key words [permeability rock experiment]

New data on permeability of the wallrock metasomatosed granites holding a polymetallic ore body were obtained (VerticalÆnaya Vein, Holst Deposit, Sadon Ore Region, North Osetia). Permeability of twelve samples was experimentally determined. The samples represent different detailed sections of the

# This study was supported by the Russian Basic Research Foundation, projects 96-05-64887 and 95-05-15354/a.

wall rocks (section length 2 m, step 2 cm) at two horizons of the deposit: Horizon VII (1182 m), Section 1 and Horizon III (1405 m), Section 2; five samples were collected at a distance up to 200 m the vein. Permeability was determined by the method of impulse damping with argon flow through a core sample 13.7 mm in diameter and 21 mm long at room temperature and impulse pressure of about 1 MPa [1]. The results of the measurements are given in the Table.

1. The permeability values obtained vary widely: from 5.9 to 860 Darcy, i.e. within a factor of more than 10². Nevertheless, the least value is 100-1000 times higher than the permeability of unaltered middle-grained granites.

2. The average permeability of rocks of Horizon VII is twice as high as the permeability of Horizon III (273 and 120 Darcy respectively). However, the rocks of Horizon III far from the ore body are more permeable than the rocks of Horizon VII (135 and 43 Darcy respectively). Both horizons are characterized by gradual decrease in permeability outward the vein (table).

3. The highest and lowest permeability values were established within the wallrock aureole nearest to the vein (0-200 cm): Horizon VII - 5.9 and 860 Darcy; Horizon III - 14 and 240 Darcy. Close to the vein the permeability values are widely scattered, especially at Horizon VII. The average permeability of Horizon VII within the 2-meter interval is three times higher than that of Horizon III (354 and 114 Darcy respectively).

4. Comparison of Zn, Pb, and Cu contents with permeability values of the same samples showed the negative correlation for Section 1 (correlation is insignificant because of small number of samples) and no correlation for Section 2.

Distribution of the elements in Sections 1 and 2 differ significantly [2,3]:

Section 1 - high Zn, Pb, and Cu contents in the inner wallrock aureole (maximum Zn up to 0.35%, Pb 0.1%; linear productivity values within the 2-meter interval are 0.08 m% for Zn and 0.04 m% for Pb); exponentially decreasing Zn content within the interval 0-100 cm; the same tendency is less pronounced for Pb; strong metasomatic alteration of wall rocks (quartzisation, sericitization, chloritization).

Section 2 - low metal contents over the whole interval studied (in some cases negative productivity values: -0.002 m% for Zn and about zero (0.0006 m%) for Pb and Cu. In the inner wallrock aureole, Zn content is lower than the background value (background 4 * 10-3%), although Zn content in the vein exceeds 5%. The maxima of the metal contents are separated in space (Pb 0.016% at 55 cm, Zn 0.08% at 48 cm, Cu 0.007% at 20 cm). The wall rocks are slightly metasomatosed.

From data on distribution of metals in 24 sections of the Holst, Arkhon, and V. Zgid deposits, different mechanisms of formation of Sections 1 and 2 can be suggested [3]: Section 1 - redeposition aureole formed through the infiltration mass exchange between the ore-bearing and ore-free fluids, which originated at different stages of ore formation, and the wall rocks. As this took place, the fracture solu -

tions partially flew off the ore-controlling tectonic structure. Section 2 - leaching aureole formed through the interaction between the solutions and wall rocks, while the porous fraction of the hydrothermal flow was drawn to the fracture channel.

Comparison of the permeability values obtained with geochemical data allows the following conclusions:

1. At a distance of the vein, the wall rocks of the Horizon III are three times more permeable than the wall rocks of Horizon VII. This fact possibly indicates that the rocks of Horizon III were originally more favorable for fluid filtration towards the fracture channel (formation of the Vertical'naya Vein).

