2 Mineral Chemistry

2.1 Carrollite has the formula Cu(Co,Ni)2S4

2 Mineral Chemistry
KEY CONCEPTS

All matter, including mineral matter, is made of

elements.
 8 to 10 elements account for most of Earth’s mass but
the elements are distributed unevenly.
The Periodic Chart of the Elements orders elements by
their atomic number and groups elements with similar
properties.
Elements with similar properties may substitute for each
other in minerals and many minerals have variable
compositions.
Ionic, covalent, or metallic bonding are most common in
minerals.
The kind of bonding affects mineral properties.
We describe mineral compositions by giving weight %s of
the elements or oxides present.
Weight % values can be normalized to give chemical
formulas.
Although 1000s of minerals are known, only a small
number are common.

2.1 Elements and Minerals

All matter is made of atoms of individual elements. For
example, the matter of the Milky Way, our galaxy, is mostly
hydrogen and helium – the same elements that make up >99% of
our sun. Earth, however, is dominated by other elements, and 8
to 10 elements account for most of Earth’s mass. Although
Earth’s crust and mantle contain the same major elements, the
proportions are not the same (see histograms below). And, the
core is completely different from the crust and mantle – it is
mostly iron with perhaps 20% nickel and lesser amounts of
other elements.
2.2 Elements in the crust and mantle

Most of the minerals and other geological materials we see

derive from the crust, but some come from the uppermost part
of the mantle. In both places, oxygen and silicon are the
dominant elements (Figures 2.2 and 2.3). Together these two
elements make up about three quarters of the crust, and two
thirds of the mantle. Other quite abundant elements in both
the crust and mantle include aluminum, iron, calcium, sodium,
potassium, and magnesium. However, the mantle contains much
more magnesium and iron, and less silicon, than the crust.
Hydrogen (0.15 wt%) and carbon (0.18 wt%) are overall minor
elements in the crust but are key components in some minerals.
And other generally rare elements are sometimes concentrated
by geological process to make exotic minerals.

The compositions of Earth’s outer layers vary somewhat

laterally and vertically. For example, the oceanic crust is
not the same composition as the continental crust. And, the
composition of the shallow crust is somewhat different from
the deep crust. Nonetheless, in most settings, we can expect
common minerals to be made of the elements shown in the
histograms above.
Some elements are common in many different minerals. Oxygen
and silicon are perhaps the best examples. Many sedimentary
rocks and nearly all igneous and metamorphic rocks are
composed of multiple minerals containing these two elements.
In contrast, because of their properties, some other elements
tend to be found in only a few distinct minerals. For example,
titanium (Ti) may occur as a minor component in biotite,
amphibole, or other minerals. In many rocks, however, Ti is
concentrated in Ti-rich minerals such as rutile (TiO 2 ),
titanite (CaTiSiO5), and ilmenite (FeTiO3). Rocks rich in Ti
always contain one of these latter three minerals. Similarly,
rocks containing significant amounts of phosphorous usually
contain apatite, Ca5(PO4)3(OH,F,Cl), or monazite,
(Ce,La,Th,Y)PO 4 .

2.2 Elements and the Periodic Table

2.3 The Periodic Table of the Elements

Here (Figure 2.3) we see the Periodic Table of the Elements.

Elements are ordered by increasing atomic number (the number
of protons in their atom nucleus), which correlates with
atomic weight and size. The first element is hydrogen (atomic
number 1) and the last is oganesson (atomic number 118).
Oganesson is an obscure synthetic radioactive element that has
only been produced a few times and in minute amounts. In this
table, elements are in numbered periods (rows) and groups
(columns) based on the configuration of their electron
orbitals.

The chart contains 118 elements in all, but only about 90

occur naturally. The other 20 or so (with atomic numbers
104-118 and 99-103) elements are artificial – they are
synthesized in nuclear reactors and some are radioactive with
very short lives. The distinction between natural and
artificial elements is, however, a bit uncertain because only
a few atoms have (speculatively) been identified for some of
the rarest natural elements. Although the Periodic Table of
the Elements has appeared in many forms, the basic
relationships are the same today as they were when Dmitri
Mendeleyev (1834-1907) devised the first version in 1870. Some
elements were unknown and omitted from the original table
during Mendeleyev’s time, but were later discovered and added.
Additionally, there has been a slight rearrangement so that
the arrangement of the elements mirrors the order in which
electrons occupy orbitals.

Chemists classify elements into different types that have

related properties (shown by different colors in the table
above). Hydrogen is a special element, but the other elements
in Group 1 are alkali metals. The elements in group 2 are
alkaline earth metals. Those elements in Groups 4 through 12
are transition metals. Group 17 elements are the halogens, and
Group 18 elements are the noble gases. Elements between the
transition metals and the halogens are called nonmetals or
other metals (but different versions of the chart have them
divided in slightly different ways). The asterisks in the
chart show where the lanthinides (also called rare earth
elements) and actinides (both considered transition metals)
were extracted from the main chart and put as separate rows at
the bottom. If we did not extract them, the chart would be too
wide to fit easily on this page. Scandium (Sc) and yttrium (Y)
are closely related to the lanthinides and so are commonly
classified with them.

2.2.1 Groups

2.4 Perthite (exsolved alkali feldspar)

Elements in the same groups (columns) have their outermost

electrons in the same kind of orbitals. This is very
significant because it means the elements have similar
chemical properties, form similarly charged ions and,
importantly, commonly substitute for each other in mineral
crystals. For example, the mineral specimen shown in Figure
2.4 is an alkali feldspar. When it first crystallized from a
magma, it was a homogeneous mixture of Na-feldspar and K-
feldspar. Na and K, both alkali elements (Group 1), commonly
substitute for each other in minerals. When this sample
cooled, however, the two different feldspar components unmixed
from each other, much the same way that chicken soup and fat
separate on cooling. The result is called exsolution (which
means unmixing) and the specimen now contains thin pinkish
veins of K-feldspar surrounded by Na-feldspar. We call
feldspar that has exsolved like this, perthite.

