Chapter 1

An Introduction to Soil Mineralogy

OARRELL G SCHULZE, Purdue University, West Lafayette. Indiana

MineraIs make up about one-halfofthe volume ofmost soils. They provide physical support for plants and create the water- and air-filled pores that make plant
growth possible. Mineral weathering releases plant nutrients that are retained by
other minerais through adsorption, cation exchange, and precipitation. Minerais are
indicators ofthe amount ofweathering lhat has taken place and the presence or absence ofparticular minerais gives c1ues to how soils fonned. The physical and chemical characteristics ofsoil minerais are important considerations in planning, constructing, and maintaining buildings, roads, and airports. Soil mineraIs can adsorb
many organic and inorganic environmental pollutants, promoting their degradation
to nontoxic fonns, attenuating their movement through the soil, or preventing their
uptake by plants and their introduction into lhe food chain. Some minerais are themselves pollutanls and can cause serious environmental problems when they are exposed to weathering at the soil surface by human activities. An understanding of
soil mineralogy is central lo understanding virtually ali facets ofhumanity's use and
misuse ofsoils and is ofien the key to solving specific environmental problems.
This chapter develops a core of concepts and tenninology needed for understanding soil mineraIs. The chemical composition ofthe Earth's crust is discussed
first to show that the most abundant elements in the crust are, not surprisingly, the
ones most likely to be encounlered in soil minerais. Then the chemical and structural c1assification ofminerals is discussed, and the major mineraIs represented in
soils are mentioned. The phyllosilicate mineraIs are covered separately because of
their major role in soils. Basic structural concepts common to ali minerais are covered at lhis point to provide the background for the discussion ofthe phyllosilicate
structures. The overall structural theme ofthe phyllosilicates is then presented, along
with a very brief summary oftheir most importanl properties. The chapter concludes
with a treatment ofsome crystallographic and x-ray diffraction concepts important
in identifYing and characterizing soil minerais.

I. CHEMICAL AND STRUCTURAL CLASSIFICATION OF MINERALS

A. Composition ofthe Earth's Crust
Most ofthe weight and volume oflhe Earth's crusl is made up ofonly a few
elements (Table 1-\). Oxygen and Si make up mosl of lhe weight, while oxygen
Copyright 2002 Cd Soil Science Society ofAmerica, 677 South Segoe Road. Madison. WI 53711, USA.
Soil Mineralogy with Environmental Applications. SSSA Book Series. no. 7.

SCHULZE

Table I-I. lhe 12 most eommon ehemieal e1ements in the Earth's erust (after Klein & Hurlbut. 1993).
Element

Crustal average

Mole fraetion

gkg- I
O
Si
AI
Fe
Ca
Na
K
Mg
Ti
H

466.0
2772
81.3
50.0
36.3
28.3
25.9
20.9
4.4

P
Mn

1.0
0.9

1.4

0.6057
0.2052
0.0626
0.0186
0.0188
0.0256
0.0138
0.0179
0.0019
0.0289
0.0007
0.0003

lonie radiust

Volume

nm

0.136 (3)
0.026 (4)
0.039 (4)
0.078 (6)
0.100(6)
0.\02(6)
0.151 (8)
0.072 (6)
0.061 (6)
~
0.017 (4)
0.083 (6)

92.88
0.22
0.23
0.54
1.15
1.66
2.89
0.41
0.03
~
<0.01
0.01

t Numbers in parentheses refer to coordination number. Radii for Fe and Mn are for lhe redueed (2+)
fonn.
~ lonie radius and volume ofW are negligible eompared wilh ()l-.

alone accounts for more than 90% of the volume. ln general, the larger O atoms
are in an approximately close-packed arrangement held together by smaller metal
atoms in the interstitial space. Most ofthe elements in the crust and in soils occur
in minerais. Thus, the elements listed in Table I-I are major constituents ofthe minerais discussed in this book.

B. Definition or a Mineral
Klein and Hurlbut (1993) define a mineral as follows: "A mineral is a naturally occurring homogeneous solid with a definite (but not generally fixed) chemical composition and a highly ordered atomic arrangement. It is usually forrned by
inorganic processes." 80th chemical composition and crystal structure (ordered
atomic arrangement) are important parts of this definition. Neither alone is sufficient to explain the properties of minerais. Minerais with similar chemical composition but different crystal structures-or, conversely, minerais with similar crystal structures but different chemical compositions-<:an be quite different from one
another despite their chemical or structural similarities.

C. Mineral Classification
Although different classification schemes could be used, mineralogists have
deterrnined that first separating mineraIs into groups on the basis oftheir chemical
composition gives classes with the greatest similarities in many other properties.
Thus, minerais are first divided into classes depending upon the dominant anion or
anionic group. The classes include: (i) native elements. (ii) sulfides, (iii) sulfosalts,
(iv) oxides and hydroxides. (v) halides, (vi) carbonates, (vii) nitrates, (viii) borates,
(ix) phosphates, (x) sulfates, (xi) tungstates, and (xii) silicates (Klein & Hurlbut,
1993). These classes are then subdivided based on chemical and structural similarities. This sarne general classification is followed in this book, but with some exceptions. Mineral classes such as the native elements, sulfosalts. nitrates. borates,

INTRODUCTION TO SOIL MINERALOGY

Table 1-2. Common nonsilicate minerais in soils.
Mineral c1ass

Mineral

Chemical fonnulat

Halides
Sulfates

Halite
Gypsum
Jarosite
Pyrite
Calcite
Dolomite
Nahcolite
Trona

NaCI
CaSO. 2H 20
Kfe)(SO.h(OH>6

Sulfides
Carbonates

Oxides and hydroxides

Aluminum
lron

Manganese

Titanium

Fe~

Soda

CaCO)
CaMg(CO)h
NaHCO]
Na2CO)oNaHCO)02H20
Na2CO)0IOH20

Gibbsite
Hematite
Goethite
Lepidocrocite
Maghemite
Ferrihydrite
Magnetite
Bimessite
Lithiophorite
Hollandite
Todorokite
Rutile
Anatase
Omenite

AI(OHh
Fe20]
FeOOH
FeOOH
Fe2 0 ]
Fej-7(OH)04H 20
fe)O.
(Na,Ca,Mn2+) Mn,o.02.8 H20
LiAI2Mn2+Mn]+ 06(OH)6
.
Ba(Mn+,Mn]+)8 0 16
(Na,Ca,K>O.]-ll.s(Mn+,Mn)+)6 0 1203.SH 20
Ti0 2
TiO,
Fe2+-TiO,

t After Klein & lIurlbun (1993), Kmpf et aI. (1999).

and tungstates that occur only rarely in soils, are covered only brietly or not at ali.
Commonly occurring mineral classes, particularly the silicates, are covered in detail.
Soil minerais are also referred to as either primary or secondary minerais. Primary minerais form at elevated temperatures. They are usually derived from igneous
or metamorphic rocks, but they can be inherited from sedimentary rocks as well.
Secondary minerais are formed by low-temperature reactions and are either inherited from sedimentary rocks or formed in soils by weathering (Jackson, 1964). The
separation ofminerals into primary and secondary mineral classes is not mutually
exclusive; some minerais can occur in both. The concept is useful however, and appears widely in the soil science literature. The major mineral classes represented
in soils are discussed below.

lo Halide, Sulfate, and Carbonate Minerais

The major soil minerais of this group are halite (NaCI), gypsum
(CaS042H20), calcite (CaC0 3), and dolomite [CaMg(C0 3 hl (Table 1-2). This
group is characterized by minerais with relatively simple structures. The halite structure is one ofthe simplest ofali minerais. It consists ofaltemating Na+ and CI- ions
arranged in cubic c10sest packing (Plate I-I). The other minerais in this group have
similar structures, with cations such as Ca 2 +, Mg 2+, or Fe 2 +altemating with anions
such as CI-, S01-, or COI-o The bonds between the cations and anions are predominantly ionic. These mineraIs are among the most soluble and the softest of ali

SCHULZE

soil minerais, and they are easily broken down by physical and chemical weathering processes. They occur mainly in soils ofarid regions or in youthful soils in more
humid regions, where weathering has been minimal.
Halite is the most soluble ofthis group and accumulates in only the most arid
environments. It is one ofthe minerais present in the salic horizon ofAridisols. Gypsum is about 100 times less soluble than halite, but it too is abundant only in arid
regions. Gypsum is a major mineral in the gypsic horizon of Aridisols. Halite and
gypsum could also occur in soils that have been contarninated by salt water, or by
soluble salts leaching from industrial stock piles or waste piles.
Calcite and doJomite are common carbonate mineraIs that occur in a variety
ofsoils. These minerais precipitate in the soil profile in arid and semiarid climates
(aridic and some ustic soiJ moisture regimes). Calcic and petrocalcic horizons
fonn ifthe accumulation is great enough. Carbonate minerais in many soils are inherited from Jimestone or other calcareous parent materiais. Carbonates are usually stable and can be found throughout the soil profile under aridic and ustic soil
moisture regimes (arid to subhumid cJimates). but are leached from the soil profile
and are generally present only in the C horizons under the udic soil moisture
regime (humid climates). Calcite and dolomite can be introduced into soils originally free ofthese minerais via lhe limestone aggregate used for road construction
in some areas.

2. Sulfides
Pyrite, FeS2' the most common mineral in this group. does not occur extensively in soils. but when it is present it causes some unique problems. Pyrite precipitates in soils on wet tidal tlats and river deltas of some coastal areas and also
occurs in some geologic fonnations originally deposited in similar environments.
Thus. pyrite ofien occurs in cJose association with coaI. Pyrite is unstable under oxidizing conditions and weathers quickly when pyritic soils are drained or when mining leaves pyritic material on the surface. The weathering products incJude the minerais jarosite, KFe3(S04h(OH)6' and gypsum and sulfuric acid, H2S04. The large
amount of acidity causes problems in using pyrite-containing soils and in revegetating mined areas and is a serious and costly environrnental problem.

