Glikin A.E. Polymineral-Metasomatic Crystallogenesis

Подождите немного. Документ загружается.

1.5 Genetic Nature of Replacement Products 39

This scheme can lead to an assumption that automorphs should have had a wider

range of formation conditions and, consequently, a more frequent occurrence in com-

parison with pseudomorphs, but this conclusion is not confirmed by the data shown

in Table 1.2. The most frequently occurring are shown to be blurred pseudomorphs

and localized automorphs that is likely to be explained by relatively close rates of the

reactant counterdiffusion and similar rates of precipitation of new formations.

Kinetic analysis in systems Ib and II requires a different approach.

Monocrystalline pseudomorphs (systems Ib) also have different morphology,

practically, in all the investigated pairs of direct and reversed reactions of Type Ib:

reactions 24, 26, 28, 30, 34 form only embossed and faceted pseudomorphs, while

reactions 25, 27, 29, 31, 35 produce spongy faceted and blurred pseudomorphs;

products in reaction pair 32–33 are indistinguishable by the shape from each other,

as are the products of pair Ia/44–45. In general, these phenomena can be formally

explained by similar distribution and interaction of the diffusion fields as described

above. However, such explanations are not illustrative, and in Type Ib systems other

effects must be given a priority, with volume effect defining differences in the

mechanisms of direct and inverse reactions being the major one (see Sect. 3.2).

When the negative products are formed in systems of Type II, diffusing the bor-

ders of the cavities is defined by the rate of lowering the temperature and by the

positions of figurative dots versus the eutonic point. In general, it can be stated that

diffusing the contours depends to somewhat extent upon the followings: protocrys-

tal dissolution rate, diffusion of its matter into the solution, crystal precipitation

from the solution, temperature gradient of solubility of precipitating substance, and

salting-in effect of the protocrystal matter. The highest degree of preserving the

protocrystal shape is achievable only either by means of extremely slow dissolution

of the protocrystal and fast crystallization of the surrounding precipitating matter,

regardless the salting-in effect, or by means of fast dissolving the protocrystal and

a very slow diffusion of its matter under the action of a strong salting-in effect.

The easiest way to obtain the negative pseudomorphs with maximum preserva-

tion of the shape is abrupt and deep supercooling with slow dissolution of the

protocrystal tightly enveloped by the precipitate; in this case the salting-in effect of

the protocrystal matter should be weak, limiting the growth of the precipitating

crystals, which border the cavity. The process is similar to a conventional crystal

encrustation followed by its dissolution, i.e., the formation of asynchronous nega-

tive pseudomorphs.

Translocated automorphs can be formed in all types of the systems following

two varying pathways having different kinetic nature (for corresponding negative

products under conditions of undersaturation cavities are formed instead of grow-

ing solid phase). The first pathway is as follows: a portion of the solution that sur-

rounds the protocrystal and is supersaturated with the newly forming phase is

transferred (translocated) by convective streams from the protocrystal to a deposi-

tion area. Following the second pathway, a temporary elevation of temperature

reduces supersaturation in the portion of the solution that surrounds the protocrystal

and prevents precipitation of a new formation, which, after cooling, crystallizes

independently, far from the former location of dissolved protocrystal. Our methods

40 1 Replacement of Monocrystals

make realization of this pathway possible, and, accidentally, a monocrystalline

translocated automorph was produced in replacement experiment conducted in

isomorphic (Co,Ni)SO

O series.

Thus, the list of taxons, which specifies the forms of replacement products

(Table 1.2) is considered to be comprehensive in accordance with the above analy-

sis of the replacement kinetic effects.

In addition to the relationships between the process kinetics and structural-morpho-

logical characteristics of the products discussed above, the following should be noted.

First, the process kinetics defines a rhythmical zonal distribution of the secondary

mineral phases forming polycrystalline polymineral pseudomorphs (reaction III/53).

Apparently, such zoning is formed in a self-oscillating diffusion mode defining

crystallization process, which is similar to formation of Liesengang rings (Liesegang

1913; Stern 1954; Carl and Amstutz 1958; Krasnova and Petrov 1997), or cyclic

capture of inclusions by a crystal (Treivus 1979; Petrov et al. 1983, and others).