2.2.2 Atomic Number and Mass

The variable Z designates an element’s atomic number, the
number of protons in its nucleus. Z is also equal to the
number of electrons orbiting the nucleus in neutral (non-
ionized) atoms, and is close to the number of electrons in
0
most ions. A neutral iron atom (Fe ), for example, contains 26
protons (Z = 26) in its nucleus and 26 electrons in an
electron cloud around the nucleus. Because the size of its
electron cloud controls the diameter of an atom, elements with
greater atomic number, with many protons and thus many
electrons, are larger than those or lower atomic number.

Atomic nuclei (except one isotope of hydrogen) contain

neutrons in addition to protons, and the number of neutrons,
designated by N, may vary. This leads to isotopes of different
mass numbers. Mass number, designated by the variable A, is
equal to the number of protons and the number of neutrons
combined: A = Z + N. Most chemical elements have several
different naturally occurring isotopes. Some isotope
varieties, however, are generally more common than others, and
some only exist in minute amounts.

2.5 Isotopes of Oxygen

Oxygen atoms, for example, may be any of three isotopes (shown

in Figure 2.5). Oxygen may be 1 6 O, 1 7 O, or 1 8 O, where the

superscript number denotes A. Examination of the equation in
the previous paragraph tells us that the three isotopes of
oxygen must have 8, 9, and 10 neutrons, respectively, because
17 18
all must have 8 protons if they are oxygen. O and O are very
16
rare; O is 99.8% of all natural oxygen.

2.2.3 Moles
A mole of an element (or of a compound) is defined as
23
containing 6.022 × 10 atoms (or molecules). The number, 6.022
23
× 10 , is known as Avogadro’s number. So, one mole of carbon
is equivalent to 6.022 × 1023 carbon atoms. The scale used to
measure atomic mass has changed slightly over time. Today, it
is standardized relative to carbon so that the mass of one

mole of 12C is exactly 12.0000. Consequently, all atomic masses

are given in atomic mass units (amus), defined as one-twelfth

the mass of 12 C. Both protons and neutrons have equivalent

mass, about one amu, and electrons have almost no mass (less

than 1/1000th the mass of protons and neutrons). So, we might

expect the mass of an atom to be equal to the mass number (A,
the total number of protons plus neutrons). But for several
reasons, not worth going into here, the mass of an atom is
generally close to, but not the same as, the mass number.

Elements are different from atoms. The atomic mass, also

called the atomic weight, of an element is the sum of the
masses of its naturally occurring isotopes weighted in
accordance with their abundances. Atomic masses/weights of
elements are molal quantities (and often given in units of
grams/mole) but they are really dimensionless numbers because
they are all calculated relative to the atomic mass/weight of
a mole of carbon. Although isotope mass numbers are always
integers, atomic weights of elements are not. For example,
many tables and charts give the atomic weight of oxygen as
15.999 and that of iron as 55.847. When elements combine to
produce a compound, the atomic weight of the compound is the
sum of the weights of the elements in the compound. FeO, for
example, has atomic weight 71.846 (15.999 + 55.847)
grams/mole.

Most elements have very small isotopic variation in nature, no

matter where they are found. Thus, most quartz (SiO2) contains
about the same relative amounts of the three natural oxygen
16
isotopes (mostly O) depicted in Figure 2.5 above.
Furthermore, isotopic variations have extremely small effects
on the properties of minerals. So, mineralogists generally do
not worry much about isotopes. Small isotopic variations,
however, may be significant to a geochemist trying to
determine the genesis of a particular mineral or rock.

● Box 2-1 What Is a Mole of Quartz?

Quartz (SiO2) is one of the most common and well-known minerals. A mole of quartz
23
is 6.022 × 10 SiO2 molecules, and that sounds like a lot. How much quartz is
that?
To answer this question, we use the atomic weights of silicon and oxygen as well
as some crystallographic data. Silicon and oxygen have atomic weights of 28.0855
and 15.9994, respectively. The atomic weight of quartz, SiO2, is therefore 60.0843
23
(= 28.0855 + 15.9994 + 15.9994). This means that a mole of quartz, 6.022 × 10
SiO2 molecules, weighs 60.0843 gm.
Crystallographers have determined that quartz crystals are made of fundamental
unit cells shaped like trigonal prisms (discussed in detail in later chapters)
containing three (Z = 3) SiO2 molecules. Each unit cell has a volume of 112.985 Å3.
So we may calculate the volume of a mole of quartz as:
V = NA × v/Z
(where V, NA, v, and Z are the molar volume, Avogadro’s number, the unit cell
volume, and the number of molecules per unit cell, respectively.)
So, V = 6.022 × 1023 SiO2/mole × 112.986 Å3/unit cell ÷ 3 SiO2/unit cell

= 2.268 × 1025 Å3 = 22.68 cm3

which is slightly smaller than a golf-ball.
We can, if we wish, then calculate the density (ρ) of quartz from the molar data:
ρ = molar weight ÷ molar volume = 60.0843 gm/mole ÷ 22.68 cm3/mole = 2.649 gm/cm3
2.2.4 Radioactive Minerals
Most common isotopes are stable isotopes. In 1896 Henri
Becquerel discovered unstable isotopes and radioactivity when
he unintentionally conducted an experiment. He put some
radioactive samples in a drawer, along with a photographic
plate, and subsequently found that the plate had recorded the
image of a key that had been sitting on top of it. Although
Becquerel didn’t know it, X-rays given off by uranium-rich
minerals had caused the image. During the following decade,
researchers including Marie and Pierre Curie and Ernest
Rutherford determined that some elements, termed
radioisotopes, emit alpha (α) and beta (β) particles, and
gamma (γ) radiation as they decay to form daughter isotopes.
So, minerals are radioactive if they contain radioisotopes,
and the amount of radioactivity depends on isotope
concentration. Most minerals that contain potassium, uranium,
and thorium are radioactive, but many other elements besides
these three can also contribute to radioactivity.

