Some considerations on the thermal stability of the calcium-iron

amphibole ferroactinolite (Ca2Fe5Si8O22(OH)2)

Jordi Delgado Martín

Albert Soler i Gil
David Arcos Bosch

Departament de Cristal.lografía, Mineralogía i Dipòsits Minerals

Universitat de Barcelona
Martí i Franqués s/n, 08028 Barcelona, Spain
ph# 343-4021342 fax# 343-4021340 e_mail: UBACMD04@EBCESCA1.BITNET

Abstract

Ferroactinolite is an important phase belonging to the Ca-Fe-Si-O-H system. Since

late sixties, only one experimental dataset constraining its stability within the P-T-fO2
volume is available (Ernst, 1966). From this work, many authors have tried to retrieve the
thermodynamic parameters of this phase although the agreement among the different
estimations is quite poor. We have compiled the highest reliable volumetric properties of
ferroactinolite (Vo298=27.65 J/bar), á%(T)=10-3E(T-293.15)+10-7F(T-293.15),
2
E=3.079;F=0.614; ß%(P)=GP+HP , G=1.472; H=-8.51; T in K and P in kbar) and we have
calculated the high temperature heat capacity (Cp(T)=A+10-3BT+102CT-0.5+104T-2,
A=193.83; B=548.52; C=116.2; D=-295.18) between 298 and 1000 K, and the 3rd Law
entropy (So298=702.1 ±1.6 J/molK) from additivity rule estimations. This data coupled with
experimental P-T brackets and the thermodynamic parameters of magnetite, quartz and
fayalite allowed us to estimate the ÄGof,298 of this mineral (-9824.1 kJ/mol). There is no
agreement between experimental brackets and theoretical calculations and the estimated
ÄGof,298 is not well constrained. Therefore, the available experimental work appears to be
highly inaccurate due probably to the bad crystallochemical characterization of reactants and
run products. So, it would be highly desirable to perform new and better constrained
experiments in order to fully characterize this important field of the Ca-Fe-Si-O-H system
and the geological implications.

Introduction

Ferroactinolite is a calcic-ferric amphibole whose formula unit is

Ca2Fe52+Si8O22(OH)2. The maximum thermal stability of ferroactinolite is lower than the
calcium-magnesium amphibole tremolite (Ca2Mg5Si8O22(OH)2) and seems to be related with
the transition between the upper Greenschist and Amphibolite Facies metamorphism.
Pure ferroactinolitic compositions are rare due to the fact that most of the common
rocks do contain additional components. So deviations from ideal stoichiometry usually
imply changes in Fe2+, Ca and Si contents by substitution with Mn, Mg, Al and Fe3+ in 4 and
6-fould coordinated sites coupled with OH substitutions by F-, Cl- or O2- and partial filling of
A-sites by big cations (Na, K) to avoid global structural charge imbalance.
Ferroactinolite is an important mineral phase in the Ca-Fe-Si-H-O system and it is of
major importance to clearly stablish their thermodynamic parameters if we have to
characterize the topology of the low temperature edge of this system. Experimental work
dealing with ferroactinolite stability relations is limited to that of Ernst (1966) who studied
the dehydration reaction

2Ca2Fe5Si8O22(OH)2 = 3Fe2SiO4 + 4CaFeSi2O6 + 5SiO2 + 2H2O

Ferroactinolite Fayalite Hedenbergite Quartz

under different fluid pressure (PF), temperature (T) and oxygen fugacity (fO2) conditions.
From this work it was concluded that over the iron-magnetite (IM) and iron-wüstite (IW)
oxygen buffers ferroactinolite was stable to temperatures below 543 oC at PF=3 kbar, and for
fO2 over the nickel-bunsenite (NNO) buffer, ferroactinolite was any longer stable (Fig. 1).

Fig. 1. Stability relations for ferroactinolite. Taken from Ernst (1966)

Hellner and Shürmann (1966) studied also the maximum stability of actinolitic
amphiboles (Ca2(Fe2+,Mg)5Si8O22(OH)2) although in their experiments they didn't record
oxygen fugacity values.

