LECTURE 9

MINERALOGY: SOLID SOLUTION, COORDINATION NUMBERS, SILICATE STRUCTURES, and IGNEOUS ROCKS

There are factors other than electrical neutrality and valence rules (last lecture) entering into why we get certain combinations of elements and we can explain some of the physical properties (e.g., cleavage and hardness) of minerals. In chemistry they have the simple situation of stoichiometric (pure) compounds, whereas in geology things are impure and one element substitutes for another. What are the rules governing this substitution (called solid solution)? Also we get certain combinations of atoms; for example silicon (Si) and Oxygen (O) characteristically form SiO₄ groups into tetrahedra with four oxygens surrounding one silicon. Why so?

Solid solution is merely the substitution of one element for another. A simple case is the mineral olivine with a formula of (Fe, Mg)₂SiO₄. Here the comma and parentheses mean that iron and Magnesium substitute freely for one another in any proportion from 100% of either to 100% of the other. If you examine the drawing of cations, you will see that magnesium and iron have similar sizes and identical charges and these facts allow the substitution. Other substitutions occur. For example there is a continuous solid solution series from NaAlSi₃O₈ to CaSAl₂Si₂O₈. Notice that Ca and Na have similar sizes, but they have different charges (+1 for Na and +2 for Ca). You would think they couldn't substitute for each other. However note the at the same rate that Ca substitutes for Na (+2 for +1) Al substitutes for Si (+3 for +4) and this exactly keeps the total charge at zero -- (add the totals for the Ca and the Na feldspar)!. On the other hand there is no continuous substitution of Na for K in KAlSi3O8 to Na AlSi3O8. The charges are right but the Na ion is considerably smaller than the K ion. Interestingly, there is some substitution but if there is too much then the strain gets too great. Specifically, when a hot granite starts to crystallize, the ions are vibrating very actively (the nature of heat) and substitution occurs. Later as the granite (and included feldspars) cool (and get less active and therefore make smaller spaces) the Na or K are forced out of the crystal structure with a kind of fine banded effect.

CaCO₃ (calcite), CaMg(CO₃)₂ (dolomite) , and MgCO₃ (magnesite) make an interesting trilogy. One is a pure calcium carbonate, another is 50/50 calcium magnesium carbonate and the other is pure magnesium carbonate. Some substitution takes place to make these ratios less than perfect, but the point is there are three minerals and not continuous substitution.

Coordination numbers were discovered by Linus Pauling (resulting in a Nobel prize). The idea refers to the concept that the number of anions (usually relatively large) that can surround a cation (usually relatively small) is based upon their relative sizes. These ratios have been exactly calculated, but for us a few comparative points will be as far as we go. If the cations are relatively large the coordination number (the number of anions that surround the cation) will be relatively large -- i.e. , up to 8. If the cation is relatively truly tiny then the coordination number is 2 , or if a little larger 3. The interesting thing turns up in silicates where the silicon ion is small but not tiny and oxygen ions are large, sort of like a baseball to basketballs. You should be able to imagine that four basketballs would if placed together would have just enough room to take a baseball in the pore made by the four basketballs being packed closely together. This analogy is like the silicon tetrahedron with a small Si⁺⁴ ion surrounded by four O^-2 ions. You can easily see that the SiO₄ tetrahedron cannot exist by itself -- it uses various cations and other tricks (to be described) to allow itself to exist, but a single SiO₄ tetrahedron has a negative electrical charge and cannot exist by itself.