Radioactive minerals are quite rare but many of them, like

those seen n Figures 2.6, 2.7 and 2.8 below, make spectacular
mineral specimens. Torbernite (copper-uranium phosphate),
autunite (calcium-uranium phosphate), and uraninite (uranium
oxide) are three of the most common examples. Uraninite is the
main ore for uranium; the specimen seen here comes from a mine
in Topsham, Maine.

2.8 Uraninite on
2.6 Torbernite 2.7 Autunite calcite
2.2.5 Periods
In atoms, electrons are in orbitals of different energies
around atomic nuclei. Electrons are always moving, and
orbitals are regions around a nucleus where electrons are
likely to be found. We designate an orbital using a number and
a letter, for example 1s or 1p – the number refers to a
specific electron shell and the letter to a subshell. S
subshells can hold up to 2 electrons, p subshells up to 6
electrons, d subshells up to 10 electrons, and f subshells up
to 14 electrons. The table below lists, from lowest to highest
energy, the orbitals and the maximum number of electrons they
can hold. s orbitals are spherical, p orbitals are dumbbell
shaped, d and f orbitals have more complicated shapes. The
shape of an atom’s outermost occupied orbital is often
important because it can affect mineral properties such as
color. But, for simplicity, in the drawings below we show all
orbitals as circular.

Electron Orbitals
orbital 1s 2s 2p 3s 3p 4s 3d 4p 5s 4d 5p 6s 4f 5d 6p 7s 5f 6d 7p
max # 2 2 6 2 6 2 10 6 2 10 6 2 14 10 6 2 14 10 6

electrons

increasing energy ➔

In a neutral (non-ionized) atom, the number of electrons is

equal to the atom’s atomic number. So, a hydrogen atom (atomic
number 1) contains one electron, a helium atom (atomic number
2) contains two electrons, a lithium atom (atomic number 3)
contains three electrons, and so forth to high atomic number.
Electrons populate orbitals sequentially, filling the lowest
energy orbitals first and proceeding to higher energy
orbitals. Thus, hydrogen has one electron in a 1s orbital.
Helium has two electrons in a 1s orbital, lithium has two
electrons in a 1s orbital and one electron in a 2s orbital,
etc.
2.9 Electron orbits in a sodium
atom

For an example of electron configuration in a heavier element,

consider sodium. This schematic drawing in Figure 2.9 shows
orbitals and electrons in a sodium atom. The electrons are
orbiting a nucleus that contains protons and neutrons (but
they are not all visible). Sodium has atomic number 11, so a
neutral atom has 11 protons and 11 electrons. Natural sodium
also has 12 neutrons. The electrons fill the 1s, 2s, and 2p
orbitals. One additional electron is in the 3s orbital.

2.10 Electron orbitals of valence electrons

We call electrons that contribute to chemical bonding valence

electrons. These electrons are generally in the outermost
electron shells (but may be in an inner shell for transition
metals). This chart (Figure 2.10) shows the orbitals of
valence electrons for different elements. Elements in the same
period (row) of the periodic table have their valence
electrons in the same s or p shell, or (for transition
elements) in a d orbital of the next lower shell.

The first period contains only two elements (H and He), while
the second and third have eight each. Electrons for second and
third period elements are in 2s, 2p, 3s, or 3p orbitals. The
fourth through seventh periods contain some elements with
valence electrons in s or p orbitals, but also contain 20
elements having valence electrons in d-orbitals. The sixth and
seventh periods contain elements with valence electrons in s,
p, d, or f orbitals. The lanthanides and actinides have
valence electrons in f orbitals. Thus, as seen in the chart
above, elements with high atomic number (compared with
elements of low atomic number) have valence electrons in
higher-energy orbitals.

2.3 Ions
Atoms are somewhat unstable if valence electrons do not
completely fill outer shells. They are very stable when the
outer shells are fully occupied. So, atoms commonly give up,
or borrow electrons, producing ions, to obtain this stability.
As a result, they may become cations, which have a net
positive charge because there are more protons than electrons,
or anions, with a net negative charge because there are more
electrons than protons.
 2.11 Forming an oxygen anion

The schematic drawings in Figure 2.11 show how a neutral

oxygen atom can gain two electrons to become an anion. The
neutral atom has 8 protons in its nucleus and 8 electrons. It
contains two 1s electrons and six electrons in its outer shell
(2s and 2p orbitals). Thus, 2 additional electrons are needed
to fill the outer shell. Commonly, two electrons move in to
stabilize the atom – and the result is an oxygen anion. The
ionic charge (-2) is the number of protons (still 8) less the
number of electrons (now 10).

Different cations and anions may have different ionic charge.

We call all ions with a charge of +1 or -1 monovalent. Those
with a charge of +2 or -2 are divalent. Trivalent and
tetravalent refer to charges of +3 or -3 and +4 or -4. The
oxygen ion shown above is a divalent anion.

Some elements ionize more easily than others. Those that

ionize most easily to become cations are called metallic
elements. The degree to which elements are metallic generally
decreases from left to right in the Periodic Table. The
elements on the right side of the table, which ionize to
become anions, are nonmetallic.

2.3.1 Typical Valences

2.12 Valences of typical ions

The chart seen here shows the most common ions for different
columns in the Periodic Table. Common cations are shaded blue
and common anions are shaded red. This table is not to suggest
that all the elements always form ions. But, when they do, for
most elements, their likely ionic form is predictable as
shown.

The red numbers below the columns are typical ionic charges.
Elements in groups on the left side of the table have “extra”
electrons in outer shells, and readily give up those electrons
to become cations. Elements in group 1 (alkali elements)
generally have an ionic charge of +1. Those in group 2
(alkaline earth elements) usually have an ionic charge of +2.

