Calculated greenschist facies mineral equilibria in the system CaO?FeO?MgO?Al2O3?SiO2?CO2?H2O

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/225239901

Calculated greenschist facies mineral equilibria in the system CaO−FeO−MgO −Al2O3−SiO2−CO2−H2O ARTICLE in CONTRIBUTIONS TO MINERALOGY AND PETROLOGY · JANUARY 1990 Impact Factor: 3.48 · DOI: 10.1007/BF00321490

CITATIONS

READS

33

45

4 AUTHORS, INCLUDING: Thomas M Will

Michel Guiraud

University of Wuerzburg

Muséum National d'Histoire Naturelle

53 PUBLICATIONS 1,181 CITATIONS

40 PUBLICATIONS 1,301 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.

SEE PROFILE

Available from: Thomas M Will Retrieved on: 04 February 2016

Contributions to Mineralogy and Petrology

Contrib Mineral Petrol (1990) 104:353-368

9 Springer-Verlag1990

C a l c u l a t e d g r e e n s c h i s t f a c i e s m i n e r a l e q u i l i b r i a in t h e s y s t e m CaO - FeO - MgO - A1203 - SiO2 - CO2 - H20 Thomas M. Will ~, Roger PowelP, Tim Holland 2, and Michel Guiraud ~ * 1 Department of Geology, University of Melbourne, Parkville, Victoria 3052, Australia 2 Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EQ, England, United Kingdom

Abstract. The kinetic problems associated with the experimental determination of reactions among complex solidsolution phases at low temperatures have hindered our understanding of the phase relations in greenschist facies rocks. In the absence of reliable experimental data, we have used the new, expanded internally-consistent thermodynamic dataset of Holland and Powell (1990), to present calculated phase equilibria for the system C a O - F e O MgO-A1203- SIO2-H20-CO2 (CaFMASCH) with quartz in excess, in the range 4000-500 ~ C at low to intermediate pressures, involving the minerals amphibole, chlorite, anorthite, clinozoisite, dolomite, chloritoid, garnet, margarite, andalusite, and calcite. By solving independent sets of non-linear equations formed from equilibrium relationships, we calculate not only the loci of reactions in pressuretemperature-x(CO2) space, but also the compositions of coexisting minerals in terms of the substitutions, FeMg_ 1 and (Fe,Mg)SiAI_tAI_~. Invariant, univariant and divariant equilibria are calculated and discussed in relation to naturally-occurring greenschist fades metabasic and siliceous dolomitic mineral assemblages. We thus avoid the use of activity-corrected curves so commonly presented in the literature as a substitute for genuine univariant phase diagram boundaries.

Introduction

Large areas of greenschist facies rocks of both Proterozoic and Phanerozoic age occur in almost all orogenic belts around the world. Although there is a close similarity in mineral assemblages, only few quantitative approaches have been attempted to unravel the conditions of formation of the common assemblages because: (i) the fine-grained nature of the minerals under greenschist facies conditions makes the rocks difficult to study, (ii) "green" greenschist facies rocks seem to be unattractive and less fascinating than high grade rocks to many metamorphic petrologists (a review of the literature proves this point), (iii) the minerals involved show complex solid solutions and are commonly involved in relatively high variance assemblages. As a consequence, direct experimental study is there* Present address: Laboratoire de Minrralogie, Museum National d'Histoire Naturelle, 61, rue Buffon, 75005 Paris, France Offprint requests to: T. Will