3. Oxides, Hydroxides, and Oxybydroxides

Primary minerais break down during weathering and release cations and anions that recombine to fonn other more stable minerais. Several elements, in particular AI. Fe. and Mn. fonn oxide. hydroxide. or oxyhydroxide minerais that are
stable in the soil weathering environment. Representative mineral species are given
in Table 1-2.
a. Aluminum. Gibbsite, [AI(OHhl is the most common AI hydroxide mineraI in soils. It is generally associated with the latter stages ofweathering when leaching ofsilica has progressed to the point that phyllosilicate minerais no longer fonn.
Gibbsite is common in highly weathered Oxisols oftropical regions. It has a very
low cation exchange capacity and contributes to the low native fertility of some Oxisols. Gibbsite is also commonly found at the weathering interface between igneous
rock and saprolite and in Andisols fonned from volcanic ash.

INTRODUCTION TO SOIL MINERALOGY

b. lron. lron oxide minerais fonn from Fe released from primary mineraIs.
Iron oxides are strong pigments and small amounts of these minerais account for
most ofthe brown and red colors ofsoils. Goethite (FeOOH) is the most common
mineral ofthis group. It accounts for the brownish to yellowish color of many soils,
although it may be present in only small quantities. Hematite (Fe203) is only
slightly less common than goethite and usually occurs in association with it.
Hematite is usually bright red and is responsible for the red color of many soils.
Goethite and hematite are stable minerais in an oxidizing environment. Large
amounts ofthese two minerais in well-aerated soils, usually in association with gibbsite and kaolinite, usually indicate an advanced stage of weathering. ln soils that
are saturated with water for at least some time during the year, the very insoluble
Fe 3+ in goethite, hematite, and other Fe oxides can be reduced to very soluble Fe 2+.
The Fe 2- can easily move with the soil water to other parts ofthe soil profile, or
even to other associated soils in the landscape, where it can then reoxidize to Fe3+
and reprecipitate as goethite, lepidocrocite, or ferrihydrite when oxidizing conditions retum. Repeated cycles of oxidation and reduction give rise to mottJes and
nodules that ref1ect an inhomogeneous distribution of Fe oxide minerais within a
soil. Soil scientists use these inhomogeneous or mottled color patterns to estimate
the depth to a soil water table and the length oftime the soil is saturated during the
year. Thus, when delineating wetlands, detennining the best location for a septic
system leach field, or mapping soils, soil scientists rely on predictable soil color
patterns that result from Fe oxide mineralogy and distribution.

e. Manganese. Manganese oxides and hydroxides (Table 1-2) are commonly found in soils as brown or black nodules or as thin coatings on the faces of
soil structural units. They are ofien associated with Fe oxides. Manganese occurs
frequentlyas birnessite or lithiophorite in soiJs. Manganese can be oxidized and reduced in the soil environment similar to Fe. Thus, the inhomogeneous distribution
ofMn into nodules is an indicator of reduction and oxidation as a result of periodic
water saturation.
d. Titanium. Rutile and ilmenite (Table 1-2) occur in soils mainly as primary
minerais inherited from igneous rocks. Anatase is less common and is generally considered a secondary mineral. Although frequently found in soils, these minerais generally do not impact soil physical or chemical properties.

4. Sitieates
The silicate mineral c1ass is an extremely large and important group ofminerals. Nearly 40% ofthe common minerais are silicates, as are most minerais in igneous rocks. Silicates constitute well over 90% ofthe Earth's crust (Klein & Hurlbut, 1993) and comprise the bulk of most soils as well. Silicates occur as both
primary minerais inherited from igneous or metamorphic rocks and as secondary
minerais formed from the weathering products ofprimary minerais.
As explained in more detail below, the fundamental unit ofall silicate structures is the Si04 tetrahedron. It consists of four 0 2- ions at the apices of a regular
tetrahedron coordinated to one Si 4+ at the center. The individual tetrahedra are linked
together by sharing 0 2- ioos to fonn more complex structures. Several different

SCHULZE

6
Table )-3. Classification ofsilicate minerais.
Silicate c1ass, unit composition,
arrangement ofSi0 4 tetrahedrat
Nesosilicates (Si0 4 )4-

Mineral

Ideal fonnulat

Olivine
Forsterite
FayaJite
Zircon
Sphene
Topaz
Gamets

(Mg.FehSiO.
Mg2SiO.
Fe2SiO.
ZrSiO.
CaTi(Si04 )
AI 2SiO.(F,OHh
X)Y 2(SiO.h. X = Ca,Mg,Fe2+,
Mn 2+. Y = AI,Fe)+'Crl+
AI 2SiOs
AI 2SiOs
AI2SiOs
Fe2Alq06(SiO.).(O,OHh

Andalusite
Sillimanite
Kyanite
Staurolite
Sorosilicates (Si 20 7)1t-

Av

Cyclosilicates (Si 60

18 )12-

Epidote

Ca2(AI,Fe)AI2(SiO.)
(Si 20 7)(OH)

Beryl
Tounnaline

Be)AI 2(Si 60 18 )
(Na,Ca)(Li,Mg.AI)(AI,Fe,Mn)6
(BO)h(Si 60 18 )(OH).

Inosilicates
(single chains) (Si03)2-

Pyroxenes
Augite
Enstatite
Hypersthene
Diopside
Hedenbergite
Pyroxenoids
Wollastonite
Rhodonite

Inosilicates
(double chains) (Si.OI\)1t-

Amphiboles
Homblende
Tremolite
Actinolite
Cummingtonite
Grunerite

(Ca,Na)(Mg.Fe,AI)(Si,Alh06
MgSiO)
(Mg.Fe)SiO)
CaMgSi 20 6
CaFeSi 20 6

(Ca,Nah.)(Mg.Fe,AI)sSi6
(Si,AI}z22(OHh
Ca2MgsSis022(OHh
Ca2(Mg,Fe)sSisOn(OHh
(Mg.Fe),Si s0 22(OHh
Fe7SisOn(OHh

(continued on next page)

arrangements ofthe Si04 tetrahedra occur, partly accounting for the large number
of silicate minerais and providing the basis for their c1assification. The tetrahedra
may be present as single tetrahedra (nesosilicates), double tetrahedra (sorosilicates),
rings (cyclosilicates), single or double chains (inosilicates), sheets (phyl1osilicates), or three-dimensional frameworks (tectosi Iicates) (Table 1-3). ln most ofthese
arrangements adjacent Si04 tetrahedra share comers; that is, they share a common
0 2-. The more common minerais from each silicate c1ass likely to be found in soils
are given in Table 1-3.
The most common mineraIs in igneous rocks are the olivines, pyroxenes, amphiboles, micas. fe1dspars, and quartzo These primary mineraIs make up most ofthe
sand and silt size fractions of soils. The clay 2 ~m) fraction of most soils consists primarily ofphyl1osilicate mineraIs. Because oftheir importance in soils and
sediments, phyl10silicates are discussed in detail below and in other chapters ofthis

book.

INTRODUCTION TO SOIL MINERALOGY

Table 1-3. Continued.

Silicate c1ass, unit composition,
arrangement ofSi0 4 tetrahedrat
PhyJlosilicates (Si~05)~-

Mineral
Micas
Muscovite
Biotite
Phlogopite
Chlorites
Clay minerais
(seJected examples)
Tale
PyrophyJlite
Kaolinite
Smectite
Vermiculite
Serpentines
Antigotite
Chrysotile

Tectosilicates (Si02)o

Feldspars
Orthoclase
Albite
Anorthite
Si0 2 Group
Quartz
Tridymite
Cristobalite
Zeolites
Analeime
Feldspathoids
Nephelene

Ideal formulat
KAI 2(AISi)OIO)(OHh
K(Mg,Feh(AISi)OIO)(OHh
KMg)(AISi)OIO)(OHh
(Mg,Feh(Si.AI)401O(OHh
(Mg,Feh(OH)(,

Mg)Si 4O IO(OHh
AI 2Si4O IO(OHh
AI 2Si 2O;(OH)4
M+o)AI2(Alo.)S3.7l0IO(OHb
M+= Ca2+, Mi+' K+, etc.
M+o7AI2(Alo.7Siu)OIO(OHh.
M+= Ca2+, Mg 2+, K+, etc.
Mg)Si 20 5(OH)4
Mg)Si 2O;(OH)4
KAISi)O.
NaAISi)O.
CaAI 2Si 2O s
Si02
Si02
Si0 2
NaAISi 20 6-H 2O
(Na.K IAISi04

t After Allen & Fanning (1983).

t Klein & Hurlbut (1993).

II. PHYLLOSILlCATE MINERALS lN SOILS

Phyllosilicate minerais strongly inf1uence both the chemical and physical
properties of soils because of their generally small particle sizes. high surface
areas, and unique cation exchange properties. A clear understanding ofthe phyllosilicate minerais is central to understanding soil c1ay mineralogy and many environmental processes.
The discovery in the early 1920s that clays are crystalline was the key to understanding many ofthe properties ofsoils and c1ays. The crystal structures ofthe
phyllosilicates define the different mineral species and account for many oftheir
unique properties. The structural schemes of these minerais are discussed below.
The purpose is to give an overview ofthe most commonly occurring structures and
the major properties of each major mineral group. Subsequent chapters will add details, ifnecessary. Emphasis will be placed on developing an understanding ofthe
various structural representations encoontered in the literature.

SCHULZE

A. Basic Structural Concepts

The following discussion will treat atoms as rigid spheres. This is an oversimplified but convenient model for developing the key concepts ofmineral structures.