Silicates (most rock forming minerals) exist as several types according to the arrangement of their tetrahedra. There are those with isolated tetrahedra that use cations to balance them electrically. One example is olivine (a mineral found in dunite or basalt). Olivine (Fe,Mg)₂SiO4 reflects its tetrahedral structure in its "formula". There are a coupe of arrangements that I am going to ignore, but what follows is a series of arrangements common in rock forming minerals. The first is the single chain silicate; so named because its silica tetrahedra share oxygens at two corners and make long chains. Pyroxenes (augite, enstatite, diopside, jadeite--the real jade of upper Burma, hypersthene etc. are all of this class of minerals) are common in such rocks as basalt and gabbro. The sharing means that each terahedron only has half of the oxygens at two corners so each tetrahedron is one silicon to three oxygens. The formula (trying to use whole numbers for the cations) is written NaAlSi₂O₆ (true jadeite). For a simple pyroxene we might get MgSiO₃ (enstatite), but all of them have 1/3 ratios even if written 2/6. Pyroxenes have this semi equidimensional chain structure (looking endwise at it) where the chains are strongly bonded along their length but somewhat more weakly bonded to each other. The result is a cleavage (planar breakage down to atomic sizes) parallel to the chains at right angles to one another.

Another organization is the so called double chain silicate. These are the amphiboles, an accessory mineral usually present in granite, more common in andesite and diorite and even mostly makes up such rocks as amphibolite (a metamorphic rock). Typical amphiboles are tremolite -- Ca₂Mg₅Si₈O₂₂(OH)₂ -- and hornblende (the amphibole of granite) -- NaAlCa₂(Mg,Fe₄)(Al₂Si₆)O₂₂(OH)₂. Examine the illustration of the double chain silicate. You will see that there are two and only two kinds of tetrahedra. One kind contacts the tetrahedra of the other chain and so has three shared oxygens and one to itself (the top one) while the other kind only has two shared tetrahedra and has two to itself. What is more, in addition to there being two tetrahedral types in a double chain silicate, these types are exactly equal in number so all we have to do to get the Si/O ratio is to add the number of silicons and oxygens in two adjoining tetrahedra along a chain. There are, of course two Si atoms. In one tetrahedron (the first above) there are three halves of oxygen and one hole one; in the other there are two halves and two whole ones. The result is two silicons to 5 halves plus 3 whole oxygens or 2Si/5.5O. This ratio is often written in the identical fractions of 4/11 or 8/22 (see formulas above for hornblende and tremolite. Notice that in hornblende, the Si/O ratio is at first glance 6/22. However, Al has replaced Si in the tetrahedra 1/4 of the time so if one adds Si and appropriate Al numbers we get a 8/22 ratio. Notice that mineralogists try to put more in a formula than simple ratios. Some Al is not included because, although present, it is not surrounded by four O ions.

Note also that endwise looking down the chains that the chains are not equidimensional. This presumable explains why amphiboles have two cleavages not at right angles. See, for comparison, the description above for pyroxene cleavage.

A further organization is the so called sheet silicate. These are the micas. White mica is muscovite -- KAl₂(AlSi₃O₁₀)(Oh,F)2 -- black mica is Biotite -- K(Mg,Fe)₃(AlSi₃O₁₀)Oh,F)₂. In sheet silicates the three oxygens in one plane all are shared with neighboring tetrahedra and, of course, the fourth oxygen is wholely owned by each tetrahedron. Micas have such a good single cleavage because the sheets are strongly bonded in one direction, but relatively weakly bonded to each other. Anyway each tetrahedron is the same and has three 1/2's of oxygen and a whole one. this is an Si/O ratio of 1/2.5. This is usually written in the identical fractions of 2/5 or 4/10 -- see formulas above. Notice again that Al is sometimes substituting for Si in being surrounded by four oxygens.

A last organization are the tectosilicates (or framework silicates). In these silicates every oxygen is shared with an adjoining tetrahedron. This means the Si/O ratio is 1/2; this thereby explains how quartz is made of silica tetrahedra and still has the formula SiO₂. Feldspars, as mentioned above are also tecto or framework silicates. They too have a 1/2 silicon to oxygen ratio, but expressed as 4/8. Furthermore, as mentioned, aluminum in feldspars is substituted for silicon.

An interesting generalization comes from examining the Si/O ratio from 1/4 (olivine) to 1/2 (quartz) through all the intermediaries. The ratio of Si to O steadily increases (to wit,1/2 is bigger than 1/4) - this relates (see later) to the classification of igneous rocks.