fore more or less impossible, particularly with the pronounced kinetic problems at such (experimentally) low temperatures. Moreover, with previously available datasets, insufficient thermodynamic data for end-members of the usual minerals were available to say much about the mineral equilibria. For example, metabasic rocks metamorphosed at low to intermediate pressures and temperatures display the common mineral association albitic plagioclase (ab) - actinolitic hornblende (amph) - clinozoisite/epidote (cz/ep) - chlorite (chl) - quartz (q) and very often calcite (cc), with minor muscovite (mu), biotite (b0, dolomite (dol), talc (ta), stilpnomelane (stilp), iron oxides and sulphides in various combinations. With such an assemblage occurring over a wide range of pressure and temperature, there is much information to be gained from mineral compositional variations, particularly FeMg_ 1 and (Mg,Fe)Si AI_ 1AI_ 1, if the mineral equilibria are well known. (FeMg_t and (Mg,Fe)Si AI_ 1AI_ ~ are exchange vectors in the sense of Thompson et al. 1982). Although some essentially qualitative work has been done (e.g., Brown 1977; Laird and Albee 1981), little is known about the dependence of mineral chemistry on pressure, temperature and fluid composition, nor even the stability limits of the mineral assemblages, let alone the quantitative locations of invariant points, univariant reactions, and divariant fields on P-T diagrams, or, more appropriately at low-P, low-T, fluid-present conditions, on isobaric T-x(CO2) projections. Qualitative attempts at this, based on field studies, for phase relations among some minerals forming in metabasic rocks at greenschist facies conditions have been made in the past (e.g. Harte and Graham 1975; Brown 1975) but were restricted to simplified systems and did not include consideration of the solid solutions which are normally encountered in metabasites and are an essential feature of these rocks. Previously, petrogenetic grids calculated from thermodynamic data were restricted to systems involving phases composed of single end-members (Bucher-Nurminen et al. 1983 for margarite-bearing assemblages; Rice 1983 for rodingites; K/ise and Metz 1980; Metz and Trommsdorff 1968 for siliceous dolomites; Watts 1973, 1974; Barron and Barron 1976 for metabasites). That no solid solutions were able to be included when the petrogenetic grids were constructed, limits the applicability of these grids to naturallyoccuring assemblages. No quantitative study has been attempted so far to construct a petrogenetic grid that is consistent in itself and covers not only most of the phases

354 relevant to rocks subjected to greenschist facies m e t a m o r phism but also tries to establish the dependence o f mineral chemistry on pressure, temperature and fluid composition. In other systems, for example Spear and Cheney (1989) on metapelites, limited experimental constraints on mineral equilibria have been combined with observations on rocks in order to generate the necessary t h e r m o d y n a m i c d a t a to calculate grids. Naturally, the resulting grid is consistent with geological observation, but there is the danger that the features ' d e s c r i b e d ' by the grid are due to elements other than those in the system for which the grid is calculated (see discussion in Powell and H o l l a n d 1989). Others have tried to generate grids using individual reactions involving end-members o f minerals and an 'activity-corrected' a p p r o a c h ; this m a y allow an estimate o f the PT at which a particular mineral assemblage is stable, but it is not possible to calculate the position of reactions along which the compositions of minerals change with this approach. It is much m o r e satisfactory to use an internallyconsistent thermodynamic datatset, for example the extensire one of H o l l a n d and Powell (1990), from which all the equilibria of interest can be calculated, without recourse to any a priori geological data. This is the a p p r o a c h followed here. This study focusses on phase relationships in the system CaO- FeO- MgO-A1203- SiOz- C02- H20 (CaF M A S C H ) , with quartz and a H 2 0 - C 0 2 fluid in excess, which is applicable to, for example, siliceous dolomites and metabasic rocks. C a F M A S C H can be considered as a subsystem o f C a N K F M A S C H T O (CaO-Na20-K20F e O -- M g O - A1203 - SiO 2 - - C O 2 - H 2 0 -- Ti02 -- 02) which would cover the greatest majority o f phases normally encountered in greenschist facies rocks. The minerals included, and their c o m p o n e n t end-members, are given in Table 1. Using the expanded internally-consistent thermodynamic dataset o f H o l l a n d and Powell (1990), it is now feasible to calculate the P-T-x location o f invariant points, univariant reactions, and divariant fields on petrogenetic grids, using the computer p r o g r a m T H E R M O C A L C (Powell and H o l l a n d 1988). The results o f such a study are presented here for the C a F M A S C H system. P-T-x pseudosections (Hensen 1971) are presented, and used to consider the influence o f bulk rock compositions on mineral assemblages, as well as to generate internal and external buffering paths on isobaric T-x(CO2) diagrams. Furthermore, the extent o f F e M g _ 1, C a F e _ 1 (Fe,Mg)SiAI_ 1A1-1 substitutions in the minerals are related to changes in intensive parameters.

The grid A 2 kbar isobaric T-x(CO2) section of CaFMASCH with excess quartz has been calculated in a temperature window from 4000-500 ~ C, Fig. 1 and Table 2. Fluid pressure has been assumed to equal total pressure. This temperature window was chosen because it encompasses the reactions relevant to greenschist facies conditions at low pressures, and the transition into amphibolite conditions if the first appearance of almandine-rich garnet is taken to be indicative. As Fig. 1 is complex, initial discussion of the grid will be in terms of phase relationships with calcite also in excess, Fig. 2. This simplifies the grid considerably and, significantly, makes the system effectively ternary, thus allowing comprehensible compatibility diagrams to be drawn. Moreover the system is a good approximation for many metabasites and low grade siliceous dolomites under greenschist facies conditions.