I. Closest Packing ofSpheres

The most efficient way of packing spheres in one plane is called hexagonal
c10sest packing (Plate 1-2). Each sphere touches six of its neighbors and a hexagon results ifthe centers ofthese six neighbors are connected. There are two types
of voids between the spheres: A voids (where the triangular outline points downward in the illustration) and B voids (where the triangular outline points upward).
Ifa second plane ofspheres is placed on top ofthe first (Plate 1-3), and ifthe first
sphere ofthe second plane is placed in the dimple formed by the three spheres surrounding an A void, then ali the remaining spheres of the second plane must be
placed over A voids as well. The spheres illustrated in Plate )-3 represent 0 2-, OH-,
or F- in the phyllosilicate structures, while positively charged cations occupy the
space between these negatively charged anions.

2. Tetrahedra and Octahedra

Another examination ofPlate )-3 shows that there are two different types of
sites into which cations can be placed. The A sites, formed over the A voids ofthe
lower plane of spheres, are called tetrahedral sites because the four 0 2- ions surrounding this site form the apices ofa regular tetrahedron. The B sites, formed over
the B voids of the lower plane of spheres, are called octahedral sites because the
six 0 2- ions surrounding each ofthese sites form the apices ofa regular octahedron
(Plate \-3). This illustrates that both octahedral and tetrahedral sites are a consequence ofthe closest packing oftwo planes ofspheres.
A cation in a tetrahedral site is in fourfold (or tetrahedral) coordination because it is bonded to four 0 2- ions whose centers define the apices (corners) of a
tetrahedron. Likewise, a cation in an octahedral site is in sixfold (or octahedral) coordination because it is bonded to six 0 2- ions whose centers define the apices of
an octahedron. Tetrahedrally and octahedrally coordinated cations-tetrahedra and
octahedra for short-are common structural elements in many mineral structures.
The approach ofdescribing mineral structures as assemblages ofpolyhedra is used
widely and, once mastered, provides an efficient way to understand even complex
mineral structures.

3. Representing Tetrahedra and Octabedra

Octahedra and tetrahedra are commonly represented using three different
types ofmodels. Each model portrays the sarne concept, but different representations are necessary to show various structural features. Plate )-3 shows two views
ofa tetrahedron and ofan octahedron drawn in these three different ways.
The sphere-packing model gives an impression ofthe space occupied by the
atoms in a structure. The sphere-packing model is particularly useful for visualiz-

INTRODUCTION TO SOIL MINERALOGY

ing the shape ofthe external surface ofa mineral, but it does not allow one to easily see the interior ofthe structure. The ball-and-stick model shows the bonds within
the crystal since the bonds are represented as sticks connecting the balls. The tetrahedra and octahedra are more difficult to visualize, but the interior ofthe structure
is easier to see because the atoms are drawn much smaller than in the close-packed
model. The polyhedral models give the best impression oftetrahedra and octahedra since the atoms are represented only as points in space or as small spheres. The
centers ofthe 0 2- ions are connected by lines to form the edges oftetrahedra and
octahedra. Thus, each apex represents the position of an 0 2- ion, while a cation resides in the center ofthe tetrahedron or octahedron. ln general, only those tetrahedra or octahedra actually occupied by cations are represented by a solid polyhedron;
sites that are not occupied by cations are not represented by solid polyhedra. One
should use care in interpreting polyhedral models, since it is easy to get the impression that a structure contains a large amount of open space or pores. The sarne
structure drawn as a c1ose-packed model will show that much ofthe space is actually occupied by the large O atoms. Other ways ofrepresenting mineral structures
are encountered as well. Sometimes, polyhedral models are drawn without spheres
to represent the oxygens and hydroxyls at the corners ofthe polyhedra (see, for example, Plate 1-9later in this chapter), while other models show only sticks to represent bonds, but no spheres to represent the atoms.
There is some ambiguity associated with each type of representation, and it
is important to understand the correspondence between, and limitations of, each of
the different structural representations. Thus, it is important to use as many different leaming aids as possible. Computer graphics programs that allow the viewer to
rotate and manipulate structural models on the computer screen offer significant
advantages over static drawings. Physical models should be studied as well, ifthey
are available. Far from being child's play, building your own models is one ofthe
best ways to learn about mineral structures. Models can be built with inexpensive
balls and some glue, or with paper polyhedra (see Moore & Reynolds [1997] for
templates).

4. Ionic Radii and Radius Ratios

Whether or not a given atom can occupy a tetrahedral or an octahedral site
depends both on the atom's charge and size. The size ofatoms and ions varies depending on their atomic number, ionization state, and coordination number (Table
I~).

The central void within a tetrahedron is smaller than the one within an octahedron. ln other words, the largest sphere that can be placed into a tetrahedral site
without pushing the 0 2- ions apart is smaller than the sphere that can be placed into
an octahedral site without pushing. Simple calculations (see, for example, Klein &
Hurlbut, 1993) illustrate that the maximum radius of a sphere that will just fit in
the tetrahedral site is 0.225 times the radii ofthe four surrounding 0 2- ions. For an
octahedral site, the maximum radius is 0.414 times the radi ofthe six surrounding
0 2- ions. The next larger type of site (not illustrated) is a cubic site consisting of
eight 0 2- ions arranged at the corners of a cube. For a cubic site, the maximum radius is 0.732 times the radii ofthe eight surrounding 0 2- ions. Assurning a radius

"

SCHULZE

Table 1--4. lonic radii and coordination ofcommon elements in phyllosilicate minerais.
lon

Jonic radiust

fP--

0.140(6)
0.133(6)
0.181(6)
0.026(4)
0.054(6)
0.065(6)
0.072(6)
0.061(6)
0.078(6)
0.083(6)
0.118(8)
0.112(8)
0.151(8)
0.142(8)
0.161(8)

Typical coordination

nm

F""

aSi'"
A13+
Fe3+

~+
Fe2+
Mn2+
Na+
cr+
K+
&2+
Rb'

Tetrahedral
Tetrahedral and octahedral
Tetrahedral and octahedral
Octahedral
Octahedral
Octahedral
Octahedral
Cubic and dodecahedral
Cubic and dodecahedral
Cubic and dodecahedral
Cubic and dodecahedral
Cubic and dodecahedral

t lonic radii vary with coordination number. Radii are for ions wilh lhe coordination nurnbers listed in
parentheses (fram Klein & Hurlbut. 1993. p. 188).

of 0.140 nm for the 0 2- ions, the limiting radius for a tetrahedral site is 0.032 nm
(0.140 nm x 0.225 = 0.032 nm), for an octahedral site it is 0.058 nm, and for a cubic
site it is 0.102 nm. The Si 4 + ion has a radius <0.032 nm (Table 1--4). Thus, Si 4 + always occurs in tetrahedral rather than octahedral sites. The Mg2 ion has a radius
that is >0.032 so Mg2 is unlikely to occur in tetrahedral coordination, but since its
radius is between 0.058 and 0.102 nm, it fits easily in an octahedral site. The A13+
ion, with a radius near to the limiting radius ofO.032 nm, can fit in both tetrahedral and octahedral sites. Cations with radii >-0.102 nm tend to occur in cubic, dodecahedral (12-fold) or higher coordination sites rather than in octahedral or tetrahedral sites (Table 1--4).
T

5. Tetrabedral and Octabedral Sbeets

Tetrahedra arranged into sheets are common to the structures of ali phyllosilicates. Octahedra arranged into sheets are present in the structures of phyllosilicates and in some hydroxide minerais as well. The structure oftetrahedral and
octahedral sheets will be discussed first. Then it will be shown how different combinations of tetrahedral and octahedral sheets give rise to the different clay mineral structures. Most ofthe diagrams in this chapter show ideal crystal structures
based on the approximate c10sest packing of spheres.
a. Tbe Octabedral Sbeet. Consider two planes of spheres representing hydroxyl (OH-) ions in hexagonal closest packing. (The W takes up very little space
and the OH- can be considered a sphere of roughly the sarne size as an 0 2- ion).
There are two ways offilling the octahedral sites depending on the valence ofthe
cation. A divalent cation, such as Mg can be placed into each octahedral site to

2"

INTRODUCTlON TO SOIL MINERALOGY

11

obtain the network of octahedra illustrated in Plate l-4a. This arrangement is

called trioctahedral because three of every three octahedral sites are occupied by
a cation. This gives a formula ofMg 3(OH)6 or Mg(OHh for the whole sheet; each
Mg 2+is surrounded by six OH-s, but since each OH- is shared equally among three
different Mg 2+atoms, each OH- contributes only one-third of its negative charge
to each Mg2+, giving the formula Mg(OHh- The sheet is electrically neutral because
the charge is balanced within the sheet.
The other possibility is to place a trivalent cation, such as AJ3+, into the octahedral sites. To preserve charge neutrality, only two trivalent cations are needed
compared with three divalent cations, giving the formula AI 2(OH)6 or AI(OHh.
Thus, only two ofevery three possible octahedral sites are filled. This arrangement
is called dioctahedral and is illustrated in Plate l-4b. The unit formula for the dioctahedral sheet can be deduced by following the sarne reasoning used above. Each
AJ3+ is surrounded by six OH-s, but since each OH- is shared equally between two
different A13+ ions (not three as for Mg2+), each OH- contributes one-halfof its negative charge to each AP+, giving the formula AI(OHh. Again the charge is balanced
and the sheet is electrically neutral.
Note how the trioctahedral and dioctahedral sheets appear in the different
types ofrepresentations. The sphere-packing model shows that both the trioctahedral and dioctahedral sheets contain the sarne dense packing ofOH-s. The pattem
ofunoccupied octahedral sites is more evident in the polyhedral model, where unoccupied octahedra appear as open spaces, and in the ball-and-stick model.
Octahedral sheets, stacked one on top of the other and held together by H
bonds. make up the structures of gibbsite (see Plate 1-7 below) and brucite
[Mg(OHhl. Gibbsite and brucite differ in that gibbsite contains AI in the octahedral sites and is dioctahedral, while brucite contains Mg in the octahedral sites and
is trioctahedral. The structures of gibbsite and brucite are the simplest in a series
of structures containing octahedral sheets.
The discussion above followed a sphere-packing approach in which we
showed that the structure of the octahedral sheets could be derived assuming that
the large O atoms were hard spheres of identical size c10sely packed in three dimensions. Thus, we assumed that the O atoms were our basic structural units and
that the cations merely served to balance the negative charge ofthe oxygens. This
approach works for some of the oxide and hydroxide minerais, but in many mineraI structures the O atoms are not as c10sely packed as strict geometric packing
would dictate. A more general way of describing mineral structures is the polyhedral approach. ln the polyhedral approach. the coordination polyhedra. primarily
octahedra and tetrahedra in the phyllosilicates, are considered the basic structural
building blocks. Thus. we can consider the octahedral sheets illustrated in Plates
1-4a and 1-4b as an assemblage of octahedra in which adjacent octahedra share
two oxygens with one another. ln other words, the octahedra share edges with one
another. ln general, coordination po1yhedra can share comers (one shared oxygen),
edges (two shared oxygens), or faces (three or more shared oxygens) with adjacent
polyhedra. The octahedral sheet could be described using either the sphere packing or polyhedral approaches. The tetrahedral sheet described below can only be
described using the polyhedral approach because the oxygens in the tetrahedral sheet
are not c1ose-packed as they are in the octahedral sheet.

