main-content

## Inhaltsverzeichnis

### Auflockerung und Verwitterung in der Ingenieurgeologie: Übersicht, Feldansprache, Klassifikation (Verwitterungsprofile) — Einleitender Beitrag

This introductory chapter deals with the significance and field Classification of weathering profiles in different important rock types. For a first description of rocks and soils in engineering geology, the distinction of 5 stages of weathering (apart from fresh parent rock) are recommended. The term “degree” of weathering should be used for more quantitative to semi-quantitative characterisations.Near a land surface lowered by denudation, all rock types are affected by stress release (fig. 1). As a result, the rock mass starts to disintegrate along pre-existing and newly developed fissures, joints, and other discontinuities. In this stage, the material of sedimentary rocks, particularly that of mudrocks, is still overconsolidated and more or less diagenetically indurated. These properties are successively lost during weathering. For rock types containing clay minerals, the “inactivation” and “reactivation” of these minerals play a great part in these processes. Therefore, kind and amount of clay minerals are used to distinguish between different rock types (fig. 2) which, for example, behave differently when they are kept under water or are exposed to repeated drying and Wetting.The weathering profiles can be subdivided into 2 groups (fig. 3). In the first one, the rock material is relatively resistent to weathering and remains hard nearly up to the land surface. In the second group, the rock is disintegrated and decomposed already at greater depth. Both groups can form either non-cohesive or cohesive matrix respectively soils in stages W4 and W5. Furthermore, one should distinguish between regulär, surface-oriented weathering profiles and irregular ones which may be caused by oxidation and dissolution of circulating groundwater and/or by special tectonic fracture zones and faults (fig. 4). Examples of surface-oriented weathering profiles including some explanations are shown for the following rock types: Granite and basalt (fig. 5), sandstones and bedded limestones (fig. 6), mudstones and tili (fig. 7).Finally, some consequences of the weathering profile for shallow foundations, permeability, angle and stability of slope cuts, “back-weathering”, development of failure planes, ground-heave, and dam construction are briefly discussed (fig. 8).

G. Einsele, K.-H. Heitfeld, Ch. Lempp, K. Schetelig

### Stabilität und Verwitterungsverhalten natürlicher und künstlicher Böschungen in Kalk-Mergelwechselfolgen

The aim of this project was to quantify the influence of lithology and dip on the inclination of natural and artificial slopes in limestonemarlstone sequences. In order to compare the results, the distinction must be made between slopes with and those without active basal erosion by a watercourse. Such erosion will lead to a limit equilibrium profile of the slope. Where basal erosion is not present, the inclination of the slope will be controlled by different weathering mechanisms (slaking in marlstones, leaching in limestones). Thus the resulting inclination reaches a critical gradient, which is controlled by weathering mechanisms.The lithology of the slope was classified by means of a limestone to marlstone ratio. This was done by dividing the slope into sections of overall uniform lithology. For each section the average thicknesses of the limestone and marlstone beds were determined. The quotient of these two parameters is the limestone-marlstone ratio.In order to have a better understanding of the influence of lithology on the inclination, only slopes lying within a quasi-homogeneous area (that is, identical dip and joint sets) were compared. In doing so, only slopes with beds dipping against the inclination of the slopes were used. Within the different quasi-homogeneous areas an exponential relationship exists between the slope gradient and the limestone-marlstone ratio(figs. 2 & 3). For equilibrium profiles not limited by basal erosion, the inclinations have been estimated mathematically by a simple model (secondary toppling) and compared to field measurements (fig. 7).Furthermore the rate of retreat of vertical walls due to weathering has been determined in abandoned quarries in limestone-marlstone sequences (fig. 9). The amount of talus accumulated at the bases of step slopes during a certain time interval can be calculated from these rates of retreat.

Dietrich Rupp

### Geotechnische Eigenschaften von überkonsolidierten Schiefertonen bei unterschiedlichen Sand- und Karbonatgehalten

The rocks of the Middle Dogger in southwestern Germany provide excellent examples of several different kinds of weathering profiles. These rocks contain numerous natural drainage bodies such as jointed sandstone or calcareous sandstone complexes. These drainages can regulate both the depth and the number of weathering profiles (fig. 3). Vertical repetition of such natural drainage bodies throughout the section can create, in effect, multiple distinct zones of weathering.The investigation of a large number of core drillings in the Middle Dogger demonstrates the existence of numerous horizontal plastic interlayers (ranging from mm to cm in thickness) which were formed as a result of horizontal stress releases and which occur mostly in zones of relatively low shear strength (fig. 4). These layers act as zones of latent weakness which can have significant mechanical effects on artificial slopes or deep excavations which may be constructed in these rocks.Weathering tests simulating various climatic conditions have demonstrated that carbonate and sand components influence the weatherability of overconsolidated shales present in the Middle Dogger section (fig. 5). For example, a granular skeleton cemented with carbonate allows only a small amount of deformation. This deformation is the result of shrinkages caused by desiccation (tab. 2) and thereby initiate deconsolidation processes (EINSELE 1983). After undergoing three successive dry/wet deterioration tests, the primary, unweathered shales with a carbonate and sand content of more than 40 — 50% demonsträte only a small amount of slaking. However, if the carbonate and sand content exceeds 70%, the amount of slaking is too small to measure (fig. 5).California Bearing Ratio (CBR) tests were also conducted with the Middle Dogger material. The results of these tests show bearing capacities which can be achieved when the investigation material is reapplied as embankment backfill. Unlike sandless shales or a mixture of Middle Dogger material consisting of 25% very sandy shale, 25% friable sandstone and 50% weathered sandless shale, the normally Consolidated, unweathered investigation material can be classified as “average” to “good” subsoil according to the CBR-norm (fig. 15). The low bearing capacity of sandless shale can be improved by adding friable sandstone material.

Clemens Ruch

### Ursachen und Mechanismen des Trocknungs-Befeuchtungs-Zerfalls überkonsolidierter Pelite

Disintegration of overconsolidated pelites is caused by dry-wet and freeze-thaw changes. For this investigation, the influence of material properties of the rock on the one hand and climatic factors on the other hand were investigated both in laboratory tests and site observations.The observations of the disintegration process are related to material properties such as grain size distribution, mineralogical composition, microstructure, porosity, specific surface and the ability to dry out with and without water supply. The influence of all these properties is examined. The climatic influence on Opalinuston (middle Jurassic) has been investigated in detail and compared with the influence of characteristic indices of the material.Disintegration is initiated by desiccation and shrinkage, which first lead to compaction and an increase of strength of the overconsolidated pelites. However, contact between the clay particles is weakened by desiccation and the structure collapses when rewetted.The extent of the initial increase in strength depends on the degree of clay particle orientation. The preferred orientation is independent of the rate of diagenesis.The decrease in strength by rewetting is decisively influenced by the degree of diagenesis. With increasing diagenesis the ability of water adsorption decreases as does the intensity of the particle disintegration.The increase of material strength resulting from shrinkage is opposed by a decrease in strength of the rock caused by fissuring. Rewetting with water or water vapour causes an additional disintegration of the already weakened material.

Irene Merklein-Lempp

### Einfluß organischer Substanzen auf die verwitterungsbedingte Entfestigung von Peliten am Beispiel der Posidonienschiefer

Organic content significantly influences strength and weathering behavior of rocks. This correlation has been demonstrated with laboratory tests and field observations for the Lower Toarcian Posidonia Shales of southwestern Germany (Lias epsilon).Two mechanisms influence the weathering processes: Physical disintegration during drying and Wetting cycles, and chemical oxidation of the long-chain organic molecules linking the mineral components. The weathering behaviour of the Posidonia Shales and other pelitic rocks with low carbonate contents is controlled by a threshold value of organic carbon. Slaking occurs if the amount of organic carbon is below approximately 4 percent (fig. 2). When the organic content is higher, slaking does not take place, but bedding-parallel exfoliation can be observed (fig. 1). Weathering profiles are not highly developed, due to the large weathering resistance of the Posidonia Shales. It is usually possible to reach almost unweathered bedrock at a depth of around 1 to 2 meters below ground surface. A distinct dependence of rock strength on organic content has been proven by uni-axial compression tests (fig. 4).

T. Irouschek

### Veränderung bodenmechanischer Parameter bei überkonsolidierten Peliten durch Verwitterung — Versuch einer Deutung anhand des Opalinustones

A quantitative method for the determination of the degree of weathering for overconsolidated claystones is presented. Plastic and variable hard claystone particles are separated in a specially designed rotation cylinder (fig. 2) by suspension. Since the degree of weathering is defined by the ratio of plastic particles to the sum of plastic and variable hard particles the determination can be easily done (fig. 1). After applying a different number of rotations the “degree of fine-fraction” is determined by evaluation of the grain-size distribution in a semilog plot. As shown in fig. 3 the latter degree is defined by the ratio of the area under the grain-size curve to the area over the grain-size curve in the ränge 0,002 mm to 0,04 mm. The plot of the so determined “degree of fine-fraction” versus the number of rotations is called “fineness-line”, which is cut by an empirically determined straight line in the ränge of total Separation of components. The intersection of “fineness-line” with the straight-line as shown in fig. 4 leads to the degree of weathering.For a certain weathering profile in a claystone (Aalenium) the relationship of soil mechanical parameters to the degree of weathering was examined. The conditions observed are described by a general qualitative model of weathering, basing on a paper of OSIPOV (1975) with the theoretical background of different types of bondings between particles in clay (fig. 9). OSIPOV distinguishes bet-ween 3 types (fig. 8) : coagulation- type of bonding with plastic behaviour, condensation type of bonding, and cementation type of bonding with a brittle behaviour. Düring Sedimentation the coagulation type of bonding is predominant, because it is the most stabil bonding type under normal conditions with respect to thermodynamics. Düring the diagenetic process the hydrated zone of the clay particles are thinned out and break at this points of contact. Thus, the bondings of coagulation at these points are converted into bondings of condensation or cementation resp. After removing the geological loading the bondings of condensation remain in a metastable condition. Weathering is to be understood as a process, which makes the phenomenon regressive. The bondings of condensation and cementation are destroyed and reversely converted into bondings of coagulation.The mechanical properties of the material at each stageof weathering are determined by the amount and the spatial arrangement of the Single bonding modes.Düring the first stage of the weathering process (W 1 - W 2) a mechanical destruction of the bondings of condensation takes place caused by deloading forces. Widening of fissures and horizontal micro-structure are the consequences. Shear strength determinations by means of large scale in situ shear tests and laboratory triaxial tests show in this stage a more remarkable decrease of shear strength parallel to Sedimentation than vertical because of the dominating widening of horizontal structure in relation to widening of vertical fissures (fig. 7).Düring the second stage (W 2 — W 4) the destruction of the bondings of condensation is spatially spread by stresses, which are produced by processes of swelling and shrinking. At the same time the reverse acting formation of coagulation bondings is starting.The third stage is characterized by an intensive formation of coagulation bondings connected to a plastification of the material. An additional increase of the water content and a decrease of the density are the consequences (fig. 5). The increased influence of oxydation and a permanent forming of acid maintain the plastifying process. The increase of base exchange capacity shows that the reactivation of clay minerals has taken place as indicated* by growing portion of coagulation bondings (fig. 6). As could be shown yet by microscopic and X-ray-refraction measurements only in this stage a spatial net of the brittly type condensation bondings is so much weakened that a decisive destruction of the micro parallel structure takes place (fig. 10). By means of the conversion of the weathering process a reverse of the originally overconsolidated claystone into an unconsolidated plastic clay takes place.

