-3.12).
In order to rapidly force nucleation, extreme concentrations were used that formed
precipitation patterns in a matter of days. Obviously the chemical composition of
concretions and timescales of formation in our laboratory bench tests are not analogous to
what is expected in the field. However, these experiments demonstrate nucleation of
periodically-distributed nucleation centers and bands in a sand-like medium and give
insight into what factors may control the morphology of mineral precipitates in such
media.

93
In our experiments, we only observed Liesegang band formation in glass beads
containing gelatin (a gel that is known to produce banded patterns), but did not observe
band formation in glass beads saturated with pore solution. Why this should be is
unclear, since 100-µm glass beads inhibit convection similar to a gel, and the grain sizes
we used are similar to quartz grains in the Navajo Sandstone matrix where Liesegang
banding is prevalent. Manley and Stern (1955) reported formation of Liesegang bands of
silver chromate in a column of 70-µm glass beads saturated with chromate solution, but
they used substantially lower concentration gradients in their experiments so that bands
took >50 days to form. In a natural system precipitation could take even longer,
depending on the local fluid concentrations and the nucleation threshold of the mineral in
question. We hypothesize that the reason Liesegang bands did not form in our 100-µm
glass bead experiments is that the high concentration gradients caused some pores to
reach the nucleation threshold very quickly. Assuming surface nucleation is favored and
that reactants are in adequate supply, those first precipitates will grow, exerting diffusive
influences on one another (Abdel-Wahab and McBride, 2001) by depleting soluble
reactants in the vicinity, leading to the formation of the distributed spheroidal precipitates
that we observed. If concentrations reach the required saturation for nucleation more
slowly, as might occur in a diffusion-dominated system where concentration gradients are
lower, precipitation might occur in all pores at a given distance from the interface to
produce a banded pattern. In numerical models of iron oxide precipitation in porous
media (Chan et al., 2007), it was found that diffusion-only systems resulted in formation
of Liesegang bands close to the fluid interface, while diffusion + advection systems

94
resulted in spheroidal concretions throughout the body of the sandstone. It is possible
that their result may be due to a similar process - advection causing some pores to reach
supersaturation very quickly, and those develop into concretions, instead of periodic
bands forming throughout. The formation of the single spherule in our experiment where
450-600-µm glass beads were mixed with agarose gel (Fig. 3.B4) may simply be an
Ostwald ripening effect where one crystal grew in a pore, consuming others in the
immediate vicinity.
The nucleation threshold itself also can vary depending on physical and chemical
properties of the medium. It is well known that the silver nitrate/potassium chromate
system produces different precipitation patterns depending on what type of gel is used in
the experiment (Henisch, 2005). Figure 3.2 shows the differences in precipitate
morphology that are caused by using different gels: silica gel produces large crystals,
agarose produces smaller crystals, and agar and gelatin produce periodic bands. Agarose
and silica gels inhibit heterogeneous nucleation of silver chromate, instead promoting
growth of existing crystals, while gelatin causes precipitation in colloidal particles that
aggregate to form periodic bands. This may be due to the amount of soluble impurities
contained in the gels. For example, gelatin, which is produced from denatured collagen,
contains negatively charged carboxyl groups that could bind to crystal surfaces and
inhibit addition of chromate, resulting in numerous small particles rather than large
crystals. Agar, a gel produced from seaweed, also contains impurities to a lesser extent
and this may explain the formation of periodic bands in Figure 3.2C. On the other hand,
agarose (also produced from seaweed but more highly purified) and silica gel (which

95
only contains anions of silicate and the acid used to make the gel) allow larger crystals to
grow.
It is possible that another factor influencing the nucleation threshold in our
experiments is the glass bead surfaces themselves. Silver and chromate do not adsorb on
glass beads to a noticeable extent (Manley and Stern, 1955) but imperfections on the
glass bead surfaces could still serve as nucleation templates. According to our
conductivity measurements, diffusion rates through different sized glass beads are not
extremely variable, so pore blocking and permeability affecting the movement of ions is
likely not a major factor in precipitate morphology. However, a tube filled with 100-µm
glass beads has about twice the available surface area as a tube filled with 150-212-µm
beads, and 6 times the surface area as a tube filled with 450-600-µm beads (using mean
radius for calculation). Smaller glass beads are more likely to have more of these
"nucleation templates" per volume, which in effect lowers the nucleation threshold in
pores where these surface templates are present - meaning on average a column of
smaller beads has more pores that will reach the nucleation threshold than a column of
larger beads. This could be one explanation for why smaller concretions form in our 100-
µm glass bead experiments, regardless of what type of gel is used. Likewise, this could
explain why heterogeneous nucleation rates are apparently low in experiments where
larger glass beads are used, leading to formation of only a few spherules, except when
gelatin is used so that its impurities lower the nucleation threshold.

96
3.5.1. Comparison to Field Examples

Studies of the Navajo Sandstone revealed many types of iron precipitate
morphologies, including mini concretions, "freckles", rind concretions, and Liesegang
banding. Different morphologies were sometimes found associated with one another, for
example larger rind concretions contained mini-concretions or "freckles" in their friable
interiors, or mini concretions formed at the edge of reaction fronts associated with
Liesegang bands.
The presence of Liesegang banding in sandstone is likely a result of diffusion-
controlled precipitation, and the self-organized distribution of spheroidal concretions also
suggests that their nucleation was controlled by reaction-diffusion processes (Chan et al.,
2005a). Once nucleated, concretions could grow by advective supply of additional
reactants. Chan et al. (2005a) determined from mass balance calculations that advection
of fluid must have aided in transporting Fe
+2
and O
2
to the site of concretion growth,
since the amount of iron in the concretions requires more Fe than could have come from
the surrounding sandstone.
The factors isolated in our laboratory experiments that affect morphology of silver
chromate precipitates included chemical components of the fluid (that were altered by the
inclusion of different types of gels), and grain size and pore size of the glass bead matrix.
Presumably similar factors (pore size, grain size, sandstone/matrix composition, fluid
composition) varying between fluid flow events can cause morphological differences
between iron precipitates in the Utah environment. The Liesegang banding we observed

97
in our laboratory experiments was a result of impurities in gelatin (most likely, peptides
with negatively charged carboxyl groups) that prevent surface nucleation. It has been
shown (Ch. 2, Ch. 4) that certain organic molecules (e.g. amino acids, acetate) can induce
periodic banding of silver chromate when added to pure gels. Though such experiments
do not simulate the chemistry of the Utah environment, the fact that soluble organic
impurities can induce periodic banding in reaction-diffusion systems is interesting, given
that the reducing fluid in the Navajo Sandstone was likely hydrocarbon-rich (Beitler et
al., 2003; Chan et al., 2000), and Liesegang bands are commonly found at fluid flow
boundaries in our field areas. The exact effects of organic compounds on Liesegang
banding in geological environments warrants further investigation.

3.6. Conclusions

Our experimental results show that self-organized precipitates can be produced in
porous media under a variety of conditions. Randomly-distributed silver chromate
spherules 1-4 mm in diameter formed throughout a column of glass beads (of typical
Navajo Sandstone grain sizes) in as little as 4-5 days. Slightly larger silver chromate
spherules were formed when the grain size of the porous medium was larger, and these
formed farther from the fluid interface than the smaller spherules. Spheroidal precipitates
nucleated under diffusion-controlled conditions, and some growth occurred although
advection was not present. Bands formed when the porous medium contained a large
amount of organic impurities in the pore space (gelatin gel). Although the chemical

98
conditions in our experiments are very different from what would be expected in natural
situations, this experimental work shows how physical and chemical properties of a
porous/permeable medium can affect the morphology of precipitates produced.
From the various iron oxide concretion forms seen in our field localities in the
Navajo Sandstone, we infer that nucleation rate, fluid concentrations, amount of organics,
and relative importance of advection vs. diffusion also play a role in determining
concretion properties. The large variety in concretion morphologies seen at our field sites
probably indicates precipitation/growth of iron oxides in multiple fluid flows with
different initial conditions (Chan et al., 2005a), similar to how changing specific
parameters in our laboratory experiments affected the physical shape of precipitate.

99

Chapter 3 References
Abdel-Wahab, A., & McBride, E. F. (2001). Origin of Giant Calcite-Cemented
Concretions, Temple Member, Qasr El Sagha Formation (Eocene), Faiyum
Depression, Egypt. Journal of Sedimentary Research, 71(1), 70-81.

Beitler, B., Chan, M. A., Parry, W. T. (2003). Bleaching of Jurassic Navajo Sandstone on
Colorado Plateau Laramide highs; evidence of exhumed hydrocarbon
supergiants?. Geology (Boulder), 31(12), 1041-1044.

Beitler, B., Parry, W. T., &Chan, M. A. (2005). Fingerprints of Fluid Flow: Chemical
Diagenetic History of the Jurassic Navajo Sandstone, Southern Utah, USA.
Journal of Sedimentary Research, 75, 547–561.