12

SCHULZE

b. The Tetrahedral Sheet. The tetrahedral sheet consists ofSi0 4 tetrahedra

arranged such that three O~- ions of each tetrahedron are shared with the three nearest neighbor tetrahedra (Plate 1-5). These shared O~- ions are ali in the sarne plane
and are referred to as basaloxygens. Note that two adjacent tetrahedra share only
one O~- between them. (The tetrahedra share apices or comers.) The fourth O~- ion
of each tetrahedron is not shared with another Si0 4 tetrahedron and is free to bond
to other polyhedral elements. These unshared O~- ions are referred to as apical oxygens. Since each basal oxygen contributes a formal charge of -I to each Si 4+ ion,
the addition ofW ions to the apical oxygens to form hydroxyls should result in an
electrically neutral tetrahedral sheet. Such individual tetrahedral sheets. although
sometimes postulated to occur as transient weathering products in aqueous solution, do not stack to form stable mineral structures on their own. Thus. tetrahedral
sheets only occur in combination with octahedral sheets as described below.
The upper portion of Plate 1-5 shows ali of the apical oxygens pointing in
the sarne direction, namely. out ofthe plane ofthe paper towards the reader. Structures in which the apical oxygens of a single sheet ali point in the sarne direction
are the most common. but structures also occur in which the apical oxygens point
altemately in opposite directions. MineraIs containing this sheet-like arrangement
ofSi0 4 tetrahedra are called phy/losilicates', sheet silicates, or layer silicates.

B. Phyllosilicate MineraIs Common in Soils

Phyllosilicates are divided into two groups, I: I and 2: I type minerais. on the
basis ofthe number oftetrahedral and octahedral sheets in the layer structure.

1. The I: I Type MineraIs

a. The I: I Layer Structure. The I: I layer structure consists ofa unit made
up of one octahedral and one tetrahedral sheet, with the apical O~- ions ofthe tetrahedral sheets being shared with (and part of) the octahedral sheet (Plate 1-6a and
1-6b). There are three planes ofanions (Plate 1-6b). One plane consists ofthe basal
0 2- ions of the tetrahedral sheet, the second consists of O~- ions common to both
the tetrahedral and octahedral sheets plus OH- belonging to the octahedral sheet,
and the third consists only ofOH- belonging to the octahedral sheet.~
The edge view of the I: I layer in Plate 1-6b is one of two possible depictions often encountered in the literature. The other view is obtained by rotating the
structure around an axis normal to the sheet and viewing the structure along a different crystallographic axis (Plate 1-6c).
1 The prefix phy/lo-. derived from the Greek word phy/loll. meaning leaf. should not be eonfused
wilh the prefixphy/o-. derived from the Greek wordphy/olI. meaning tribe. Phy/lo- is pronouneed 'fiI-.
while phylo- is pronouneed 'fi-l. Thus. the proper pronuation for phyllosilieate is 'fil--sil-a-kl.
1 The usage ofthe terms plane. sheet. and /ayer follows lhe recommendations ofthe AIPEA Nomenc1ature Committee. They refer to inereasingly thieker arrangements sueh as a single p/alie of atoms. a
tetrahedral or oetahedral sheet. and aI: I or 2: I /ayer. A sheet is a eombination of planes. and a layer is
a eombination of sheets. Layers ean be separated by various illler/ayer materiais. sueh as eations. hydrated eations. organie moleeules. and hydroxide octahedral groups and sheets. The total assembly of
a layer plus interlayer material is ealled a unil structure (AIPEA Nomenclature Committee, 1980).

Ii'iTROO eTlo TO OILMI ERALOGY

13

Plme I I. The strueturc ofhalite ( ael).

"A" voids

"B" voids

Plate I ' Ilexagonal close!>l packing of sphcre!> in a plane.

Face

Edge

"A" site
Sphere-Packing
Model

Ball-and-Stick
Model

Polyhedral
Model
Tetrahedron

Octahedron

Tetrahedron

Octahedron

Plate I 3. Octahedra and tetrahedra as a con!>equence oft\\o planes ofclose-packed spheres and three
ways of repre!>enling octahedra and telrahedra.

I~

H LZE

The Octahedral Sheet (trioctahedral)

The Octahedral Sheet (dioctahedral)

Plate I

. The octahedral heet. (a) A tnoctahedral heet. (b) A dioctahedral sheet.

I TROD

T10N TO O/L .\J1. ERALOGY

HULZE

16

b
PLANES OF 10NS

SHEETS, LAYERS

=-o,...........",...".-.,.,.foooII""""~~.,.."""" ......... ",...~ }tetrahedral 1

O's r
tetrahedralbasal
cations
OH's & apical O's _
octahedral cations _

sheet
octahedral
sheet

OH's-

tetrahedralbasalO's
cations f""

~~~~~f~;~~:
OH's & apical O's _

1'1
laYer

} letrahedral }

} oct:~:~1
}

sheel
tetrahedral cations 't..
tetrahedral
basal O's _ rl~....~"'"",,~--I'~""'~~r...~""''''>''l'''''H'''''.... sheet

2:1
layer

c
Alternate
edge view

Plate I 6. (a) Oblique view of the I: I layer slrueture illuslrating the relationshlp belween the tetrahedral and oelahedral sheet. ote that adjaeent apieal oxygens ofthe tetrahedral sheet (arrows a) also
define edge;, of oelahedra in the oetahedral sheet. Arrows b point to Olls lhat lie direetly in the eenler ofthc hexagonal rings oflelrahedra. although they appear off-eenter in this obliquc \ iew. (b) Edge
vie\\ ofthe I: I and 2: I layer struclures illu;,lrating phyllosilicate nomenclature. (c) An aItemate edge
\ ic,," that arises \\ hen the 2: I structure above is rotated n mlal to lhe plane of the layer.

17

I 'TRODUCTIO 'TO OIL I\IINERALOGY

Gibbsite
0.46nm [
(4.6 A)

Kaolinite

0.72 nm [
(7.2 A)

~~~~~~~~

~1Itl1J'll1llAl~~'-A~IIIAIP_

Si~+

_AI 3 +

CWltAlIfliI"~~IIAIPwl1JP _ Si~+
_AI3+

Pyrophyllite
""'""""'V''V''........................"'V'v-..,.-".

',;; A~ [~~~~~~~~~

'1Il1lflillAlPt. .~1Ifl~AIP _

AI~ ~

'1I<liIPI......

Si~+

_AI 3 +

..

Mica (muscovite)
""~"'V''V'' .....................''''V"'V"''........

~1rlW1~ttI~1rlW1~t>_ Si~+,AI3+

_AI 3+
1.00 nm
(10.oA)

Plate I 7. Structural schemc of soilmincrals bascd on octahcdral and tetrahcdral shects.

18

CH LZE

Vermiculite
Q8W'-.AlAiAII"'~IIIIltIIA~

'Y

'Y.

'Y.

.... .... -... ....