K. F. Henke, M. Hiller

### Zustandsbeschreibung und Verwitterbarkeit des eozänen Messeler Ölschiefers

In the abandoned oil shale pit “Grube Messel” fossils are excavated and rubbish ought to be deposited (fig. 1). The diggers and the deposition are threatened by slope failures. In order to investigate the slope stability the soil physical parameters of primary indisturbed core samples were determined (table 1 and 4).The material can be divided into a bituminous clay/silt mixture. The elutriation analyses or the combined sieve- and elutriation analyses proved that all layers have a fluctuate portion of clay between 57 – 8% (fig. 2). The water absorption of a desiccated sample is very slow; at first after 24 h 0,4 were obtained and after a month 0,74 (fig. 4). By reiterated desiccation and moistening the water absorption increased, after ten repeated interactions a water content of up to 0,50 was obtained. In vacuum the absorption capacity increases faster and stronger 0, 61 (fig. 5). In spite of a high content of active clay minerals the water content will not increase after ten repeated remoistening phases because the montmorillonites will be masked by kerogenes. The falling into pieces prevails.The laboratory tests for weathering of the material were accomplished in six series as pointed out (fig. 6): 1)already after one desiccation an evident Separation of beds is distinguishable,2)the influence of frost action is of less importance, it increased under vacuum conditions,3)the series of investigation “desiccation-moistening” is the decisive criterion for desintegration,4)the series of investigation “A — B in vacuum” (at 3 0 mbar residual pressure) lead to a comparable stronger comminution.The uniaxial shear test yielded strongly deviating values for the angle of internal friction in relation to bedding and shear planes. In the case of steeper inclined layers the angle of internal friction obtained an angle of 46° and in the case of horizontal bedding an angle of 26°. That is important because the layers have a steep inclination. The compression tests in the case of horizontal layers at 80 N/cm2 obtained a final settlement of 0,39 mm (tab. 3). The pressure of overlying strata (here max. 150 m) is of no distinguishable influence on the geological preloading and on all the soil physical parameters.

### Mineralbestand, Gefüge und bodenmechanisches Verhalten tropisch verwitterter Vulkanite

The investigation of tropically weathered andesites from Costa Rica and of basalts from the Vogelsberg area, subject to tropical-subtropical weathering during the younger tertiary shows changes in mineral composition with different degrees of weathering (tab. 1, 3).A sequence of Smectite/Allophane ➛ Halloysite/Metahalloysite ➛ Kaolinite Gibbsite was found which might be modified especially during the early phase. The weathering process results in decreasing base Saturation and an increase of sorbed Aluminium-ions and crystalline Al-Oxides (tab. 1, 3). The relative enrichment of iron oxides causes the change to a brick-like colour in the gibbsite stage or even earlier.The change in mineral composition is accompanied by the development of a different rock texture. The jointed rock bodies can no longer be distinguished in the kaolinite stage of weathering or might only be left as relic structures. Even in the smectite/allophane stage where the texture seems to be intact the strength decreases dramatically.According to the appearance of rock bodies the outcrops in Costa Rica must be assigned to weathering stages W 2 and W 3, the two outcrops in the Vogelsberg to W 3 — W 5 and W 2 (fig. 6). W 2 and W 3 representing the smectite/allophane and halloysite/metahalloysite stage respectively, W 4 the kaolinite and W 5 the gibbsite stage.In all weathering stages the rock bodies or ground mass is characterized by an aggregate structure. Even though the material consists mostly of clay minerals it decomposes usually to a sandy-silty system. This behavior is a result of the Al-bonding and the positive Charge of Fe- and Al-Oxides and allophane caused by the low pH-values. The aggregates (fig. 3) as cellular-spongy structures show a high water binding capacity, which results in high liquid and plastic limits and a low plasticity index (fig. 4, 7); The aggregate structure and the stability of the aggregates cause relatively high friction angles. For the Costa Rica samples they are between 30° and 40° (tab. 2). The Vogelsberg samples show lower values but higher cohesion and commonly higher clay mineral content as well (tab. 4).The material is exposed to extremely high precipitation in Costa Rica. Street embankments usually show very steep slopes which prevents immediate infiltration. Landslides commonly occur where water is seeping into the embankments through flow Channels.

K. Knoblich

### Bestimmung eines Parameters für den Verwitterungsgrad von Tonsteinen

New parameters to determine quantitatively the degree of weathering and the engineering properties of I-lesozoic mudstones and marls:The engineering properties of mudstones and marls of Keuper and Jura in Southern Germany have been observed to be also governed by the degree of weathering as it was determined earlier in England (see e.g. CHANDLER 1969). Although,the influence of weathering on the engineering properties is very important for foundation engineering analyses, there has neither been an exact quantification of the degree of weathering nor any quantitative relation between the degree of weathering and the engineering properties so far. Therefore, this paper is concerned with the determination of a quantitative parameter for the degree of weathering, which may be useful to evaluate engineering properties, especially shear strength parameters.With the research project reported here, a great number of index properties, used in former time to determine the degree of weathering, have been determined such as natural water content, plastic and liquid limit, clay fraction, aggregation ratio, Fe(III)/Fe (II) ratio, FeOOH content, quantitative mineralogical composition, effective shear strength parameters, etc. If there has been a homogenous mineralogical composition in an originally unweathered profile, e.g. Opalinuston, some of these parameters (fig. 2) can be used to deterimine quantitatively five weathering zones and it is also possible to relate characteristic shear strength parameters to each weathering zone in a weathering profile of the same geological history. But these parameters do not apply, if the mineralogical composition is inhomogenous, e.g. Gipskeuper, or if weathering profiles and engineering properties of different geological histories shall be compared.For this purpose, new parameters are introduced. With the “Quotient PF” the most important variations both in terms of mineralogical composition and gradation are described quantitatively. For this purpose the actual value is (1)$${\text{PF = }}\frac{{{\text{clay fraction}}}}{{{\text{solid compounds - clay fraction}}}}{\text{ }}$$ and the value for the theoretical end of weathering (2)$${\text{PF}}_{{\text{END}}} {\text{ = }}\frac{{{\text{clay - mineral content}}}}{{{\text{quartz content ( + feldspar)}}{\text{.}}}}$$The “Zustandsindex TW” is a mean value of three figures which are related to either the natural water content, the plastic limit, and the liquid limit. It is standardized such that IW=0 applies to wP=wL and Iw=5 to wP=w respectively, see fig. 1. The correlation which was determined is given by the following equation: (3)$${\text{I}}_{\text{W}} \;{\text{ = }}\;{\text{1/3}}\;\;{\text{\{ (w}}_{\text{L}} {\text{ - 10)}}\;{\text{/7}}{\text{.9}}\;{\text{ + }}\;{\text{(w}}_{\text{P}} \;{\text{ - }}\;{\text{10)}}\;{\text{/2}}{\text{.4}}\;{\text{ + }}\;{\text{w/4}}{\text{.47\} }}$$Then a significant correlation is found between IW and PF such as (4)$${\text{I}}_{\text{W}} \;{\text{ = }}\;{\text{6}}{\text{.6}}\;{\text{PF}}^{{\text{1/2}}} \;{\text{ - }}\;{\text{0,75}}\quad {\text{and}}\quad {\text{Iw}}_{{\text{END}}} \;{\text{ = }}\;{\text{6}}{\text{.6}}\;{\text{PF}}^{{\text{1/2}}} \;{\text{ - }}\;{\text{0}}{\text{.75}}\;{\text{.}}$$By means of these parameters, an “index of weathering VM” can bee defined as (5)$${\text{V}}_{\text{M}} \;{\text{ = }}\;{\text{I}}_{\text{W}} {\text{/Iw}}_{{\text{END}}} \quad \quad {\text{or}}$$(6)$${\text{V}}_{\text{M}} \; = \;\frac{{1/3\;\left\{ {({\text{w}}_{\text{L}} {\text{ - 10}})/7.9 + ({\text{w}}_{\text{P}} {\text{ - 10}})/2.4\; + \;{\text{w/4}}{\text{.47}}} \right\}}}{{6.6\quad \quad \quad \frac{{{\text{clay}}\;{\text{mineral}}\;{\text{content}}\quad \quad {\text{1/2}}_{{\text{ - 0,75}}} }}{{{\text{quartz}}\;{\text{content}}\;{\text{( + }}\;{\text{feldspar)}}}}}}$$By this index it is possible to compare the engineering properties of diagenetic mudstones and marls having defferent mineralogical compositions and different geological histories.The actual index of weathering can also be used to define a quantitative parameter of the “degree of weathering VR” by reference to limit values of VM:(7)$${\text{V}}_{\text{R}} \;{\text{ = }}\;{\text{100}}\;{\text{(V}}_{\text{M}} {\text{ - V}}_{\text{B}} {\text{)/(V}}_{\text{E}} {\text{ - V}}_{\text{B}} {\text{)}}\quad {\text{\% ,}}$$where VB is the index of unweathered matter and VE is the index of completely weathered matter. Both values are to be determined for a given weathering profile.The new parameters are also appropriate to describe quantitatively weathering zones in a homogenous weathering profile (fig. 3), but more over, they are also fitted to calculate correlations between the degree of weathering and the shear streght parameters (tab. 1 and 2) for the actual state and the possible state for the time when the process of weathering is finished.

P. Keller, K. Mack

### Verwitterungsabhängige geochemische, mikrostrukturelle und geomechanische Eigenschaften halbfester Tonsteine der oberen Röt-Folge (Rhön)

The weathering State of lower Triassic shales (Upper Bunter, Rhoen mountains, Germany FR) was characterized by mineralogical, geochemical, micro structural and geomechanical methods comparing materials of near-surface weathering profiles and cores from a 50 m deep bore hole (fig. 1). The clay fraction increases towards the Earth’s surface. The predominant clay minerals are illite, chlorite and corrensite. The mobility of major cations, determined by selective extraction methods, increases with increasing weathering intensity. Due to the dissolution of (dolomitic) carbonates and of iron Compounds the stability of intergranular bonds is loosened and is not compensated by newly formed minerals. It results in a mixture of unweathered mineral aggregates and plastified clay matrix The depth distribution of the stability of the major minerals and the tendencies of mineral transformations were confirmed by use of the thermodynamic Computer model WATEQF (fig. 2) and microprobe analysis.The interrelations between the changes of geomechnical properties (e.g. cohesion, compressibility, shear strength, creep and failure behavior (tab. 1 and 2)) and the complex mineralogical, geochemical and microstructural processes (fig. 3 – 5) during weathering are described and their causes are discussed.

K. Gronemeier, G. Mattheß, W. Brunhof, W. Ohse, A. Pekdeger, D. Schenk, A. Winkler

### Ingenieurgeologie aufgelockerter Granite Verwitterung und Auflockerung — Einführung und Versuch einer Klassifikation

The degree of disintegration of rock masses and rock materials defines their use in construction foundations, excavations, underground cavities as well as the extraction of construction materials. Within the framework of the research programme “Engineering Geology” of the German Research Society (Deutsche Forschungsgemeinschaft), an interdisciplinary working group at Karlsruhe University investigated various aspects of weathering and disintegration of granitic rocks. One of the primary objectives of this research project was to quantify disintegration and its effect on the geotechnical properties of such rocks.In light of the results obtained, it seems necessary to distinguish between mechanical disintegration and chemical alteration. In addition, it is appropriate to differentiate mechanical disintegration in regard to its macro- and micro-scale features, that is the system of discontinuities in a rock mass as opposed to the fabric of micro-fractures in the rock material.Two tables of characteristics, used to determine the extent of mechanical disintegration and chemical alteration of rocks, are proposed for a general field Classification. These tables allow Classification on the basis of 10 characteristics for both the rock mass (table 1) and the rock material (table 2) with average numerical values between zero (indicating no sign of disintegration or decomposition) and five (representing completely disintegrated and decomposed material). Certain stages of weathering (“W 0” to “W V”) D are also recommended for usage in conjunction with these numerical values.In addition the degree of weathering or disintegration of granites can be expressed quantitatively using three diagnostic characteristics: a)frequency of joints in rock massb)frequency of micro-fractures in rock materialc)extent of chemical weathering (proportion of newly formed clay minerals in rock material) The relative importance of these three characteristics may vary depending on the application. Therefore a general system, classifying degrees of weathering with a quantitative scale based on a), b) and c), seems not very useful. Such a general Classification system should be restricted to a more qualitative and general description of a rock mass, as previously described.

J. Brauns, H. Hötzl, K. Kast, Ch. Lempp, F. Metzler, W. Smykatz-Kloss

### Der chemische Verwitterungsgrad von Gesteinen als Maß für ihre ingenieurgeologische Verwendbarkeit

Chemical weathering of rocks is mainly due to hydrolysis. That is, the weathering cycle includes processes of dissolution of primary rock constituents (e.g. see fig. 1) and of precipitation from weathering solutions. These precipitation products (see figs. 2 and 3) include (OH) — ions or water or both. Thermal methods of analysis are well suited to characterize the influence of hydrolysis on rocks, e.g. by determining the amount of newly formed weathering products (sheet Silicates, hydroxi des). In the present paper a differential thermal analytical method is briefly described which allows the determination of the degree of chemical weathering (= the amount of neoformations, that is the formation of new minerals, expressed as a percentage of the weight of the whole rock) of granitic and basaltic rocks. The method uses the fact that all observed neoformations include (OH). Therefore, a weathering index can be calculated from the determination of all dehydroxylation effects as seen in Standard DTA curves (see fig. 4). For this, the sum of the peak areas multiplied by the sum of the values of the peak heights (= △T) of all dehydroxylation effects is used (figs. 4 – 6). A scaled relationship of clay minerals (e.g. kaolinite in granitic and nontronite in basaltic rocks) to the weathering index is a suitable measure of the degree of chemical weathering of siliceous rocks (such as granite, basalt, gabbro, gneiss) which are poor in primary OH-bearing minerals. The method offers a quick and reliable tool for engineering geologists to estimate the applicability of siliceous rocks.