Beitler Bowen, B., Martini, B. A., Chan, M. A., & Parry, W. T. (2007). Reflectance
spectroscopic mapping of diagenetic heterogeneities and fluid-flow pathways in
the Jurassic Navajo Sandstone, American Association of Petroleum Geologists
Bulletin, 91(2), 173–190.

Blakey, R. C., Peterson, F., & Kocurek, G. (1988). Synthesis of late Paleozoic and
Mesozoic eolian deposits of the western interior of the United States: Sedimentary
Geology, 56, 3–125.

Blakey, R.C. (1994). Paleogeographic and tectonic controls on some lower and middle
Jurassic erg deposits, Colorado Plateau, in Caputo, M. V., Peterson, J. A., and
Franczyk, K. J., eds., Mesozoic Systems of the Rocky Mountain Region, USA:
SEPM, Rocky Mountain Section, 273–298.

Busigny, V. and Dauphas, N. (2007) Tracing paleofluid circulations using iron isotopes:
A study of hematite and goethite concretions from the Navajo Sandstone (Utah,
USA). Earth and Planetary Science Letters, 254, 272-287.

Chan, M. A., Parry, W. T., & Bowman, J. R. (2000). Diagenetic Hematite and
Manganese Oxides and Fault-Related Fluid Flow in Jurassic Sandstones,
Southeastern Utah. American Association of Petroleum Geologists Bulletin, 84(9),
1281-1310.

Chan, M.A., Beitler, B., Parry, W. T., Ormö, J., & Komatsu, G. (2004). A possible
terrestrial analogue for haematite concretions on Mars. Nature, 29, 731-734.

100
Chan, M. A., Beitler, B., Parry, W. T., Ormö, J., & Komatsu, G. (2005a). Red rock and
red planet diagenesis: Comparisons of Earth and Mars concretions. GSA Today,
15(8), 4-10.

Chan, M. A., Parry, W. T., Park, A. S. (2005b). Hematite "Blueberry" Concretion
Doublet and Triplets on Mars: Iron Oxide Twin Analogs From Utah. American
Geophysical Union Fall Meeting Supplement Abstract #P21A-0139.

Chan, M. A., Johnson, C. M., Beard, B. L., Bowman, J. R. & Parry, W. T. (2006). Iron
isotopes constrain the pathways and formation mechanisms of terrestrial oxide
concretions: A tool for tracing iron cycling on Mars? Geosphere, 2(7), 324 – 332.

Chan, M. A., Ormö, J., Park, A. J., Stich, M., Souza-Egipsy, V., Komatsu, G. (2007).
Models of iron oxide concretion formation: field, numerical and laboratory
comparisons. Geofluids, 7, 1-14.

Cochran, K. A.,& Elmore, R. D. (1987). Paleomagnetic dating of Liesegang bands.
Journal of Sedimentary Petrology, 57(4), 701-708.

Eichhubl, P., Taylor, W. L., Pollard, D. D., & Aydin, A. (2004). Paleo-fluid flow and
deformation in the Aztec Sandstone at the Valley of Fire, Nevada—Evidence for
the coupling of hydrogeologic, diagenetic, and tectonic processes, Geological
Society of America Bulletin. September/October, 116(9/10), 1120–1136.

Garcia-Ruiz, J. M. (1994). Inorganic Self-Organisation in Precambrian Cherts. Origins
of Life and Evolution of the Biosphere, 24, 451-467.

Heaney, P. J., & Davis, A. M. (1995). Observation and Origin of Self-Organized
Textures in Agates. Science, 269, 1562-1565.

Henisch, H. K. (2005). Crystals in Gels and Liesegang Rings. Cambridge: Cambridge
University Press.

Lagzi, I., & Ueyama, D. (2009). Pattern transition between periodic Liesegang pattern
and crystal growth regime in reaction-diffusion systems. Chemical Physics
Letters, 468, 188-192.

Liesegang, R. E. (1896). Naturwiss. Wochenschr., 11, 353.

Lifshitz, I. M., & Slyozov, V. V. (1961). The Kinetics of Precipitation From
Supersaturated Solid Solutions. Journal of Physics and Chemistry of Solids,
19(1/2), 35-50.

101
Lindquist, S. J. (1988). Practical characterization of eolian reservoirs for development:
Nugget Sandstone, Utah-Wyoming thrust belt, Journal of Sedimentary Geology,
56, 315-339.

Manley, D. R., & Stern, K. H. (1955). Liesegang rings in inhomogeneous media:
powdered glass. Journal of Colloid Science, 10, 409-412.

Marko, F., Pivko, D., Hurai, V. (2003). Ruin marble: a record of fracture-controlled fluid
flow and precipitation. Geological Quarterly, 47(3), 241-252.

Muller, S. C., & Ross, J. (2003). Spatial Structure Formation in Precipitation Reactions.
Journal of Physical Chemistry A, 107, 7997-8008.

Nielsen, G. B., & Chan, M. A. (2006). Colorful diagenetic facies and fluid-related
alteration features of the Jurassic Navajo Sandstone, Snow Canyon State Park,
Utah, Geological Society of America Abstracts with Programs, Vol. 38, No. 7,
p. 518.

Ortoleva, P. (1982). Solute Reaction Mediated Precipitate Patterns in Cross Gradient Free
Systems. Zeitschrift für Physik B: Condensed Matter, 49, 149-156.

Ostwald, W. (1897). Zeitschrift für Physikalische Chemie (Leipzig), 23, 356.

Parry, W. T., Chan, M. A., & Beitler, B. (2004). Chemical bleaching indicates episodes
of fluid flow in deformation bands in sandstone., American Association of
Petroleum Geologists, 88(2), 175-191.

Seiler, W. M. (2008). Jurassic Navajo Sandstone of Coyote Buttes, Utah/Arizona:
Coloration and Diagenetic History, Preservation of a Dinosaur Trample Surface,
and a Terrestrial Analog to Mars, M.S. Thesis, University of Utah, Salt Lake City,
Utah, USA.

Smith, D. A. (1984). On Ostwald's supersaturation theory of rhythmic precipitation
(Liesegang's rings). Journal of Chemical Physics, 81(7), 3102-3115.

Stern, K. H. (1954). Liesegang rings in inhomogeneous media: partially coagulated
gelatin. Journal of Colloid Science, 9(4), 329-337.

Stern, K. H., & Shniad, H. (1958). Diffusion in inhomogeneous media: capillary
diffusion in beds of glass beads. Journal of Colloid Science, 13, 24-31.

Stone, D. A., & Goldstein, R. E. (2004). Tubular precipitation and redox gradients on a
bubbling template. Proceedings of the National Academy of Sciences, 101(32),
11537-11541.

102
Squyres, S. W., Grotzinger, J. P., Arvidson, R. E., Bell III, J. F., Calvin, W., Christensen,
P. R., Clark, B. C., Crisp, J. A., Farrand, W. H., Herkenhoff, K. E., Johnson, J. R.,
Klingelhofer, G., Knoll, A. H., McLennan, S. M., McSween Jr., H. Y., Morris, R.
V., Rice Jr., J. W., Rieder, R., & Soderblom, L. A. (2004). In Situ Evidence for
an Ancient Aqueous Environment at Meridiani Planum, Mars. Science, 306,
1709-1714.

Toramaru, A., Harada, T., & Okamura, T. (2003). Experimental Pattern Transitions in a
Liesegang system. Physica D, 183, 133-140.

103
Chapter 4. Effects of Amino Acids on Morphological Transitions in
Inorganic Precipitates

Summary

Morphological transitions between crystals and banding were observed in a silver
nitrate / potassium chromate system in purified agarose gel, with amino acids added in
solution. When amino acids were added to purified agarose, silver chromate crystal
growth was inhibited with increasing amino acid concentration, and at high concentration
periodic spaces were observed between bands of macroscopic crystals. When N-acetyl
amino acids were added to the agarose gel, crystal growth was suppressed and only
rounded particles formed, which aggregated to form Liesegang banded patterns. Silver
chromate typically forms Liesegang bands in gelatin gels and larger crystals in agarose
gels; our results suggest this may be due to the types of carboxyl-terminated chains
present in gelatin.

