METAMORPHIC REACTIONS
The word metamorphism means transformation. Metamorphism of rocks in nature involves
reactions among minerals. Because minerals are chemical compounds, metamorphism involves
chemical reactions.
Type of metamorphic reactions
Polymorphic phase transformation
The simplest metamorphic reaction involves polymorphic transformation of minerals as P or T
changes. The most important metamorphic polymorphs are in the Al2SiO5 system, the minerals
andalusite, kyanite, and sillimanite. The transformation of calcite to aragonite with pressure in
subduction zones is also an important metamorphic transformation.
Solid-solid net-transfer reactions
These reactions involve the formation of a new mineral assemblage from an old mineral
assemblage, without the involvement of a volatile phase. An example is the reaction:
NaAlSi3O8 (albite)  NaAlSi2O6 (jadeite) + SiO2 (quartz),
that occurs in subduction zones with increase in pressure.
Another example is the reaction
Mg3Si4O10(OH)2 (talc) + 4 MgSiO3 (enstatite)  Mg7Si8O22(OH)2 (anthophyllite)
This reaction has hydrous minerals, but no volatile phase is produced or consumed.
Devolatilization reactions
Most metamorphic reaction involve the production or consumption of a volatile phase. Because
they also involve a change in modal proportion of minerals, they are also net-transfer reactions.
An example is:
KAl2AlSi3O10(OH)2 + SiO2  KAlSi3O8 + Al2SiO5 + H2O .
Because devolatilization reactions involve production of a high-volume volatile phase (the
reactions have high positive ∆V), the temperatures that these reactions occur at are highly
dependent on pressure and the proportion of the relevant volatile component in the fluid phase,
because the fluid phase may contain additional component, such as CO2.
Figs. 26-2 and 26-4 from Winter, 2001
Continuous reactions
Rocks that contain minerals that are solid solutions usually undergo continuous reactions over a
broad temperature range. Continuous reactions are particularly important in pelitic rocks in that
minerals have variable Fe/Mg ratios. Consider the reaction
Chl + Qtz  Grt + H2O .
Chlorite, garnet and biotite are all solid solutions containing Fe and Mg end-members. This
implies that the reaction will occur over a given temperature range, depending on the Fe/Mg ratio
of the rock. This reaction can be illustrated using the following schematic phase diagram, that
looks just like the olivine-melt or plagioclase-melt binary diagrams that we covered in igneous
petrology:
During the reaction, the first garnet that will form will be more Fe-rich than chlorite. With
increasing T, more garnet will form at the expense of chlorite as both will become more Mg-rich.
The reaction will be completed when all chlorite disappears.
Ion-exchange reactions
Ion-exchange reactions involve the exchange of two ions between two minerals as a function of
temperature and/or pressure. An ion-exchange may or may not involve a continuous reaction as
above. An example of an ion-exchange reaction in which no mineral is consumed or produced is
the Fe-Mg exchange between biotite and garnet. The exchange reaction is written in the
following forms:
annite + pyrope = phlogopite + almandine
or
Ion-exchange reactions form the basis for geothermometry and geobarometry to be covered later.
Oxidation/reduction (redox) reactions
These reaction involve the addition or removal of oxygen from the rocks. They primarily involve
oxide minerals. Two examples are:
4 Fe3O4 + O2  6 Fe2O3 (MH)
or
3 Fe2SiO4 + O2  2 Fe3O4 + 3 SiO2
(FMQ)
(These reactions also occur in igneous rocks). At a given pressure, these reactions are univariant.
Therefore, when minerals on both sides of the reaction are present, at a given temperature they
fix (buffer) the partial pressure of oxygen (fugacity) in the rock or magma.
Fig. 26-10 from Winter, 2001
Metasomatic ion-exchange reactions
These reactions involve the exchange of ions between fluids and minerals. Therefore these
reactions can be thought of as diagenetic reactions. Some examples include:
NaAlSi3O8 + K+Cl- (in fluid)  KAlSi3O8 + Na+Cl- (in fluid)
[occurs at high T]
or
2 KAlSi3O8 + 2 H+ (acid) + H2O  Al2Si2O5(OH)4 (kaolinite) + SiO2 + 2 K+ [true diagenesis].
APPLICATION OF THE PHASE RULE TO METAMORPHIC ROCKS
In igneous petrology we covered the Gibbs phase rule:
F = C - + 2 .
In magmatic systems, both pressure and temperature can change during evolution of a magma,
hence the “2” at the end of the phase rule equation. Consider, however, a rock that equilibrated at
a given metamorphic facies at fixed P-T conditions. In that case, the phase rule reduces to:
F=C-
This means that for a mineral assemblage to be in chemical equilibrium, the number of phases
(minerals) must be equal to or is less than the number of components. The above form of phase
rule is called the Goldschmidt’s mineralogic phase rule or simply mineralogic phase rule.
How many possible minerals can there be in a metamorphic rock that is in a state of equilibrium?