It is interesting to note that in the above case III/53 the diffusion front was expand-

ing from a protocrystal periphery to its center.

Second, microimpurities, which can be present in the solution (including acidic–

alkaline admixtures: reaction Ia/5, III/42), and the process temperature (Ia/5) have

different effects upon morphology of the new formations due to their unequal influ-

ence upon the diffusion ratios of different reactants and also due to variation of the

adsorptive effects, which, by modification of the growth and dissolution kinetics

(Treivus 1979; Petrov et al. 1983, and others), may totally alter distribution of dif-

fusion fields and, consequently, location and width of the reaction zone.

Third, metasomatic crystallogenesis differs from conventional crystallization

occurring at lowering temperature, pressure drop, etc., as it always proceeds under

profoundly nonstationary conditions determined by variations in solution composi-

tion (Dolivo-Dobrovol’skii 1967, and others). It was already mentioned (Glikin and

Sinai 1991; Glikin 1996a) that this is the reason of formation of case-like pseudo-

morphs and occurrence of protocrystal relics in replacement products. Nevertheless,

this fact has not been properly investigated yet, and so discussion of some effects

demonstrating its significance is presented in Chapter 4.

1.5.4 Imperfection of Protocrystals as a Factor

of Replacement Process

Influence of defects present in protocrystals upon replacement process has

already been mentioned in the previous sections, but it deserves a thorough

and careful study. Growth and dissolution of crystals is generally known to

depend to a considerable extent upon the crystal defects (Petrov et al. 1983;

Chernov 1984, and others). Genetic role of inclusions, twins, dislocations, etc.

is being extensively studied. It is obvious that imperfection of a protocrystal

affects the product morphology and replacement kinetics, its influence upon

morphology being dual.

1.5 Genetic Nature of Replacement Products 41

On one hand, imperfection is inherited and can be seen in some features of

the new formations. They include such well-known phenomena as inheritance of

polytypic motives during transformations of layered silicates in hydrothermal

solutions (Frank-Kamenetskii et al. 1981), and retention of solid inclusions in

the form of shade pseudomorphs at their initial places in protocrystals

(Chesnokov 1974, p. 67). This list can be extended by addition of the above-

mentioned examples of autoepitaxial excrescences formed on fractures and

outcrops of dislocations in the course of monocrystal replacement: see reactions

Ib/24 (Fig. 1.4), 26, 28, 30 (Fig. 1.3), 32, 33 (Glikin and Sinai 1983, 1991), of

replacement product inheriting inclusions localized in a protocrystal: see reac-

tions III/44, 45, 47 (Fig. 1.18 a–c) (Sinai and Glikin 1989; Glikin and Sinai 1991),

of pseudomorph inheriting the attributes of the initial twin structure (Glikin and

Sinai 1983): see reactions Ib/35 (Fig. 1.19).

On the other hand, under replacement conditions imperfection of a protocrystal

promotes loss of information about the protocrystal. For example, monocrystalline

pseudomorphs formed under volume-excess conditions and bearing a large number

of autoepitaxial excrescences (Fig. 1.4) lose the initial face relief (reactions Ib/26, 28,

30, 32, 33), while the products of interaction between the surrounding solution and

liquid inclusions of protocrystal may completely overlap and hide the new formation,

which replaces a solid matter of the protocrystal (reaction III/47 – Fig. 1.18c).

Our experimental results are represented below.

Interaction of the substances contained in the protocrystal fluid inclusions with a

metasomatic solution takes place after opening the inclusions as the protocrystal dis-

solves. Under these conditions precipitation of new formations is facilitated by salt-

ing-out or ion-exchange reactions (reactions of Types Ia and III). This results in

formation of solid products of inclusion replacement, which are genetically and mor-

phologically similar to the crystal replacement products (Glikin and Sinai 1988,

1991; Sinai and Glikin 1989). This analogy is supported by a well-known opinion of

the fluid inclusions being regarded as negative crystals (Laemmlein 1929, 1973).

Some examples of the reactions shown in Table 1.1 are given below.