4.1. Introduction

Gel diffusion experiments have been performed for over a century to study the
mechanisms behind the formation of self-organized periodic patterns known as Liesegang
bands (Henisch, 2005). These precipitation patterns commonly form when a
concentrated solution of one reactive ion diffuses into a gel containing another reactive
ion in dilute solution, precipitating in concentric rings (when the experiment is done in a
petri dish) or rhythmic bands (when performed in a test tube). This phenomenon of

104
periodic distribution of precipitate in a diffusion-controlled system was first reported in a
silver dichromate precipitate (Liesegang, 1896), and has since been observed in silver
chromate as well as many other salt systems (Henisch, 2005, and references therein). In
addition the silver chromate reaction system can be applied to studying pattern formation
in biology, since the chromate ion (CrO
4
2-
) is similar to phosphate in structure but forms
dark red easily visible precipitates, and the silver cation (Ag
+
) has the ability to form
precipitates with many types of anions.
Many types of precipitation structures have been observed in gel diffusion
experiments, ranging from dendritic crystals to periodic bands to continuous precipitates
(e.g. Henisch, 2005, and references therein; Lagzi and Ueyama, 2009; Toramaru et al.,
2003). In general, the positions of periodic bands can be calculated from a solution of
Fick's laws of diffusion, assuming proper boundary conditions (Henisch, 2005, Ortoleva
1994) and bands that form in gel diffusion experiments tend to exhibit some predictable
behavior. For example, bands may follow a spacing law such that the ratio of the
distances of consecutive bands from the top of the gel is a constant (known as the spacing
coefficient (Jablczynski, 1923)). However, the physical and chemical factors that
influence formation of periodic bands are complex. For example, previous investigations
have shown that pattern transitions between band and crystal formation can be achieved
by changing factors such as gel concentration or gel type (Lagzi and Ueyama, 2009;
Toramaru et al., 2003). Presence of soluble impurities may also be a factor. We
observed in a previous study (Ch. 2) that slight banding of silver chromate crystals occurs
in untreated agarose gel, but this periodic pattern does not form if the agarose gel is first

105
extracted with water to remove soluble ionic impurities. Impurities may affect the
growing precipitate either by poisoning the surface of the crystal from further growth or
by altering the threshold for heterogeneous nucleation (Hillson, 1961; Lagzi and Ueyama,
2009), and this may be one explanation for the observation that periodic bands of silver
chromate and dichromate commonly form in gelatin, a gel containing many organic
impurities, but tend to form larger crystals in agarose (Ch. 2, Henisch, 2005).
We hypothesize that the periodic banding of silver chromate and dichromate in
gelatin that is so often seen in experimental work, is a result of the interaction of silver-
chromate precipitate with charged organic impurities that are present in gelatin gel.
Gelatin is composed of denatured collagen, and contains long carboxyl-terminated side
chains which could interact with the growing precipitate. We test this hypothesis by
examining the effects of added impurities (amino acids and N-acetyl amino acids) on
precipitation patterns of silver chromate in purified agarose, a gel which contains little to
no soluble ionic impurities.

4.2. Experimental

4.2.1. Gel Preparation

Agarose gels (~ 0.5 weight %) were made to contain 5 mM potassium chromate
and varying concentrations of amino acids by the following procedure:

106
(1) The agarose powder (Molecular Biology Certified, Shelton Scientific) was purified
before use by extracting with deionized water, then centrifuging for 5 minutes and
pouring off the water containing the gel impurities. This extraction was repeated 5 times
to purify the agarose before it was used in the procedures described below.
(2) For each set of experiments, two 0.5 weight % agarose gel solutions were prepared:
one containing 5 mM potassium chromate and 5 mM biochemical (solution A), and one
containing just 5 mM potassium chromate (solution B).
(3) These gels were mixed proportionally in 20 ml flasks to obtain gels with biochemical
concentrations of 0.2, 0.4, 0.6, 0.8, 1.0, 2.0, 3.0, 4.0 and 5.0 mM. For example, to make a
gel with 3.0 mM biochemical concentration we mixed 12 ml of solution A + 8 ml of
solution B, and to make a gel with 0.4 mM biochemical we mixed 1.2 ml of solution A +
13.8 ml of solution B.
(4) Three 6 ml test tubes were poured from each 20 ml flask.
(5) After gels hardened at room temperature, 150 µl of 6 M silver nitrate solution was
added to the top of the gel.
The biochemicals used were glycine, N-acetyl glycine, L-alanine, N-acetyl L-
alanine, L-leucine, and N-acetyl L-leucine (Fig. 4.1). Agarose controls (containing only
solution B) were also prepared. pH measurements were made on solutions containing 5
mM potassium chromate and 5 mM of each amino acid, to determine the maximum
acidity that can be expected to be present in these gels as precipitation occurs (Table 4.1).

107
4.2.2 Analysis of Precipitates

Gels were photographed in test tubes and density profiles of precipitate were
obtained digitally. Analysis was performed on a Macintosh computer using the public
domain NIH Image program (developed at the U .S. National Institutes of Health and
available on the internet at http://rsp.info.nih.gov/nih-image/). Band widths and distances
from the gel-air interface were measured.
After density profiles were made, gels were removed from the tubes and ~1-mm
slices of gel were placed on microscope slides. Photographs of silver chromate
precipitates were taken at 100x magnification.

108

Biochemical pH
Glycine 8.13
Alanine 8.07
Leucine 8.01
N-acetyl glycine 5.17
N-acetyl alanine 5.46
N-acetyl leucine 5.10

TABLE 4.1: pH measurements. pH values are for solutions containing 5 mM potassium chromate and 5
mM biochemical.

109

FIGURE 4.1: Amino acids used in this study. Glycine, alanine, and leucine are neutral compounds that
become zwitterions when dissolved in water, containing a positively charged amine group and negatively
charged carboxyl group. N-acetyl glycine, N-acetyl alanine, and N-acetyl leucine are negatively charged
ions when dissolved (the negative charge on the carboxyl group is noted).

110
4.3. Results

4.3.1. Gel Diffusion Experiments

Precipitation of silver chromate in purified agarose in the absence of added
biochemical produced randomly oriented crystals that did not display any rhythmic
banding. Periodic patterns formed, apparently because amino acids suppressed crystal
growth. In this work we observed two general kinds of banding of silver chromate
precipitates, which we define here as "Type 1" and "Type 2". Type 1 represents bands
composed of macroscopic crystals, in which the crystals were large enough to be visible
yet precipitated with defined spaces in between "bands" of crystals, and Type 2
represents bands of tiny particles with larger spaces in between, that appear similar to
typical Liesegang patterns. Examples of density plots for both types of banding are
shown in Figure 4.8.
Banding was generally not visible in most glycine experiments, although some
banding was observed in the 5.0 mM glycine concentration (Fig. 4.2). In alanine
experiments banding did not occur until about 1.0 mM alanine was added, after which
banding became stronger and sharper with increasing concentration (Fig. 4.3). Increasing
concentrations of leucine caused crystals to become smaller, and at 0.8 mM leucine,
banding became visible and band sharpness increased with leucine concentration (Fig.
4.4).
The addition of N-acetyl amino acids to the gel induced sharper, more well-
defined bands. In agarose containing N-acetyl glycine, randomly oriented silver

111
chromate crystals formed at concentrations below 1.0 mM, although crystals became
slightly smaller as concentration increased. At 2.0 mM N-acetyl glycine, crystals were
larger and showed a slight banding pattern, and bands of small particles begin to form at
3.0 mM concentration. Banding patterns were not reproducible in the 3.0 mM N-acetyl
glycine experiments - although particles always exhibited the same small rounded
morphology, periodic banding occurred in one tube and not in two others. However, at
4.0 mM N-acetyl glycine, bands were reproducible. The highest concentration of N-
acetyl glycine (5 mM) showed no banding, only continuous precipitation of fine particles
(Fig. 4.5).
Addition of N-acetyl alanine caused crystals to become smaller with increasing
concentration, and below 0.8 mM crystals formed with bands of empty space in between
("Type 1" bands). Around 1.0 mM concentration, small crystals precipitated that did not
show any banding pattern, and distinct "Type 2" banding was observed in the 2.0 through
5.0 mM concentrations (Fig. 4.6).
N-acetyl leucine caused "Type 1" banding of crystals to occur in concentrations
from about 0.6 mM to 3.0 mM, and around 4.0 mM, particles became smaller but no
more banding was produced (Fig. 4.7).

112

FIGURE 4.2: Silver chromate precipitation in agarose containing glycine. At low concentrations only
crystals form, similar to those that form in pure agarose. At 5.0 mM concentrations "Type 1" banded
patterns were observed.

113

FIGURE 4.3: Silver chromate precipitation patterns in agarose gel containing Alanine. "Type 1"
banding patterns become clearer with increasing concentration.

114

FIGURE 4.4: Silver chromate precipitation patterns in agarose gel containing Leucine..

115

FIGURE 4.5: Silver chromate precipitation in agarose gel containing N-acetyl glycine. As
biochemical concentration increased, crystal size became smaller and when particles are small enough they
clumped to form "Type 2" bands. At the highest concentration (5 mM) particles were very small but bands
did not form.

116

FIGURE 4.6: Silver chromate precipitation patterns in agarose gel containing N-acetyl alanine.
Some "Type 1" banding appeared at lower concentrations, and at 2.0 mM particles abruptly became very
small and aggregated to form periodic "Type 2" bands.

117

FIGURE 4.7: Silver chromate precipitates in agarose gel containing N-acetyl leucine. Some "Type 1"
banding appeared in the 2.0 and 3.0 mM concentrations, but at higher concentrations only homogenous
precipitation of small particles occurred.

