Download SiO2 Al2O3 Fe2O3 Na2O K20 TiO2

IV. COMPOSITION OF CRUST, MANTLE, AND CORE
4.1 CRUSTAL ROCKS
Lets say a little bit about the rocks that form the earth’s crust and those we infer are in the mantle - again
you will hear more next quarter, and in later courses! (Petrology is this field - the study of rocks)
Since we’re interested in gross earth structure - we focus on igneous rocks those derived by cooling from a
magma (whether on the surface or at depth). We‘re not interested in sedimentary or metamorphic rocks
which are derived by reworking or altering igneous rocks. After all, the crust is a very small (volumetric)
portion of the earth and in this class we’re not interested in purely crustal processes since we really want to
extrapolate to the mantle.
The classification of igneous rocks is a complex and involved business, lets just look at some simple ideas.
Rocks can be classified a number of ways: the same chemical composition can have different names
depending on how it was formed.
two names: extrusive (formed above surface - volcanic)
intrusive (formed at depth - plutonic)
Composition of Igneous Rocks in Weight %
INTRUSIVE
(Extrusive)
SiO2
Al2 O3
Fe2 O3
FeO
MgO
CaO
Na2 O
K2 0
TiO2
GRANITE
(Rhyolite)
71
15
2
2
1
2
4
4
0.4
DIORITE
(Andesite)
58
17
3
4
4
7
3
2
1
GABBRO
(Basalt)
49
18
3
6
8
11
2
1
1
These make up most of the earth’s crust: granite is the most common intrusive and basalt the most common
extrusive. Naturally, many more rock types are named to provide finer distinctions.
The SiO2 (silica) content is the key index here!
Note - basalt has much more iron and magnesium than granite, much less SiO2 !
Rocks really aren’t discrete types they’re continuous and their names indicate ranges of compositions.
-2-
The upper continental crust can be thought of as mostly granite (grano-diorite)
The lower continental crust is considered different intermediate to basic rocks (i.e., less SiO2 than the
upper crust). There is no sharp transition between the upper and lower crust, so the boundary is unclear.
Oceanic crust - mostly basalt (gabbro)
These compositions tend to fit the seismic data
(P velocity, km/s)
(g/cm3 )
granite
6.0
gabbro
6.8
2.65
3.0
-3-
-4-
4.2 THE MANTLE
What about the mantle? The approach petrologists took was to find rocks that
a) occur on the surface so their properties are known
b) have the physical properties know, from seismology, to exist
( ~8 km/s, ~3. 3g/cm3 )
c) could be the source of the basalt that makes up the oceanic crust.
in
the
upper
mantle
ASIDE - later we’ll see that the oceanic crust is formed by volcanic activity at (spreading centers, midocean ridges) - so the mantle must be a source for the oceanic crust.
The most likely possibilities were proposed
I. Peridotite-ultra basic rock
= 7.8-8.0 km/s = 3. 3g/cm3
60 - 80 % olivine (Mg, Fe)2 SiO4
pyroxene - enstatite MgSiO3 ; diopside CaMgSi2 O6
garnet/spinel/plagioclase - contain the Al2 O3
II. Eclogite-basic rock
= 8.0-8.2 = 3.4-3.5 g/cm3
garnet
pyroxenes-diopside CaMgSi2 O6
-jadite NaAlSi2 O6
Eclogite is chemically identical to basalt and results from higher pressure.
plag feldspar + pyroxene + olivine → eclogite (garnet + pyroxene + quartz)
The Moho would be a different type of boundary for each case
The peridotite model is generally accepted now, based on laboratory experiments (experimental petrology)
for details take one of Bina’s classes. It also fits seismological data better (Bott).
One interesting idea - If the Moho were a phase boundary, its depth would depend on the thermal gradient.
However, this doesn’t seem to be the case. The seismological data shows that the uppermost mantle
velocity (Pn-headwave velocity) depends on direction:
-5-
This anisotropy would be expected for olivine rich mantle, since olivine crystal structure is strongly
anisotropic, but not for a garnet - pyroxene mantle, so it favors the peridotite model.