On the other side of the chart, helium and other elements in

Group 18 of the Periodic Table have completely filled outer
shells and, consequently, do not ionize. Elements in Groups 16
and 17 have room for a small number of additional electrons in
their outer shells and consequently accept extra electrons to
become anions. Elements in group 17 (halogens) are lacking a
single electron to fill their outer shells completely. So,
they typically become monovalent anions (charge of -1) because
they acquire an extra electron to fill the shell. Oxygen and
other elements in Group 16 typically form divalent anions
(-2). Elements in Group 13, Group 14, and Group 15 typically
ionized to form cations with charge of +3, +4, and +5,
respectively, although other valences are possible.

Ionization of the transition metals in the central portion of

the table (groups 3 through 12) is less predictable than
ionization of elements near the table’s sides. When ionized,
elements in groups 3, 4, and 5 commonly have valences of +3,
+4, and +5, respectively. The rest of the transition metals
exist in a number of ionic states, typically +2 or +3.

The three most abundant elements in the Earth’s crust (oxygen,

silicon, and aluminum) are on the right side of the Periodic
Table, in groups 13, 14, and 16. Oxygen typically has a charge
of -2, silicon of +4, and aluminum of +3.

2.4 Bonding in Minerals

2.4.1 Ionic Bonds
Negatively and positively charged particles attract each
other. Protons (positively charged) attract electrons
(negatively charged) in atoms. Similarly, positively charged
cations attract negatively charged anions, producing ionic
bonds in minerals.

2.13 Ionic bonding to form halite (NaCl)

We call the process of losing electrons to become a cation

oxidation. Gaining electrons, and becoming an anion, is
reduction. The formation of cations and anions go hand in
hand. Metals becoming cations donate electrons to non-metals
becoming anions. The drawing shown in Figure 2.13 is a
schematic showing how sodium and chlorine can combine to make
the mineral halite (NaCl). First sodium (Na) and chlorine

(Cl) react to produce the ions Na+ and Cl– (when a 3s electron
from sodium is added to the 3p orbital in chlorine). Na is
oxidized as Cl is reduced. These two ions then bond, producing
NaCl.

Oxidation and reduction also occur when iron and oxygen

0
combine to form hematite (Fe2O3). Two Fe atoms each donate 3
0
electrons that are picked up by three O atoms. So, the iron
atoms become Fe3+ while the oxygen atoms become O2-. The ions
may then bond ionically to produce (Fe2O3). Alternatively, iron
2+
sometimes oxidizes to Fe and combines with oxygen to form the
mineral wüstite (FeO). A third iron oxide mineral, magnetite

(Fe 3 O 4 ), contains both Fe 2+ and Fe 3+ . Wüstite, hematite and

magnetite are all minerals, but wüstite is not a common
mineral on Earth (although it is found in meteorites) because
under normal Earth surface conditions, iron easily oxidizes to
become trivalent. The bonds in all three iron oxides, however,
are not entirely ionic; they are partially covalent (discussed
later in this chapter).

2.4.1.1 Common Cations and Anions

Common Cations,
Anions, and Anionic
Groups
anionic
cations anions groups
monovalent

Na+ Cl– (NO3)–

 K+ F– (OH)–

H+

divalent

Fe2+ O2- (SiO4)4-

Ca2+ (SO4)2-

Mg2+ (CO3)2-

(WO4)2-

other ions

Al3+ (BO4)5-

Fe3+ (PO4)3-

Si4+

C4+

The table seen here lists the most common cations and anions
in Earth’s crust. For the most part, these are the same
elements that we discussed at the beginning of this chapter
when we considered the composition of the crust and mantle.

While thinking of individual atoms bonding together to form

minerals seems straightforward, in reality atoms are seldom
unbonded to others. Single atoms are very reactive. They tend,
when possible, to bond to other atoms to form molecules and
often compounds. Sometimes, they bond to other atoms of the
same element. For example, N2, composed of molecules containing
two nitrogen atoms, dominates the Earth’s atmosphere. Small
atoms with several valence electrons, such as silicon or
carbon, are especially reactive. They seldom exist by
themselves, readily combining with oxygen, and sometimes other
elements, to form strongly bonded anionic units called anionic
groups, also called molecular ions or polyatomic ions.
Molecular ions are so strongly bonded that they behave like
individual anionic units in many minerals. So when minerals
dissolve, molecular ions such as (CO3)2- will not dissociate in
water.

The right-hand column of the table lists the most common and
important anionic groups. They include, from top to bottom,
the nitrate, hydroxyl, silicate, sulfate, carbonate,
tungstate, borate and phosphate groups. As pointed out in the
previous chapter, we generally classify minerals based on
their anion species because the properties of minerals with
the same anions or anionic groups are generally very similar.
So, we often write mineral formulas with parentheses to
emphasize any anions or anionic groups that are present.

Any of the cation species listed in the table can combine with
the anions or molecular anions to produce a long list of
different ionic compounds. For example, all these minerals

contain Ca2+:

lime fluorite larnite anhydrite calcite scheelite

CaO CaF2 Ca2SiO4 CaSO4 CaCO3 CaWO4

Lime, fluorite, larnite, anhydrite, calcite, and scheelite

belong, respectively to the oxide, halide, silicate, sulfate,

carbonate, and tungstate mineral groups. Additional Ca 2 +

minerals with more complicated formulas include sinjarite
(CaCl 2 •2H 2 O), colemanite (Ca 2 B 6 O 1 1 •5H 2 O), nitrocalcite

(Ca(NO 3 ) 2 •4H 2 O), and apatite (Ca 5 (PO 4 ) 3 (OH)). Thus, Ca 2+ can
combine to make minerals with all except one of the anion and
anionic species listed in the table above. The lone exception

is nitrate, (NO 3 ) – . But, synthetic calcium nitrate (CaNO 3 ),

called Norwegian saltpeter is sometimes manufactured for
industrial use.

2.4.2 Covalent Bonds

Rather than giving up or gaining electrons to become ions,
atoms with incompletely filled energy levels may become
stabilized by sharing electrons. The sharing of pairs of
electrons between atoms produces a covalent bond. If the
sharing is complete, the bond is 100% covalent. This is the
case for diatomic gases such as N 2 , but does not occur in
common minerals except for diamond.