2. Interaction of the outflowing solutions with wall rocks (Section 1) resulted in strong metasomatic alteration of the granites and deposition of ore minerals. This process is manifested as sharp permeability variations (within more than a factor of 10²).

3. When the leaching aureoles formed (Section 2), the solution was relatively equilibrated with the wall rock as for all the components and did not cause significant metasomatic alteration. This is manifested as moderate permeability variations (within a factor of ten) even in the immediate vicinity of the fracture channel.

4. The permeability minima correlate well with maxima of ore deposition. This was evidently due to welding of microcracks during the disseminated ore formation (Section 1, Horizon VII).

1. Vitovtova V.M. (1989) Permeability of rocks at high temperature and pressure. Cand.Thesis, Chernogolovka.
2. Borisov M.V. and Goreva Yu.S. (1994) Rudy I metally, N.1, pp. 30-37.
3. Borisov M.V., Goreva Yu.S., and Shvarov Yu.V. (1995) Water-Rock Interaction, Balkema, Rotterdam, pp.717-720.

key words [fluid inclusion artificial]

Quartz is usually used as a matrix for synthetic fluid inclusions, but sometimes other minerals are preferable. For example, if one deals with fluids with low mole fraction of water, it is very difficult to obtain sufficiently large inclusions because of very low rate of crack welding. The experiments with H₂S-bearing fluids, at low temperatures (below 500^oC), etc. also are the cases when quartz is discouraged. Fluorite and calcite were chosen for the fluid inclusion synthesis, because these minerals are more soluble, and the crack welding must occur more readily and faster. Fluorite and calcite are transparent and easily available. The samples for the experiements can be obtained by simple cleaving rather than sawing out of a monomineral block. The perfect cleavage of these minerals makes it possible to obtain inclusions confined to the planes parallel to the cleavage directions.

The samples were subjected to thermoshock in alcohol for better subsequent drying the cracks. The temperature of thermoshock was adjusted so that the sample was densely cracked, but did not disintegrate. The optimal thermoshock temperatures were 260 and 390^oC for fluorite and calcite, respectively.

The cracks were welded in 1M and 2M NaCl and KCl solutions and water-carbonate fluid in autoclaves for 2-7 days. The inclusions obtained - flat and volumetric, up to 30-40 ´m in size - were studied by cryometry. The compositions of the samples corresponded to the run conditions.

It was shown that fluorite and calcite matrixes can be successfully used in synthesis of fluid inclusions.

key words [gold fracture zones]

There is no as yet satisfactory physico-chemical description of gold-bearing formations which differ markedly in mineral and chemical composition of metasomatites and veins. Our models for gold transfer and deposition are based on experimental data on hydroxyl, chloride, and hydrosulfide complexes of gold. From combination of the solubility diagram of gold and stability diagram for mineral assemblages, the role of Au(HS)₂^- is expected to be predominant under the conditions of formation of the main types of metasomatites and ores. Gold solubility is 10^-6 to 10^-7 M at 300^oC and 1 kbar and 10^-5 to 10^-6 at 450^oC and 3 kbar; these data are in agreement with fluid-inclusion studies. When sulfide environment changes to sulfate, solubility of gold decreases by a factor of 10²-10³ as a result of hydrolysis of Au(HS)₂^-. This mechanism is most probable for deposition of native gold in low-temperature subsurface oxidation and argillization processes (gold-silver formation), but unacceptable for abyssal formations, which are characterized by stable redox conditions.

Behavior of Au at temperature- and pressure-gradient conditions was considered in open and closed systems. A negative temperature dependence

of Au solubility with a maximum at 150-170^oC dominates in the open system. Deposition of gold in veins is possible only on further cooling. At higher temperatures, the more effective factor is the positive pressure dependence of Au solubility, which manifests itself at T>300^oC in the closed system as well. Deposition of gold is possible on decompression together with quartz and associated quartz-feldspar metasomatites typical of gold-quartz formation.