118

FIGURE 4.8: Density plots of silver chromate precipitates. Top: 5.0 mM Alanine and 0.6 mM N-
acetyl Alanine show examples of "Type 1" banding, in which the bands are thicker than the spaces in
between. Bottom: 3.0 mM N-acetyl Glycine and 4.0 mM N-acetyl Alanine show "Type 2" banding, in
which particles aggregate into bands leaving spaces that are larger than the bands themselves, visually
resembling the typical "Liesegang band" phenomenon (Henisch, 2005).

119
4.3.2. Band Measurements - Spacing and Width Laws

In experiments where bands were clearly visible, density plots were made (Fig.
4.7) and the positions and widths of bands were measured to calculate the spacing
coefficient p (Jablczynski, 1923):

p = X
n+1
/X
n

and the exponential constant α in the width law (Droz et al., 1999):

W
n
∼ X
n
α
,

where W
n
is the width of band n, and X
n
is the distance of a band from the junction point
of the two reactants (i.e. the gel-air interface). α = 1 corresponds to linear proportionality
between the width of a band and its distance from the gel-air interface.
For alanine, N-acetyl glycine, N-acetyl alanine, and N-acetyl leucine, the spacing
constant p showed an increasing trend with biochemical concentration (Table 4.2, Fig.
4.9). At the highest concentration (5 mM), spacing constants for all amino acids fell
between 1.073 and 1.086. In some of our experiments we observed that the spacing
coefficient linearly depended on the concentration of biochemical. The slopes of these
lines are a rough indicator of the ability of each biochemical to induce banding, since a
higher slope would imply that the sharpness and definition of periodic banding increases

120
quickly with additional biochemical present. For example, the linear dependence of p
with concentration for N-acetyl alanine has a slope of 0.041, whereas for N-acetyl
glycine the slope is 0.026 (Fig. 4.9). This implies that N-acetyl alanine is more effective
at inhibiting crystal growth and inducing banding than N-acetyl glycine.
Spacing coefficients for alanine and N-acetyl leucine also showed a linear trend
with increasing biochemical concentration. (Leucine spacing coefficient measurements
did not show any linear trend.) However, both of these biochemicals produced bands that
were "Type 1", i.e. oscillatory zoning of small crystals rather than thin bands (Figs. 4.3,
4.7). Although at all concentrations N-acetyl leucine had a higher p value than alanine,
indicating that N-acetyl leucine is a better inducer of banding, the dependence of p on
concentration for N-acetyl leucine and alanine is almost equal (each producing a slope of
0.003; Fig. 4.9).
The width law exponent α was calculated for all experiments that produced
banding, and values between ~0.6-1.3 were obtained (Table 4.3). No definitive trends in
α values were observed.

121

Biochemical concentration (mM) 0.4 0.6 0.8 1.0 2.0 3.0 4.0 5.0
p (glycine) 1.073
p (alanine) 1.072 1.078 1.083 1.084
p (leucine) 1.084 1.075 1.072 1.082
p (N-acetyl glycine) 1.062 1.087
p (N-acetyl alanine) 1.092 1.096 1.045 1.086
p (N-acetyl leucine) 1.078 1.082 1.087 1.086

TABLE 4.2: Dependence of the spacing coefficient on biochemical concentration. Concentrations of
the precipitating ions (Ag
+
and CrO
4
2-
) are 6 M and 5 mM, respectively. p values are given for those
experiments that produced banding.

122

FIGURE 4.9: Graph of spacing constants as a function of concentration for all biochemicals.
Alanine, N-acetyl glycine, N-acetyl alanine, and N-acetyl leucine show a linear relationship between
biochemical concentration and the spacing coefficient p. See text for further explanation. (Note: N-acetyl
alanine produced Type 1 banding at concentrations below 1.0 mM, and these experiments were not
included in the linear fit analysis because they show a different slope than the Type 2 banding produced at
higher concentrations. Glycine values were not included because p was only observable at one
concentration.)

123

Biochemical concentration (mM) 0.4 0.6 0.8 1.0 2.0 3.0 4.0 5.0
α (glycine) 1.091
α (alanine) 0.961 0.879 1.136 1.219
α (leucine) 1.025 0.882 1.039 1.025
α (N-acetyl glycine) 0.680 0.718
α (N-acetyl alanine) 0.995 1.537 1.013 0.956
α (N-acetyl leucine) 1.120 0.960 1.105 1.313

TABLE 4.3: Dependence of the width law exponent α on biochemical concentration.

124
4.3.3. Crystal Structure

In purified agarose without the addition of biochemical, silver chromate crystals
were randomly oriented and grew in twinned and branching forms (Fig. 4.10). Addition
of amino acids suppressed crystal growth and caused crystals to become smaller and
more branched, and in the case of N-acetyl amino acids, caused precipitation to occur in
non-crystalline particles.
In the glycine experiments, crystals formed that appeared similar to those formed
in pure agarose gel, except for some slight Type 1 banding at 5.0 mM glycine
concentration in which the bands were composed of smaller crystals. In agarose
containing alanine or leucine, at low amino acid concentration (below about 0.6 mM)
crystals did not show banding patterns. However increased concentrations of amino acid
suppressed crystal growth and by about 3.0 mM, when periodic precipitation patterns
began to appear, overall crystal size was diminished. Different crystal forms were
observed in the band and interband regions in 5.0 mM leucine and alanine experiments;
the "band" region contained small branched crystals and the "interband" region contained
larger rod-like crystals (Figs. 4.11C, 4.11D).
N-acetyl amino acids were much more effective at suppressing crystal growth and
inducing banding than pure amino acids. In agarose containing 0.6 mM N-acetyl glycine,
large crystals formed, but by 2.0 mM, concentration crystal branching was suppressed
and only rod-like crystals formed (Fig. 4.12). In the N-acetyl glycine experiments
periodic banding became visible at 3.0 mM concentration, and the bands at 3.0 mM N-

125
acetyl glycine were composed of small elongated crystals (Fig. 4.12D) while at 4.0 mM
the bands were composed of large amorphous particles (Fig. 4.12E). At 5.0 mM N-acetyl
glycine, small rounded particles formed that did not aggregate to form bands (Fig. 4.12F).
When silver chromate precipitated in agarose containing N-acetyl alanine, Type 1
bands were formed at lower concentrations, and these bands were composed of small
branched crystals (with larger crystals in the interband regions) that resembled the
crystals in high-concentration alanine and leucine experiments. At higher concentrations
of N-acetyl alanine rounded particles formed instead of crystals, and at 2.0 mM the
particles sometimes had protruding crystal arms (Fig. 4.13C) while at 3.0 mM and above,
crystal growth was suppressed. Banded patterns only appeared in the 4.0 and 5.0 mM N-
acetyl alanine experiments, and these bands were composed of elongated rounded
particles that did not show crystal structure (Figs. 4.13E, 4.13F).
N-acetyl leucine was slightly less effective at suppressing crystal growth
compared to N-acetyl glycine and N-acetyl alanine. Increasing concentrations of N-
acetyl leucine in the gel caused silver chromate crystals to become smaller until at 4.0
mM concentration, rounded particles with crystal arms formed. At the highest
concentration of N-acetyl leucine tested (5.0 mM), small rounded particles formed that
still had protruding crystal arms (Fig. 4.14F). Except for some slight Type 1 banding in
2.0 and 3.0 mM N-acetyl leucine experiments, bands did not otherwise form.

126

FIGURE 4.10: Silver chromate crystals in purified agarose (no biochemical added).

127

FIGURE 4.11: Silver chromate precipitate in agarose containing alanine. A: 1 mM alanine. B: 3 mM
alanine. C and D: 5 mM alanine. crystals in the band (C) and interband (D) regions. Bands are composed
of small branched crystals, and interband crystals are larger and either branched or rod-shaped. Crystal
forms were similar to these in all amino acid experiments that produced Type 1 banding.

128

FIGURE 4.12: Silver chromate precipitates in agarose containing N-acetyl glycine. A: 0.6 mM N-
acetyl glycine. B) 1.0 mM N-acetyl glycine. C) 2.0 mM N-acetyl glycine. Crystal branching is prevented
and rod-like crystals begin to show some periodicity. D) 3.0 mM N-acetyl glycine. Precipitation occurrs in
small rod-like crystals that aggregate to form periodic bands (see Figure 4). E) 4.0 mM N-acetyl glycine.
Image shows large particles that compose the periodic bands near the bottom of the tube in Figure 4. F) 5.0
mM N-acetyl glycine; very small particles of precipitate form and banding no longer occurs.