We have only touched on mineral physical properties. Physical properties are determined by how the atoms are arranged and attached to each other. Diamond, for example, is made of pure carbon (C) as is graphite. Diamond's carbon atoms are very strongly attached in three dimensions so diamond is very hard. Nonetheless, the geometry of the arrangement creates (within this generally hard mineral) some weaknesses and diamond cleaves in six directions (the most of any mineral and the same as sphalerite --ZnS). Graphite is strongly bonded in two directions, but the layers are weakly bonded. This gives graphite great softness and one excellent cleavage. Oddly, at a microscopic level, graphite can be considered very strong because each layer is strongly held together.

IGNEOUS ROCKS

Minerals are the components of rocks, which is why minerals were discussed before rocks. Traditionally, igneous rocks are considered to be the primary rocks with sediments naturally derived from them and the metamorphic rocks, by there very nature, derived from something. Curiously, many a paper has been written on the "oldest" rock. The present winner is the Isua complex which is close to 4 billion years old.. This rock is mostly somewhat metamorphosed sediments. These rocks contain carbon and it is conceivable that life existed in Isua times.

Further curiously, if the earth grew by collision with meteorites, then sedimentation, in the broad sense, was the process whereby the earliest rocks formed. Isua, by the way is after that time, and so far we have no rocks from Earth's first half billion years.

Anyway, at least we now know igneous rocks are formed from melted materials that originate below the ground. This point alone suggests that igneous rocks weren't primary. The great puzzle of igneous rocks is the following. The deep earth (actually a topic for the future) -- the source of igneous rocks -- is relatively homogeneous while the earth's surface is intensely heterogeneous. Something must differentiate these things (igneous rocks) from a simple start.

We can first classify igneous rocks and then seek an explanation:

K and Na FELDSPAR, QUARTZ, and a little BIOTITE and AMPHIBOLE	NO QUARTZ, Ca AND Na FELDSPAR, PYROXENE and AMPHIBOLE	NO QUARTZ Ca FELDSPAR, PYROXENE, SOME OLIVINE	NO QUARTZ or FELDSPAR, PYROXENE and/or OLIVINE
GRANITE	DIORITE	GABBRO	ULTRAMAFICS-dunite, peridotite and pyroxenite
RHYOLITE	ANDESITE	BASALT or dolerite or diabase	RARE or NEVER
OBSIDIAN	RARE	NEVER or VERY RARE	NEVER

This classification is, in effect, a graph. The vertical axis is grain size. You should wonder right away why we get essentially two grain sizes of igneous rocks (ignoring, for the moment, the rare glasses) and not a continuum. Well, of course the reason is there are two and only two processes for making igneous rocks -- intrusion and extrusion. Intrusions are coarse grained because they cool slowly because of all the insulation above them. Deep intrusions and large intrusions will be coarser than small or shallow intrusions. This knowledge does somewhat muddy the two grain sized classification because big deep intrusives are coarser the shallow small intrusives. Even big extrusives are coarser (although still fine grained) than small extrusives. And, of course, intrusives and extrusives are both finer grained at their contacts. The bottom line of the classification is the glasses, the finest grained of igneous rocks. Glasses form because the lava cooled and solidified so fast that the ions never had a chance to move by electrical attraction to one another and form mineral grains. In effect, glasses are "solid" liquids. Most glasses are granitic in composition because such lavas are very stiff (viscous) and this (in addition to fast cooling) inhibits the formation of even tiny mineral grains. Rocks of andesitic or basaltic composition rarely or very nearly never in the basalt or beyond (to the right) range because all these lavas are runny (unviscous) and so small mineral grains form rapidly. You will note in the library reading of Hall that pieces of basalt melted in foundries and allowed to cool form glass (unlike rocks). Hall, you will recall, showed that this was the result of the fast cooling (of a small amount). This formation of glass in the lab does not contradict its rareness in nature. The point is lava flows are big enough so the cooling generally avoids being fast enough to for glass.

