fluids in metamorphism

Read: Guiraud et al., 2001, JMG 19, 445–454

I along with solid minerals, metamorphic rocks commonly carry
a fluid phase
I in most silicate rocks, this is dominantly H2 O, but more
complex fluids also occur
I H2 O–CO2 –CH4 –SO2 mixes
I fluids with dissolved ionic species, e.g. Cl− , Na+ , Si4+ , Ag+
I many minerals carry -OH bonded in their structure
I e.g. micas, chlorite, amphiboles
I the breakdown of these minerals releases this -OH as free H2 O
I calcite, dolomite, siderite etc. break down to release CO2
I organic material release CH4 and SO2
I fluid is a mobile phase
I can enter or leave a rock → open system process
I it is this irreversible process that allows metamorphic
assemblages to be preserved
fluid escape mechanisms

I fluids are produced by dehydration reactions during

metamorphism
I e.g. chlorite + muscovite → biotite + staurolite + H2 O
I H2 O on high-T side, so H2 O released with increasing T
I rocks have negligible porosity, so fluids must escape
I requirements for fluid escape:
I low viscosity (ability to flow)
I water has a very low viscosity
I a potential to move (the need to flow)
I pressure gradients, such as from the lower to the upper crust
I a pathway to move along (permeability)
I permeability occurs on several scales
I percolation: small-scale intragranular flow
I channel flow: fractures, faults & shear zones
I porosity waves: fluid pressure creates its own porosity which
moves through the crust
what evidence do we see of fluid flow?
I mostly provided by things
precipitating out of the fluid
I e.g. quartz veins, liesegang
structures
I multiple, overprinting veins
indicates long-lived fluid flow
I can also be inferred by the
effect of fluid on the rocks it
passes through
I fluid can leach elements
what evidence do we see of fluid flow?
I certain types of seismic activity
seem to be correlated with
large-scale dehydration reactions
I dehydration of subducting slab
is focused at lawsonite- and
glaucophane-chlorite breakdown 0.5
20

Cascadia
0.6

Shikoku
I coincides with the occurrence of 0.7
non-volcanic tremor 0.8
30
0.9

Pressure (GPa)
I could this be large amounts of

Depth (km)
1
freshly-generated fluids 1.1
fracturing their way out? 40
1.2
1.3
1.4
1.5 50
1.6
1.7
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
Fluid released (mol fraction)
the effect of fluid on metamorphism

I if no water is added along 700

g st
AFM (+ mu + q)
the retrograde path, the bi
sill
bi
sill 6 kbar
675 H2O H2O sill + g + H2O
higher-grade assemblage T (°C) st + bi

will be preserved as the 650

H2O saturation line

rock cools 625
g
bi
g
st
sill bi
sill st
I this is why we see 600 st bi
st
chl chl chl
bi H 2O bi
sill bi H2O
high-grade peak H 2O H 2O
575
assemblages preserved in st
bi
550
rocks at the Earth’s st
chl
bi bi
surface 525
chl

H2O in minerals
I if water is added, the 500
40 50 60 70 80 90 100
M
high-grade assemblage H2O
Prograde path taken by rock
can be retrogressed to a Retrograde path taken by rock
H2O saturation line
lower-grade one
asymmetry
700
g st
bi bi AFM (+ mu + q)
sill sill 6 kbar
675 H2O H2O sill + g + H2O
st + bi
T (°C)
I case 1: heating to peak with fluid loss, 650

then cooling → peak assemblage 625

g g
H2O saturation line

bi st
preserved sill bi
sill st st
600 st bi chl
chl chl
bi H 2O bi
sill bi H2O
I case 2: cooling with high-T fluid addition 575
H 2O H 2O

→ retrogressed to lower-grade assemblage st

bi
550
st
chl
I case 3: cooling with low-T fluid addition 525
bi bi
chl

→ retrogressed to lower-grade assemblage H2O in minerals

500
40 50 60 70 80 90 100
M
H2 O
Prograde path taken by rock
Retrograde path taken by rock
H2O saturation line
implications of fluid loss

28 26 24 22 20 18 16 14 12

I the “peak” assemblage preserved in 10

subsolidus rocks represent the P–T 8

conditions of minimum water Peak P-T 6%

content
Min H2O
I fortunately, the P-T of this
minimum water content commonly
coincides, or is close to, the peak 14 12 10 8 6%
P–T conditions. 16

I H O contours are usually 18 Min H2O

2
20
steep 22
I and decrease up T 24 Peak P-T
26
I but this generalisation is not always 28
true: e.g.
complex contours: 3-D

(a)
21 (b)
o gl hb law
P (kbar)

100
o
water
g o gl hb law
gl g
20 hb o
B g o gl hb content
law gl 95
hb B
law C
A
b gl
19 law cz h
90
hb gl
law cz