129

FIGURE 4.13: Silver chromate precipitates in agarose containing N-acetyl alanine. A) 0.6 mM N-
acetyl alanine. Type 1 bands are composed of small branched crystals and interband regions contain longer
rod-like crystals. B) 1.0 mM N-acetyl alanine. Only small branched crystals form and no periodic patterns
are seen (Figure 5). C) 2.0 mM N-acetyl alanine. Crystal growth is mostly suppressed and round particles
form. Inset: particle with crystal arms, observed in the same region. D) 3.0 mM N-acetyl alanine. Crystal
growth is completely suppressed. E) 4.0 mM N-acetyl alanine. Elongated particles form, that aggregate to
form periodic bands as shown in Figure 5. F) 5.0 mM N-acetyl alanine. Only small particles form; image
shows particles that compose bands near the top of the tube in Figure 5.

130

FIGURE 4.14: Silver chromate precipitate in agarose containing N-acetyl leucine. A) 0.6 mM N-
acetyl leucine. B) 1.0 mM N-acetyl leucine. C) 2.0 mM N-acetyl leucine. D) 3.0 mM N-acetyl leucine. D)
4.0 mM N-acetyl leucine. Crystals are smaller and growth of crystal arms begins to be suppressed at this
concentration. E) 5.0 mM N-acetyl leucine. Only small particles with few crystal arms form.

131
4.4. Discussion
In purified agarose, silver chromate precipitated in large crystals with twin or
branched morphologies (Fig. 4.10), but addition of biochemicals (simple amino acids or
their N-acetylated derivatives) caused precipitates to be smaller and sometimes led to the
formation of banded patterns. Increasing concentration of amino acids led to a change in
morphology from twinned crystals to small branched crystals (e.g. Fig. 4.11C), and
finally in some cases to rounded particles that aggregated to form bands (e.g. Fig 4.13E).
This change can be explained in terms of ionic interaction. Crystal growth rate in our
experiments is limited by supply of chromate, so crystal surfaces should have a net
positive charge due to the abundance of silver ions. The amino acids used in this study
have negatively charged groups and can bind to the positively charged silver ion (Ag
+
) on
the surface of a crystal, interfering with the addition of chromate. Initially the
biochemical only caused the silver chromate crystals to precipitate in periodic fashion,
but at high enough concentrations crystal growth became suppressed altogether and only
small particles could form.
It is clear from our results that certain biochemicals are more effective at inducing
banding (or, inhibiting crystal growth) than others. Pure amino acids were somewhat
effective in diminishing crystal size and at the highest concentrations of alanine and
leucine, clear periodicity was observed in the precipitate. Amino acids are zwitterions
that when dissolved in water will have a positively charged amine group and a negatively
charged carboxyl group, and we hypothesize that the carboxyl group binds to the surfaces
populated by silver ions, interfering with the addition of chromate. 5 mM glycine

132
induced some periodicity, but alanine and leucine were more effective at causing
banding. Alanine and leucine both have neutral side chains, and the side group of leucine
is hydrophobic. These side groups may affect precipitate morphology because the
carboxyl group can bind to the crystal surface, and a longer neutral side chain may be
able to maintain itself on the surface without being displaced by a chromate ion.
Although both were more effective than glycine, alanine and leucine gave similar
banding patterns, indicating that leucine is not much more effective than alanine at
inducing banding despite its longer side chain. This may be because, at certain molecular
orientations, leucine's more hydrophobic side chain may reduce the chance of the
molecule interacting with the crystal surface. Further studies need to be done to establish
how orientation of neutral or hydrophobic side groups affect surface interactions.
N-acetyl amino acids were even more effective than pure amino acids at
suppressing crystal growth and inducing banding, probably because they are negatively
charged molecules and can bind to silver ions more strongly. This may explain why
crystals were generally larger in the pure amino acid experiments than the N-acetyl
amino acid experiments. Of the three N-acetyl amino acids tested in this study, N-acetyl
alanine was most effective at suppressing crystal growth and inducing banding, perhaps
for similar reasons as stated above for pure amino acids. Higher concentrations of N-
acetyl amino acids led to decreased particle size, and initially this may have been an
effect of surface inhibition as crystal surfaces were coated with biochemical and
prevented from growing. However at the highest N-acetyl amino acid concentrations in

133
our experiments, the low pH of around 5 (Table 4.1) will convert most of the CrO
4
2-
ions
to HCrO
4
-
ions:

H
2
CrO
4
⇔ H
+
+ HCrO
4
-
⇔ H
+
+ CrO
4
2-
(pK
1
= 0.74, pK
2
= 6.49)

and HCrO
4
-
ions can react to form dichromate:

2HCrO
4
-
⇔ Cr
2
O
7
2-
+ H
2
O.

which can also react with silver to form an insoluble precipitate. The decrease in particle
size observed as N-acetyl amino acid concentrations increase up to 5 mM may therefore
be an effect of low pH inhibiting precipitation of silver chromate, rather than increased
ionic interaction between biomolecules and crystal surface. Silver dichromate crystals
are visually distinct from silver chromate crystals (see Fig. 2.1) and we did not observe
any crystals that were easily identifiable as silver dichromate. However, that does not
preclude the possibility of silver dichromate becoming incorporated in the rounded
particles that are produced at 5.0 mM N-acetyl amino acid concentrations.
It is commonly seen that in silver chromate or dichromate precipitation
experiments of this nature, the type of gel greatly affects the pattern morphology. For
example, silver chromate usually forms small crystals in agarose, silver dichromate forms
larger dendritic crystals in agarose, and both silver chromate and dichromate form
periodic bands in gelatin (Chapter 2, Henisch, 2005; Lagzi and Ueyama, 2009). It has

134
been seen in previous investigations that increasing gel concentration is one factor in the
pattern transition from crystals to bands, and this may be partly due to the increased
number of nucleation sites in the gel matrix which would lead to a higher number density
of particles (Lagzi and Ueyama, 2009;, Toramaru et al., 2003). We suspect that there is
an important contribution from the chemical composition of the gel, independent of gel
concentration. Gelatin, a gel which is known to give strong banding patterns, is
composed of protein and likely contains long carboxyl-terminated hydrophilic side
chains, and this may be one explanation for its propensity to cause banding of silver
chromate and dichromate, similar to how amino acids with neutral side chains induced
banding in the experiments reported here.
One property of periodic banding that becomes apparent from our results is that
distinct banding only occurs in a specific range of experimental parameters. This was
also observed by Lagzi and Ueyama (2009) in silver dichromate precipitation
experiments in agarose, where the addition of increasing amounts of gelatin first
suppressed crystal growth and led to periodic band formation, but when gelatin
concentration exceeded a certain point, only small particles precipitated in a homogenous
fashion. Similarly, increasing amounts of N-acetyl glycine and N-acetyl alanine in our
experiments first led to band formation, but at very high concentrations only homogenous
precipitation occurred. When N-acetyl leucine was used the transition from crystals to
homogenous precipitation was achieved without Type 2 bands ever forming at all.
Agarose gels promote a lower heterogeneous nucleation rate compared to gelatin (Lagzi
and Ueyama, 2009) and when surface nucleation is favored, larger crystals can grow.

135
The addition of soluble impurities that inhibit surface nucleation (and hence, crystal
growth) may cause nucleation of new particles to be favored again, and lead to a higher
overall number density and diminishing particle size. The suppression of silver chromate
crystal growth by the negatively charged biomolecules binding to the Ag
+
ions could be
supplemented by a surface charge neutralization by the amino acid side chains. Perhaps
at an intermediate stage, diminishing strength of surface charges leads particles to
aggregate into periodic bands, but when enough biochemical is present that the particle
surface charges are completely neutralized, homogenous precipitation is favored.

4.5. Conclusions
Precipitation patterns of silver chromate in agarose gels were investigated, and
morphological transitions were observed due to the addition of various biochemicals.
The inclusion of soluble biochemicals in the purified agarose gel inhibited crystal growth
of silver chromate, and 1) N-acetyl derivatives of amino acids were more effective than
pure amino acids because of their negative charge, and 2) of the amino acids tested in this
study, alanine was most effective at inducing banding. We consistently observed that
increasing concentrations of biochemical led to a greater number density of particles;
however, the formation of periodic banding only occurred in a narrow concentration
range, and sometimes not at all for certain compounds. The effect of diminishing crystal
size was reproducible, although sometimes the specific banded patterns were not. At
certain concentrations banding was more likely to occur while at higher or lower
concentrations banding was never observed.

136
The binding of soluble impurities to the growing crystal surface may be one
mechanism for Liesegang band formation in systems where crystals nucleate and grow.
There are other mechanisms that can lead to periodic banding, and there is no reason why
impurities that induce banding cannot be inorganic as well. However, our results show
that certain biomolecules are more capable of suppressing crystal growth than others, and
the nucleation regime where periodic pattern formation is favored may vary greatly
depending on which organic molecules are present in solution. This may have
implications for the formation of periodic patterns along diffusion gradients in biological
systems. More work needs to be done to establish the features of biochemicals that are
most effective in inducing periodic pattern formation, to better understand the
mechanisms that contribute to the self-organization of structures in biology. Future
investigations should include determining whether amino acids with charged and/or polar
side groups can induce banding, in addition to further study of non-polar, neutral amino
acids.