Thus, we think of the mantle as having a peridotite chemical composition: a common model is called
"pyrolite"
Above the transition zone (depth < 400km) the composition is
-6-
Uppermost Mantle Composition
Olivine (Fo89 )
(Mg, Fe)2 SiO4
Wt %
57
Orthopyroxene
(Mg, Fe)SiO3
17
Clinopyroxene
(Ca, Mg, Fe)2 Si2 O6 − NaAlSi2 O6
12
Pyrope-rich garnet
(Mg, Fe, Ca)3 (Al, Cr)2 Si3 O12
14
Note: Olivine is 90% Mg, 10% Fe, Pyrope is the Mg-garnet, Mg3 Al2 Si3 O12 with increasing pressure,
these minerals undergo changes to higher pressure phases.
To illustrate this, consider a common model: For a mantle composition, note that at a pressure
corresponding to the depth (400 km discontinuity) the olivine-spinal phase change occurs. The pyroxene
components go to a garnet structure.
With depth, further phase changes occur, and by the 660 km discontinuity another set of phase changes
occur:
Mg2 SiO4 → MgSiO3 + MgO
spinel structure → perovskite structure + rock salt structure
-7-
Similarly, other components have phase changes. Hence the rapid velocity changes in the 400-700 km depth range
reflects phase changes.
The smooth increase in velocity with depth in the lower mantle below this, presumably, represent density increase
with depth due to self-compression.
A yet unresolved question: Does the lower mantle have the same chemical composition as the upper mantle or is it
enriched in iron?
-8-
4.3 THE CORE: CURRENT PREJUDICES
Given the size of the earth’s core, we don’t know that much about it-but we ought to say something!
Facts: there are significant velocity and density changes even if the details are unclear at the transitions.
The dramatic changes in physical properties at the CMB strongly suggest a major compositional change rather than
a phase change. The core is thought to be primarily iron for a number of reasons:
1) High pressure experiments suggest a relationship between mean atomic number, density, and

√
K
(called bulk
sound velocity - P wave velocity for zero rigidity material) It is clear that the mantle and core must be very different
with the core close in composition to iron.
-9-
2) Iron is the only heavy element found in volume in meteorites and the solar system.
3) the earth’s magnetic field is thought to be generated in the core by fluid motions - it is to hot to be a permanent
magnet. Iron is the only suitable material.
But-the data show the core is lighter (less dense) than it would be if it were pure iron. In meteorites, some nickel
(Ni) is found with the iron so the core is probably Fe-Ni, but Ni is heavier than iron! The core must thus be mostly
Fe, some NI and an unknown light element.
A number of light elements have been proposed - silicon and sulphur are the most popular but others are possible.
4.4 THE INNER CORE
In contrast to the outer core, the inner core is denser than pure iron, so it is thought to be iron-nickel without a light
element. Thus it is chemically different from the outer core, as well as being solid rather than liquid.
WHY? One possible explanation for the differences can be obtained from the chemistry of the Fe-FeS system at
high temperature and pressure.
-10-
The phase diagram is extrapolated from the laboratory.
-pure iron (Fe) melts at ~4000°
-pure FeS melts at ~3000°
-BUT about 45% FeS (~16% S) mixture melts at a much lower temperature ~1800 °
(this temperature is called a eutectic temperature: material melts into a solid
of the same composition : the eutectic composition)
If it were not at the eutectic - start off with a liquid that is 33% FeS at a temperature of 4000 ° and this cools. Eventually it hits the phase boundary, at this point solid Fe starts to freeze out. As the Fe freezes out of the melt the melt
becomes richer in FeS. With continued cooling we get a solid Fe material and a liquid FeS melt. The material melts
into a solid and liquid of different composition!
This may be what is happening in the core-so the inner Fe core is growing with time as the core cools. It is thought
that the nickel would go preferentially with the iron into the inner core.
If so (model only!), we can say:
Outer core: 30% earth’s mass: fluid ~12 % S, 86 % Fe, 2% Ni
Inner core: 2% earth’s mass: ~80 % Fe, 20% Ni
These numbers are subject to change without notice!
How did it get there? We’ll see that current theories of planetary formation assume the earth heated up to the melting
point of iron which sank to form the core. If so, the inner core is a result of the slow cooling of the original core.