Introduction to Silicates

As discussed in Chapter 5 of the book, the 8 elements O, Si, Al, Fe, Ca, Na, K, Mg make up nearly 99% of the crust of the earth (by weight). The actual numbers are:

Composition of the Earth's Crust

Element	Weight %	Atomic %	Volume %
O	46.60	62.55	94
Si	27.72	21.22	6 in total
Al	8.13	6.47
Fe	5.00	1.92
Ca	3.63	1.94
Na	2.83	2.64
K	2.59	1.42
Mg	2.09	1.84
Totals	98.59	100.00

Looking at the atomic percent column in the table, it should be no surprise that most of the minerals crystallized from melts in the outer parts of the earth are silicates. (For those who like trivia - the fact that by volume the crust is 94% oxygen is a neat statistic.)

Strict radius ratio analysis places the Si:O bond (0.26/1.36 = 0.19) in the trigonal packing group (0.155-0.225) however the high charge on the Si ion coupled with the orbital hybridization (sp3) generates 4-fold packing - a tetrahedron. The boundary between tetrahedral and octahedral coordination lies at Rc:Ra of 0.414 and thus there is little chance that silicon will be found in octahedral coordination near the surface of the earth.

Under greater pressures however the 'fluffy' oxygen ions can be compressed, reducing their radii and eventually packing 6 of them around a silicon cation. This occurs in the mantle of the earth and the familiar quartz framework structure can be converted to the rutile-structure (stishovite). It is likely that all silicon is in 6-fold coordination below about 670 km. (Think about this in terms of understanding the mineralogical makeup of the mantle and the geophysical properties of the earth.)

If a silicon atom bonds to 4 oxygen atoms then its (the silicons) valence is satisfied but each oxygen has only satisfied half its charge needs. Thus the oxygen can form a bond either with another silicon or with one or more other cations. An oxygen bonded to two silicon atoms is termed a bridging oxygen.

Because of the crustal abundance of aluminum, it is important to consider its role in silicate structures. The ionic radii of Al allows it to fit easily in either tetrahedral or octahedral sites in an oxygen structure. In silicate minerals up to half the tetrahedral sites can be occupied by Al. Other cations which can substitute to some degree for Si include Boron, Beryllium, Titanium and ferric Iron.

With the exception of some surfaces and local defects, crystal structures must be charge balanced and thus substitution (for silicon) of cations with a charge other than +4 (Al+3, B+3, Fe+3) must be charge balanced by the addition of other cations into near or adjacent sites in the mineral's structure.

In principle, silicate minerals can consist of tetrahedra that are:

1. isolated
2. share 1, 2, 3, or 4 corners
3. share edges or
4. share faces.

The high charge of the Si+4 cation mitigates against either 3. or 4. actually occurring - only one mineral with edge sharing tetrahedra (of Si) is known. No face sharing Si tetrahedra minerals have been found in nature. Thus adjacent silica tetrahedra can effectively only have one common oxygen atom.

The bonds between the tetrahedrally coordinated cations and their anions (usually oxygen) are commonly the strongest in the structure. These cation groupings (i.e. rings or single chains or ...) form the backbone of the minerals structure and the nature of the groupings determines physical properties like cleavage and hardness.

We generally classify the silicate minerals based on the numbers of these shared or bridging oxygen atoms. Because of these bridging oxygens we say that the silicate structures are polymerized. This classification scheme has the further bonus that it relates directly to the minerals composition. Thus increasing polymerization corresponds to an increase in the Si:O ratio between the limits of 1:4 and 1:2.

Figure 13.3 on pages 442-3 in your text shows most of the possible combinations of silica tetrahedra starting with the

• isolated tetrahedra with a Si:O ratio of 1:4. Minerals with such tetrahedra are variously called 'nesosilicates', 'orthosilicates' or 'island' silicates. The most important example of this structure type is olivine.
• Two tetrahedra sharing a corner form a 'sorosilicate' or 'bow-tie' silicate and have a Si:O ratio of 2:7 (1:3.5).
• Three, four or six tetrahedra can each share 2 corners and form a variety of ring structures grouped under the term 'cyclosilicate'. In all these cases the reduced formula has a 1:3 Si:O ratio although the mineral formula is more useful if it is written indicating the actual nature of the grouping (i.e. Si₃O₉, Si₄O₁₂, or Si₆O₁₈).
• Infinite single chains of tetrahedra (each of which shares 2 corners) are known as 'inosilicates' or 'single chain' silicates and have a reduced formula of 1:3 although here again the formula is more useful if reported as Si₂O₆. The most common examples of this type of structure are found in the pyroxenes such as diopside or enstatite. The pyroxenoid minerals (pyroxene-like) have again reduced formulas of Si:O = 1:3 but in fact have 3, 5, 7 and 9 tetrahedra repeat distances and have 'correct' formulas which report this number of silicate tetrahedra (e.g. pyroxmangite has a formula built around Si₇O₂₁).
• Double chain 'inosilicates' where each single chain is cross-linked to a parallel single chain by sharing an oxygen are best typified by the amphiboles. The Si:O ratio of this construction is 4:11 (= 1:2.75) and the infinite double chains are held together by a variety of mono-and divalent cations in sites whose coordination polyhedra range from 6- to 12-fold. Amphiboles are also generally hydrous with structurally integral OH (or possibly Cl, F) groups.
• If all 3 of the basal largely coplanar oxygens are shared between adjacent tetrahedra, sheet or phyllosilicates result. The Si:O ratio here is 2:5 (or 1:2.5) and this group of silicates includes the micas and the clay minerals.
• The final step involves sharing the apical oxygens in a continuous 3-D network of interlinked tetrahedra - the framework or tectosilicates. If the tetrahedra are all occupied by Si atoms then the resulting Si:O ratio of 1:2 is also completely charge balanced and the resulting mineral is one of the polymorphs of SiO₂ - e.g. quartz. Substitution of Al for one out four Si atoms (and the coupled addition of a +1 cation like Na or K) results in one of the feldspar minerals - the most common mineral in the crust of the earth.

Over the next few weeks we will work our way down this list beginning with the isolated tetrahedra minerals.

Examination of the drawings on pages 442-3 of your text show that the basic silicate backbone of each different grouping of tetrahedra has some inherent symmetry. For example the 3-, 4- or 6-fold rings in the sorosilicates have, ideally, 3-, 4- or 6-fold rotational symmetry respectively. If the rings are warped during the process of linking them to each other by additional cations (or they are not stacked directly over one another) then the symmetry can be lowered. For another example consider the sheet silicates - the figure shows inherent 6-fold symmetry in the rings. There are in fact hexagonal sheet silicates known in nature but most phyllosilicates are either monoclinic or triclinic with orthorhombic and trigonal examples common as well.

Thus our study of the silicate minerals will bounce back and forth between chemical and structural views of the common silicates and you must begin to learn the mineral formulas and have an appreciation of the common structure types.