137
Chapter 4 References
Droz, M., Magnin, J., & Zrinyi, M. (1999). Liesegang patterns: Studies on the width law.
Journal of Chemical Physics, 110(19), 9618-9622.

Henisch, H. K. (2005). Crystals in Gels and Liesegang Rings. Cambridge: Cambridge
University Press.

Hillson, P.J. (1961). Liesegang phenomenon - the importance of impurities in periodic
precipitation. Transactions of the Faraday Society, 57, 1031-1034.

Jablczynski, K. (1923). Bulletin de la Société Chimique de France, 33, 1592.

Lagzi, I., & Ueyama, D. (2009). Pattern transition between periodic Liesegang pattern
and crystal growth regime in reaction-diffusion systems. Chemical Physics
Letters, 468, 188-192.

Liesegang, R. E. (1896). Naturwiss. Wochenschr. 11, 353

Ortoleva, P. (1994). Geochemical Self-Organization. Oxford Monographs on Geology
and Geophysics No. 23, Oxford University Press.

Toramaru, A. Harada, T. & Okamura, T. (2003). Experimental Pattern Transitions in a
Liesegang system. Physica D, 183, 133-140.

138
Chapter 5. Conclusions

This dissertation work is a preliminary step toward a more detailed understanding
of how initial chemical and physical conditions affect periodic pattern morphology.
The laboratory results described in the preceding chapters display a wide variety of
self-organized precipitates that can be formed through simple diffusion processes in
abiotic systems, in a matter of days to weeks.
It is clear that the diffusion medium itself plays a large role in determining the
morphology of precipitate, through physical and chemical mechanisms. In Chapters
2 and 3 we showed how carrying out the same reaction in different types of gels
produced very different patterns, which indicates that either soluble impurities and/or
interaction with the gel lattice itself may be significant in determining pattern
morphology. Chapter 3 showed the physical effects of a sand-like diffusion medium
on precipitation, and precipitation patterns were produced that closely resemble those
often seen in porous sandstone. Precipitation experiments in sand-gel mixtures
confirmed that chemical properties of the gel greatly affect pattern morphology
despite the physical effects of the sand column.
Previous work has suggested that the nucleation thresholds are at least partly
determined by the chemical properties of the gel (Henisch, 2005; Lagzi and Ueyama,
2009; Toramaru et al., 2003), and our experiments with silver chromate precipitation
in gelatin vs. agarose confirm this. Gelatin clearly promotes a higher number density
of particle nucleation and formation of bands, rather than the growth of crystals that

139
commonly occurs in agarose, and our work in Chapter 4 indicates that the reason for
this may be (at least in part) due to the presence of soluble organic impurities.
Dissolved biochemicals in purified agarose caused silver chromate crystals to become
smaller as biochemical concentration increased, and periodic bands similar to gelatin
experiments were formed when amino acid concentration was high enough.
Interestingly, out of the biochemicals we tested, N-acetyl glycine and N-acetyl
alanine were the most effective at inducing banding. Use of N-acetyl amino acids
rather than pure ones simulates the effect of peptide chains ending in those amino
acids (since the negatively charged carboxyl group in the C-terminal position of a
peptide can interact with the positively charged silver chromate crystal surface in the
same way as an N-acetyl amino acid). This may explain why gelatin, which is
composed of denatured collagen, commonly produces banding patterns: collagen has
a composition of ~30% glycine and ~11% alanine (Garrett and Grisham, 2005) and in
gelatin gel, soluble impurities containing C-terminal glycine or alanine should be
abundant.
This study has several important implications for studies of periodic patterning in
geological systems. In Chapter 2 it was seen that the reactants present in the system
have an effect on the morphology of precipitate, independent of gel characteristics.
Iron hydroxide, carbonate, and phosphate all precipitated in agarose gels, but in
distinctly different patterns. Direct mineral replacement (similar to fossilization) did
not change the periodic banding produced by iron phosphate; however, oxidation of
initially homogeneous precipitates produced layered bands that were not previously

140
visible. This suggests that patterning in geology may be able to serve as a diagnostic
tool to identify possible reactants that may have produced that precipitate, as well as
the degree of oxidation that may have occurred. Spheroidal precipitates similar to
concretions are also easily produced in our laboratory experiments, as well as
Liesegang banding in sand-like media, and the size and distance of precipitate from
the reactant interface was shown to be related to factors such as grain size and
presence of impurities (e.g. different types of gel). This type of experiment could be
potentially useful for geological studies of formations such as the Navajo Sandstone,
UT that contain self-organized precipitates, since it permits observation of how
individual parameters affect pattern morphology.
The mechanisms that cause banding in diffusion gradients in our gel experiments
could also be one cause of periodic patterning in biology. We have shown that
periodic pattern formation in diffusion gradients can be induced by the presence of
organic impurities, specifically amino acids and N-acetylated amino acids, although
we think it is likely that other biochemicals could have this property as well. In
embryonic development where gene expression is determined by the concentration
gradient of a morphogen substance, the fact that simple amino acids can induce
periodicity over length scales at least up to ~10 cm (the length of our test tube) has
interesting implications. The results of Chapter 4, indicating that biochemicals can
induce banding in inorganic precipitates, also has implications for the study of
biosignatures. Future work in this area should include a detailed study of how
various amino acids, nucleic acids and other important biochemicals affect band

141
periodicity and/or crystal growth in naturally occurring minerals, as this would
provide an insightful tool for determining the past presence of organic material in the
rock record.

142
References

Abdel-Wahab, A. & McBride, E. F. (2001). Origin of Giant Calcite-Cemented
Concretions, Temple Member, Qasr El Sagha Formation (Eocene), Faiyum
Depression, Egypt. Journal of Sedimentary Research, 71(1), 70-81.

Beitler, B., Chan, M. A., & Parry, W. T. (2003). Bleaching of Jurassic Navajo Sandstone
on Colorado Plateau Laramide highs; evidence of exhumed hydrocarbon
supergiants?. Geology (Boulder), 31(12), 1041-1044.

Beitler, B., Parry, W. T., & Chan, M. A. (2005). Fingerprints of Fluid Flow: Chemical
Diagenetic History of the Jurassic Navajo Sandstone, Southern Utah, USA.
Journal of Sedimentary Research, 75, 547–561.

Beitler Bowen, B., Martini, B. A., Chan, M. A., & Parry, W. T. (2007). Reflectance
spectroscopic mapping of diagenetic heterogeneities and fluid-flow pathways in
the Jurassic Navajo Sandstone, American Association of Petroleum Geologists
Bulletin, 91(2), 173–190.

Blakey, R. C., Peterson, F., & Kocurek, G. (1988). Synthesis of late Paleozoic and
Mesozoic eolian deposits of the western interior of the United States. Sedimentary
Geology, 56, 3–125.

Blakey, R. C. (1994). Paleogeographic and tectonic controls on some lower and middle
Jurassic erg deposits, Colorado Plateau, in Caputo, M. V., Peterson, J. A., and
Franczyk, K. J., eds., Mesozoic Systems of the Rocky Mountain Region, USA:
SEPM, Rocky Mountain Section, p. 273–298.

Busigny, V. & Dauphas, N. (2007). Tracing paleofluid circulations using iron isotopes: A
study of hematite and goethite concretions from the Navajo Sandstone (Utah,
USA). Earth and Planetary Science Letters, 254, 272-287.

Chan, M. A., Parry, W. T., & Bowman, J. R. (2000). Diagenetic Hematite and
Manganese Oxides and Fault-Related Fluid Flow in Jurassic Sandstones,
Southeastern Utah. American Association of Petroleum Geologists Bulletin, 84(9),
1281-1310.

Chan, M. A., Beitler, B., Parry, W. T., Ormö, J., & Komatsu, G. (2004). A possible
terrestrial analogue for haematite concretions on Mars. Nature, 429, 731-734.

Chan, M. A., Beitler, B., Parry, W. T., Ormö, J., & Komatsu, G. (2005). Red rock and red
planet diagenesis: Comparisons of Earth and Mars concretions. GSA Today,
15(8), 4-10.

143

Chan, M. A., Parry, W. T., Park, A. S. (2005). Hematite "Blueberry" Concretion Doublet
and Triplets on Mars: Iron Oxide Twin Analogs From Utah. American
Geophysical Union Fall Meeting Supplement Abstract #P21A-0139.

Chan, M. A., Johnson, C. M., Beard, B. L., Bowman, J. R. & Parry, W. T. (2006). Iron
isotopes constrain the pathways and formation mechanisms of terrestrial oxide
concretions: A tool for tracing iron cycling on Mars? Geosphere, 2(7), 324 – 332.