Reaction III/53 (replacement of NiSO

·7H

O crystals by nickel chromates in

CrO

solution). The protocrystals possess prismatic shapes having tetrahedral

heads and the sizes of about 1 cm. They contain dissipated isolated inclusions with

the size of about 1 mm or less having isometric shapes or slightly elongated along

the longest dimension of the crystal. The products of replacement of such crystals

are represented by faceted or faceted-embossed polycrystalline pseudomorphs,

composed of fine-grain aggregates of nickel chromates and K

CrO

. The products

of the inclusion replacement are visible on the pseudomorph fractures. They are

isolated fragments having similar phase composition, but differing in grain size;

generally their sizes and shapes correspond to those of initial inclusions, but precise

definition has not been carried out. These products crystallize directly in the vol-

ume of the inclusions, either at the moment of their opening, when the front of

crystal replacement touches the inclusion, or in advance owing to solution penetrat-

ing the crystal through the microcracks and reaching the inclusions in accordance

with the mechanism defined formerly (Grigor’ev 1961, p. 173). Reaction Ia/5

42 1 Replacement of Monocrystals

Fig. 1.18 Fine-grained aggregates of BaSO

– products of replacing the inclusions contained in

the initial crystals a, b – tubular shapes (localized automorphs): replacement of potassium alum

in Ba(NO

)

solution (a – tube and a circular automorph, reaction III/44) and replacement of

Ba(NO

)

in potassium alum solution (b – a tube cluster, reaction III/45; photo by N.I. Krasnova);

c – dense precipitate (dissipated automorph): tube coalescence during replacement of BaCl

·2H

in potassium alum solution (the upper part contains the protocrystal contour, reaction III/47)

1.5 Genetic Nature of Replacement Products 43

(replacing KAl(SO

)

·12H

O crystals predominately with nickel chromates in

CrO

solution) is analogous to the previous one. Pseudomorphs formed on the

inclusions have radiate-fibrous or laminated structures, and, in contrast to the previ-

ous case, it is difficult to find a unique correspondence between their shapes and

the shapes of the initial inclusions.

Reaction III/44 (replacement of KAl(SO

)

·12H

O crystals with barite in

Ba(NO

)

solution). The protocrystals have octahedral shapes with weakly devel-

oped cubic faces and the size of about 1 cm. Inclusions are confined within the

sectors of the cube growth.

Reaction III/45 (replacement of Ba(NO

)

crystals with barite in KAl(SO

)

solu-

tion). The protocrystals having the size of about 1 cm are faceted by positive and

negative tetrahedrons and cube and flatted along the {111} axis. Multiple point-like

inclusions are confined within the cube sectors. Both above reactions result in

gradual development of the localized BaSO

automorphs containing weakly bonded

grains with the size up to 0.1 mm. This process is accompanied by much faster

growth of tubular barite aggregates having complex shapes, which start developing

from the tetrahedron sectors saturated with inclusions (the crystal size up to

0.1 mm) (Figs. 1.18a, b). The tubes have diameter of about 0.5–1 mm and length up

to several centimeters; they are composed of the liquid contents of opened inclu-

sions, which has already reacted with the solution medium, and the rate of their

elongation is about 0.5–1 mm/s.

The tubes define localization of inclusions in protocrystals. Successive opening

the inclusions, which are frequently arranged in chains directed from the crystal

periphery to its center, results in formation of the tubes. The tubes promote diffu-

sive transportation of the inclusion solution to the solution surrounding the crystal

Fig. 1.19 Preservation of initial boundaries between the individuals of the triplet during volume-

deficit monocrystal replacement of K

CrO

with K

(Cr,S)O

(reaction Ib/35). Crossed nicols.

Absent and bloomed individuals are visible

44 1 Replacement of Monocrystals

and, thus, facilitate growth of the tube end faces. Breaking an inclusion chain stops

growth of the tube. Several tubes can grow simultaneously, being initiated at differ-

ent time. The growth is fast as the both reactants are in dissolved state.