Table i. Mineral end-members and their compositions as used in text and figures. The abbreviations of the mineral end-members are those used in Holland and Powell (1990) amph tr fir hb

fhb

amphibole tremolite ferro-tremolite hornblende ferro-hornblende

Ca2Mg3Mg2Si4Si4022(OH)2 CazFe3Fe2Si4Si4022(OH)2 Ca2Mga[MgxAll][Si3AI~] Si,*O22(OH)2 Ca2Fe3[FelAll][Si3All]

Si4022(OH)2 chl clin

chlorite clinochlore

daph

daphnite

ames fame

amesite ferro-amesite

Mg4[MglAll][Si,All] Si201o(OH)s Fe4[Fe~AI~][SilAll] Si2Olo(OH)s Mg4[AU[A12]Si201 o(OU)s Fe4[A12][A12]Si2Olo(OH)s

ctd mctd fctd

chloritoid magnesium-chloritoid ferro-chloritoid

MgA12SiOs(OH)2 FeA12SiOs(OH)2

dolomite dolomite ferro-dolomite

CaMg(C03)2 CaFe(C03)2

talc talc ferro-talc Tschermak's talc ferro-Tschermak's talc

Mg2Mg[Si2]Si201 o(OH)2 Fe2Fe[Si2]Si20~o(OH) 2 Mg2AI[Si~AI~]Si201 o(OH)2 Fe2AI[SilAlllSi201 o(OH)2

garnet almandine grossular

FeaAI2Si3012 Ca3A12Si3012

epidote epidote clinozoisite

Ca2A12[Fe3+]Si3Olz(OH) Ca2A12[A1]Si3012(OH)

andalusite margarite anorthite albite calcite quartz wollastonite phyrophyllite prehnite diopside forsterite

A12SiO5 CaAlz[A12Si2]O10(OH)2 Ca[A12Si2]O8 Na[A1Si3]Os CaC03 SiO2 CaSiO3 A12Si4Olo(OH)2 CaA12Si3Olo(OH)2 CaMgSi206 Mg2SiO4

do! do! fdol ta ta fta tats ftat

g alm gr el; ep cz and ma an ab ce

q wo pyhl pre di fo

In order to ensure the validity of the grid, and especially its validity at the low and high temperature boundaries of the temperature window, phases additional to those appearing in the grid have been considered. These phases are wollastonite (wo), pyrophyllite (phyl), diopside (di), and prehnite (pre). In projection from quartz and calcite, the stability field of prehnite is confined to equilibrium with a very H20-rich fluid at very low temperatures. The limiting reactions for the stability of prehnite are pre + C02 = cz + cc + q + H20 and p h i l + cc =pre + 1 + C02. The stability of pyrophyllite is limited by the latter reaction and by phyl + cc = cz + q + H2 0 + C02 at very low x(CO2), p h y l + c c = m a + q + C O a + H 2 0 at x(CO2) values up to 0.7, and by p h y l = a n d + q + H 2 0 at an even higher x(CO2). At intermediate fluid compositions pyrophyllite ( + quartz, +calcite) is not stable at temperatures higher than 330~ and is restricted to even lower temperatures at low or high values of

355

T(~ ~\

500

9~ 4 9 0 I-

I

\

\..

and '

~

\'

\ ' ~

~

\

~

3 '~ \

~

\

\

CaFMASCH / at fixed P = 2 k b a r

oh, \='Zl

+q

480

470

460

cz

450

cc

k I ~176

440

Idol all

430

420 IC~'~/~ 1 / " 0f I ~g~l I

/// / /~da II I , , /

~8 ] I/

n ....

~

/

m~ ~

,2

410

c.,~o

/

400 |1 /

I/Ill

/ 0.1

+ cC

t/ 0.2

I 0.3

i 0.4

R

doI

/[/ 0.5

I

I 0.6

III 0.7

i i i ~ , . . . . . . . . . II,~,l 0.8 0.9

x(co2) Fig. 1. T-x(COz) petrogenetic grid for the system C a O - F e O - M g O - - A 1 2 0 3 - S i O 2 - C O 2 - H20 at 2 kbar projected from quartz. Bold lines are C a F M A S C H reactions, except for the reaction e t d = g + a n d + c h l + H 2 0 (+quartz) which is in F M A S C H ; light lines are CaFASCH and CaMASCH reactions; CMSCH and CFSCH reactions are given by light lines with the phases in italics. Reactions involving f d o l in CaFASCH are metastable with respect to reactions involving calcite and siderite. For labelling of invariant points see Table 2; for the abbreviations of the phases, Table 1. Tremolite labelled in italics as tr is the pure Mg-endmember, whereas tr (along the C a F M S C H reaction dol + q + H 2 0 = ta + tr + C 0 2 ) is an ( F e - Mg)-tremolite