85 g hb gl cz
g
18 o g pa gl hb cz g o gl h
b cz
gl
80
hb
l hb
cz g o pa g C
hb
17 550 g o gl
g o pa gl hb
g o pa hb
T(°C) 20
600 P (kbar)
g hb gl cz 18
16
g o pa hb
650 16
A
520 540 560 580 600 620 640
T(°C)
retrogression

I in many cases water is added back into the system as the rock
is exhumed back to the surface after peak metamorphism
I this causes retrogression
I usually the late fluid is restricted over time and space

I time: fluid addition is a specific

ssion
event - does not occur continuously
during retrograde history

ogre

it
I space: unlike the prograde fluid,

ndu
l retr

retr tense :
2O

ion
the retrograde fluid is not

id co
ima

of H

ess
pervasively distributed

/ flu
: min

ogr
lots
in

ture
I it is externally sourced, so is often

H2O

frac
channelled little
I rocks close to fluid conduit are
strongly retrogressed, but rocks
further away are unaffected
polymetamorphism

I rocks that have been through a metamorphic cycle will be

dehydrated
I and remain dry unless fluid is added from an external source
I if we metamorphose such rocks a second time, very little will
happen to them because
I the rock will remain fluid-absent
I old peak metamorphic assemblage will remain stable & the
rock will not grow new equilibrium minerals
I mineral compositions will be sluggish to equilibrate
I unless
I the rock is forced to recrystallise & re-equilibrate through e.g.
deformation
I or until P–T conditions of the second metamorphic event
exceed those of the first
I at this point the rock will become fluid-bearing and all the
usual prograde processes will operate again
 polymetamorphism
14

prograde to peak M1 retrograde 12

M1
14 minimum H2O retrograde
14
g chl10 g bi H2O
g bi H2O
g chl H2O 12
H2O
g chl H 2O
12 chl
g chl g8
chl g chl bi
g bi
10 10 g ky bi H2O
bi g ky bi H2O
6

P (kbar)
chl H2O
P (kbar)

chl H2O st g st g
8 8 bi H2O
bi H2O g chl bi
4
g bi g bi
sill H2O O sill H2O
6 st bi H2O 6 i H2
st b 2
sill bi H2O chl bi sill bi H2O
520
4 4
chl bi H2O chl bi H2O
and bi H2O and bi H2O
525 550 575 600 625 650 675 700 525 550 575 600 625 650 675 700
T (°C) T (°C)
 polymetamorphism

M2 lower-grade than M1 M2 higher-grade than M1

14 14
g chl g bi H2O g chl g bi H2O
12
H2O
g chl H 2O 12
H2O
g chl H 2O
chl chl
g chl g g chl g
bi bi
10 g ky bi H2O 10 g ky bi H2O
P (kbar)

P (kbar)
chl H2O st g chl H2O st g
8 bi H2O 8 bi H2O
g chl bi g chl bi
g bi g bi
O sill H2O O sill H2O
6 i H2 6 i H2
st b st b
chl bi sill bi H2O chl bi sill bi H2O
4 4
chl bi H2O chl bi H2O
and bi H2O and bi H2O
525 550 575 600 625 650 675 700 525 550 575 600 625 650 675 700
T (°C) T (°C)
polymetamorphism without H2 O addition

I initial peak metamorphism

formed an assemblage of
opx–sill–bi–q
I subsequent metamorphism
occurred without fluid addition,
and at much lower P–T
I all that happened was that the
old peak metamorphic minerals
were recrystallised and
re-equilibrated
 polymetamorphism without H2 O addition
(a) NCKFMASHTO (+ q + ru) S009
10
1 0.8

0.6
P (kbar) @ 850 oC

opx cd sill bi liq

opx sill bi pl liq
11
9
opx cd sill bi pl liq
opx cd
10
bi liq

P (kbar)
0.4
8 M1
9 1.7 Ga
0.2 opx cd
7.5; 825 opx cd bi pl liq
sill bi pl opx cd

ky
8

sill
bi pl
800
0.2 opx ged cd bi pl liq 2
7 M2
775 0.6
1.65 Ga
0.8
opx sill ged cd
bi ksp pl opx ged 6
750 bi liq
T (oC) @ 7 kbar

cd bi pl ged cd
bi pl liq
725 0.4 5
ged cd bi pl
opx sill bi pl

3
opx ged cd sill bi pl and
700 4 4
7 400 500 600 700 800 900
5 ged cd bi pl H2O T (oC)
675 sill
ky
opx ged ky bi pl 8 6
9
650
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
MH2O
another example: garnet coronae in mafic rocks