Reaction III/47 (replacement of BaCl

·2H

O crystals with barites in KAl(SO

)

solution) is partially similar to the previous one. The protocrystals having the sizes

of about 1 cm are impregnated with multiple point-like and relatively large (up to

0.5 mm) inclusions, which do not group in any obvious arrangement. In contrast to

the previous case, the tubes are quickly merged to form a dense precipitate, which

overlaps the automorph replacing the solid part of the crystal (Fig. 1.18c).

Such new formations comply with suggested basic classification of replacement

products of monocrystals (Sect. 1.4): the above products of the inclusion replacement

are similar to the pseudomorphs deposited within the contours of the initial inclusions

(reactions III/53 and Ia/5), and to localized automorphs, if the products indicate location

of the inclusions (reactions III/44, 45), and to dissipated automorphs, when this location

cannot be determined (reaction III/47). It should be noted that both pseudomorphic and

automorphic replacements of the solid part of the crystals are accompanied by the same

type of inclusion replacement. This is to be expected as morphology of the products is

defined by the ratio of reactant diffusion rates in the solution (Sect. 1.5.3). However, it

should be kept in mind that replacement of the crystal matter is complicated by crystal

dissolution and the above regularity may not hold true for all reactions.

Unfortunately, scarcity of the investigated products does not allow us to draw a

complete analogy between the classification of the crystal replacement products and

that of inclusion replacement products, since it is difficult to estimate a probability

of the other structural (monocrystalline and amorphous) and morphological types to

occur in the process concerned. In any case, there is no a priori prohibition for the

majority of taxons of the basic classification. However, classification of pseudo-

morphs according to their inclusions into “embossed,” “faceted” and “blurred”

pseudomorphs is hardly appropriate, since regarding the inclusion as negative crys-

tals cannot be universal, and lack of the faces and microreliefs in majority of inclu-

sions cannot promote their division into corresponding morphological types.

Therefore, structural-morphological classification of the replacement products

may be represented in the form of Table 1.3. Outward similarity between classifica-

tion of replacements of monocrystals and that of inclusion replacements does not

mean genetic resemblance of all the corresponding processes. Of course, polycrys-

talline replacement of the inclusions is certainly quite similar to replacement of

monocrystals of Types Ia or III. However, a hypothetic monocrystalline replace-

ment of inclusions normally reflects the natural growth of individual, which totally

or partially fills up the inclusion volume, and monocrystalline isomorphic replace-

ment (Type Ib) is absolutely impossible in this case. In should be noted, however,

that formation of automorphs on inclusions (reactions III/44, 45 and 47) was

observed in clusters of inclusions, but not on individual inclusions; i.e. such auto-

morphs can be closer related to some products of replacement of certain crystalline

aggregates, although this phenomenon has not been thoroughly studied.

During monocrystalline replacement of a twin, the individuals preserve their

initial borders (Fig. 1.19), but monoclinic and triclinic crystals change orientation

1.6 Mineralogical Aspects of the Problem of Crystal Replacement 45

of the optical indicatrix. In those cases there must exist a discrepancy between the

twinning angles and the composition, which either can be a sign of isomorphic

replacement process (for instance, for plagioclases) or the source of errors when

optical methods are used for isomorphic analysis of mineral composition.

Influence of protocrystal imperfection upon the reaction kinetics was not thor-

oughly investigated; however, some specific phenomena have been observed in direct

and inversed monocrystalline replacements in isomorphic K(Al,Cr)(SO

)

·12H

alum series (reactions Ib/28 and 29).

Replacement rates observed for protocrystals having similar appearance

(4–5 mm), which have been grown together, differ significantly in spite of location

near each other in the same reacting solution. For instance, spongy pseudomorphs

containing at intermediate stages various ratios of isomorphic components are pro-

duced under volume-deficit conditions (Ib/29). On the contrary, under volume-

excess conditions (Ib/28) the replacement front is very uneven with threefold to

fourfold variation in the thickness of the newly forming rim. Both phenomena can

be easily distinguished as new formations differ in color. The protocrystals and/or

their sectors are likely to contain different amount of defects. Clusters of fluid

inclusions should effectively influence these processes owing to accelerated trans-

portation of the substances by means of the liquid phase; under conditions of

sponge replacement dislocation clusters should accelerate dissolution and growth

of the new formation, while under conditions of continuous replacement they

should accelerate the solid phase diffusion (see Sect. 3.2).