2.14 Covalent bonding to

make hydrogen gas

The simplest example of a covalent bond involves two hydrogen

atoms combining to produce H 2 gas, shown in Figure 2.14.
Hydrogen atoms have one proton in their nucleus and one
electron in orbit. If two hydrogen atoms share electrons, this
effectively puts two electrons in the outer orbits (1s) of
each atom, completely filling the orbits and making a very
stable diatomic molecule, H 2 . Fluorine, chlorine, and other
halogen elements form covalently bonded diatomic gasses (F2,
Cl2, etc.) like hydrogen.

2.15 Covalent bonding to

make oxygen gas

Sometimes covalent bonds involve more than a single pair of

shared electrons. Figure 2.15 shows how two oxygen atoms can
share four electrons to form covalent bonds, and produce O 2
gas. By sharing four electrons, each atom has a total of 8
electrons in its 2s and 2p orbitals, making the outer shells
fully occupied and stable. Because there are two pairs of
electrons shared, we call this kind of covalent bond a double
bond.

2.16 Simplified view of

bonding in quartz

For a mineralogical example of covalent bonding, let’s

consider quartz (SiO2). The drawing shown in Figure 2.16 is the
atomic arrangement in quartz reduced to two dimensions. Four
oxygen atoms surround each silicon atom, and each oxygen atom
bonds to two silicon atoms. A lone silicon atom has 4
electrons in its outer shell (3s and 3p orbitals). But, each
of the surrounding oxygen atoms can share an electron with the
silicon atom, bringing the total number of electrons to 8,
filling the outer shell, and making it stable. And, each
silicon atom can share one electron with each of its four
surrounding oxygen atoms. So, oxygen, which begins with 6
electrons in its outer level (2s and 2p orbitals), gains 2
electrons, bringing the total to 8, which stabilizes the atom.

Many elements form both covalent and ionic bonds. As mentioned

previously, Cl– forms ionic bonds with Na+ to make the mineral
halite. But, two Cl atoms can also bond covalently to form a
stable gas, Cl 2 . The two types of bonds are not exclusive –
ionic bonds and covalent bonds form a spectrum, and in nature,
100% ionic and covalent bonds do not exist. In nearly 100%
ionic bonds, valence electrons are almost exclusively
associated with one atom. (But, even the highly ionic NaCl
bond has a slight covalent character.) In nearly 100% covalent
bonds, electrons are shared more or less evenly between two
atoms. Diatomic gases such as H2 and N2 are extremely close to
being 100% covalent.

So, many minerals (and other substances) contain bonds that

are partly ionic and partly covalent. For example, the
preceding discussion of covalent bonding in quartz was an
oversimplification. In actuality, silicon-oxygen bonds are not
completely covalent – there is still some ionization and
electrons are more localized around oxygen than silicon. The
overall bonding is about half covalent and half ionic. And,

although they behave as anions, (SiO4)2-, (SO4)2-, (CO3)2-, (OH)–

and all other molecular ions (listed in the table above in
Section 2.4.1.1) are held together by partially covalent
bonds. This explains why they act as individual ionic units in
ionic crystals.

2.4.3 Metallic Bonds

 2.17 Metallic bonding

In covalent bonding, pairs of atoms share electrons. In a

third kind of bonding, metallic bonding, many atoms share the
same electrons. Individual atoms give up their valence
electrons and the delocalized electrons are free to move and
interact with all the positively charged ions in the structure
(Figure 2.17). Metallic bonding, is especially common in
minerals involving transition metals. Gold, silver, and copper
are examples of minerals with metallic bonds. Because valence
electrons move easily throughout the structure, metallically
bonded compounds are good conductors of heat and electricity.
Electrons are easily transferred along wires, for example.
Minerals with metallic bonds may be malleable and have only
low-to-moderate hardness, reflecting the loose nature of their
metallic bonds.

2.4.4 Other Kinds of Bonds

Besides the three kinds of bonds just discussed, some minerals
include other types of bonds such as van der Waals bonds and
hydrogen bonds. These bonds do not involve valence electrons
but instead result from weak electrostatic forces due to
uneven charge distribution in a crystal structure. Very weak
van der Waals bonds are important in graphite and some clay
minerals, for example. This explains why graphite is much
softer than diamond, which has the same composition but
covalent bonds. In the mineral brucite, combinations of
hydrogen and van der Waals bonds hold sheets of MgO6 molecules
together. Clay minerals have excellent cleavage because
covalent and ionic bonds create strongly bonded layers, but
weak van der Waals and hydrogen bonds hold the layers
together. So, the minerals split easily into sheets. Because
most mineral properties are explained by ionic, covalent, or
metallic bonds and we can ignore van der Waals and hydrogen
bonds for most purposes.

2.4.5 Ionic, Covalent, and Metallic

Crystals

2.18 Different kinds of bonds in

different minerals

As shown in the triangular diagram (Figure 2.18), bonding in

most minerals is neither 100% ionic, 100% covalent, nor 100%
metallic, but some come close. Most minerals contain
combinations of ionic and covalent bonding. Metallic ore
minerals such as pyrite (FeS 2), stibnite (Sb2S3), and copper
(Cu) generally have little ionic character. Most of them,
especially those in the sulfide and sulfosalt groups, contain
combinations of covalent and metallic bonding. Metallic and
ionic bonds do not often combine, although galena (PbS) may be
one example in which they do to a limited extent.

The degree to which a bond is ionic depends on both elements

involved. For example, because alkali elements (group 1) have
a very strong tendency to become cations, and halogens (group
17) have an equal tendency to become anions, halite (NaCl) and
other alkali halides form crystals in which bonds are nearly
100% ionic. Fluorite (CaF2), too, is nearly completely ionic.
Alkaline earth oxides such as periclase (MgO) or lime (CaO),
involving cations from group 2 and oxygen from group 16, are
about 75% ionic. Many other oxides too, are mostly ionic.
Silicates, the most common kind of mineral in Earth’s crust,
are generally about 50% ionic and 50% covalent. And, bonds in
diamond (C) are entirely covalent.