W. Smykatz-Kloss, J. Goebelbecker

### Mechanische Eigenschaften von Störungen und Verwitterungszonen im Granitgebirge und deren genetische Charakterisierung

The Hercynian granites of the Schwarzwald area in southwestern Germany display distinct variations in the amount of disintegration and weathering. Different stages of transition ranging from unweathered rock to cohesionless material can be observed in the same level of erosion. The process of disintegration and weathering is accompanied by a continuous decrease in strength. Several material properties, such as dry-density or porosity (fig. 2, 3), creep behaviour (fig. 4, 5) and elasticity (fig. 5, 6) change in a characteristic manner correlating to the decrease in strength. Non-elastic properties of the granite become dominant during disintegration. The granites of the Schwarzwald area can be classified into six stages of weathering, where each stage is characterized by certain mechanical properties (fig. 7).This Classification in terms of mechanical properties combines the following parameters: 1. uniaxial peak strength, ßD, 2. dry-density, ρd, 3. strain rate ratio, εlateralaxial, 4. creep strain rate per 2 minutes under one half of the uniaxial peak stress, 5. percentage of peak strain rate, % εBR which depends on uniaxial stress in a nonlinear manner, 6. modulus ratio, Ereloading/E1st loading.Tentatively the problem of determination of rock mass strength is investigated by the aid of laboratory tests. The results of direct shear tests on natural joints of the granites (fig. 8) were used to estimate a lower limit of the strength of the granitic rock mass. Discontinuities as well in the mineral structure as in the rock mass structure cause a similar effect on the p-wave-velocity (fig. 9). In connection with this result it appears plausible that a disintegrated specimen may react like a jointed rock mass. Accordingly the result of multi-step triaxial tests on rock specimens (fig. 10, 11, 12) were used to define an upper limit of rock mass strength of the Schwarzwald granites. The strength of the gran itic rock mass is also classified depending on the mechanical properties of the rock and its discontinuities (fig. 13).The decrease in strength and the change in mechanical properties are Chiefly caused by increasing microcrack density, disintegration of miner al structure and increasing porosity. Mineralogical and chemical changes especially formation of phyllosilicates, promote disintegration but in all types of granite in the Schwarzwald area occur conditions of only slight chemical weathering intensity.Disintegration and decrease in strength are mainly the results of mechanical, tectonic and thermal stresses.

Ch. Lempp, O. Natau

### Durchlässigkeits- und Injektionsverhalten aufgelockerter Granite

Within the present research project investigations were made regarding the groutability of disaggregated granites in the Black Forest, Southwest Germany. For comparison and better interpretation of the results general geological methods as well as hydrogeological, engineering-geological and engineering-geophysical methods were applied. A preliminary step was the characterization of rock mass disaggregation with special regard to its importance for the permeability and water routing. For the registration of the direction of propagation special tracing and permeability tests as well as seif-potential measurements have been carried out. Exact information on the preferred water paths. has been obtained by joint tracing with highly adsorptive dyes (e.g. rhodamine). The application of this method, however, is limited, since the Observation of dyed joint planes requires exposure of the affected rock. To detect permeability anisotropy caused by the joint pattern additional geoelectrical seif-potential measurements proved to be most effective. This method is based on the location of increased flow phenomena thus indicating preferred underground passages. Furthermore, the direction of water propagation can be proved by watergauge observations during water pressure tests.Water absorption was investigated by numerous water pressure tests after LUGEON in three selected test areas showing all transitional stages from solid rock to granitic gruss. Injection pressure and corresponding flowquantities have been registered synchronously. Due to the resulting p,q-diagrams and due to water-gauge and surface observations it was well possible to identify erosional processes of joint fillings. These phenomena were found in 50% of 128 performed packer tests. Injection pressure has been measured both in water pressure tests and grouting experiments directly in the test section.The grouting experiments were carried out on the analogy of water pressure tests, i.e. with Single packer in steps from bottom to top by means of a variable adjustable Mohno pump. The quantity measurements were carried out by an automatic dosage reservoir and a magnetic inductive flow meter. The water/cement mixtures were prepared in a colloid mixer (C-Mix). Cement mixtures with W/C factors of 0.6, 0.8, 1.0 and 2.0, each with the addition of 3 percent bentonite, were applied. In all grouting tests an oil shale cement (ÖZ 4 5 F) providing a Blaine index of 6047 (± 124) cm2/g and a grinding size of R = 0.09 mm was used. The rheological parameters of the individual cement mixtures and the Sedimentation behaviour were previously determined under laboratory conditions.The relatively wide-jointed and unweathered first granite type showed heterogeneous LUGEON values, due to individual, particularly hydraulic active joints with larger widths (0.3 – 2.5 mm). Mean values of water and grout absorption were 21.6 LUGEON and 30.1 kg/m. The mean ränge of cementation was 1.90 m.The second granite type, characterised by a higher joint density and reduced widths (0.2 – 0.5 mm) presented a slightly more homogeneous distribution of the LUGEON values. Average water absorption was 20.1 LUGEON with a grout take of 15.3 kg/m and a mean ränge of cementation of 1.65 m.For the third granite type an almost homogeneous distribution of water absorption across the whole tested area was found. Due to weathering processes leading to a granite gruss the joint pattern was almost completely removed. Besides this, the influence of granular disintegration zones and a mylonitized fault were found to be relevant factors for local deviations of the LUGEON values. Here the mean value of water absorption and grout take amount to 11.2 LUGEON and 13.4 kg/m at a mean ränge of cementation of 0.65 m.By combination of the mentioned hydro- and engineering geological insitu tests with additional engineering geophysical measurements it was possible to obtain relevant parameters permitting a geotechnical Classification of disaggregated granites.