Chan, M. A., Ormö, J., Park, A. J.;, Stich, M., Souza-Egipsy, V., & Komatsu, G. (2007).
Models of iron oxide concretion formation: field, numerical and laboratory
comparisons. Geofluids, 7, 1-14.

Chopard, B., Luthi, P., & Droz, M. (1994a). Reaction-Diffusion Cellular Automata
Model for the Formation of Liesegang Patterns. Physical Review Letters, 72(9),
1384-1387.

Chopard, B., Luthi, P., & Droz, M. (1994b). Microscopic Approach to the Formation of
Liesegang Patterns. Journal of Statistical Physics, 76(1/2), 661-677.

Chopard, B., Droz, M., Magnin, J., Rácz, Z., & Zrinyi, M. (1999). Liesegang Patterns:
Effect of Dissociation of the Invading Electrolyte. The Journal of Physical
Chemistry, 103, 1432-1436.

Cochran, K. A., & Elmore, R. D. (1987). Paleomagnetic dating of Liesegang bands.
Journal of Sedimentary Petrology, 57(4), 701-708.

Crick, F. (1970). Diffusion in Embryogenesis. Nature, 225, 420-422.

Dee, G. T. (1986). The Patterns Produced by Precipitation at a Moving Reaction Front.
Physica 23D, 340-344.

Droz, M., Magnin, J., & Zrinyi, M. (1999). Liesegang patterns: Studies on the width law.
Journal of Chemical Physics, 110(19), 9618-9622.

Eichhubl, P., Taylor, W. L., Pollard, D. D., & Aydin, A. (2004). Paleo-fluid flow and
deformation in the Aztec Sandstone at the Valley of Fire, Nevada—Evidence for
the coupling of hydrogeologic, diagenetic, and tectonic processes, Geological
Society of America Bulletin. September/October, 116(9/10), 1120–1136.

Feeney, R. E., Petersen, I. M., & Sahinkaya, H. (1956). “Liesegang-like” rings of growth
and inhibition of bacteria in agar caused by metal ions and chelating agents.
Journal of Bacteriology, 73, 279-283.

144
Feeney, R., Schmidt, S. L., Strickholm, P., Chadam, J., & Ortoleva, P. (1983). Periodic
precipitation and coarsening waves: Applications of the competitive particle
growth model. Journal of Chemical Physics, 78(3), 1293-1311.

Flicker, M., & Ross, J. (1974). Mechanism of chemical instability for periodic
precipitation phenomena. Journal of Chemical Physics, 60(9), 3458-3465.

Fujikawa, H. (1992). Periodic growth of Bacillus subtilis colonies on agar plates. Physica
A, 189, 15-21.

Garcia-Ruiz, J. M. (1994). Inorganic Self-Organisation in Precambrian Cherts. Origins
of Life and Evolution of the Biosphere, 24, 451-467.

García-Ruiz, J. M., Rondón, D., García-Romero, A., & Otálora, F. (1996). Role of
Gravity in the Formation of Liesegang Patterns. The Journal of Physical
Chemistry, 100, 8854-8860.

García-Ruiz, J. M., Carnerup, A., Christy, A. G., Welham, N. J., & Hyde, S. T. (2002).
Morphology: An Ambiguous Indicator of Biogenecity. Astrobiology, 2, 353-369.

Garrett, R. H., & Grisham, C. M. (2005). Biochemistry, 3rd Edition. Belmont: Thomson
Brooks/Cole.

George, J., & Varghese, G. (2002). Formation of periodic precipitation patterns: a
moving boundary problem. Chemical Physics Letters, 362, 8-12.

George, J., Paul, I., Varughese, P. A., & Varghese, G. (2003). Rhythmic pattern
formations in gels and Matalon-Packter law: A fresh perspective. Pramana,
60(6), 1259.

George, J., & Varghese, G. (2005). Intermediate colloidal formation and the varying
width of periodic precipitation bands in reaction-diffusion systems. Journal of
Colloid and Interface Science, 282, 397-402.

Golding, I., Kozlovsky, Y., Cohen, I., & Ben-Jacob, E. (1998). Studies of bacterial
branching growth using reaction-diffusion models for colonial development.
Physica A, 260, 510-554.

Gregor, T., Bialek, W., de Ruyter van Steveninck, R. R., Tank, D. W., & Weischaus, E.
F. (2005). Diffusion and scaling during early embryonic pattern formation.
Proceedings of the National Academy of Sciences, 102(51), 18403-18407.

Heaney, P. J. & Davis, A. M. (1995). Observation and Origin of Self-Organized Textures
in Agates. Science, 269, 1562-1565.

145

Henisch, H. K. (2005). Crystals in Gels and Liesegang Rings. Cambridge: Cambridge
University Press.

Hillson, P. J. (1961). Liesegang phenomenon - the importance of impurities in periodic
precipitation. Transactions of the Faraday Society, 57, 1031-1034.

Jablczynski, K. (1923). Bulletin de la Société Chimique de France, 33, 1592.

Kozlovsky, Y., Cohen, I., Golding, I., & Ben-Jacob, E. (1999). Lubricating bacteria
model for branching growth of bacterial colonies. Physical Review E, 59(6),
7025-7035.

Krug, H. J., Brandtstädter, H., & Jacob, K. H. (1996). Morphological instabilities in
pattern formation by precipitation and crystallization processes. Geologische
Rundschau, 85, 19-28.

Krug, H. J. & Brandtstädter, H. (1999). Morphological characteristics of Liesegang Rings
and Their Simulations. The Journal of Physical Chemistry A, 103, 7811-7820.

Lagzi, I. & Izsák, F. (2004). Stabilization and destabilization effects of the electric field
on stochastic precipitate pattern formation. Chemical Physics, 303, 151-155.

Lagzi, I. & Ueyama, D. (2009). Pattern transition between periodic Liesegang pattern and
crystal growth regime in reaction-diffusion systems. Chemical Physics Letters,
468, 188-192.

Lander, A. D., Nie, Q., & Wan, F. Y. M. (2002). Do Morphogen Gradients Arise by
Diffusion? Developmental Cell, 2, 785-796.

Lander, A. D. (2007). Morpheus Unbound: Reimagining the Morphogen Gradient. Cell,
128, 245-255.

LeVan, M. E., & Ross, J. (1987). Measurements and a Hypothesis on Periodic
Precipitation Processes. The Journal of Physical Chemistry, 91, 6300-6308.

Liesegang, R. E. (1896). Naturwiss. Wochenschr., 11, 353.

Lifshitz, I. M. & Slyozov, V. V. (1961). The Kinetics of Precipitation From
Supersaturated Solid Solutions. Journal of Physics and Chemistry of Solids,
19(1/2), 35-50.

146
Lindquist, Sandra J. (1988). Practical characterization of eolian reservoirs for
development: Nugget Sandstone, Utah-Wyoming thrust belt, Journal of
Sedimentary Geology, 56, 315-339.

Loughnan, F. C., & Roberts, F. I. (1990). Composition and origin of the 'zebra rock' from
the East Kimberly region of Western Australia. Australian Journal of Earth
Sciences, 37, 201-205.

Manley, D. R., & Stern, K. H. (1955). Liesegang rings in inhomogeneous media:
powdered glass. Journal of Colloid Science, 10, 409-412.

Marko, F., Pivko, D., & Hurai, V. (2003). Ruin marble: a record of fracture-controlled
fluid flow and precipitation. Geological Quarterly, 47(3), 241-252.

Matalon, R., & Packter, A. (1954). The Liesegang Phenomenon: I. Sol Protection and
Diffusion. Journal of Colloid Science, 10(1), 46-62.

McBride, E. F., Abdel-Wahab, A., & El-Younsy, A. R. M. (1999). Origin of spheroidal
chert nodules, Drunka Formation (Lower Eocene), Egypt. Sedimentology, 46,
733-755.

Mimura, M., Sakaguchi, H., & Matsushita, M. (2000). Reaction-diffusion modeling of
bacterial colony patterns. Physica A, 282, 283-303.

Morse, H. W., & Pierce, G. W. (1903). Diffusion and Supersaturation in Gelatine.
Proceedings of the American Academy of Arts and Sciences, 38(22), 625-648.

Morse, J. W., & Casey, W. H. (1988). Ostwald Processes and Mineral Paragenesis in
Sediments. American Journal of Science, 288, 537-560.

Msharrafieh, M., & Sultan, R. (2005). Patterns with High Rhythmicity Levels in
Multicomponent Liesegang Systems. ChemPhysChem, 6, 2647-2653.

Muller, S. C., & Ross, J. (2003). Spatial Structure Formation in Precipitation Reactions.
Journal of Physical Chemistry A, 107, 7997-8008.

Nasreddine, V., & Sultan, R. (1999). Propagating Fronts and Chaotic Dynamics in
Co(OH)
2
Liesegang Systems. The Journal of Physical Chemistry A, 103, 2934-
2940.