2.4.5.1 Bonding and Mineral Properties

The nature of chemical bonding in minerals controls many
properties. The table below makes some comparisons. Covalent
bonds are stronger than ionic bonds, which in turn are
stronger than metallic bonds. Consequently, covalent minerals
are hard and tenacious, while metallic ones are usually not,
and ionic minerals fall between. Melting temperatures follow a
similar pattern (covalent compounds generally melt at the
highest temperatures due to their stronger bonds), while
solubility in water is greatest for ionic crystals (because of
their weak bonds and easy ionization).

Bond type also affects crystal symmetry. Both ionic and

metallic bonds are nondirectional so bonding can occur equally
in all directions. In contrast, covalent bonds involve pairs
of atoms and are linear. Consequently, metallic and ionic
minerals generally have high symmetry compared with covalent
ones.
 Characteristics of Minerals Dominated by Different Kinds of
Bonds
property ionic bonds covalent bonds metallic bonds

from opposite sides

common elements close together in the
of the transition metals
involved Periodic Chart
Periodic Chart

electrical
low low high
conductivity

thermal
low low high
conductivity

solubility in water high low very low

melting temperature moderate to high high variable

hardness medium to hard very hard often malleable

brittle, good brittle, common

ability to break variable
cleavage fractures

very high
crystal symmetry high symmetry low symmetry
symmetry

fully to partially partially transparent

transparency opaque
transparent or opaque

most nonmetallic most metallic

how common? some minerals
minerals minerals

halite (NaCl); diamond (C); copper (Cu);

examples
calcite (CaCO3) sphalerite (ZnS) silver (Ag)

2.5 Compositional Variation in

Minerals
 2.19 Cleavelandite (albite-
rich plagioclase)

Most natural minerals are formed from combinations of many

elements. Unlike mechanical mixtures (such as two things being
ground up together) the elements are ordered and intimately
bonded together. The resulting crystalline solutions are solid
solutions. Common plagioclase (a feldspar), for example, is
mostly a solution of CaAl 2 Si 2 O 8 (anorthite) and NaAlSi 3 O 8
(albite). It may also contain a small amount of orthoclase
(KAlSi3O8). Anorthite, albite, and orthoclase are plagioclase
end-members. End-member formulas describe ideal components
that make up solid-solution minerals. The plagioclase sample
seen in Figure 2.19 contains blades of a variety of
plagioclase called cleavelandite accompanied by a few small
flakes of silver-gray muscovite. Cleavelandite has composition
close to end-member albite.

Some minerals, such as hornblende, contain many elements, and

are complex solid solutions that have long and complicated
formulas. Other minerals, including plagioclase and other
feldspars, have limited compositional variations. Still
others, such as fluorite (CaF 2 ) or quartz (SiO 2 ), have
relatively simple formulas and vary little from their ideal
compositions.
We may broadly classify the elements that comprise a mineral
in one of three categories: major elements, minor elements,
and trace elements. Major elements are those fundamental to a
mineral; they control a mineral’s basic atomic structure and
gross properties. They must be in the mineral or the mineral
has been misidentified. Minor elements are those present in
small amounts, usually as replacements for a major element.
Such elements, perhaps in amounts up to a several weight %,
may affect color and a few other properties, but the basic
atomic arrangement of a mineral is controlled by its major
element chemistry. Minerals also contain extremely small
amounts of elements called trace elements. Trace elements are
in all minerals and provide valuable information for
geologists attempting to determine how, when, and where
specific minerals formed. They have little effect on most
mineral properties. A notable exception to this is sometimes
color; even trace amounts of some elements can have major
effects on a mineral’s color.

Mineralogists can acquire mineral analyses in many ways. In

the past, most chemical analyses were determined by titration
and other “wet chemical” techniques. Today we use
sophisticated analytical instruments, including atomic
absorption spectrophotometers and electron microprobes. We
normally report analytical results by listing oxide weight
percentages. We must convert weight percent values to mole
value if we wish to have mineral formulas.
 2.20 Crystals of olivine

As an example of chemical variability in minerals, let’s

consider olivine. An example of olivine is shown in Figure
2.20, above, and the table below lists analyses of six
olivines from different geologic environments. The top part of
the table presents the analyses in terms of element weight %;
the middle part presents the same information in terms of
oxide weight %; the bottom part presents the composition in
terms of the number of atoms in an olivine formula. The three
parts of the table are redundant; values from one can be
converted to the others by an arithmetical process called
normalization (described later in this chapter). Most
mineralogists and petrologists prefer to consider oxide weight
% (for reasons explained in a later chapter), or numbers of
atoms (because numbers of atoms directly translate into
mineral formulas), instead of element weight %.

Olivine Analysis
NewX EastXX
Source: Burma Finland Minnesota Germany
Zealand Greenland