1.6 Mineralogical Aspects of the Problem

of Crystal Replacement

The proposed replacement concept based on experimental data includes known

mineralogical classifications of the numerous natural replacement forms and

defines genetic position of the products (Blüm 1843; Strunz 1957; Betekhtin et al.

1958; Grigor’ev 1961; Polovinkina 1966; Kostov 1968; Zhabin and Rusinov 1973;

Amstutz 1981c, and others). Moreover, the concept includes the forms, which are

Table 1.3 Classification of the inclusion replacement products

Shape

Structure

Monocrystalline Polycrystalline Amorphous

Pseudomorphs Hypothetic

2 (Ia/5, III/53)

Hypothetic

Automorphs

Localized Hypothetic

2 (III/44, 45)

Hypothetic

Dissipated Hypothetic

1 (III/47)

Hypothetic

Note: The number of systems where the corresponding products were observed

is printed in bold; examples of the systems are given in brackets.

46 1 Replacement of Monocrystals

not normally paid attention to, and, by doing this, extends the list of products of

potentially metasomatic genesis.

Conventional types of replacement products, normally described in published

literature, are arranged according to the taxons of our classification represented in

Table 1.4. It is remarkable that despite a certain conditional character of such dis-

tribution it has been possible not only to represent pseudomorphs as a whole, but

also to correlate many products with the new taxons, which include modifications

of both pseudomorphs and automorphs. The list of pseudomorphs indicates that

polycrystalline replacement products are described more frequently and in more

details. However, their generalized list has been considerably extended by including

some new formations, which were put into the table on the basis of developed

genetic concepts. The debatable cases are reviewed below.

The crystal ensembles of metasomatic veins represent the products of transfor-

mations of minerals, which were located outside those veins. The educts and the

products are spatially separated and degree of this separation is not particularly

limited. All this corresponds to a definition of translocated automorphs. The crystal

ensembles of metasomatic rocks are the products of transformation of minerals,

which were located within a massif. In this case the educts and the products are also

spatially separated; however, if they are at least uniformly distributed in the initial

and transformed massifs, the areas of their distribution overlap and the products can

be regarded as dissipated automorphs. Secondary porosity in metasomatic rocks

Table 1.4 Some traditional types of natural replacement products within the new classification

scope (Table 1.2)

Shape

Structure

Monocrystalline Polycrystalline Amorphous

Pseudomorphs

In general Monocrystalline Continuous Hollow

Homoaxial Box-shaped (case-like, enveloping, hollow) Replica

Ion-exchanging Shadowed

Rims

Borders

Embossed and

faceted

Formed in situ

Formed with filling up

Perimorphs

Blurred Epimorphs

Formed with transportation (displacement)

Automorphs

Localized Rims, borders, coronas, coronites, drusites

Crystal clusters in a uniform matrix

Liesengang rings

Orbiculars

Dissipated Ensembles of crystals in

metasomatic rocks

Secondary

porosity

Translocated Ensemble of crystals in metasomatic veins

1.6 Mineralogical Aspects of the Problem of Crystal Replacement 47

can be regarded similarly. This approach can be supported by reliable observation

of diffusing the pseudomorphs by means of product transfer beyond the boundaries

of the initial monocrystal (Grigor’ev 1961). We propose a combined approach to all

the above phenomena, taking into account that the distance of the matter transfer

may be great enough. Of course, metasomatic nature of the process is defined by

cause–effect relationships, which control dissolution of the educt, transfer of the

matter, and deposition of the product; however, correlation between the above proc-

esses cannot be proved in each particular case even for ordinary pseudomorphs.

Analysis of data on replacement of monocrystals demonstrates that metasomatic

crystallogenesis can be applied to a significantly greater phenomenological range in

comparison with the diapason limited by traditional views adopted in mineralogy,

especially considering the absence of the term “automorph” in mineralogical method-

ology. In other words, mineralogical definition of metasomatic replacement consists in

simultaneous dissolving the protocrystal and growing the new formation, which does

not permit essential deviations from the simultaneous proceedings of elementary proc-

esses and significant outgrowing the boundaries of the initial crystal.