356 Table 2. P-T-x(CO2) locations and mineral compositions of (T-x(CO2) invariant points, x, y, and z are XFe= Fe/(Fe+ Mg) in amph, chl, dol, ctd, ta, XAI,M2 in amph, chl, ta, and F e / ( F e + C a ) in garnet, respectively; for more details see Table 3. The numbers 1-11, ml-m8, and fl-fl7 correspond to those given in the Figures. The list of phases following the invariant point label are the phases not involved at the invariant point

CaFMASCH 1

(amph, ctd, chl, dol, ta, cz, g) P (kbar) 2.0 5.0

T (~ 398 496

x (CO2) 0.82 0.717

(amph, ta, and, g, cz) P (kbar) 2.0 5.0

T (~ x (CO2) 409 0.59 metastable

x (ctd) 0.977

x (chl) 0.895

y (chl) 0.759

x (dol) 0.888

x (CO2) 0.74 0.61

x (ctd) 0.918 0.836

x (chl) 0.692 0.554

y (chl) 0.775 0.720

x (dol) 0.678 0.545

x (CO2) 0.82 0.625

x (ctd) 0.949 0.841

x (chl) 0.794 0.564

y (chl) 0.727 0.716

z (g) 0.902 0.840

x (ctd)

x (chl)

y (chl)

x (dol)

0.945

0.800

0.698

0.793

x (ctd)

x (chl)

y (chl)

z (g)

x (dol)

0.959

0.848

0.689

0.761

0.842

x (ctd)

z (g)

x (dol)

0.964

0.766

0.861

x (chl)

y (chl)

z (g)

x (dol)

0.736

0.661

0.740

0.729

x (amph)

y (amph)

x (chl)

y (chl)

z (g)

0.715

0.160

0.683

0.628

0.713

x (amph)

y (amph)

x (chl)

y (chl)

z (g)

x (dol)

0.692

0.161

0.659

0.629

0.725

0.652

y (amph) 0.0827

x (ta) 0.039

y (ta) 0.054

x (chl) 0.060

y (chl) 0.460

(amph, ta, g, cz, cc) P (kbar) 2.0 5.0

T (~ 412 515

(amph, ta, ma, cz, cc) P (kbar) 2.0 5.0

T (~ 431 517

x (dol) 0.785 0.556

(amph, ctd, dol, chl, ta, and, g) P (kbar) 2.0 5.0

T (~ 391 499

x (CO2) 0.06 0.308

(amph, ta, and, g, an) P (kbar) 2.0 5.0

T (~ x (CO2) metastable 490 0.225

(amph, ta, an, ma, and) P (kbar) 2.0 5.0

r (~ x (CO2) metastable 495 0.24

(amph, ta, chl, an, and) P (kbar) 2.0 5.0

T (~ x (CO2) metastable 498 0.295

(amph, ctd, ta, ma, and) P (kbar) 2.0 5.0 10

(dol, ctd, ta, ma, and) P (kbar) 2.0 5.0

11

T (~ x (CO2) metastable 516 0.235

(ctd, ta, ma, and, cz) P (kbar) 2.0 5.0

12

T (~ x (CO2) metastable 508 0.268

T (~ x (CO2) metastable 519 0.29

(ctd, g, ma, and, cz) {for activity of anorthite = 0.40} P (kbar) 2.0

T (~ 451

x (CO2) 0.58

x (amph) 0.0728

x (dol) 0.058

357 Table 2 (continued) CaFASCH fl (fta, fctd, fdol, and, ma, g) P (kbar) T (~ x (COz) 2.0 419 0.049 5.0 metastable f2

f3

(amph, fta, P (kbar) 2.0 5.0

f6

f7

f8

t"12

f13

y (amph) 0.133

y (chl) 0.614

x (CO2) 0.770 0.673

T (~ x (CO2) 424 0.83 metastable

(fta, amph, P (kbar) 2.0 5.0

and, ma, cz, cc) T (~ x (CO2) 431 0.75 metastable

(fta, fctd, and, ma, cz, cc) P (kbar) T (~ x (CO2) 2.0 450 0.63 5.0 metastable

z (g) 0.900

y (chl) 0.712

z (g) 0.879

y (amph) 0.135

y (ch0 0.617

y (amph) 0.128

z (g) 0.808

(fta, fctd, chl, and, ma, cz) T (~ x (CO2) 458 0.70 metastable

(amph, fta, fdol, and, g, cz) T (~ x (CO2) 410 0.41 metastable

y (chl) 0.753

(amph, fta, chl, and, g, cz) P (kbar) 2.0 5.0

fll

T (~ 406 505

P (kbar) 2.0 5.0

P (kbar) 2.0 5.0 fl0

r (~ x (CO2) 448 0.61 metastable

(amph, chl, fta, ma, cz, cc)