Across the top (in bold) is the chemical -- here mineralogical -- variation of igneous rocks. In the first column (left), granite, rhyolite and obsidian are all made of the same material - there are some variations but typically granites etc. are made of 70% feldspar (shared in various amounts between K- orthoclase and Na - plagioclase feldspars), 20% quartz, and less than 10 % amphibole and biotite. There might be some magnetite and other rare accessory minerals, but these are trivial. The obsidian would have the same composition of ions, but would have no mineral grains (it's a glass)

The diorite column would lack quartz and would be mostly feldspars. Feldspar would be more than 50% and would be generally plagioclase (now with significant Ca in addition to some Na -- remember solid solution) . There also would be significant dark minerals (amphibole -- mostly hornblende, and pyroxene -- mostly augite). The gabbro. column would have no quartz and would be mostly pyroxene (more than half) and the feldspars (plagioclase) would be rich in Ca. Some olivine would occur.

The ultramafic column includes rare rocks that are rich in iron and magnesium and generally lower in silicon (although they are generally composed of silicates). They emphatically have no quartz or feldspar. They are composed of pyroxene, olivine or in some cases amphibole.

The general scheme of the chemical classification is K, Na and Si are concentrated to the left with a sharp diminution of K and Na to the right. Si diminishes much less sharply to the right. In contrast Fe, Mg are low and Ca practically nonexistent to the left. They all increase to the right. This explains most of the mineralogical differences in that, to emphasize K and Na feldspar along with quartz are to the left and the dark minerals and Ca feldspar are to the right.

Before going on, let me comment on volcanos (the surface expression of igneous activity). Volcanos mostly are basalt emitters if oceanic (and that is most of them) . These volcanos generally put out a runny hot liquid that is not explosive (although it may be deadly). Along the Pacific margins the volcanos are mostly andesitic. Andesite is stiffer when a lava than basalt and so may trap gasses and these often result in explosive eruptions (Mt. St Helens). There are some (rare) rhyolitic volcanos and these are very explosive. In fact the hot gasses may be so charged with ash and other solid volcanics that they become heavy and race down hill as a dense hot gas flow. When they stop the solids settle and solidify and even though they have a mechanical origin there appearance is to fool people into interpreting them as rhyolites.

There is a very coarse rock hitherto not mentioned. These are p[egmatites. These are granitic in composition and result as follows. The crystallizing granite removes some water and other volatiles (Cl etc.). However the amount used is well less than that present. The result is the last melt is extraordinarily rich in water etc. and is very runny (unviscous). Consequently, huge crystals 12 -15 feet or longer form. Mostly these are K feldspar, quartz, and biotite but the volatiles cause many oddities (Li mica, Beryl, Tourmaline etc).

The explanation of why earth is so homogeneous (a future topic) at depth (in the upper mantle) while so heterogeneous in the crust. is relatively complex chemistry, but we can make some progress. The Palisades sill crops out in across the Hudson River from New York City. It is a former tabular injection of magma between the layers of the late Triassic (200,000,000 years ago) of that region and now exposed by erosion. This layer of former melted igneous rocks is about 200 feet thick. The top and bottom few feet are a chilled zone because they came into immediate contact with the cold (by igneous standards) sedimentary country rock. This caused them to crystallize immediately into a very fine grained chilled zone and to bake the surrounding country rock into contact metamorphism. Because of the immediate solidification, this chilled zone is pretty much a sample of the entire intrusive rock body except it is finer grained.

The rest of the liquid stayed liquid for quite a while as it crystallized so the sequence of crystallization can be observed because the first things sank. Examination of the rock body just above the chilled zone reveals a layer a few feet thick that is rich in olivine as compared to the rest of the sill. The reasonable interpretation is that olivine formed first (when temperature was highest) and settled to the bottom. Pyroxene is more concentrated in the bottom half of the sill than plagioclase and so we can interpret that to mean pyroxene starts crystallizing before feldspar but their formation overlaps. This explains the lower amount of pyroxene near the top of the sill;l and the higher amount of feldspar.

Anyway, the point is rock bodies can become differentiated by minerals forming at different temperatures and by settling or deformation being separated from the remaining liquid. Other forms of differentiation might include reaction to the country rock and its incorporation, and immiscible fluids (some compositions like oil and water won't mix ).

Last modifie July 8, 1997