;]l

Si 4 +,AI 3+

___ A1 3+

~~~ l{

;]l

...

...

Smectite

--- H20
>1.8 nm
(>18A)

exch. Ca 2+,
Mg 2+, ele.

Hydroxy-interlayered
Vermiculite and Smectite

A1 3 + poly..
;- exch. Ca 2+,
Mg 2 +, etc.

Chlorite

(11 r 1
L 11M

( I

j,

r
~[

1 1 1 I ] I )
.lI II ~1

I( ~I ~

l(

JI

jJ j

1.4 nm
(14 A)

:.a
I
)j

Plate I 7.

ontinued.

:a

I II
.liI.UI
..:a .iII
:I! ilL,,),. },
.ilI

." ".

.I:

1 l'l
lT li
jJ j
,)

,j }

. :.:'. :. ' .:. :.:.:,.:

INTROD eTlo TO OIL 11NERALOGY

f ] IJ

~~,iI

Palygorskite

r ] 1J

I II ~r I
.1 III
( }(Atl
.1 li j II
] rI
I l:~
.l1 MI

ti

. [ }( j II

,I

.li ti
~(Ji
.1
;.r,J

,1d~

1 T

19

1 I

11

.M .lf.ll
.r }I }.11

'1It.,::l~~~~~~AlAIIP
__ Si4+
__ Mg 2+ ,AI3 +
'tl"IIII'tI"!:l.

H20

_-M.,---H 20
(not shown)

~1AiIA~~~;:wtIpfI.

LiJ TJ

Clinoptilolite (a zeolite)
fl"'1I~~~~"""~~

Imogolite

H20
(nol shown)

1.0 nm
(10 A)
Plate 1-8. lructural models ofrepresentatives ofthree other alumino ilicate mineral groups thal occur
frequently in soils. Ali three are drawn to the same scale.

00

Structural Details Common to Phyllosilicate Minerais

Tetrahedral Sheet

Octahedral Sheet

occupied
octahedra
smaller lhan
unoccupied

shared edges
shorter lhan
unshared

r--

corrugated
basal plane

LPlate I 9. Struetural detaib of phyllosilieates as illuslralcd by lhe octahcdral and tctrahcdral shects of muscovite. Figures \Vcre prcparcd from the single-cryslal structural
refinel11cnt data of ROlhbaucr ( 1971).

n
:t
c::
rN

C"'l

INTRODUCTlON TO SOIL MINERALOGY

21

b. Kaolinite. The structure of kaolinite consists of I: I layers stacked one

above the other. Kaolinite is dioctahedral and contains AP+ in the octahedral sites
and Si4+in the tetrahedral sites (Plate 1-7). The I: I layer is electrical1y neutral and
adjacent layers are held together by hydrogen bonding between the basal oxygens
of the tetrahedral sheet and the hydroxyls of the surface plane of the adjacent octahedral sheet.
Kaolinite is a common mineral in soils and is the most common member of
this subgroup. It tends to be particularly abundant in more weathered soils such as
Ultisols and Oxisols. Kaolinites have very little isomorphous substitution in either
the tetrahedral or octahedral sheets and most kaolinites are close to the ideal formula AI~Si~05(OH)4' The I: 1 layer has little or no permanent charge because ofthe
low amount ofsubstitution. Consequently. cation exchange capacities and surface
areas are typical1y low. Thus. soils high in kaolinite are generally less fertile than
soils in which 2: I clay minerais dominate.
Kaolinite can form in soils from AI and Si released by the weathering ofprimary and other secondary minerais. For example. feldspars often weather to kaolinite in soils formed from igneous rocks. Kaolinite can also be inherited from the
soil parent material (i.e. clayey sediments). Kaolinite mined from geologic deposits
is used widely in a variety of industrial applications such as fil1ers for plastics, in
ceramics, and as a fil1er and coating for paper.
e. Halloysite. Hal10ysite has ai: 1 layer structure similar to kaolinite except
that the I: 1 layers are separated by a layer of H~O molecules when fully hydrated
(Plate 1-7). Most clay silicates occur as thin plates, but halloysite often occurs as
tubular or spherical particles.
Hal10ysite is usually found in soils formed from voleanic deposits. particularly voleanic ash and glass. It is a common clay mineral in the Andisol soil order.
Hal10ysite forms early in the weathering process, but it is generally less stable than
kaolinite and gives way to kaolinite with time.
2. The 2: 1 Type Minerais
ln contrast to the I: I minerais, which are mostly represented in soils by onty
two major minerais, the 2: I minerais are structurally more diverse and are represented by severaI mineral species.
a. The 2: 1 Layer Strueture. The 2: I layer structure consists oftwo tetrahedral sheets with one bound to each side ofan octahedral sheet (Plate 1-6b). There
are four planes of anions. The outer two consist of the basal oxygens (Os) of the
two tetrahedral sheets, while the two inner planes consist of oxygens common to
lhe octahedral sheet and one ofthe tetrahedral sheets, plus the hydroxyls (OHs) belonging to the octahedral sheet.
b. Tale and Pyrophyllite. The simple structures oftale and pyrophyllite are
good starting points for discussing 2: 1 structures. 80th minerais consist of2: I layers stacked one above the other. Tale has Mg~+ in the octahedral sites and is trioctahedral. while pyrophyllite (Plate 1-7) has AIJ+ in the octahedral sites and is dioctahedral. The tetrahedral sheets of both minerais contain only Si 4+. giving ideal
formulas ofMg JSi40 IO(OHh for tale and AI~Si401O(OHh for pyrophyl1ite. ln each

22

SCHULZE

case the charge is balanced within the 2: I layer, making the layer electrically neutral. Adjacent 2: I layers are held together only by weak van der Waals forces.
Tale and pyrophyllite occur only rarely in soils, usually only when they are
inherited from low-grade metamorphic rocks. Tale and pyrophyllite are used industrially as ingredients in paints, ceramics, plastics, paper, and cosmetics. The occurrence oftale in river and estuarial sediments when there is no geological source
is a reflection of industrial activity in the watershed.
c. Micas. Mica minerais have the 2: I layer structure described for tale and
pyrophyllite but with two important differences. First, instead ofhaving only Si 4 +
in the tetrahedral sites, one-fourth ofthe tetrahedral sites are occupied by AI3+. Because ofthis substitution, there is an excess of one negative charge per formula unit
in the 2: I layer. Second, this excess negative charge is balanced by monovalent
cations, commonly K+, that occupy interlayer sites between two 2: I layers. This
gives an ideal formula ofKAI 2(AISi 3)OIO(OHh for a mica mineral with AI in the
octahedral sites.
Just as in tale and pyrophyllite, the octahedral sheet can be either dioctahedral (Plate 1-7) or trioctahedral. There are several different mica species because
Fe 2 +and Fe3+can substitute for Mg 2+and AI'+ in the octahedral sheet and Na+ and
Ca2+ can substitute for K+ in the interlayer.
Mica in soils is usually inherited from the parent rock and is likely to occur
in soils derived from various igneous and metamorphic rocks, as well as from sediments derived from these rocks. Muscovite, biotite, and phlogopite are the three
most common mica group minerais in rocks, and consequently in soils. Allthree
contain K in the interlayer (Table 1-3), but they differ in the composition ofthe octahedral sheet and whether they are di- or trioctahedral.
Mica in the clay fraction of soils and sediments differs somewhat from the
macroscopic muscovite mica it most closely resembles, This clay-size mica is
often referred to as i/lire. Glallconite is another mica mineral that is similar to illite, but it contains more Fe and less AI in its octahedral sheet than illite.
Micas weather to other minerais, particularly to vermiculites and smectites
and the K+ released during weathering is an important source of K for plants. As a
rule, the dioctahedral micas, such as muscovite, are more resistant to weathering
than trioctahedral micas. Thus, muscovite tends to be the most common mica minerai found in soils. Micas are used commercially in paints and cosmetics.
d. Vermiculites. Vermiculite has a 2: 1 layer structure as described for mica,
but instead ofhaving a layer charge of -I per formula unit and K+ in interlayer positions, vermiculite has a layer charge of 0.9 to 0.6 and contains exchangeable
cations, primarily Ca and Mg, in the interlayer (Plate 1-7). The high charge per formula unit gives vermiculites a high cation exchange capacity and causes them to
have a high affinity for weakly hydrated cations such as K+, NH 4 +, and Cs+. Fixation of K+by vermiculites can be significant in soils high in vermiculite.
Vermiculites in soils are believed to form almost exclusively from the weathering ofmicas and chlorites. The weathering ofmicas to vermiculite (or smectite)
is believed to occur by replacement of K+ in the interlayer sites with hydrated exchangeable cations. The integrity ofthe 2: I layer is preserved, but there is a reduction
in the layer charge. Vermiculite does not swell as extensively as smectite, and this

INTRODUCTION TO SOIL MINERALOGV

13

is illustrated in Plate 1-7 by the presence of only two planes ofwater molecules
surrounding the hydrated cations in the interlayer space. Some commercial uses of
vermiculites include horticultural poning media and thermal insulating materiais.
e. Smectites. The smectite group consists ofminerals with the 2: I structure
already discussed for mica and vermiculite, but with a stilllower charge per formula weight, namely 0.6 to 0.25. As in vermiculite, the interlayer contains exchangeable cations (Plate 1-7).
As for the micas and vermiculites, dioctahedral smectites are more common
in soils than trioctahedral smectites. The most common smectite minerais range in
composition between the three endmembers: montmorillonite, beidellite, and nontronite. Ali are dioctahedral, but they differ in the composition of the tetrahedral
and octahedral sheets. Smectites do not fix KTas readily as do vermiculites because
smectites have a lower layer charge, but smectites swell more extensively than vermiculites. This is illustrated in Plate 1-7 by the larger spacing between the 2: I layers.
Smectites are important minerais in temperate region soils because oftheir
high surface area and their adsorptive properties. Smectites shrink upon drying and
swell upon wening. This shrink-swell behavior is most pronounced in the Vertisol
order and in vertic subgroups ofother soil orders. The shrink-swell properties lead
to cracking and shifting problems when houses, roads, and other structures are built
on smectitic soils. Smectites mined from almost pure geologic deposits are used
widely as catalysts, adsorbents for spills, cat liner, sealants for ponds and abandoned
wells, in drilling tluids for oi! wells, and in liners for landfills.

f. Chlorites. Like mica. chlorite minerais have a 2: I layer structure with an

excess of negative charge. ln contrast to mica. however, the excess charge is balanced by a positively charged interlayer hydroxide sheet (Plate 1-7), rather than
K+. The interlayer hydroxide sheet is an octahedral sheet as illustrated in Plate 1-4
and can be either di- or trioctahedral. Instead ofbeing e1ectrically neutral as in brucite
or gibbsite, the hydroxide sheet has a positive charge caused by substitution ofhigher
valence cations for lower valence ones, for example, Mg2AI(OH)6+' Either octahedral sheet-the one that is part of the 2: I layer or the interlayer hydroxide