Nielsen, G. B., & Chan, M. A. (2006). Colorful diagenetic facies and fluid-related
alteration features of the Jurassic Navajo Sandstone, Snow Canyon State Park,
Utah. Geological Society of America Abstracts with Programs, 38(7), 518.

147
Ortoleva, P. (1982). Solute Reaction Mediated Precipitate Patterns in Cross Gradient Free
Systems. Zeitschrift für Physik B: Condensed Matter, 49, 149-156.

Ortoleva, P. (1984). From Nonlinear Waves to Spiral and Speckle Patterns:
Nonequilibrium Phenomena in Geological and Biological Systems. Physica 12D,
305-320.

Ortoleva, P. (1984). The Self Organization of Liesegang Bands and Other Precipitate
Patterns. In: G. Nicolis and F. Baras (eds.), Chemical Instabilities: Applications
in Chemistry, Engineering, Geology, and Materials Science, 289-297.

Ortoleva, P. (1994). Geochemical Self-Organization. Oxford Monographs on Geology
and Geophysics No. 23. Oxford: Oxford University Press.

Ostwald, W. (1897). Zeitschrift für Physikalische Chemie – Leipzig, 23, 356.

Panjarian, S., & R. Sultan (2001). Crystal selection and Liesegang banding in dynamic
precipitate systems. Collection of the Czechoslovak Chemical Communications,
66, 514–541.

Parry, W. T., Chan, M. A., & Beitler, B. (2004). Chemical bleaching indicates episodes
of fluid flow in deformation bands in sandstone. American Association of
Petroleum Geologists Bulletin, 88(2), 175-191.

Prager, S. (1956). Periodic Precipitation. Journal of Chemical Physics, 25(2), 279-283.

Raso, D. S., Greene, W. B., Finley, J. L., & Silverman, J. F. (1998). Morphology and
Pathogenesis of Liesegang Rings in Cyst Aspirates: Report of Two Cases With
Ancillary Studies. Diagnostic Cytopathology, 19(2), 116-119.

Rauprich, O., Matsushita, M., Weijer, C. J., Siegert, F., Esipov, S. E., & Shapiro, J. A.
(1996). Periodic Phenomena in Proteus mirabilis Swarm Colony Development.
Journal of Bacteriology, 178( 2), p. 6525-6538.

Ripszam, M., Nago, A., Volford, A., Izsak, F., & Lagzi, I. (2005). The Liesegang eyes
phenomenon. Chemical Physics Letters, 414, 384-388.

Seiler, W. M. (2008). Jurassic Navajo Sandstone of Coyote Buttes, Utah/Arizona:
Coloration and Diagenetic History, Preservation of a Dinosaur Trample Surface,
and a Terrestrial Analog to Mars. M.S. Thesis, University of Utah, Salt Lake City,
Utah, USA.

148
Shreif, Z., Mandalian, L., Abi-Haydar, A., & Sultan, R. (2004). Taming ring morphology
in 2D Co(OH)
2
Liesegang patterns. Physical Chemistry Chemical Physics, 6,
3461-3466.

Smith, D. A. (1984). On Ostwald's supersaturation theory of rhythmic precipitation
(Liesegang's rings). Journal of Chemical Physics, 81(7), 3102-3115.

Squyres, S. W., Grotzinger, J. P., Arvidson, R. E., Bell III, J. F., Calvin, W., Christensen,
P. R., Clark, B. C., Crisp, J. A., Farrand, W. H., Herkenhoff, K. E., Johnson, J. R.,
Klingelhofer, G., Knoll, A. H., McLennan, S. M., McSween Jr., H. Y., Morris, R.
V., Rice Jr., J. W., Rieder, R., & Soderblom, L. A. (2004). In Situ Evidence for
an Ancient Aqueous Environment at Meridiani Planum, Mars. Science, 306,
1709-1714.

Stern, K. H. (1954) Liesegang rings in inhomogeneous media: partially coagulated
gelatin. Journal of Colloid Science 9, 4, 329-337.

Stern, K. H. & Shniad, H. (1958). Diffusion in inhomogeneous media: capillary
diffusion in beds of glass beads. Journal of Colloid Science, 13, 24-31.

Stone, D. A. & Goldstein, R. E. (2004). Tubular precipitation and redox gradients on a
bubbling template. Proceedings of the National Academy of Sciences, 101(32),
11537-11541.

Swami, S. N. & Kant, K. (1967). Effect of Light on the Formation of Liesegang Rings of
Copper Chromate in Agar Agar Gel. Colloid and Polymer Science, 215(1), 60-
61.

Toramaru, A., Harada, T., & Okamura, T. (2003). Experimental Pattern Transitions in a
Liesegang system. Physica D, 183, 133-140.

Turing, A. M. (1952). The Chemical Basis of Morphogenesis. Philosophical
Transactions of the Royal Society of London. Series B, Biological Sciences,
237(641), 37-72.

Tuur, S. M., Nelson, A. M., Gibson, A. M., Neafie, R. C., Johnson, F. B., Mostofi, F. K.,
& Connor, D. H. (1987). Liesegang rings in tissue: How to distinguish Liesegang
rings from the giant kidney worm, Dioctophyma renale. The American Journal of
Surgical Pathology, 11(8), 598-605.

Volford, A., Izsak, F., Ripszam, M., & Lagzi, I. (2007). Pattern Formation and Self-
Organization in a Simple Precipitation System. Langmuir, 23(3), 961-964.

Abstract (if available)

Abstract Self-organized patterns can be formed in diffusion experiments where two interdiffusing electrolytes react to form an insoluble precipitate in a medium (such as gelatin) that permits diffusive motion of ions, but prevents product particles from moving from their site of formation. Various pattern morphologies such as periodic bands, dendritic crystals, and continuous precipitates can be formed in many types of diffusion media. We hypothesized that interfering ions that are not part of the dominant precipitation reaction could affect the formation of self-organized patterns in a reproducible fashion, and that the same precipitation reaction occurring in different diffusion media would produce different pattern morphologies (depending on the physical and chemical properties of that medium).

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Carbonate geochemistry in primary, diagenetic and biological systems

PDF

New perspectives on ancient microbes and microbialites: from isotopes to immunology

PDF

Bioturbation in Cambrian siliciclastic shelf strata: paleoecological, paleoenvironmental, and temporal patterns

PDF

Non invasive detection of microbes in natural environments using deep UV native fluorescence spectroscopy and hyperspectral imaging

PDF

Shewanella spc. 16S rDNA signal attenuation due to UVC, gamma and cryogenic lab conditions

PDF

Stromatolites in the ancient and modern: new methods for solving old problems

PDF

Self-assembly and self-repair by robot swarms

PDF

Multiscale and multiresolution approach to characterization and modeling of porous media: From pore to field scale

PDF

Evolution & ecology of Mesozoic birds: a case study of the derived Hesperornithiformes and the use of morphometric data in quantifying avian paleoecology

PDF

Psychological correlates of certainty strength: measuring the relative influence of evidence, opinion of personal contacts & importance to self-identity

PDF

Paleoenvironments and the Precambrian-Cambrian transition in the southern Great Basin: Implications for microbial mat development and the Cambrian radiation

PDF

The geobiology of fluvial, lacustrine, and marginal marine carbonate microbialites (Pleistocene, Miocene, and Late Triassic) and their environmental significance

PDF

Paleoecology of Upper Triassic reef ecosystems and their demise at the Triassic-Jurassic extinction, a potential ocean acidification event

PDF

Stromatolites as biosignatures and paleoenvironmental records: experiments with modern mats and examples from the Eocene Green River Formation

PDF

Unraveling mass extinctions: Permian to Early Jurassic onshore-offshore trends of marine stenolaemate bryozoans

PDF

Social support, self-efficacy, and gender in treatment adherence of heart failure patients

PDF

Assessing the quality of the fossil record using a phylogenetic approach

PDF

Synthesis and mechanical evaluation of micro-scale truss structures formed from self-propagating polymer waveguides

PDF

Ages, origins and biogeochemical role of water across a tropical mountain to floodplain transition

PDF

The early Triassic recovery period: exploring ecology and evolution in marine benthic communities following the Permian-Triassic mass extinction

Asset Metadata

Creator Barge, Laura Marie J. (author)

Core Title Self-organized chemical precipitates: laboratory and field studies

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Geological Sciences

Publication Date 10/16/2009

Defense Date 06/03/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Liesegang bands,OAI-PMH Harvest,silver chromate

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Nealson, Kenneth H. (committee chair), Bottjer, David J. (committee member), Corsetti, Frank A. (committee member), Hammond, Douglas E. (committee member), Petruska, John A. (committee member)

Creator Email barge@usc.edu,laurie.barge@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2672

Unique identifier UC1124929

Identifier etd-Barge-3330 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-268824 (legacy record id),usctheses-m2672 (legacy record id)

Legacy Identifier etd-Barge-3330.pdf

Dmrecord 268824

Document Type Dissertation

Rights Barge, Laura Marie J.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu