B. Velde

Si+4 Content of natural phengites

The chemical compositions of white micas separated from adjacent rocks of glaucophane and greenschist facies are compared with respect to their Si+4 content. The micas are predominantly phengitic, i.e. between muscovite, K[Al2Si3AlO10(OH)2] and celadonite, K[(R+2R+3)Si4O10(OH)2] in composition. Constancy of Si content in micas coming from rocks of different bulk chemical composition but closely similar physical conditions indicates that the silica content of a potassic dioctahedral mica can be used to indicate the pressure and temperature conditions of its formation. This conclusion is in part based upon previous experimental data obtained for synthetic phengites.

The origin of clay minerals in soils and weathered rocks

Fundamentals of Clay Mineral Crystal Structure and Physicochemical Properties.- Basics for the Study of Soil and Weathered Rock Geochemical Systems.- The Development of Soils and Weathering Profile.- Clay Mineral Formation in Weathered Rocks: Water/Rock Interaction.- Plants and Soil Clay Minerals.- Clays and Climate - Clay Assemblages Formed under Extreme Humidity Conditions.- Physical Disequilibrium and Transportation of Soil Material.- The Place of Clay Mineral Species in Soils and Alterites.- The Place of Clay Mineral Species in Soils and Alterites.

CLAY MINERALS IN THE MEUSE-HAUTE MARNE UNDERGROUND LABORATORY (FRANCE): POSSIBLE INFLUENCE OF ORGANIC MATTER ON CLAY MINERAL EVOLUTION

A clay-rich Callovo-Oxfordian sedimentary formation was selected in the eastern Paris Basin (MHM site) to host an underground laboratory dedicated to the assessment of nuclear waste-disposal feasibility in deep geological formations. As described initially, this formation shows a mineralogical transition from an illite-smectite (I–S) mixed-layered mineral (MLM), which is essentially smectitic and randomly interstratified (R = 0) in the top part of the series to a more illitic, ordered (R ⩾ 1) I–S in its deeper part.This description has been challenged by using the multi-specimen method developed by Drits et al. (1997a) and Sakharov et al. (1999). It is shown that all samples contain a physical mixture of an unusually (?) illitic (∼65% I) randomly interstratified I-Exp (illite-expandable MLM) and of a discrete smectite, in addition to discrete illite, kaolinite and chlorite. Structural parameters of the different clay phases vary little throughout the series. According to the proposed model, the mineralogical transition corresponds to the disappearance of smectite with increasing burial depth.Comparison with clay minerals from formations of similar age (Oxfordian-Toarcian) throughout the Paris Basin shows that the clay mineralogy in the deeper part of the series originates from a smectite-to-illite transition resulting from a low-temperature burial diagenesis. The anomalous lack of evolution of clay minerals in the upper part of the series is thought to be related to specific interactions between organic matter and clay minerals.

New algorithms in 3D image analysis and their application to the measurement of a spatialized pore size distribution in soils

We introduce a skeletization method based on the Voronoi diagram to determine local pole sizes in any porous medium. Using the skeleton of the pore space in a 3D image of the porous medium, a pole size value is assigned to each voxel and a reconstructed image of a spatialized local pore size distribution is created. The reconstructed image provides a means for calculating the global volume versus size pore distribution. It is also used to carry out fluid invasion simulation which take into account the connectivity of and constrictions in the pore network. As an example we simulate mercury intrusion in a 3D soil image

Fractal analysis of fractures in rocks: the Cantor's Dust method

Abstract The Cantors Dust method of fractal analysis has been used to describe the regularity of fracture patterns in geological materials seen on a two-dimensional surface. This linear method of analysis is well adapted to a determination of two-dimensional phenomena. The determination of the fractal dimension, D, indicates that it varies regularly as a function of the orientation of the analysis direction. This is taken to indicate that there is a relation, unexplained as yet, between the regularity of the fracture pattern and the forces which produced it. The use of the Cantors Dust method on naturally fractured materials ranging in dimension from 3 mm to 45 m in maximum dimension indicates the fractal nature of the fractures found in natural materials. The analysis method can also be used to identify local variations in the stress field which caused failure, such as the fractures in a feldspar compared to those in the host granite, etc. The method appears to be a sensitive descriptive measure of the failure patterns in rocks.

Compaction trends of clay-rich deep sea sediments

Abstract Analysis of published porosity data for clay-rich (> 50%) sediments from the Ocean Drilling Project, the Deep Sea Drilling Project and deep wells in Recent and Tertiary sediments in Japan and Italy indicate that there is no single, simple compaction (porosity-depth) law for these sediments at all depths. The deep sea, clay-rich sediments have an initial porosity of 75–90% decreasing to near 40% at about 500 m depth. Deeper clay-rich sediments show a reduction in porosity at a different rate from 40 to 10% porosity between 500 and 4000 m. These published data show significant scatter in the shallow samples ( The difference in compaction law for the upper sediments is considered to reflect the resistance of the clay flocculation structure formed upon or just after sedimentation.

Structure of surface cracks in soil and muds

Abstract Image analysis of photographs of surface cracks in cultivated soils for a variety of types (Vertisol, Andosol, Mollisol), and surface cracks in mud deposits shows various similar geometric characteristics. Analysis of the skeletal structure of these crack networks indicates that the relations of intersections and numbers of crack segments show a tendency to lie between those of square and hexagonal networks while the relations between particles and the number of bounding crack segments which define particle contours indicates the presence of more segments than found in a hexagonal network. Crack length distribution appears to be symmetric on log length–linear frequency co-ordinates. Area distributions of pore size or crack width indicate that the cultivated soils have a regular, symmetric distribution on log width–linear area percent occupied co-ordinates while muds have skewed distributions. Particle size abundance (defined by crack boundaries) indicates a non-linear cumulative frequency distribution on a log frequency–log size analysis that suggests non-fractal relations. Measurement of the fractal dimension of crack patterns using box counting indicates that as the porosity increases the regularity of the distribution of the cracks increases. Cultivated soils show the greatest irregularity for a given porosity. Each type of crack structure (soils, muds) has a linear relation on a porosity–fractal dimension plot. The surface area spatial characteristics of the crack network and the particles defined by them appear to be different.

Convenient technique for estimating smectite layer percentage in randomly interstratified illite/smectite minerals

The validity of the saddle/001 method for estimating the percentage of smectite layers in randomly interstratified illite/smectite (I/S) minerals as a routine laboratory technique has been examined with respect to the effects of the crystallite size distribution (N = number of layers) of I/S and the degree of preferred orientation of crystallites in the prepared specimen. X-ray powder diffraction experiments of I/S clays indicated that the crystallite size distribution was 3 < N < 12; these values were supported satisfactorily by the variation of the d(002) value of the samples. An analysis of the Lorentz factor concerning the degree of preferred orientation of crystallites indicated that a calibration curve calculated using the random powder Lorentz factor and the above crystallite size distribution fit the data better than an assumption of perfect orientation. Consequently, if a calibration curve of the saddle/001 ratio is used to estimate the percentage of smectite layers in I/S, an error of 10–15% should be expected from the variable crystallite size distribution of actual samples, in which I/S dominates over other phyllosilicate phases that give reflection between 10–14 Å. This method is useful, however, in estimating the relative percentage of smectite layers in randomly interstratified I/S for samples examined under identical experimental conditions.

Evolution of structural and physical parameters of clays during experimental compaction

Abstract To provide a better understanding of sedimentary basin geological history, it is important to describe correctly the evolution of the various physical, mechanical and hydraulic properties of clayey rocks as a function of burial depth. As a contribution to this field, a programme of experimental studies on reworked clay samples compacted under various load pressures in oedometric conditions has been set up. The evolution of samples under compaction was followed with microscopic and macroscopic measurements. In a specially designed oedometric cell, samples are compacted under different total stresses from 0.1 to 50 MPa. In this cell, cylindrical cores are submitted to progressive loading from both ends under controlled pore pressure conditions. The symmetrical loading allows more even deformation about the midplane of the sample. This device allows the evolution of hydraulic pressure, radial stress, displacement and expelled pore fluid to be followed as a function of time. In a first step, kaolinite was chosen because it retains a high permeability (compared with other clays such as illite or smectite), which allows compaction tests to be performed within a few days. A complete set of measurements was performed after the tests. These measurements are: (1) micro-structures investigated by means of transmission electron microscopy (TEM), mercury porosimetry, water removal under low water vapour pressure, granulometry and specific area measured by ethylene glycol adsorption; and (2) various physical parameters measured including hydraulic conductivity and thermal conductivity. TEM gives an understanding of the arrangement of particles. It was found that each particle is composed of several crystal units, each unit formed by ≈25 individual kaolinite layers. During compaction, these particles remain undeformed, but are rotated. The angular distribution of grain orientation is a function of the applied effective stress. This reorientation is in agreement with the observed decrease in porosity and pore size. It also explains the occurrence of a strong anisotropy in the thermal conductivity and hydraulic permeability. The combination of these experimental results allows a qualitative and quantitative understanding of the behaviour of kaolinite with respect to parameters such as permeability, porosity, mechanical and thermal properties, the knowledge of which are necessary for basin modelling.

Fractal patterns of fractures in granites

Abstract Fractal measurements using the Cantors dust method in a linear one-dimensional analysis mode were made on the fracture patterns revealed on two-dimensional, planar surfaces in four granites. This method allows one to conclude that: (1)|The fracture systems seen on two-dimensional surfaces in granites are consistent with the part of fractal theory that predicts a repetition of patterns on different scales of observation, self similarity. Fractal analysis gives essentially the same values of D on the scale of kilometres, metres and centimetres (five orders of magnitude) using mapped, surface fracture patterns in a Sierra Nevada granite batholith (Mt. Abbot quadrangle, Calif.). (2)|Fractures show the same fractal values at different depths in a given batholith. Mapped fractures (main stage ore veins) at three mining levels (over a 700 m depth interval) of the Boulder batholith, Butte, Mont. show the same fractal values although the fracture disposition appears to be different at different levels. (3)|Different sets of fracture planes in a granite batholith, Central France, and in experimental deformation can have different fractal values. In these examples shear and tension modes have the same fractal values while compressional fractures follow a different fractal mode of failure. The composite fracture patterns are also fractal but with a different, median, fractal value compared to the individual values for the fracture plane sets. These observations indicate that the fractal method can possibly be used to distinguish fractures of different origins in a complex system. It is concluded that granites fracture in a fractal manner which can be followed at many scales. It appears that fracture planes of different origins can be characterized using linear fractal analysis.