sheet---.an be di- or trioctahedral, and can contain Mg2+, Fe 2+, Mn 2+, Ni 2+, AP+,
Fe 3 +, and cr3+, giving a large number of different mineral species.
Chlorite minerais in soils are often primary minerais inherited from either
metamorphic or igneous rocks. They may also be inherited from sedimentary rocks
such as shales, or from hydrothermally altered sediments. Chlorites are rather infrequent minerais in soils and when they do occur they generally make up only a
small amount ofthe soil. Chlorites weather to form vermiculite and smectite and
the ease with which they break down makes them sensitive indicators ofweathering.
g. Hydroxy-Interlayered Vermiculite and Smectite. Hydroxy-interlayered vermiculite and smectite can be considered a solid solution with vermiculite
or smectite as one endmember and chlorite as the other. Hydroxy-interlayered
minerais form as AP+ released during weathering hydrolyzes and polymerizes to
form large polycations with a postulated formula of AI 6(OH)fs+ (or similar) in the
interlayers of vermiculite and smectite. These polycations balance some of the

SCHULZE

charge ofthe 2: 1 layer. The combination of a 2: 1 layer with hydroxy AI in the interlayer gives a structure similar to that of chlorite (Plate 1-7). Thus. these mineraIs are also called secondary ch/orites. The degree offilling ofthe interlayer with
hydroxy AI can vary from none to almost complete. with properties ofthe c1ay varying accordingly. The interlayer hydroxy AI is not exchangeable; therefore. it lowers the cation exchange capacity ofsmectite or vermiculite almost Iinearly as a function ofthe amount of AI adsorbed in the interlayer.
Interlayer hydroxy AI prevents smectite from shrinking and swelling as it
would normally. ln vermiculite. it reduces K+ fixation by lowering the exchange
capacity and by preventing the interlayer from collapsing around the K+. The positively charged hydroxy interlayers also provide potential sites for anion adsorption.
Hydroxy-interlayered vermiculite and smectite are most common in Alfisols
and Ultisols. Within a given profile they tend to be most abundant near the soil surface. Hydroxy-interlayered smectites produced industrially are used as catalysts in
the chemical industry.
h. InterstratificatioD in Phyllosilicates. Phyllosilicates in soils do not always
occur as discrete particles ofmica. vermiculite. smectite. chlorite. or kaolinite. For
example. instead of being made up of a stack of identical 2: I vermiculite layers.
one physically discrete particle may consist of a mixture Df both mica and vermiculite layers instead. Such minerais are referred to as mixed-/ayer or interstratified
minera/s.
Ditferent types ofinterstratified mineraIs have been identified. Two-component systems include: mica-vermiculite. mica-smectite. mica-ehlorite. chlorite-vermiculite. chlorite-smectite. chlorite-swelling chlorite. and kaolinite-smectite. Three-component mixed layer systems can also occur. The sequence oflayers
can be either regular Dr random. A regularly interstratified mineral consisting of
two types of layers denoted by A and B could have a sequence Iike ABABAB...,
or ABBABBABB ... Dr any other repeating sequence. ln a randomly interstratified
mineral the sequence oflayers is random. for example. ABBABAABBAAA. ... Random interstratification oflayer-silicates is more common in soils than regular interstratification, though regular interstratification. especially in weathering micas,
is not rare.
Partial removal of interlayer K from micas or of interlayer hydroxide from
chlorite is one way that interstratified minerais can form in soils. Other possibilities include (i) fixation of adsorbed K+ by some vermiculite layers to give micalike layers and (ii) the formation ofhydroxide interlayers to produce chlorite-like
layers.
i. Palygorskite and Sepiolite. Palygorskite and sepiolite are considered
phyllosilicates. but are distinct structurally from the typical I: I and 2: I layer structures. 80th minerais have continuous tetrahedral sheets. but adjacent bands of
tetrahedra within one tetrahedral sheet point in two opposite directions rather than
in one direction as in the I: I and 2: I structures. The result is a structure that can be
described as ribbons of 2: I layers joined at their edges. as illustrated in Plate 1-8.
Water molecules occur in the spaces between the ribbons. The 2: I ribbons are wider
in sepioJite than in palygorskite.

INTRODUCTION TO SOIL MINERALOGY

Palygorskite and sepiolite are often found in soils ofarid and semiarid environments. Both minerais have a fibrous morphology in contrast to the platy morphology of most I: I and 2: I minerais. Because of their fibrous morphology, suspensions ofthese clays can fonn thick gels even at low solid concentrations. Thus,
palygorskite and sepiolite are used industrially as gelling agents to keep other
solids in suspension.

C. Structural Details of Pbyllosilicates

Plates I-I through 1-8 illustrate ideal mineral structures based on the regular close packing of spheres. As a consequence, ali of the octahedra in Plate 1-4
are exactly the sarne size and ali are regular (undistorted) solids. Likewise, in Plate
1-5, ali ofthe tetrahedra are regular solids, the six-member tetrahedral rings fonn
a regular hexagon, and the basal oxygens alllie in exactly the same plane. Crystallographic structure refinements of real phyllosilicate minerais show distortions
from the ideal structures illustrated up until now in this chapter. These distortions
are similar for ali ofthe phyllosilicate minerais; although they vary somewhat in
magnitude.
There are two reasons why real minerais deviate from these ideal structures.
First, atoms are not hard spheres whose closest approach is detennined solely by
the sum oftheir radii. Second, ifone does try to join octahedral and tetrahedral sheets
built ofhard spheres like those illustrated in Plates 1-4 and 1-5, geometry alone
dictates that they will not fit together ifthe spheres representing the O atoms touch
each other in both the octahedral and tetrahedral sheets. The reason becomes apparent by examining the sphere-packing model ofthe tetrahedral sheet in Plate l-S.
Since the apical oxygens are common to both the tetrahedral and octahedral sheets,
two adjacent apical oxygens in the tetrahedral sheet also define the edge of an oetahedron in the octahedral sheet (Plate l--6a). Note that, although basal oxygens
touch neighboring basal oxygens, apical oxygens do not touch neighboring apical
oxygens (Plate I-S). Thus, the octahedral sheets illustrated in Plate 1-4 cannot share
apical oxygens with the tetrahedral sheet illustrated in Plate 1-5 because the tetrahedral sheet is larger than the octahedral sheet. This problem was overcome in Plates
1--6 to 1-8 by increasing the 0-0 distance in the octahedral sheet to match the 0-0
distance ofthe apical oxygens ofthe tetrahedral sheet. Thus, the octahedra remain
undistorted, but they are larger than they would be ifthe spheres representing oxygen touched.
ln the real phyllosilicate structures, this mismatch between the octahedral and
tetrahedral sheets is compensated, not so much by an expansion ofthe octahedral
sheet, but more so by distortions in the tetrahedral sheet. The net effect is that, the
tetrahedra are rotated from the ideal arrangement shown in Plate I-S such that the
hexagonal shapes outlined by the six-membered tetrahedral rings in Plate I-S become the ditrigonal shapes illustrated in Plate 1-9. ln addition, the tetrahedra are
tilted such that the basal oxygens no longer lie in the sarne plane, resulting in a corrugated basal plane in which some rows of atoms lie slightly below and some lie
slightly above the average plane ofthe basal oxygens. These corrugations are best
seen in the lowennost edge view in Plate 1-9 where spheres representing the O
atoms have not been drawn. The implications ofthese distortions are that molecules

26

SCHULZE

interacting with phyllosilicate surfaces "see", not the ideal surface iIIustrated in Plate
1-6a, but the more distorted surface illustrated in Plate 1-9.
The octahedral sheet is distorted from the ideal structure as well. First, because ofthe attraction of each central octahedral cation to its six surrounding oxygens, octahedra occupied by cations are smaller than octahedra in which the centraI sites are unoccupied (Plate 1-9). ln addition, when two octahedra share two
adjacent oxygens (the octahedra share edges), the 0-0 distance along this shared
edge is shorter than the 0-0 distance along an edge that is not shared with an adjacent oeeupied octahedron (Plate 1-9). Thus, the oetahedra are distorted from the
ideal Euclidean solids in very predietable ways. ln general, trioctahedral phyllosilieates show less struetural distortion and more closely resemble the ideal struetures than do the dioctahedral phyllosilieates.
The structural distortions just discussed probably aecount for relatively minor
differences in the properties ofthe phyllosilicates, with the major differences deterrnined mainly by whether the mineral has a I: I or 2: I layer structure and by the
layer charge. On the other hand, since dioctahedral phyllosilicates are more common in soils than trioctahedral phyllosilicates, water, ions, and molecules interaet
predominately with surfaces like those illustrated in Plate 1-9. rather than surfaces
with the more regular arrangement illustrated in Plates 1---4 and l--6a.

III. OTHER ALUMINOSILICATE MINERALS COMMON lN SOILS

A. Zeolites
Zeolites are a large group ofminerals that consist structurally ofSi04 tetrahedra arranged in ways that result in large amounts of pore space within the crystais (Plate 1-8). Aluminum substitutes for Si in the tetrahedral sites ando as a result. the (Si,AI)04 framework has a net negative charge. The charge is balanced by
eations that reside in the channels and pores, along with water molecules. Because
the cations are exchangeable, zeolites have cation exchange properties similar to
the phyllosilicates. but because the tetrahedral framework ofthe zeolites is rigid
and the size ofthe pores is fixed, small cations can move into and out ofthe pores
freely. while larger cations are excluded. Thus, zeolites are often referred to as molecular sieves because oftheir very selective cation exchange properties.
Zeolites are relatively rare in soils because they weather easily in humid regions, but they do occur in some soils in arid regions. Zeolites are widely used in
industry and agriculture. for example, as catalysts and adsorbents in the chemical
industry. as additives in animal feeds. and as ingredients in laundry detergents. Thus,