The gold-sulfide formation is characterized by concentration of gold in pyrite or arsenopyrite. The experimental study of this process at 200-600^oC and varying alkalinity of the solution showed that maximum gold concentrations (up to 0.08%) are attained in the pyrite synthesized at the parameters of H₂S=HS^- transition, when the gold solubility is maximum. The most Au-rich pyrites were the most fine-grained; this fact indicates significant supersaturation of the solution. These data suggest a sorption mechanism of gold trapping during pyrite crystallization. The conditions favorable for the formation of supersaturated solutions with pyrite and undersaturated with gold can be attained on decompression in cataclasis zones. In this case gold mineralization is associated with sericite metasomatites enriched in carbonaceous matter and sulfides (gold-sulfide formation).

key words [amphibole phlogopite fluorine substitution]

An attempt was made to reveal crystallochemical criteria for anionic isomorphism of amphibole asbestos and micas in order to obtain new formulations of these species.

This paper presents the results of studies of hydrothermal systems with natural talc and alkali-metal-fluoride solutions at different temperatures and concentrations and pressures up to 100 MPa.

We found talc to interact with hydrothermal NaF solution (0.5-5 wt %) to yield fibrous amphibole-richterite at 400-500^oC and 30-100 MPa. A part of hydroxyls in the anionic sublattice of this compound is substituted for fluorine. The fraction of asbestos in the products of talc transformation and the degree of OH -F substitution in its structure are dictated by the NaF concentration of the hydrothermal solution. Richterite-asbestos was obtained as a monomineral phase when talc was exposed to 2-3%-NaF solution at 450-500^oC and 70-100 MPa.

Richetrite-asbestos with a half of its hydroxyls substituted for fluorine was first synthesized under hydrothermal conditions. The crystal chemistry formula of the asbestos calculated from its chemical analysis is

Na_1.78Mg_5.90Fe³⁺_0.20[Si_7.78Fe³⁺_0.22O₂₂](OH_0.98F_1.02).

The optical characteristics of the amphibole-asbestos obtained are n_g=1.593-1.598; n_p=1.579-1.583; n_g-n_p = 0.014-0.015. CN_g = 12^o. The increase in thermic stability of the richterite was attained thanks to the OH ¬ F substitution. The temperature of decomposition of richterite is 850-880^oC, while that of hydroxyl-richterite is 780-800^oC.

In the system talc-NaF-solution-SiO₂, the formation of fibrous F-substituted Na-Mg-triple-chain silicate Na₂Mg₄Si₆O₁₆(OH,F)₂ was established in addition to the formation of richterite. This mineral crystallized in NaF solutions with concentration of 0.3-2.5 wt % at 350-500^oC and P³ 30 MPa.

When talc is exposed to KF solutions with concentration of 0.15-0.75 wt % at 300-500^oC and P³ 30 MPa, trioctahedral phlogopite with partial substitution OH ¬ F (0.2-0.3 formula units) forms. Up to now we have failed to obtain K-richterite and Na-phlogopite by hydrothermal processing of talc in the temperature, pressure, and fluoride concentration ranges studied.

Structural and geometric factors and, especially, alkali-metal cation size are responsible for the preferential formation of band and chain silicates in the system talc-NaF-solution and layered silicates (phlogopite) in the system talc-KF-solution.

key words [fluid nitrogen ore mineralization]

A number of recent reports have shown that dissolution of nonpolar gases, such as N₂, CO₂, CH₄ in aqueous solutions leads to a decrease of the dielectric permeability of the mixture and, consequently, to a shift of the equilibria of the association reactions towards association products. Said volatile components are persistently found in the composition of fluid inclusions being relicts of natural mineral-forming solutions. In the review by O.F.Mironova et al on an investigation of nitrogen-containing melt and fluid inclusions in minerals encompassing more than 110 natural objects from different regions of the globe it has been concluded that nitrogen participates actively in natural mineral formation. It has been

shown that the amount of nitrogen in fluid inclusions can vary within two orders of magnitude and the molar CO₂/N₂ ratios can be <1, since nitrogen is often the principal volatile component of natural fluids.