Selection of the thermodynamic database

The thermodynamic data used in the retrieval calculations of this work together with
the corresponding sources are compiled in Table 1. The data have been selected according
to its reliability inferred from internal consistency tests from recent independent experimental
works (Anovitz and Essene, 1987; Haselton et al., 1987; Hemingway, 1987; Hemingway,
1990; Krupka et al., 1985; Moecher et al., 1988; Moecher and Chou, 1990).
There is a important lack of information regarding the thermochemical parameters of
ferroactinolite, so our first priority is to set this data (see Table 1 for numbers).

Molar volume: Ernst (1966, 1968) published volume cell data useful to calculate the
molar volume of ferroactinolite. Because some troubles exist with the amphibole volume cell
compilation of Ernst (1968), Ernst's (1966) data for the synthetic amphiboles used in his
experiments seems more reliable. As this material supposed to be 'on composition' Vegard's
Law corrections hasn't been applied.

Thermal expansion and compressibility factors: In the low P-T range, volume
corrections involving thermal expansion and compressibility factors may be disregarded
because they tend to cancel each other. Anyway, it seems appropriate to use them if we wish
to obtain accurate estimations for higher P-T conditions. There is no currently available data
on the variation of the cell parameters with changes in P and T and they have to be
approximated from the closest chemical-structurally posible mineral: tremolite.

Heat capacity and 3rd Law entropy: Measurements on the low or high temperature
heat capacity are not available for ferroactinolite. From additivity rule considerations
(Helgeson et al., 1978) it is posible to approximate the value of the entropy and a polynomial
expressing the variation of heat capacity with temperature. We have used entropy, heat
capacity and volume data from tremolite, hedenbergite and diopside. The results are shown
in Table 1.

Standard Gibbs free energy of formation (ÄGf,298): It is posible to calculate the

standard Gibbs free energy of formation of any mineral phase provided that a equilibrium
point for a reaction involving that mineral (i.e. a phase equilibrium point within an
experimental bracket) is known and the thermodynamic parameters for all but the problem
phase are perfectly stated. In our case, Ernst's (1966) experiments provide the equilibrium
point and we only need to solve the Gibbs-Duhem equation:

leaving as unknown
the ÄGf,298 of 0
ferroactinolite.
Water fugacities
were taken from the data of Burham et al. (1969). The resulting ÄGf,298 value appears to be
-9824.1 kJ/mol.

Discussion of results

In figure 2 we show the results of the calculations performed with the

thermodynamic data discussed in the previous section.

Fig. 2. Maximum thermal stability curves for ferroactinolite and ilvaite compared with Ernst's (1966)
experimental brackets. Curve 1 is calculated with the thermodynamic data from Table 1. Curve
3 is calculated from Holland and Powell's (1990) database. Curve 2 represents ilvaite maximum
stability (Delgado and Essene, in prep.)
 It is clear the bad agreement between theoretical calculations and experimental work.
There are potentially two posible explanations to justify the error: the thermodynamic data
used in the calculations is inadequate or the experimental brackets are poorly constrained. As
we discussed before, we have to be highly confident with the thermochemical data of
hedenbergite, fayalite, quartz and water as their consistency has been proved by independent
experiments and measurements. Ferroactinolite data, even questionable, has been extracted
according to methods that have been succesful for many other minerals (Helgeson et al.,
1978). So, if we assume that there is no problem with the thermodynamic data used, the
inconsistency has to be due to the experimental work.

Ernst (1966) performed a poor chemical characterization of the reactants and run
products as he only did x-ray diffraction patterns and reffractive index measurements. It is
posible that what he thought to be minerals 'on composition' actually could have been 'off
composition'. This problem may affect not only ferroactinolite but hedenbergite also.

Moreover, recent high resolution transmission electron microscopy (HRTEM and

TEM) studies have shown that synthetic amphiboles may have stacking fault defects
(Graham et al., 1989). Such a kind of uncertainties influence both entropy and enthalpy and,
in minor amount, volume of the amphibole being used in the experimental work.
Consequently, these efects should be contemplated in order to modify the shape and the
Clapeyron's slope of the equilibrium curve. It is not posible to evaluate the importance of
these effects over the experimental equilibrium brackets without inspection of the reactants
and run products. Hedenbergite activity correction is posible from the experimental work of
Lindsley and Munoz (1969). Nevertheless, the lowest temperature recorded in their
unreversed experiments is 600 oC. Extrapolation to lower temperatures is unjustified because
of the lack of constraint on the shape of the solvus between hedenbergite and ferrosilite at the
temperatures of our interest.