I many isobarically-cooled mafic

rocks contain garnet coronae
I garnet was not part of the
prograde or peak assemblage
I only introduced during
retrogression
I the loss of H2 O during prograde
metamorphism dehydrated the
rock, allowing garnet to be
stabilised during cooling
another example: garnet coronae in mafic rocks
(a) NCFMASHTO (+ pl + q + ilm) 515161 @ 7.5 kbar
900

opx di hb mt H2O [q]

opx di hb mt [q]
opx di hb mt H2O
850
opx di hb H2O b
di h opx di hb mt
opx

g
op
800

xd
ih
b
di hb H2O
T (oC)

g opx di hb mt

ne
750

tion li g di hb
satura

700
water

gd
di hb sph H2O

ih
bm
650 di hb ep

t
h

sph H2O
sp

[ilm]
ep

g di
hb

hb ep
di

600
5.62 5.07 4.52 3.96 3.41 2.86 2.31 1.76 1.20 0.65 0.1
MH2O (mol. %)
another example: garnet coronae in mafic rocks
prograde–to–peak M (mol. %)
peak–to–retrograde
H2O

(b) NCFMASHTO (+ q + pl + H2O) (c) NCFMASHTO (+ q + pl) @ MH2O = 3.18

13 13
di hb ep g di g di
ru [pl] hb ru hb ilm ru g di hb ilm ru
12 di hb ep ru 12

b ru

ru
11 11

di h

ilm
di hb

hb
ep sph lm
bi
h

di
10 10 ih
b sp

gd di
hb
di h

ep lm hb
9 sph 9 hbi ilm
di

P (kbar)
P (kbar)

px
go
8 8
ph

di hb ilm
ilm s

7 7

ne
di hb

opx di hb ilm

tion li
h
sp

di
hb

6 6

atura
hb
ilm
b ilm

rs
H 2O
5

wate
5
di h
opx

4 4
600 700 800 900 600 700 800 900
T (oC) T (oC)
yet another example: eclogitisation of gabbro
I patchy replacement of metagabbro by eclogite has been
suggested as due to fluid infiltration
I gabbro is dry & retains metastable igneous assemblage during
subduction
I fluid infiltration at depth causes direct transformation of
gabbro to eclogite
I HOWEVER, eclogite is still dry → H2 O is only a catalyst for
the transformation, not a reactant
yet another example: eclogitisation of gabbro
(a) NCKFMASHTO (+ hb + q) @ 650°C hb-eclogite, sample GB1 (b) @M(H2O) = 3.0
SiO2: 52.84 Al2O3: 9.13 CaO: 13.95 MgO: 11.73 FeO: 8.24 K2O: 0.04 Na2O: 2.14 TiO2: 1.11 O: 0.83 1.0
20 ilm q H 2O
opx

Modal abundance
o g ep ru H2O ru 0.8 sph ru
19 ep o g ep 5 pl g

g-amphibolite
g

hb ‘granulite’
18
0.35 0.8 o ky ksp
d 0.6
dio
ru
17 0.6 0.4
o
0.30 o g ep
ru 0.4
16 dio g ep ru H2O ep ky ru ep ru 0.2 hb
g g
o
15 0.25 dio di sp ep
k
14 ky 2 5 10 15 20
0.2 4
0.20 Pressure [kbar]
13 1
Pressure [kbar]

2
6

g h (c) @M(H2O) = 0.5

12 o p dio g pl ep ru 3
9 di p s 1.0 q
11 e
opx ru
dio g pl

Modal abundance
sph ru
0.8

gabbro / granulite
dio g pl ep sph ilm + mt ky
10 g

true eclogite
dio g pl ru 0.6
9 pl
dio g
dio pl pl sph 0.4
8 10 8
sph H2O 7
7 0.2 dio
11 dio g pl ilm hb ep o
dio g opx pl ilm
6
2 5 10 15 20
5 12 Pressure [kbar]
dio pl
4 dio pl sph H2O

dio opx pl
sph ilm (d) @18 kbar

ilm mt
b c 1.0 q
3 dio opx pl ilm H 2O ky

Modal abundance
2 0.8 ru g ksp

hb-cz ‘eclogite’

true eclogite
0.6

hb eclogite
5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 dio
0.4
M(H2O) [mol %] o
hb

dio
0.2
1 hb dio g pl ep ky ru q 5 dio g ep ky ksp ru q 9 hb dio ep sph q H2O
2 hb dio g pl ky ksp ru q 6 hb dio g ep sph ru q 10 hb dio pl sph q ep
3 hb dio g pl ky ru q 7 hb dio g pl sph ilm q 11 hb dio pl sph q 5 4 3 2 1 0
4 hb dio g ky ksp ru q 8 hb dio g pl ilm ru q 12 hb dio pl ilm q M(H2O) [mol%]