P (kbar) 2.0 5.0 f9

y (chl) 0.742

(ha, amph, chl, g, cz, cc) P (kbar) 2.0 5.0

f5

y (chl) 0.580

(fctd, fta, and, ma, g, cz) P (kbar) 2.0 5.0

f4

and, ma, g, cz) T (~ x (CO2) 413 0.600 metastable

y (amph) 0.115

T (~ x (CO2) 409 0.66 metastable

(amph, fctd, fta, chl, and, cz) P (kbar) 2.0 5.0

T (~ x (CO2) metastable 501 0.51

z (g)

(amph, fta, P (kbar) 2.0 5.0

chl, and, an, cz) T (~ x (COz) metastable 496 0.43

z (g)

(amph, fdol, fta, chl, and, an) P (kbar) T (~ x (CO2) 2.0 metastable 5.0 495 0.265

z (g)

0.793

0.791

0.766

z (g) 0.817

358 Table 2 (continued) f14

(fctd, fta, and, ma, an, cz) P (kbar) 2.0 5,0

f15

y (chl)

z (g)

0.109

0.537

0.791

r (~ x (CO2) metastable 484 0.18

y (chl)

z (g)

0.670

0.785

y (chl)

z (g)

0.672

0.747

y (amph)

y (chl)

z (g)

0.127

0.579

0.671

y (amph) 0.106 0.152

y (ch0 0.533 0.602

y (amph) 0.107 0.086

y (ta) 0.0705 0.0605

y (amph) 0.125 0.155

y (chl) 0.567 0.606

y (amph) 0.127 0.161

y (ta) 0.0845 0.123

(amph, fdol, fta, and, ma, an) P (kbar) 2.0 5.0

1"17

y (amph)

(amph, fta, and, ma, an, cz) P (kbar) 2.0 5.0

f16

T (~ x (CO2) metastable 495 0.15

T (~ x (CO2) metastable 482 0.12

(fdol, fctd, fta, and, ma, an) P (kbar) 2.0 5.0

T (~ x (COz) metastable 479 0.056

CaMASCH ml (ta, mctd, dol, and, ma, g) P (kbar) 2.0 5.0 m2

m6

m7

T (~ 461 525

x (CO2) 0.600 0.295

T (~ 463 536

x (CO2) 0.63 0.45

(mctd, amph, ta, g, cz, co) P (kbar) T (~ x (CO2) 2.0 413 0.735 5.0 514 0.608 (amph, mctd, ta, and, g, cz) P (kbar) r (~162 x (CO2) 2.0 410 0.555 5.0 rnetastable

y (chl) 0.775 0.721

y (chl) 0.757

(amph, mctd, ta, and, g, ma) P (kbar) 2.0 5.0

m8

x (CO2) 0.56 0.12

y (chl) 0.527 0.451

(mctd, ma, and, g, cz, cc) P (kbar) 2.0 5.0

m5

T(~ 457 496

(mctd, ta, ma, and, g, cz) P (kbar) 2.0 5.0

m4

x (CO2) 0.046 0.21

(mctd, an, ma, and, g, cz) P (kbar) 2.0 5.0

m3

T (~ 427 520

T (~ x (CO2) metastable 511 0.257

(amph, mctd, ta, and, g, an) P (kbar) T (~ x (CO2) 2.0 metastable 5.0 485 0.185

CaMSCH m' (mctd, clin, P (kbar) 2.0 5.0

an, ma, and, g, cz) T (~ x (CO2) 463 0.59 503 0.128

y (chl) 0.649

y (chl) 0.700

y (chl) 0.571 0.614

359

T(~ CaFMASCH

5O0

at

fixed

P

=

2 kbar

+q

490

+CC

/

480

,m3

',\ _/

\

470

m2

11,

460

450

440

430

420

410

m6 c h~h~t d m/~

~ct fdol

400

L 0.1

0.2

0.3

0.4

0.5

0.6

i

I 0.7

0.8

0.9

x(CO2) Fig. 2. T-x(CO2) petrogenetic grid at 2 kbar as projected from quartz and calcite. The traverse is discussed in the text

x(CO2). Wollastonite is stable in the window but is restricted to fluid compositions of x(CO2)