We think that the concept of space-time superposition of dissolution and growth

processes should be extended. The data shown in Table 1.2 and analysis of the

respective cause–effect relationships (Sects. 1.4–1.5) demonstrate that complete

superposition was observed only during volume-deficit formation of embossed

pseudomorphs and also during formation of products, in which deficiency of physi-

cal volume was accompanied by the total volume compression. At the same time,

the borders in the series of taxons ranging from embossed pseudomorphs to translo-

cated automorphs are rather conventional and changes from one taxon to another are

stipulated by monotonous alteration of some kinetic parameters of physicochemical

systems. Alterations in volume effect are also determined by arbitrary variations in

relative solubility of components in various systems. The essential conclusion is that

all these products are of common nature depending upon a similar driving force

(salting-out for reactions of Types I and III or salting-in for reactions of Type II).

Therefore, in general, an extent of time-space superposition of elementary proc-

esses of growth and dissolving is very important. It has a maximum in processes of

monocrystalline isomorphic replacement. This maximum is likely to have a physical

meaning defined by the sizes of diffusion fields surrounding adjacent points of growth

and dissolution. The extent of time-space superposition is gradually reducing (both in

time and space) when transferring to the processes of formation of other products, and

completely absent in the case of translocated automorph formation. Obviously, such

transition causes the products to gradually loose the metasomatic attributes, and in the

cases of translocated automorphs they cannot be distinguished from the products of

conventional direct growth. This fact probably explains a conventional understanding

of metasomatism as a process proceeding within a limited volume.

Classical division of replacement products into various types of pseudomorphs

seems to be insufficient, as the products are divided into the groups according to their

structural-morphological attributes regardless to their genesis. The commonly used

genetic definitions (pseudomorphs of displacement and pseudomorphs of mixing, syn-

chronous and asynchronous pseudomorph types, ion-exchange types, etc.) are incom-

plete and are not related to morphological attributes or physicochemical factors.

48 1 Replacement of Monocrystals

Our classification do not contradict the classical one, but makes it more detailed,

streamlined, and elaborated, since it takes into consideration genetic peculiarities

of the taxons. It seems convenient and easy to visualize, because taxons are defined

in accordance with known hierarchy of discrete and readily detected structural and

morphological signs. Therefore, our classification can be recommended as a basis

for description and genetic analysis of natural products.

The classification can be modified for numerous individual peculiarities of the prod-

ucts to be taken into consideration. On one hand, it can include additional taxons. For

example, intermediate (faceted-blurred pseudomorphs etc.) or combined (various extent

of preserving different protocrystal faces) morphological modifications can be added.

Structural species can include intermediate products, transitional from mono- to poly-

crystalline and comprising graphical (Figs. 1.6c, d) or epitaxial (Glikin and Kaulina

1988) textures. It might have been appropriate to divide polycrystalline products into

twins, textured, and irregular formations, and classify amorphous formations into dis-

persed, glass, and fluid structures. It could be interesting to use the same approach to

review metasomatic replacements of liquid phases in fluid systems and glasses. On the

other hand, the basic classification can be elaborated to include additional attributes,

such as volume ratio between a protocrystal and a new formation (and related dividing

the products into continuous, loose, sponge and hollow ones), the number and space

distribution of the newly formed phases, overlapping the protocrystal replacement prod-

ucts and inside inclusions. Those attributes can be considered as additional coordinates

of the classification transforming it into a multi-dimensional concept.

An example of classification modified to comprise products of inclusion replace-

ment in crystals is presented in Sect. 1.5.4. We succeeded in finding the only analogous

result in published mineralogical literature describing some synthesized products.

It shows a schematic representation of replacement front containing “advancing sites”

(Fig. 1.20) corresponding to pseudomorphs formed on inclusions described by us.

Fig. 1.20 Replacement front containing “advancing sites” for pyroxene replacement with amphi-

bole (According to data obtained by Grigor’ev 1961)