Our studies of fluid inclusions in quartz of ore-bearing associations and scheelite from the Olimpiada Au-(Sb-W) deposit have also shown a substantial concentration of nitrogen in their composition. Therefore, it appeared necessary to carry out calculations of the equilibria in the systems Au(I)-S(II)-Cl-O-H; Sb(III)-S(II)-O-H; Ca-W-O-H with account taken of the proceeding of all possible complex formation reactions in the presence of nonpolar gas W2 in order to elucidate its influence on the formation of Au-Sb-W mineralization.

Earlier the modelling of the influence of a nonpolar volatile component on the solubility of individual mineral was carried out by F.Gibert (solubility of scheelite in 1M NaCl in the presence of N₂) and G.R.Kolonin (solubility of gold in chloride-carbon dioxide fluid) who noted its change as compared with the system without a volatile component. Our consideration was based on a model of conjugate base and degassing of a high-temperature chloride aqueous nitrogen fluid saturated with respect to gold, antimonite, and scheelite. The calculations were performed under the conditions characteristic of natural ore formation: the temperature range was 150-360^oC, the pressure was 1 kbar, the concentration of NaCl 1.7 M, of sulphide sulphur 10^-2 M. A modified version of the program of isobaric potential minimization BALANCE designed by N.N.Akinfiev and intended for a study of equilibria ion multisystems was used.

The calculations show that nitrogen differently affects the mineral solubility (and so the fluid transportability), and that the solubility in a nitrogen-containing fluid differs from that in a nitrogen-free one. So the solubility of antimonite is weakly dependent on the presence of nitrogen in a solution. The solubility of scheelite and gold is decreased for a mixed solvent, however, the magnitude of this effect is much smaller against the one expected from the formation reactions equilibrium of multicharged ions AuCl^2- , Au(HS)Cl^2-, HWO^-₄ and WO^2-₄ . This is because along with a change of the dielectric permeability of a mixed fluid leading to a shift of the equilibrium of the reaction towards the products there takes place a decrease in the concentration of the ligands Cl^-, HS^-, and H⁺ due to the association reaction, and a total change of the solubility is moderate.

Our consideration suggests that mixing of a fluid enriched in Au, Sb, with a nitrogen-containing one can lead to a simultaneous precipitation and separation of gold and tungsten from antimony. This is confirmed by an analysis of natural objects. The segregation of antimonite is largely controlled by the temperatural factor.

Although preliminary, our calculations nevertheless show that the allowance for the effect of nonpolar volatile component makes it possible to reveal singularities of transport and ore-generating possibilities of a fluid.

key words [water vapor polymerization constants ]

Based on P-V-T data for water vapor the polymerization con- stants have been calculated for water vapor in the temperature range 25-800^oC (T-temperature in Kelvin):

pK[2H₂O=(H₂O)₂]=-726.5/T+3.74 + 1/(exp(1421/T - 0.4076)-1)

pK[3H₂O=(H₂O)₃]=-1393/T+7.047 + 1/(exp(1154/T - 0.6645)-1)

pK[4H₂O = (H₂O)₄]=-2443/T+10.94+ 1/(exp(1363/T - 0.832)-1)

Associates of the water vapor molecules are real particles which could be considered as separate chemical components. Determination of their individual thermodynamic properties may be sometimes useful and simply curious because one would like to understand and make a quantitative description of the processes which lead to the nonideality of real gases. In order to calculate the polymerization constants it will suffice to make the following assumptions whose validity would hardly leave doubt at low density of vapor.

1. Water vapor is an ideal mixture of ideal particles. However, it should be mentiond that: the 'ideality' in this case means that we disregard the size of particles, that the activity and volatility coefficients of all the particles are assumed equal to 1, and the expression PV=nRT is true for all forms of gas mixture. Otherwise, it is suggested that when taking a complete account of water vapor complexation, the nonideality of the given system is completely considered (which, naturally, becomes untrue at high pressures).