Element Weight %
 Si 19.50 19.21 19.15 14.22 14.09 13.94

Ti 0.00 0.03 0.01 0.72 0.12 0.00

Al 0.00 0.29 0.11 0.26 0.04 0.00

Fe 0.86 3.39 6.11 44.79 50.84 54.01

Mn 0.00 0.18 0.10 0.00 0.78 0.22

Mg 34.88 32.60 31.27 4.93 0.63 0.00

Ca 0.00 0.00 0.11 0.94 1.56 0.00

Na 0.00 0.00 0.01 0.00 0.00 0.00

O 45.43 45.20 44.56 33.86 32.06 31.41

Total 100.67 100.90 101.43 99.72 100.12 99.58

Oxide Weight %

SiO2 41.72 41.10 40.97 30.42 30.14 29.82

TiO2 0.00 0.05 0.02 1.20 0.20 0.00

Al2O3 0.00 1.10 0.42 0.98 0.15 0.00

FeO 1.11 4.36 7.86 57.62 65.40 69.48

MnO 0.00 0.23 0.13 0.00 1.01 0.28

MgO 57.84 54.06 51.85 8.18 1.04 0.00

CaO 0.00 0.00 0.15 1.32 2.18 0.00

Na2O 0.00 0.00 0.03 0.00 0.00 0.00

Total 100.67 100.90 101.43 99.72 100.12 99.58

Atoms per 4 Oxygen

Si 0.978 0.968 0.979 0.957 1.001 1.011

Ti 0.000 0.001 0.000 0.028 0.005 0.000

Al 0.000 0.015 0.006 0.018 0.003 0.000

Fe 0.022 0.086 0.157 1.516 1.817 1.970

Mn 0.000 0.005 0.003 0.000 0.028 0.008

Mg 2.022 1.899 1.848 0.383 0.052 0.000

Ca 0.000 0.000 0.004 0.044 0.078 0.000

Na 0.000 0.000 0.001 0.000 0.000 0.000

Total 3.022 2.974 2.998 2.946 2.984 2.989

As the analyses above show, Mg, Fe, and Si are major elements
in olivines. Olivine may also contain Ca, Mn, Ti, and
sometimes Ni or Cr, but except in extremely rare
circumstances, these elements are minor or trace elements.

The compositions of these six olivines are all different.

Silicon, magnesium, and iron are especially variable. How can
we make better sense of it all? The answer is to use the atom
numbers in the bottom third of the table above to write
formulas for each olivine. The ideal stoichiometry (ratios of
elements) of olivine is (R)2SiO4, where R is usually Fe, Mg,
Mn, or Ca. Thus, a general formula is (Fe,Mg,Mn,Ca)2SiO4 and a
good olivine analysis should produce a formula close to this.
Formulas for each of the six olivines – seen in the table
below – come out to be very close to ideal. Apparent
deviations from ideal stoichiometry (for example, the numbers
of Si in the formulas are not exactly equal to 1) can mostly
be attributed to inclusions of other minerals within the
olivine, or to analytical error.

Normalized Olivine Formulas

% %
specimen
formula forsterite fayalite
source
Mg2SiO4 Fe2SiO4

Burma
(Mg2.022Fe0.022)Si0.978O4
Finland 98.94xx 1.06xx
New (Mg1.899Fe0.086Mn0.005Al0.015Ti0.001)Si0.968O4 95.67xx 4.33xx

Zealand (Mg1.848Fe0.157Mn0.003Ca0.004Al0.006)Si0.979O4 92.16xx 7.84xx

Minnesota (Mg0.383Fe1.516Ca0.044Al0.018Ti0.028)Si0.957O4 20.19xx 79.81xx

East (Mg0.052Fe1.817Mn0.028Ca0.078Al0.003Ti0.005)Si1.001O4 2.77xx 97.23xx

Greenland (Fe1.970Mn0.008)Si1.011O4 0.00xx 100.00xx

Germany

Chemical substitutions in some minerals involve many elements

and are complex. In olivine the substitutions are relatively
simple. We can describe the compositions of most natural
olivines as combinations of the end members (represented by
ideal formulas) called forsterite (Mg2SiO4), fayalite (Fe2SiO4),
tephroite (Mn2SiO4), and calcio-olivine (Ca2SiO4). In contrast
with olivine, many minerals (for example amphiboles and micas)
have lengthy formulas and have many elements substituting in
their structures. For such minerals, choosing useful end
members is difficult and arbitrary.

The six olivines considered above contain very little besides

forsterite and fayalite. This is typical for most olivines
and, if we ignore the other components, we get the percent
values in the two columns on the right in the table. The %
forsterite and % fayalite were calculated, respectively, from
the ratio of Mg/(Mg+Fe) and the ratio of Fe/(Mg+Fe) in each
olivine. The binary (2-component) diagram below (Figure 2.21)
shows the compositions plotted on a line anchored by pure
forsterite on the left, and pure fayalite on the right. The
sample from Burma is nearly 100% forsterite (Mg 2SiO4), while
that from Germany is 100% fayalite (Fe 2 SiO 4 ). The other
olivines fall between.

2.21 Comparing olivine compositions

The degree to which elements may substitute for each other

depends on the elements and on the mineral. In olivine, Fe and
Mg mix freely, so any composition between fayalite and
forsterite is possible. Olivines can incorporate only minor
amounts of Ca, however, so no compositions midway between
calcio-olivine and forsterite are found in nature. In
contrast, Ca, Mg, and Fe mix freely in some other minerals,
garnets for example. Natural garnets can have any composition
between end members grossular (Ca 3 Al 2 Si 3 O 1 2 ), almandine
(Fe3Al2Si3O12), and pyrope (Mg3Al2Si3O12).
2.5.1 How to Normalize a Mineral Analysis
We normally report mineral analyses in values of oxide weight
%, but mineral formulas are written in terms of numbers of
atoms. Normalization is the process of converting an analysis
into a formula. Normalization is a tedious but straightforward
arithmetical operation when done by hand. Fortunately there
are Excel spreadsheets and programs to do it for you.

The table below gives the chemical analysis of a feldspar from

Grorud, Norway. Column A lists the oxides and column B gives
the results of the analysis. Columns C, D, E, and F give
chemical data for the oxides.

Th
e Grorud feldspar contains 65.90 wt % SiO2, 19.45 wt % Al2O3,
1.03 wt % Fe2O3, 0.61 wt % CaO, 7.12 wt % Na2O, and 6.20 wt %
K2O. To convert from % grams of oxides to moles of oxides, we
divide the oxide weight % values (column B) by oxide atomic
weight (column C); column G shows the results. The values in
column G are relative values only; they total to 1.486, which
has no scientific meaning.

To calculate relative numbers of cations, we multiply the

values in column G (moles of oxides) by the number of cations
in each oxide (column E). To calculate moles of oxygen, we
multiply the values in column G by the number of O in each
oxide (column F). Columns H and I give the results of these
calculations.