Another interesting point of reflection is the stability of ilvaite (Delgado and Essene,
in prep.). If we compare the experiment adjusted dehydration reaction of ilvaite with the
experimental brackets of ferroactinolite by Ernst (1966) it is stricking the almost full
agreement between curve and reversals (Fig. 2). Ilvaite stability was studied by Gustaffson
(1974) and Ghazi-Bayat et al. (1987) both them performing unseeded experiments. So it is
intriguing why ilvaite was not found during Ernst's (1966) experiments.

Comparison with other estimations

There have been previous estimations of the thermodynamic parameters of

ferroactinolite and its upper temperature stability. These estimations are recorded in Table 2.
Most of the estimations share the only experimental work available (Ernst, 1966) and in that
way they are limited by the same uncertainties. Holland and Powell (1990) offer
 thermodynamic data for ferrotremolite 1 based on exchange reactions between well
characterized P-T-x natural occurrences of two or more amphiboles. Holland and Powell's
data have been plotted in figure 2 and may be compared with our estimation and the
experimental brackets by Ernst (1966). The very high dehydration temperature for
ferroactinolite seem to us highly questionable, according to natural occurences.

Conclusions

Thermodynamic data extraction for ferroactinolite is seriously hampered by the

experimental data available. Therefore, thermodynamic data presented in our study must be
taken as highly preliminar. New and better constrained experiments on the stability of
ferroactinolite should be performed with the aid of HRTEM and TEM techniques as well as
conventional SEM-EDS and EPMA routinary analysis.

Acknowledgements

This study has been financed by the CICYT project GEO90-0507.

Bibliography

ANOVITZ, L.M. and ESSENE, E.J. (1987): 'Compatibility of geobarometers in the system
CaO-FeO-Al2O3-SiO2-TiO2 (CFAST): Implications for garnet mixing models' Jour. Geol.
95, 633-645

BABUSKA, V.; FIALA, J.; KUMAZAWA, M. and OHNO, I. (1978): 'Elastic properties
of garnet solid-solution series' Phys. Earth Plan. Int. 16, 157-176

BENNINGTON, K.O.; BEYER, R.P. and BROWN, R.R. (1984): 'Thermodynamic

properties of hedenbergite, a complex silicate of Ca, Fe, Mn and Mg' U.S. Bur. Mines Rept.
Inv. 8873, 19 pp.

BIRCH, F. (1966): 'Compressibility: elastic constants' in S.P. Clark (Ed.) Handbook of

physical constants, Geol. Soc. America Mem. 97, 97-174

BURHAM, C.W.; HOLLOWAY, J.R. and DAVIS, N.F. (1969): 'Thermodynamic

properties of water to 1000 °C and 10000 bars' Geol. Soc. Am. Spec. Paper 132, 96 pp.

CAMERON, M.; SUENO, S.; PREWITT, C.T. and PAPIKE, J.J. (1973): 'High
temperature crystal chemistry of acmite, diopside, hedenbergite, jadeite, spodumene and
ureyite' Am. Mineral. 58, 594-618

1
Ferroactinolite should be used instad ferrotremolite, according to the IMA guidelines (Leake, 1978).
CHASE, M.N.Jr., DAVIES, C.A., DOWNEY,J.R.Jr., FRURIP, D.J., MCDONALD,
R.A. and SYVERUD, A.N. (1985) 'JANAF Thermochemical tables', 3ª ed., J. Phys. Chem.
Ref. Data 14, suplem. 1 and 2, 1856 pp

COMODI, P.; MELLINI, M.; UNGARETTI, L. and ZANAZZI, P.F. (1991):

'Compressibility and high pressure structure refinement of tremolite, pargasite, and
glaucophane' Eur. Jour. Miner. 3, 485-499

ERNST, W.G. (1966): 'Synthesis and stability relations of ferrotremolite' Am. Jour. Sci 264,
37-65