F. Metzler, A. Blinde, H. Hötzl

### Ingenieurgeophysikalische Erfassung der Auflockerung und des Durchlässigkeitsverhaltens von Graniten

The engineering geophysical investigations were designed to determine geophysical parameters permitting to classify the disaggregation of rock and rock masses. The investigations described in this study were performed in close interdisciplinary Cooperation with the project “Classification criteria for soil improvement by grouting of disaggregated rocks” (see METZLER et al., this volume). The three test sites are situated in the Middle Black Forest, SW-Germany. There the engineering geophysical measurements were carried out simultaneously with water pressure tests and grouting experiments in order to obtain details about soil improvement by grouting.A major aspect in the performed seismic investigations was to detect and classify the joint-depending anisotropy of the rock masses with regard to direction and depth by means of travel-time measurements (see fig. 1 and fig. 2) and by attenuation of the seismic waves (fig. 8). A formula, derived from the time-average-relation, concerning the velocity in unweathered rocks (matrix; Vg), the measured mean velocity (Vm) and the velocity in the cleavage space (Vf), permitted the determination of a joint coefficient $$(\tilde K)$$ for the measuring ranges in the three different test fields (see fig. 2). Additionally, the Vp and Vs velocities were P s measured at borecores. This helped reveal the relations existing between the anisotropy of rock and rock mass.It was tried by comparative measurements before and after grouting (see fig. 4, 5, 6, 7) to determine in more detail the mechanical degree of improvement in the rock mass by means of seismic measurements.The increasingly applied electric self-potential measurements carried out simultaneously with the water permeability and grouting experiments showed (see fig. 3 and 9) that this technique proves the best differentiation of the oriented permeability for water and grouting material depending on the joint pattern. Self-potential measurements, performed on a dam model in the laboratory (see fig. 10 and 11), helped to work out theories and data concerning the quantification of the self-potentials (filtration potentials) in the field, in water pressure tests and grouting experiments.Due to experimental working procedures, the engineering geological and engineering geophysical investigations helped to improve the correlation of the relevant parameters concerning the Classification of granite weath ering and disaggregation.

G. P. Merkler, A. Blinde, H. Hötzl

### Verdichtungs-, Verformungs- und Sättigungsverhalten von Schüttungen in Abhängigkeit von der geologischen Gesteinsentfestigung

The object of this research project was to determine the influence of geological disintegration of rock material on the mechanical properties of granitic rockfills. As a first step an appropriate method to quantify the degree of geological disintegration of granites was developed. Comparative investigations led to a characterization based on micro-mechanical and chemical properties, i.e. on the frequency of microfractures and the content of newly formed clay minerals in the rock material (fig. 2).Numerous oedometric and triaxial compression tests on various granite fills with different degrees of disintegration (ranging from sound to severely decomposed) revealed the influence of disintegration on the strength and stress strain characteristics.The preparation of realistic samples of grain breakage sensitive rockfill materials with the help of dynamic compaction was used in this project. The method and the intensity of compaction has a great influence on the mechanical behaviour of such materials.The Settlements of fills due to first flooding, an important aspect in embankment dam engineering, and the necessary pre-compaction to avoid such Saturation Settlements can both be estimated using the methods developed in this research project.The main conclusions from this research project may be summarized as follows: a)Dependent on the geological disintegration, rockfill materials show a characteristic particle breakdown behaviour during compaction (fig. 5). The structure of such a fill is mainly determined by the process of breakdown and this structure governs the mechanical behavior of such fills.b)Under consolidation or shear stresses, dry rockfills and rockfills with natural moisture content exhibit a stiffer behaviour and a higher shear strength than those which are watersaturated (fig. 8). Saturation of a dry sample, under a certain constant load results in a Saturation strain. This strain was found to be equal to the difference between the strains obtained from a dry and a saturated test at the same load (fig. 7).c)Saturation strains, mainly caused by particle breakdown effects during Saturation, are considerably larger under shear than under isotropic compression for the same mean stress (fig. 11).d)Strains due to Saturation can be anticipated by “breakdown compaction”. The necessary specific compaction energy to achieve this, increases as the rock grains become more disintegrated (fig. 6).e)By applying the developed test technique, it is possible to estimate the necessary compaction intensity to avoid Saturation strains (fig.6).The results of this investigation show that it does not seem to make much sense to improve the existing numerical solutions for the calculation of stresses and strains in rockfill dams unless the techniques for testing the material behaviour under conditions as close as possible to those in situ are also improved at the same time.

K. Kast, A. Blinde, J. Brauns

### Statistische Untersuchungen von Rutschungen im Nordbayerischen Deckgebirge

The North-Bavarian cover rock area is a cuesta landscape in the south of Germany consisting of land-, clay- and limestones of mesozoic age (fig. 1 and 2). Every year it is haunted by a great number of land slides.The claystones of the Triassic and the Jurassic are of a special significance in relation to these slides (table 1). The Statistical evaluation of the landslides which were recorded during the past 30 years indicates that most of the slides are exposed to the south or east (fig. 3). This is partly the consequence of an eastward oriented inclination of the layers, but also of the intensive exposure to sunlight which results in a faster weathering of the layers in the ground. Most of the slides were found at the margin of the Frankenalb (fig. 2) or in regions with higher frequency of tectonical faults. The slides are restricted to only a few stratigraphic units. It is noticeable that the majority of the landslides is bound to reactivated sliding planes which were created during the Pleistocene. Neglecting the cohesion, a comparison of characteristics from back-calculations and laboratory tests on slip-surface-material (table 2) indicates a rather good agreement. The examination of climatic and human influences indicated among others, that many slides were found at slopes where man had changed the morphology or the Vegetation for example by the construction of roads or by the clearing of forests.