zeolites may be introduced into soils and sediments in unexpected ways.

B. Allophane and Imogolite

The aluminosilicate mineraIs discussed above have three-dimensional crystal structures with atoms packed together in a more or less regular manner over relatively long distances (tens of nanometers). They exhibit long range order. Two
other aluminosilicates. allophane and imogolite. exhibit short-range (or local)

INTRODUCTlON TO SOIL MINERALOGY

27

arder. Structures with short-range order exhibit order over several nanometers, but
on a larger scale the structure is disordered.
Allophane is a material consisting chemically of variable amounts of ()2-,
OH-, A13+, and Si 4+ and characterized by short-range order and a predominance of
Si-o-AI bonds. It consists of small (3.5-5.0 nm) spheres, the structure of which
has not been determined. The spheres clump together to form irregular aggregates.
Imogolite consists oftubes several micrometers long with an outer diameter
of2.3 to 2.7 nm and an inner diameter of-1.0 nm. The tubes consist ofa single dioctahedral sheet with the inner surface OH replaced by SiO,OH groups (Plate 1-8;
Farmer et aI., 1983). Several individual tubes are arranged in bundles lOto 30 nm
across to give thread-like particles several micrometers long.
Allophane and imogolite usually occur as weathering products of volcanie
ash and are important minerais in the Andisol soil order. lmogolite has also been
identified in the Bs horizons of Spodosols. Allophane and imogolite can specifieally adsorb many inorganic and organic compounds. Andisols, for example, usually fix large amounts of phosphate, making it unavailable to plants, and the large
amounts of organic matter common in Andisols may be due, in part, to adsorption
of organic molecules by allophane and imogolite. Soils containing large amounts
of allophane and imogolite usually have unique physical properties such as a low
bulk density, high water holding eapacity, high liquid and plasticity 1imits, and a
thixotropic consistence.

IV. SOME CRYSTALLOGRAPHIC CONCEPTS

ln 1912 Max von Laue conducted an experiment to see ifthe x-rays diseovered 17 yr earlier by Wilhelm Conrad Rntgen could be diffracted by a crystal. This
one experiment had two very important results: (i) it proved that x-rays behave as
waves, and (ii) it showed that crystals are made up ofa regular array ofatoms in
space, and can therefore serve as a grating to diffract the x-rays. Within a few years
ofvon Laue's discovery, much ofthe mathematieal theory deseribing the diffraetion ofx-rays by crystals had been developed and the atomic structures of some simpie crystals had been determined (Azroff, 1968; Klug & Alexander, 1974). As early
as 1923 and 1924, severa1clays were shown to be crystalline by x-ray diffraetion,
but it was not until 1930 that some ofthe first strueture determinations ofclay minerais were made. About the sarne time, two independent papers, one by S.B. Hendrieks and w'H. Fry in 1930 and the other by w'P. Kelly, w'H. Dore, and S.M.
Brown in 1931, gave the first x-ray diffraction evidence that soil materiaIs, even
those in the finest fraetions, were crystalline (Marshall, 1949; Grim, 1968).
X-ray diffraction remains the most important method for identif)ting soil minerais. A working knowledge of certain concepts of crystallography and x-ray diffraetion is necessary for understanding day mineralogy. A brief introduction to some
ofthese coneepts follows.

A. Periodicity in Crystals
ln a erystal, a particular pattem or arrangement ofatoms is repeated over and
over in three dimensions. Repeating pattems oceur not only in erystals but also in

SCHULZE

.,l. ..

.
........

..

.t....-';Ir-...........,..-~_
Fig. I-I. (a) A lattice array af leaves and (b) a plane lattice of points.

many familiar places such as wallpaper, tile floors. and brick walls. Concepts that
apply to crystals can be developed by starting with these familiar examples.
Figure I-Ia illustrates a pattem made up ofleaves much like the pattem on
a piece ofwallpaper. The periodicity ofthis pattem can be characterized with three
parameters. Along the X axis the pattem is repeated at intervals of a. We also say
that the pattem is translated (moved) along X at increments of a. The translation
interval along the Yaxis is designated b and the angle between the X and Yaxes is
designated by the angle y. The array ofleaves in Fig. I-Ia consists ofreal objects.
namely. leaves. We can represent the periodic nature ofthis pattem by an array of
points in space. each point having identical surroundings (Fig. I-I b.). Such an array
is called a lattice. A lattice of points is imaginary since each point is an imaginary
infinitesimal spot in space. The concept of a lattice is useful because it allows us
to represent the periodic nature of a pattem regardless of the actual object or configuration ofobjects at each lattice point. The lattice depicted in Fig. I-I b is called
a plane lattice because it has only two dimensions. The concept can be extended
to three dimensions to give a space lattice. A space lattice is a regular and unlimited distribution ofpoints in space (Fig. 1-2). The concept ofa space lattice is used
in describing crystal structures because crystals are three dimensional objects. 3 The
directions ofthe crystallographic axes in three dimensions are designated by X. Y.
and Z, and a. b, and c are used for the repeat distances along these axes (AIPEA
Nomenclature Committee. 1980). The angles between the axes are designated a
between Yand Z, p between X and Z, and y between X and Y.

B. Tbe Unit Cell

Joining the points of a space lattice produces a series of parallel-sided unit
cells (Fig. 1-2). Each unit cell contains a complete unit of the crystal pattem because the complete pattem is reproduced at each lattice point. Note that there are
altemative ways of outlining the unit cell. The choice of the "best" unit cell for a
mineral is made when the structure is deterrnined. A unit cell is chosen for conve] The tenns fal/ice and structure are ofien misused by authors and speakers. The two tenns are not
synonymous. Lal/ice is the mathematical concept ofan infinite unifonn distribution ofpoints in space
(e.g., the 14 Bravais lattices). Structure refers to the actual physical assemblage of atoms (e.g., the 2: 1
slructure) (AIPEA Nomenclature Committee, 1980). The concept of a lattice is used to describe the periodicity in crystal slructures.

29

INTRODUCTION TO SOIL MINERALOGY

Fig. )-2. A space lanice with several altemative unit cells outlined (after Klug and Alexander, 1974).

nience in visualizing the symmetry and in making mathematical calculations. Different researchers will occasionally choose different unit cei Is to describe the sarne
mineral structure.
There are 14 unique ways of arranging points in three-dimensional space.
These are known as the 14 Bravais lattiees. They forro the geometrieal basis for ali
possible unit eells. Each Bravais lattiee belongs to one of six erystal systems depending on the symmetry ofthe lattiee (Table 1-5). The eubie system is most symmetrie. Ali axial translations are equal (a = b = c) as are ali axial angles (a. = ~ =
y). The triclinie system is most general (and least symmetrie). Ali axial translations
and axial angles are different. The unit eell dimensions and the axial angles are important parameters in deseribing the erystal strueture of a mineral.
C. Miller Indices

A consistent and concise notation is used when planes of atoms in a crystal

are diseussed. Different planes are generally referred to by their Miller indices, designated as (hkl). Forexample, we may speak of(OOI) planes, (110) planes, and (431)
planes. The eoneept of Miller indiees follows from the lattice eoneept diseussed
above. Consider the lattice ofpoints in Fig. 1-3a and a plane that is parallel to c
and passes through any two lattice points (represented in an edge view by the heavy
line). An identieal plane must pass through eaeh and every lattiee point (Iight
lines) beeause the lattiee is periodie and eaeh point is the sarne as any other. This
family ofparallel planes euts the a dimension ofthe unit cell into two parts and the
b dimension into three parts. The c dimension is not eut at ali (zero parts) because
Table l-S. Axial ratios and angles between crystal axes for lhe six crystal symmetry systems (after Klug
and Alexander. 1974).
System

Axial ratios

Cubic (isometric)
Hexagonal
Tetragonal
Orthorhombic
Monoclinic
Triclinic

a=b=c

a=~=y=90

a=b~c

a = ~ = 90. Y= 120

a=b~c

a=~=y=90
a=~=y=90
a=y=90,~>90
a#~#y=90

a~b~c
a~b~c

a#h#c

Angles between crystal a.xes

SCHULZE

30

Fig. )-3. (a) Planes with Miller indices 230 c axis is out ofthe plane ofthe figure) and (b) planes with
different Miller indices on an orthogonallanice.

the planes are parallel to Z. The Miller indices are thus (230). ln tenns of(hkf), the
planes cut the a dimension ofthe unit cell into h parts, the b dimension into k parts,
and the c dimension into I parts. Additional examples of Miller indices are given
in Fig. 1-3b.

D. X-ray Diffraction
X-ray diffraction (XRD) is the main analytical technique used both to identify unknown mineral phases and to detennine crystal structures. Soil minerais are
usually studied using the powder diffraction method. A powdered sample, with particles typically <50 Ilm in diameter, is placed in a diffractometer and irradiated with
x-rays. A plot of diffracted x-ray intensity vs. the diffraction angle (i.e., 28) is obtained. A simple mathematical relationship called the Bragg Law is then used to relate the peaks on the x-ray chart with the distances between the diffracting planes
of atoms within the crystals.

1. The Bragg Law

X-rays are electromagnetic waves with wavelengths on the order ofthe spacings between the planes ofatoms in crystals. Thus, crystals diffract x-rays. Understanding the additive relationships ofwaves that are in phase or out ofphase is important to understanding x-ray diffraction. Fig. l--4a illustrates three different sine
waves, ali ofwhich begin at the sarne point in their cycles. The waves are in phase
because the maxima of each wave coincide. The sum of ali three waves is a much
larger wave with the sarne period (distance between successive maxima) as the originai waves but with three times the intensity. The three waves on the right are identical to those on the left, except that they are out ofphase. Each wave begins at a
different point in its cycle relative to the others. The maxima ofthe three waves no
longer coincide, and thus the sum, which takes into account the positive and negative aspects ofthe three individual waves, approaches zero.
ln XRD, constructive interference occurs when many waves are in phase. The