ERNST, W.G. (1968) Amphiboles: Crystal chemistry, phase relations and occurrence,
Springer-Verlag, New York, 125 pp

FISHER, J.R. and ZEN, E-An (1971): 'Thermochemical calculations from hydrothermal
phase equilibrium data and the free energy of H2O' Am. Jour. Sci. 270, 297-314

GHAZI-BAYAT, B.; ANTHAUER, G.; SHÜRMANN, K. and HELLVER, E. (1987):

'Synthesis and characterization of the mixed-valence iron silicate ilvaite:
CaFe3[Si2O7/O/OH], Miner. Petrol. 37, 97-108

GRAHAM, C.M.; MARESCH, W.V.; WELCH, M.D. and PAWLEY, A.R. (1989):
'Experimental studies on amphiboles: A review with thermodynamic perspectives' Eur. Jour.
Miner. 1, 535-555

GUSTAFFSON, W.I. (1974): 'Stability of andradite and hedenbergite and related minerals
in the system Ca-Fe-Si-O-H' Jour. Petrol. 15, 455-496

HASELTON Jr., H.T.; ROBIE, R.A. and HEMINGWAY, B.S. (1987): 'Heat capacities
of synthetic hedenbergite, ferrobustamite and CaFeSi2O6 glass' Am. Mineral. 51, 2211-2217

HAZEN, R.M. and PREWITT, C.T. (1977): 'Effects of temperature and pressure on
interatomic distances in oxygen-based minerals' Am. Mineral. 62, 309-315

HELGESON, H.C.; DELANY, J.M.; NESBITT, H.W. and BIRD, D.K. (1978):
'Summary and critique of the thermodynamic properties of rock-forming minerals' Am. Jour.
Sci. 278-A, 1-229

HELLNER, E. and SCHÜRMANN, K. (1966): 'Stability of metamorphic amphiboles: The

tremolite-ferroactinolite series' Jour. Geol. 74, 322-331

HEMINGWAY, B.S. (1987): 'Quartz: Heat capacities from 340 to 1000 K and revised
values for the thermodynamic properties' Am. Mineral. 72, 273-279

HEMINGWAY, B.S. (1990): 'Thermodynamic properties for bunsenite, NiO, magnetite,

Fe3O4, and hematite, Fe2O3, with comments on selected oxygen buffer reactions' Am.
Mineral. 75, 781-790
HOLLAND, T.J.B and POWELL, R. (1990): 'An enlarged and updated internally
consistent thermodynamic dataset with uncertainties and correlations: The system
K2O-Na2O-CaO-MgO-MnO-FeO-Fe2O3-Al2O3-TiO2-SiO2-C-H2-O2' Jour. Met. Geol. 8,
89-124

JOHNSON, J.W. and NORTON, D.N. (1985): 'Theoretical prediction of hydrothermal

conditions and chemical equilibria during skarn formation in porphyry copper systems' Econ.
Geol. 80, 1797-1823

KRUPKA, K.M.; HEMINGWAY, B.S.; ROBIE, R.A. and KERRICK, D.M. (1985):
'High temperature heat capacities and derived thermodynamic properties of antophyllite,
diopside, dolomite, enstatite, bronzite, talc, tremolite and wollastonite' Am. Mineral. 70,
261-271

LEAKE, B.E. (1978): 'Nomenclature of amphiboles' Am. Mineral. 63, 1023-1052

LEVIEN, L. and PREWITT, C.T. (1981): 'High pressure structural study of diopside' Am.
Mineral. 66, 315-323

LINDSLAY, D.H. and MUNOZ, J.L. (1969): 'Subsolidus relations along the join
hedenbergite-ferrosilite' Am. Jour. Sci. 267-A, 295-324

MOECHER, D.P.; ESSENE, E.J. and ANOVITZ, L.M. (1988): 'Calculation and
application of clynopyroxene-garnet-plagioclase-quartz geobarometer' Cont. Min. Pet. 100,
92-106

MOECHER, D.P. and CHOU, I.M. (1990): 'Experimental investigation of andradite and
hedenbergite equilibria employing the hydrogen sensor technique, with revised estimates of
ÄfGom,298 for andradite and hedenbergite' Am. Mineral. 75, 1327-1341

ROBIE, R.A., HEMINGWAY, B.S. and FISHER, J.R. (1979) 'Thermodynamic

properties of minerals and related substances at 298.15 K and 1 bar (105 Pascals) pressure
and at higher temperatures', U.S. Geol. Surv. Bull. 1452, 456 pp.