Let us consider composition of an andesite that may have undergone metamorphism:
SiO2 57.9
TiO2
0.87
Al2O3 17.0
Fe2O3 3.27
FeO
4.04
MnO 0.14
MgO 3.33
CaO
6.97
Na2O 3.48
K2O
1.62
P2O5
0.05
There are eleven of these oxides, which at first view may suggest that we could have 11 minerals
in the rock. Remember, however, that minerals are made of multiple oxides and we have to
reduce the number of components to a minimum for the phase rule to work. In an andesite, we
know that
1) All silicates have SiO2 plus there is quartz, so we do not need to consider it.
2) Na2O and K2O occur essentially only in feldspar where they are always tied-up with
aluminum. So we only have to adjust Al2O3 for these two components and eliminate them
from consideration.
3) FeO, MgO, and MnO usually substitute for each other in ferromagnesian minerals, so we
can combine them into one component.
4) Fe2O3 can be low, but if it is high, it usually behaves as Al2O3, so we treat is the same as
alumina.
5) P2O5 occurs mostly in apatite, so we correct CaO for its presence.
6) We ignore TiO2, because there is only little of it. However, it can cause the presence of
ilmenite or rutile, which are oxides.
So we end-up with these three components on molar basis:
A:
C:
F:
(Al2O3 + Fe2O3) – (Na2O + K2O)
CaO – 3.3*P2O5
FeO + MgO + MnO
Thus, from the many-component system that we started with, we have reduced it to the three
component system, which we can depict on a flat piece of paper. This means that in addition to
quartz, alkali feldspar, and apatite, we can have three additional minerals at a fixed pressure and
temperature, or a given facies. Because in the reduction of the rock composition to the three
components we implicitly considered quartz, alk. feldspar, and apatite, we say that the
compositionn of the rock or include minerals are projected from these minerals or their
components - Na2O, K2O, SiO2, P2O5, SiO2.
CHEMOGRAPHIC REPRESENTATION OF METAMORPHIC FACIES
Mineral assemblages in projected metamorphic systems can be depicted on ternary diagrams.
The basic ACF and AKF diagrams were devised by Eskola in 1915.
ACF diagram
The minerals which fall in the ACF system can be represented in a ternary diagram. Their
position in the diagram is calculated with the equations shown above (see Fig. 24-4). At a given
facies, a rock of a given composition will have a set of three (or two) ACF minerals, in addition
to quartz, alk. feldspar, and apatite. (Fig. 24-5). At another facies (P-T) conditions, the set of
minerals will be different. The ACF diagram is most useful for illustrating mineral assemblages
in mafic, calcareous, and calc-silicate rocks.
Figs. 24-4 and 24-5 from Winter, 2001
AKF diagram
The AKF diagram is most useful for illustrating reactions in pelitic and mafic rocks. It is
a projection from plagioclase, alkali feldspar and quartz. The three components are calculated as
follows:
A:
K:
F:
(Al2O3 + Fe2O3) – (Na2O + K2O + CaO)
K2O
FeO + MgO + MnO
Note that in calculation of the A component, we are interested in Al2O3 that occurs in excess of
the amount tied-up in plagioclase and alkali feldspar. The positions of the most common AKF
minerals and example of assemblages in the low-P amphibolite facies are shown in Figures 24-6
and 24-7.
Figs. 24-4 and 24-5 from Winter, 2001
AFM (AKFM) diagram
In 1957, James B. Thompson recognized that there is a deficiency in the ACF and AKF
diagrams, in that they do not consider the important solid-solution properties of Fe and Mg. The
problem of lumping Fe and Mg together is especially critical for pelitic rocks, because minerals
can have variable Fe/Mg ratios.
Thompson recognized that virtually all pelites have muscovite (or K-feldspar at granulite
facies), quartz. Another assumption he made is not usually true, is that PH2O = Ptotal, which
implies that all hydrous minerals contain only OH on their “hydroxyl” sites. (This is not always
true, however.) The derive the diagram, first he reduced the components K2O, Al2O3, and SiO2
into two components, by projecting all appropriate minerals from SiO2 (left diagram below).
Then he projected rest of the minerals from either muscovite or K-feldspar unto the Al2O3-FeOMgO plane (right diagram below).
As an example, the right diagram shows the projected position of biotite from muscovite. Note
that the projected muscovite falls and negative A-components. That is OK.
Mathematically, the AFM coordinates are calculated as follows:
A:
F:
Mg:
Al2O3 – 3*K2O
Al2O3 – K2O
FeO (+ MnO)
MgO
(if projected from muscovite)
(if projected from K-feldspar)
Figure 24-19 shows the position of most common AFM minerals as projected from muscovite.
Figure 24-20 shows an AFM phase diagram appropriate for the lower amphibolite facies. Note
the invariant and divariant fields.
Chemography of Reactions
Phase-appearance/disappearance discontinuous reaction
Switching tie-line discontinuous reaction
Continuous reaction