energy from the large summation wave is detected by the diffractometer. Destructive interference occurs when the waves are out ofphase. ln this case no appreciable energy is detected. The peaks ofan XRD pattem correspond to areas where con-

31

INTRODUCTION TO SOIL MINERALOGV

Waves in Phase

Waves out of Phase

Fig. 1-4. (a) ln phase and oul ofphase waves and (b) geometry oflhe Bragg reflection analogy for diffraclion ofx-rays by crystals (after AzrotT, 1968).

structive interference occurs. The Bragg Law relates the position ofthese peaks to
the distances between planes of atoms in the crystal. This information. along with
the intensities of the peaks is then used for identif)'ing mineral phases and determining crystal structures. The following derivation ofthe Bragg Law will show how
it works.
Consider the parallel planes of atoms with indices (hkl) separated by a dislance d depicted in Fig. l--4b. A parallel (or collimated) beam ofx-rays ofwavelength impinges on the (hkl) planes at an angle e. The incident x-rays are in phase
with one another. Two waves. one traveling along OA and being retlected off the
upper plane of atoms and traveling along AP. and a second wave traveling along
O'C and being retlected offthe second plane ofatoms and then traveling along Cp'
may. under specific conditions. be in phase when they reach points P and p'. respectively. ffthe waves are n phase at fine AB. when will they be in phase at line
AD and thus at PP'? A wave along CP' is in phase with a wave along AP only ifthe
distance BC + CD is an integral number ofwave\engths. Thus
BC + CD = n. where n = 1,2,3,...

[I]

Since BC = CD, substituting into Eq. [I] gives

2BC =n..

[2]

32

SCHULZE

From Fig. 1-4b,

BCld,

[3]

BC=dsine.

[4]

2BC = 2d sin e.

[5]

n = 2d sin e.

[6]

sin e

which can be rewritten as

Multiplying both sides by 2 gives

Substituting Eq. [2] into Eq. [5],

This expression is the Bragg Law. Rewritten as

dln = 1(2 sin e),

[7]

the Bragg Law relates the perpendicular distance between diffracting planes, d, to
the diffraction angle e. Since the order of diffraction, n, is not always known, n is
usually set to 1 and the quantity dln is referred to as the d-value ofa diffraction line.
Note that the quantity needed in Eq. [7] is in units ofe, while the data from
x-ray diffractometers is in units of2e. This is a consequence ofthe physical construction ofthe diffractometer and division by 2 is necessary to obtain e from 2e
data before using Eq. [7] to calculate d-values. The d-values can be easily calculated with a pocket calculator or a computer programo Tables for determining d-values from 2e data are also given in Brindley and Brown (1980) for severaI common
x-ray wavelengths.
The distances between the diffracting planes in the crystal, din Eq. [7], are
fixed by the structure ofthe crystal. However, x-ray wavelengths (s) used for diffraction experiments are not always the sarne. If in Eq. [7] changes, e must change
because dln is a constant fixed by the crystal structure ofthe sample. X-ray line positions expressed in 2e can only be compared when the x-ray wavelength is lhe sarne.
Calculating the d-values for the diffraction lines allows comparison ofd-values regardless ofthe wavelength.
The d-values are directly related to the unit cell dimensions and the Miller
indices. The d-value of a line with Miller indices (hkf), dhkl, in the orthorhombic
system is given by the relationship:

[8]
where a, b, and c are the unit cell dimensions. Relationships for the other crystal
systems are given elsewhere (e.g., Klug & Alexander, 1974).
The dln values obtained from the Bragg Law can, in most cases, be related
to actual distances between planes of atoms in the crystals. Thus, the d-values between successive phyllosilicate layers listed in Plate 1-7 show up in x-ray ditfrac-

INTRODUCTION TO SOIL MINERALOGY

tion pattems of soi! c1ays and are used to identify the specific c1ay minerais present in unknown samples 4 . ln some cases, however, in particular for randomly interstratified minerais or cases where diffraction lines are very broad due to the small
size of the crystals, d/n values from the Bragg Law are not directly related to the
distances between planes ofatoms in the crystal. To explain these effects more inclusive diffraction theory utilizing such concepts as the reciprocallattice, structure
factor, and atomic scattering factors must be applied.

2. Tbe Scberrer EquatioD

Once a mineral phase has been identified based on its diffraction peaks, additional information can ofien be obtained from diffraction line widths. Well-crystallized mineraIs of sand and silt size give sharp diffraction lines whose widths are
determined only by broadening caused by the x-ray diffractometer itself. Clay-size
particles show broader lines caused by diffraction effects from the small particles.
The smaller the particles are, the broader the diffraction lines. X-ray line width and
particle size are related by the Scherrer equation:
L hkl = (K)/(~ cos e),

[9]

where Lhkl is the mean crystallite dimension perpendicular to the diffracting planes
with Miller indices (hkl), K is a constant equal to 0.9 ifthe particles are cubes, . is
the x-ray wavelength, ~ is the width (in radians) at one-halfpeak height corrected
for instrumental broadening, and e is the diffraction line position. The application
ofthe Scherrer equation involves a considerable number of assumptions (Klug &
Alexander, 1974; p. 634-642,687-704 >, and an understanding ofthese assumptions
is essential. ln addition, Lhkl measures the size ofthe coherently diffracting domains
within the crystals. If the crystals consist of several coherently diffracting domains, then L hkl will be smaller than the actual physical size ofthe crystals observed
by electron microscopy (Schulze & Schwertmann, 1984).

V.SUMMARY
This chapter has developed some ofthe terminology used later in this book.
lt has shown how the abundance of aluminosilicate minerais is a consequence of
the abundance ofSi, AI, and O in the Earth's crust. lt has introduced the chemical
and structural c1assification ofminerals, mentioned the major minerais that will be
encountered in soils, and mentioned some ofthe environmental implications ofsome
ofthe minerais. lt has developed the major structural themes ofthe phyllosilicate
minerais, and it has explained some of the crystallographic and x-ray diffraction
terminology used in later chapters.
The chapters that follow will discuss in more detail the many facets of soil
mineralogy, particularly as it relates to current environmental concems.
4 Strict usage ofSI units requires that d-spacings be given in nanometers (nml. Angstroms (Al, however, are used in the older literature and are still used widely in lhe current crystallography literature
and in many ofthe databases used for mineral identification. There are 10 Ain [ nm (I nm = [O AJ.

SCHULZE

34

VI. ACKNOWLEDGMENTS
I thank the many students over the years who have provided feedback on this
chapter.

VII. QUESTIONS AND EXERCISES

I. Draw sketches from memory of an octahedron and a tetrahedron using the
sphere-packing, ball-and-stick. and polyhedral representations.
2. Draw sketches from memory of sphere-packing models of a tetrahedral sheet,
a dioctahedral sheet, and a trioctahedral sheet. (A coin makes a good template
for a circle.)
3. Draw three sketches of a unit cell and label the cell dimensions a, b, and c as
shown in Fig. 1-3b. On the first cell sketch in the planes with Miller indices (002),
on the second cell sketch planes with indices (lI I ), and on the third cell sketch
planes with indices (021).
4. (a) Calculate the d-values ofx-ray diffraction lines that occur at 8.84, 12.36, and
26.68 28 for a sample x-rayed using CuKa radiation (. = 0.15418 nm). (b) Calculate the 28 positions for these sarne three lines when the sarnple is x-rayed using
CoKa radiation (. = 0.179026 nm).
Answers: (a) 1.000,0.716, and 0.334 nm, respectively. (b) 10.27, 14.36, and
31.09 28, respectively.
5. A goethite sample was x-rayed with CoKa radiation. The III diffraction tine
was centered at 24.70 28, and its width at half-height (corrected for instrumental
broadening) was found to be 0.35 28. Use the Scherrer equation to calculate
the size ofthe crystals (more exactly, the size ofthe coherently diffracting domains) perpendicular to the (110) planes. Assume that K = 0.9.
Answer: 27 nm.

VIII. SUGGESTED READING

For more information on general mineralogy, see introductory texts such as
Klein and Hurlburt (1993) or Nesse (1999). Azroff{ 1968) and Moore and Reynolds
(1997) provide good, readable introductions to crystallography and x-ray diffraction theory.

REFERENCES
AIPEA Nomenclature Comminee. 1980. Summary of recommendations ofAIPEA nomenclature comminee. Clays Clay Miner. 28:73-78.
Allen, B.L., and D.S. Fanning. 1983. Composition and soil genesis. p. 141-192./n L.P. Wilding et aI.
(ed.) Pedogenesis and soil taxonomy I. Concepts and interactions. Elsevier. New York, NV.
Azroff, L.v. 1968. Elements ofx-ray crystallography. McGraw-Hill Book Co., New York, NV.
Brindley, CiW., and Ci Brown (ed.) 1980. Crystal structures ofclay mineraIs and their x-ray identification. Mineralogical Society, London, UK.

INTRODUCTlON TO SOIL MINERALOGY

Fanner, v.c., M.J. Adams, A.R. Fraser, and F. Palmieri. 1983. Synthetic imogolite: Properties, synthesis,
and possible applications. Clay Miner. 18:459--472.
Grim, R.E. 1968. Clay mineralogy. 2nd ed. McGraw-Hill Book Co., New York, NY.
Jackson, M.L. 1964. Chemical composition ofsoils. p. 71-141./n F.E. Bear (ed.) Chemistry ofthe soi!.
Reinhold Publ., New York. NY.
Kmpf, N., A.C. Scheinost. and D.G Schulze. 1999. Oxide minerais. p. FI25-FI68./n M.E. Sumner
(ed.) Handbook ofsoil science. CRC Press, Boca Raton, FL.
Klein, C., and C.S. Hurlbut, Jr. 1993. Manual ofmineralogy (after James D. Dana). 21st ed. John Wiley
& Sons, New York. NY.
Klug, H.P., and L.E. Alexander. 1974. X-ray diffraclion procedures for polycrystalline and amorphous
materiais. 2nd ed. John Wiley & Sons, New York. NY.
Marshall, C.E. 1949. The colloid chemistry ofthe silicale minerais. Academic Press. New York, NY.
Moore, D.M., and R.C. Reynolds. Jr. 1997. X-ray diffraction and identification and analysis ofc1ay minerais. 2nd ed. Oxford University Press, New York. NY.
Nesse, w'D. 1999. Introduclion lo mineralogy. Oxford University Press, New York, NY.
Rothbauer. R. 1971. Untersuchung eines 2M 1-Muskovits mit Neutronenstrahlen. N. Jahrbuch f. Mineralogie. Monatshefte 1971: 143-154.
Schulze, D.G, and U. Schwertmann. 1984. The influence ofaluminium on iron oxides: X. Propenies
of AI-substituted goethites. Clay Miner. 19:521-539.