ROBIE, R.A.; HEMINGWAY, B.S. and TAKEI, H. (1982): 'Heat capacities end
entropies of Mg2SiO4, Mn2SiO4 and Co2SiO4 between 5 and 380 K' Am. Mineral. 67,
470-482

ROBINSON, P.; SPEAR, F.S.; SHUMACHER, J.C.; LAIRD, J.; KLEIN, C.; EVANS,
B.W. and DOOLAN, B.L. (1982): 'Phase relations of metamorphic amphiboles: natural
occurrence and theory' in P.H. Ribbe (Ed.) Amphiboles: Petrology and experimental phase
relations, Reviews in Mineralogy 9b, 1-211

SKINNER, B.J. (1966): 'Thermal expansion' in S.P. Clark (Ed.) Handbook of physical
constants, Geol. Soc. America Mem. 97, 75-96

SUENO, S.; CAMERON, M.; PAPIKE, J.J. and PREWITT, C.T. (1973): 'The high
temperature crystal chemistry of tremolite' Am. Mineral. 58, 649-664
TAYLOR, D. (1984a): 'Thermal expansion data: I. Bynary oxides with the sodium chloride
and wurtzite structures, MO' British Cer. Trans. Jour. 83(2), 5-9

TAYLOR, D. (1984b): 'Thermal expansion data: III. Sesquioxides, M2O3, with the
corundum and the A-, B- and C-M2O3 structures' British Cer. Trans. Jour. 83(2), 92-98

TAYLOR, D. (1985a): 'Thermal expansion data: VI. Complex oxides AB2O4: The spinels'
British Cer. Trans. Jour. 84(5), 121-127

TAYLOR, D. (1985b): 'Thermal expansion data: VII. Complex oxides AB2O4: The spinels'
British Cer. Trans. Jour. 84(5), 149-153

TORNOS, F. (1989): 'Los skarns y mineralizaciones asociadas del Sistema Central Español.
Modelo de caracterización petrológica, geoquímica y metalogénica' Tesis Doctoral, Univ.
Complutense de Madrid, 487 pp.

UCHIDA, E. (1983): 'Grunerite from the Shinyama ore deposit, Kamaishi Mine, Japan'
Canad. Mineral. 21, 517-528

VAIDYA, S.N.; BAILEY, S.; PASTERNAK, T. and KENNEDY, G.C. (1973):

'Compressibility of fifteen minerals to 45 kilobars' Jour. Geoph. Res. 78, 6893-6898
Table 1. Thermodynamic data used in the calculations (So in J/molK; ÄGof,298 in kJ/mol; Vo in J/bar)

So298 Ref a b c d Range Ref ÄGof,298 Ref

Nickel (Ni) 29.87 (2) 7.276 8.796 6.194 -18.069 298-1726 K (2) - -

Oxygen (O2) 205.15 (2) 49.511 -1.073 -4.473 5.422 298-1800 K (2) - -

Hematite 87.40 (3) 258.258 116.588 -39.292 19.476 298-950 K (3) -744.27 (3)

Magnetite 146.14 (2) -3001.043 1624.250 591.291 -673.948 298-846 K (3) -1012.57 (3)

Bunsenite 37.99 (3) 148.227 21.684 -20.950 1.372 573-1800 K (3) -211.10 (3)

Quartz 41.46 (5) 97.920 -3.356 -7.745 -6.351 298-1400 K (5) -856.29 (5)

Forsterite 94.10 (6) 227.910 3.434 -17.430 -8.968 298-1800 K (2) -2133.00 (7)

Fayalite 152.13 (6) 58.015 71.155 20.576 -58.404 298-1400 K (3) -1377.10 (8)

Diopside 142.70 (15) 303.798 3.182 -17.788 -31.390 298-1600 K (15) -3025.00 (7)

Hedenbergite 173.59 (16) 310.459 12.570 -20.399 -18.460 298-1600 K (17) -2676.60 (10)

Wollastonite 81.69 (15) 163.657 -1.343 -11.685 -8.367 298-1400 K (15) -1549.00 (7)

Tremolite 548.90 (2) 160.523 501.582 129.254 -359.826 298-1000 K (15) -11578.55 (2)

Ferroactinolite 702.10 (12) 193.827 548.524 116.199 -295.178 298-1000 K (12) -9824.13 (12)

Cp(T)= a+10-3bT+102cT-1/2+104T-2
Table 1. Continuation.