Using the numbers in column H and the oxygen total of 2.977,

we could write the formula of the feldspar as
Ca0.011Na0.230K0.132Fe0.013Al0.382Si1.097O2.977, but normal feldspar analyses
are written with 8 atoms of oxygen. So, in the last column (J)
we have multiplied all the atom numbers in columns H and I by
a constant factor so that the number of atoms of oxygen is
exactly 8. The values in column J give us the normalized
formula for the feldspar (Ca0.029Na0.617K0.354)(Fe0.035Al1.025Si2.948)O8.
The parentheses remind us which elements substitute for each
other.

As pointed out earlier in this chapter, most feldspars are

primarily solutions of three end members: anorthite
(CaAl2Si2O8), albite (NaAlSi3O8) , and orthoclase (KAlSi3O8). So,
ideal feldspar formulas have stoichiometry described by
(Ca,Na,K)(Al,Si)4O8; the Grorud feldspar comes very close to
this. Any discrepancy is due to analytical imprecision.

If we ignore the small amount of Fe that is present, the

Grurud feldspar composition can be described as:

XAn = 0.029 CaAl2Si2O8 (= 2.9% anorthite)

XAb = 0.617 NaAlSi3O8 (= 61.7% albite)
XOr = 0.354 KAlSi3O8 (= 35.4% orthoclase)

The numbers 0.029, 0.617, and 0.354 are called mole fractions;
they are equivalent to percent divided by 100.

Above we plotted olivine compositions on a line – on a binary

diagram. Lines work for depicting solid solutions with two end
members but not for solutions, like feldspars, that have
three. We use triangular (ternary) diagrams for that purpose.
And, for solutions with more than three end members, we are
out of luck because we cannot easily draft diagrams in more
than two dimensions.
 2.22 Ternary feldspar diagram

The diagram seen in Figure 2.22 is a ternary feldspar diagram

with end members anorthite, albite, and orthoclase at the
corners. The red numbers label lines of constant anorthite
content, the purple numbers label lines of constant orthoclase
content, and the green numbers label lines of constant albite
content. Thus, any feldspar made of combinations of albite,
anorthite, and orthoclase can be plotted on diagrams like
this.

The yellow region shows a range of compositions that includes

most common feldspars. Except in very high temperature igneous
rocks, feldspars are either alkali feldspars (solutions
dominated by orthoclase and albite) or plagioclase (solutions
dominated by albite and anorthite). Solutions between
anorthite and orthoclase do not exist in nature. The white dot
shows where the composition of the Grorud feldspar plots; it
is an alkali feldspar that is 62% albite.
2.6 Common Minerals
We classify rock-forming minerals in many ways. Often,
geologists contrast primary minerals and secondary minerals.
Primary minerals are those that are present from the time a
rock first forms. Secondary minerals form later by chemical or
physical reaction within the rock. Often, such secondary
reactions involve H 2 O or CO 2 and occur during weathering,
diagenesis, or other low-temperature alteration of a
preexisting rock. The distinction between primary and
secondary minerals is not exact. A primary mineral in one rock
may be a secondary mineral in another.

We further divide primary minerals into essential minerals and

accessory minerals. Essential minerals are those whose
presence is implied by the name of the rock. All limestones,
for example, contain calcite or dolomite, and all granites
contain quartz and K-feldspar. Essential minerals, for the
most part, control rock properties.

Accessory minerals are generally present in minor amounts and

do not affect most rock properties. These minerals may be made
of common elements such as iron in magnetite (which is a
common accessory mineral). Accessory minerals also commonly
contain incompatible elements, elements that are not easily
incorporated into essential minerals. Zirconium (Zr), for
example, often concentrates in zircon, ZrSiO 4 , a minor
accessory mineral in many rocks. Phosphorus (P) may lead to
formation of phosphate minerals such as apatite,
Ca5(PO4)3(OH,F,Cl), also a common accessory mineral. In most
igneous and metamorphic rocks, silicates make up the essential
minerals, while oxides, sulfides, and other kinds of minerals
make up the accessory minerals.

2.6.1 Common Elements Make Up the Common

Minerals
Although more than 90 natural elements exist on Earth, many
exist in very small amounts. Occasionally they may be
concentrated in accessory minerals, but more often they are
just minor diluents for major elements. It is estimated that
14 elements (O, Si, Al, Fe, Ca, Na, K, Mg, Ti, H, P, Mn, Ba,
C) make up more than 99.7 wt % of Earth’s crust. We can expect
the common minerals to be made from these elements. And, we
can expect minerals made of other elements to be rare.

2.23 Common minerals in the

crust

Mineralogists have identified more than 5,000 minerals. Of

these, fewer than 200 are common, and fewer than 50 are common
enough to be considered essential minerals. Abundant oxygen
and silicon, together making up nearly three quarters of
Earth’s crust, naturally lead to an abundance of silicate
minerals. It is no wonder that other types of minerals, or
minerals that include rarer elements as key components, are
less common. Of the known minerals, about 92% are silicates.
Thus, all the nonsilicates together only make up about 8% of
the total.
 2.24 Plagioclase, a
feldspar

The pie diagram above in Figure 2.23 shows estimated

abundances of different minerals in the crust. Plagioclase,
like the plagioclase shown in this photo (Figure 2.24), is the
most abundant mineral (40% of the total) in Earth’s crust
because it comprises more than half of the basalt that makes
up the oceanic crust. Alkali feldspar and quartz, the major
components in granite, make up about another 25%. The other
minerals, combined, account for 37% of the total.

blank line

●Figure Credits
Uncredited graphics/photos came from the authors and other primary
contributors to this book.
2.1 Carrollite, Robert M. Lavinsky, Wikimedia Commons
2.4 Perthite, Enrico Zimbres, Wikimedia Commons
2.6 Torbernite, Didier Descouens, Wikimedia Commons
2.7 Autunite, Robert M. Lavinsky, Wikimedia Commons
2.8 Uraninite, Robert M. Lavinsky, Wikimedia Commons
2.19 Cleavelandite, Marie Lan Taÿ Pamart, Wikimedia Commons
2.20 Crystals of olivine, James St. John, Wikimedia Commons
2.24 Plagioclase, USGS, Wikimedia Commons