2. The following equilibria are realized in water vapor: 2H₂O = (H₂O)₂, 3H₂O = (H₂O)₃, 4H₂O = (H₂O)₄ etc.

3. The real water vapor pressure is a sum pressures of a monomer and all water polymers: P = P(1) + P(2) + P(3) + P(4) +.....

4. An ideal pressure is a pressure attained in a system when all the water polymers anyhow affected

are disintegrated: P_id = RT/V = P(1) + 2P(2) + 3P(3) + 4P(4) +....

5. Taking into account the first item, one can conclude that in the given approximation the pressure of a monomer is in a full accord with the determination of the volatility function because it shows the pressure of ideal gas particles, corresponding to to the given formula (H₂O) at a given total pressure of water vapor. Therefore, when pressures are not too high, and the size of a particle itself does not effect gas properties, the volatality of a component equals the pressure of a monomer of this component. The dependences of real and ideal pressures on volatility were used to calculate the constants:

P = f + Kiif2 + KIIIf3 + KIYf4 , and

Pid= RT/V== f + 2KIIf2 + 3KIIIf3 + KIYf4.

Practically, the range of the P-V-T data used appeared to be limited in pressure up to 150 bar. The principle of calculation is rather simple and can be reproduced graphically.

To determine the constant of dimerization reaction a diagram has been plotted in coordinates y=lg(RT/V-P), x=lg(f). Subtraction of real pressure from the ideal one is undertaken to get rid of the monomer pressure. The difference remained after the pressure drop should approach the ideal square dependence on water volatility, since all this difference is approaching the pressure of dimer. Actually, the dependence given in logarithm coordinates is a line with a slope precisely equal to two at low pressures and increasing to 2.2-2.5 at elevated pressures. The logarithm of the dimerization constant can be determined by extrapolation of this dependence to lg(f)=0.

Similarly, the dependence y=lg(RT/V-P-KIIf2)/2, x=lg(f) is suitable to be used to find the constant of the trimerization reaction. The diagram of this dependence is a line which at low pressures has a slope equal to 3.

The constant of the tetramerization reaction is found in much the same way. To this end, one can use a function y=lg(RT/V - P - KIIf2 - 2KIIIf3)/3), x=lg(f), with a virtually ideal slope of 4. Polymers of higher polymerization degrees are found in the temperature range 300-600^oC at elevated pressures. However, the corresponding constants are difficult to be found in the same way because of two factors: the error after using already determined constants becomes more significant and the size of particles has an appreciable effect ( a volume effect of polymerization reaction decreases). By the reasons mentioned, it is rather difficult to identify the associates given.

Below are the constants calculated by the method described:

The uncertainty in calculation of the dimerization constants is 0.01, for the trimerization constants -0.1 and for the tetramerization ones - 0.3.

Only the dimerization constants can be calculated with high accuracy because significant volume effects corresponding to the reaction given are observed at extremely low densities of vapor. The volume effects conforming to trimerization and tetramerization are observed at relatively high densities of vapor where the approximation used is less effective. That is why the values cited are only a matter of estimates.

The temperature dependence of polymerization constants is approximated by the function:

pK = A/T + B + 1/(exp(C/T - D) - 1)

Such a somewhat "strange" dependence is not chosen by merely chance.

The point is that the change in heat capacity in the reactions considered is poorly described by the conventional polynom. A curious phenomenon takes place: at a certain temperature, something like degeneration of energetic states corresponding to various water polymers occurs. The approximation equations obtained are given above.