Vo298 Ref e f Ref g h Ref

Nickel (Ni) 0.659 (2) 4.165 14.341 (18) 6.194 -18.069 (19)

Oxygen (O2) - - - - - - - -

Hematite 3.028 (1) 2.188 17.804 (35) 0.642 -7.947 (1)

Magnetite 4.452 (2) 0.466 17.161 (20) 0.553 1.012 (1)

Bunsenite 1.097 (2) 0.876 3.757 (34) 0.600 -1.000 (21)

Quartz 2.269 (2) 8.722 -41.552 (18) 2.706 -24.000 (19)

Forsterite 4.379 (2) 3.004 8.052 (21) 0.770 0.000 (19)

Fayalite 4.615 (10) 0.864 1.606 (33) 0.767 -0.979 (1)

Diopside 6.611 (27) 3.474 -0.853 (28) 0.891 -2.212 (27)

Hedenbergite 6.785 (28) 2.335 5.258 (28) 2.039 -26.112 (29)

Wollastonite 3.979 (1) 3.585 -11.074 (1) 1.465 -9.819 (29)

Tremolite 27.273 (30) 3.079 0.614 (31) 1.472 -8.510 (30)

Ferroactinolite 27.650 (32) 3.079 0.614 (31) 1.472 -8.510 (30)

% expansion = 10-3e(T-293)+10-7f(T-293)2

% compression = gP+hP2

(1) Robinson et al. (1982); (2) Robie et al. (1979); (3) Hemingway (1990); (5) Hemingway (1987); (6) Robie et al. (1982); (10) Moecher and Chou,
(1990); (12) This study; (15) Krupka et al. (1985); (16) Bennington et al. (1984); (17) Haselton et al. (1987); (18) Skinner (1966); (19) Birch (1966);
(20) Taylor (1985a); (21) Hazen et al. (1977); (27) Levien and Prewitt (1981); (28) Cameron et al. (1973); (29) Vaidya et al. (1973); (30) Comodi et al.
(1991); (31) Sueno et al. (1973); (32) Ernst (1966); (33) Taylor (1985b); (34) Taylor (1984a); (35) Taylor (1984b)
Table 2. Thermodynamic parameters for ferroactinolite according different sources.

ÄGf,298 (kJ/mol) ÄHf,298 (kJ/mol) S298 (J/molK) V298 (J/bar) Notes

Helgeson et al., 1978 - - 684 28.28 (1)
Uchida, 1983 -9806.84 -10467.53 736 28.28 (2)
Johnson and Norton, 1984 - - 687 28.28 (3)
Holland and Powell, 1990 -9857.04 -10527.10 705 28.28 (4)
Tornos, 1989 -9803.90 -10464.60 736 28.28 (5)
Tornos, 1989 -9829.59 -10504.63 687 28.28 (6)
This study -9824.12 -10484.81 702 28.27 (7)

(1) Heat capacity and entropy from additivity rule estimations. Enthalpy and Gibbs free energy not referenced.
(2) Entropy and formation enthalpy derived from graphic adjustment to Ernst's (1966) experiments. Molar volume from Helgeson et al. (1978). Heat capacity from additivity rule
(tremolite, grunerite and antophilite).
(3) Preliminar data from Ernst's (1966) experiments. Enthalpy and Gibbs free energy not referenced.
(4) Data derived from exchange reactions between two or more amphiboles in well characterized natural mineral assemblages.
(5) Gibbs free energy from entropy and heat capacity data from Uchida (1983) and a experimental equilibrium point by Ernst (1966).
(6) Idem. (5) but with entropy and heat capacity data from Jonhson and Norton (1984).
(7) See discussion in the text.