When analysing heat effects of polymerization reactions one can notice that they approximately relate as 1:2:3. This suggests the idea that water polymers have a linear structure with the energy of hydrogen bond of about 14 kj/mol.

key words [fluorite solubility experiment calculation]

A method of weight loss of a single crystal has been used to study the solubility of synthetic fluorite in individual solutions 1m - 34.7 m NaCl (m is mole/kg H₂O) at 400, 500 and 600^oC, 0.5-5 m CaCl₂ at 500 and 600^oC, and in mixed solutions NaCl-CaCl₂ with m_totCl=10 at 500^oC. The solubility isotherms, taken in logarithmic scale, show a linear dependence for all the studied chloride concentration ranges, except the 400^oC isotherm. In the mixed solutions the curve of the Cl constant values exhibits a minimum at Na/Na+Ca ratios in the solutions close to 0.95. The solubility in the CaCl₂ solutions is more than half an order higher against the NaCl solutions of the same concentrations.

The high solubility values in NaCl and CaCl₂ individual solutions are indicative of the formation in the solution of particles of a complex composition, i.e., ionic pairs or associates. This conclusion is confirmed independently by the fact that solubility in chlorides is high, and it is particularly high in CaCl₂ solutions even thought CaCl₂ solutions have a common ion with the crystalline phase.

The general principles of the experimental data processing (analysis) to identify the particles forming in a solution involve a comparison of theoretically expected solubility isotherms slopes as a function of the chloride concentration for a number of hypothetic dissolution reactions (Sl_ther) with the experimental slopes (Sl_exp), and subsequent selection of the reaction in accord with the criterion of the best fit of theoretical and experimental slopes. It is supposed, herewith, that associated neutral particles are predominant in the solution.

Earlier, based on this method for processing the 600-800^oC data, the authors proposed the following particles: NaF × NaCl^o (Na₂FCl)^o for NaCl solutions, and CaCl₂ × CaF₂^o solutions [1].

In addition to the above forms Ca₂Cl₃F^o, Na₂CaF₄^o and NaCaCl₃^o particles were also proposed using an independent method of thermodynamic modelling [2]. For reasons of parallelity of the 600, 700 and 800^o solubility isotherms for NaCl solutions and the 600, 700 and 800^oC ones for CaCl₂ solutions, the above proposed particle forms can be related to all said temperatures.

For the 500^oC isotherm of fluorite solubility in NaCl solutions, for which Sl_exp=0.8+0.1, the following reactions that accompany the dissolution with the formation of the corresponding particle forms may be regarded as possible.

CaF_2(c) + NaCl^o = CaF₂ × NaCl = NaCaClF₂^o,

2CaF_2(c)+2NaCl=CaClF^o+NaCl × NaF^o=
=CaClF^o+Na₂ClF^o,

2CaF_2(c)+ 2 NaCl= CaCl₂+ CaCl₂ × 2NaF^o =
= CaCl₂ + Na₂CaCl₂F₂^o,

For all the three reactions Sl_ther =1.

For the 500^o isotherm of fluorite solubility in CaCl₂ solutions with Sl_exp =1.1+0.1 one reaction and the corresponding particle form may be proposed:

CaF_2(c)+ CaCl₂^o = CaF₂ CaCl₂ = × Ca₂Cl₂ × F₂^o

with Sl_theor =1 that ought to be related to the 600^oC isotherm too.

As earlier discussed in [1] this method for particle identification does gives no unambiguous answer concerning their particular form, however it allows an understanding of the type of the forming particles as neutral associates of a complex mixed composition. This conclusion is based on the linear character of the dependence of the isotherms in a broad range of the chloride concentrations.

1. Malinin S.D., Kurovskaya N.A. (1992) Experimental study of fluoride solubility in aqueous solutions of Na, Ca chlorides and their mixtures at 600-800^oC and 2 kbar. // Geochem., N.11, pp.1473-1482.
2. Ryzhenko B.N., Knyazeva S.N., Malinin S.D., Kurovskaya N.A. (1994) Estimation of the form of dissolved particles in the system CaF₂-NaCl-CaCl₂-H₂O at 800^oC and 2 kbar using a method of thermodynamic modelling. // Geochem., N.4, pp. 467-475.