M. Kany, H. Hammer

### Beobachtungen und Berechnungen zur Böschungsentwicklung bei der Talbildung im wechselhaften veränderlich festen Gestein

Since 1970 landslides in the Isar river valley south of Munich have been observed by the Geological Survey. One landslide spreads over 700 m along the eastern riverside near Gruenwald (fig. 1). The failures usually take place in Miocene marls consisting of silt (siltstone), sand (sandstone) and fissured clay of high plasticity. The marls are overlain by gravel, which is 30 to 40 m thick (fig. 2). The river is deeply cut into these alternating layers by erosion. At a critical depth of the cut the failures start by sinking and spreading and continue as retrogressive slides (fig. 3).This mechanism can be observed also in similar geological formations in many Bavarian river Valleys north of the Alps. The investigations by the Geological Survey will increase knowledge about the failures and will allow to predict them and their development. The applied methods of measuring, their accuracy and the type of results are described in this paper (tab. 1).The displacements are compared with the Variation of the overburden, caused by fluctuations in the river or in the groundwater, by precipitation and by displacement of the center of gravity in the sliding mass. The distribution of the horizontal displacements along the eastern riverside (fig. 1) gives a good impression of the gradual expansion of the sliding movement.The displacement vectors evaluated in several cross sections enable the determination of the failure mechanism, which must be adopted for the calculations. Figure 4 presents several mechnisms of failure and their hodographs, which agree or seem to agree with the observed displacement vectors. They all show a vertical sinking movement in the upper parts of the slope and a horizontal movement at its foot. But there are many cross sections showing very different states of sliding, which influence each other and which must be considered when the mobilized angles of internal friction are computed.In order to simplify the computation each cross section is divided into three parts (fig. 5). Table 2 shows the results of computation at several stages of sliding and different water tables. At one stage (April3 0th, 19 76, low water table) the distribution of the computed driving and breaking forces is plotted in fig. 6.At the beginning of a new failure, the mobilized angles of internal friction were computed between 17° and 21° (tab. 3). The division of the cross sections is shown in fig. 7. But it is not possible to predict the beginning failure exactly enough. It is easier to determine the safety of the slopes by investigating the structures of the sliding mass in the neighbourhood. The best way to find out when a new failure will begin is to measure the movements at the foot of an undisturbed slope.

H. J. Baumann

### Hangstabilitäten im Mainzer Becken

In the Tertiary sediments of the Mainz Basin west of the Rhine slope stabilities have been studied and mapped. Mineralogical investigations of the slip horizons have been made.The immediate cause of the landslides in 1982 were extreme weather conditions. The development of the mass movements is analysed in-situ and by soil mechanical, hydrogeological and air-photogeological investigations. 90 percent of the recent landslides are reactivated old slides.Diverse measurements and fieldwork have been made to delineate the kinematic patterns of the landslides and the forms of their slip planes.For the first time in the Federal Republic of Germany a slope stability map at a scale of 1: 50 000 has been prepared. There is potential danger of mass movements in 8 percent of the studied area. The occurrence of 197 landslides in 1982 demonstrates the necessity of this mapping.A proposal is given for the “semi-quantitative” appraisal of labile slopes.Grain size distribution, chemical and mineralogical composition from samples of slip horizons of recent landslides have been examined quantitatively. The material is very poorly sorted and the samples are characterized by a homogeneous distribution of grain size fraction in the ränge of 0,2 — 20 μm. The clay — silt — ratio is about 1: 1, coarser grain sizes are rare.Considering the clay minerals their chemical composition of the fraction < 2 μm has been analysed and a normative mineral composition was calculated. Hüte in concentrations of 40 – 50 wt. -percent is the dominant clay component. Montmorillonite was found in all samples with 20 – 30 wt.-percent whereas the values of chlorite and kaolinite were smaller than 7 wt.-percent. The calculated mineral content agrees with the data from X-ray determination. Compared with the total sample there are only 33 wt.-percent of clay minerals while the rest consists of calcite/dolomite ≧ quartz ≫ goethite > feldspar. The exceptional good crystallinity of the clay minerals, their grain size distribution and mineral composition indicate that the slip horizons correspond to sedimentary and diagenetical structures.For the development of slip planes and slip horizons the grain size distribution and the mineralogical composition of the clay fraction are important: the silt fraction facilitates rapid water resumption whereas the clay fraction tends to retain the water over longer periods. When sufficient water is supplied this interaction leads to a greater fixation of water and to swelling of the sediment, which greatly reduces its shear strength.

E. Krauter, H. von Platen, A. Queisser, K. Steingötter

### Massenbewegungen im Saar-Nahe-Gebiet unter Berücksichtigung des Verwitterungsverhaltens der Pelite

Mass movements are geologic processes which cause a lot of damage. The object of this project was the investigation of mass movements in Rotliegendes rocks of the Saar-Nahe-region, SW-Germany. Field and laboratory work was employed to achieve a better understanding of the processes.The field investigations showed the following results: Two types of mass movements occur: fall (13%) and slide (87%).The most important cause of mass movements are the geologic conditions, e.g. the parallel run of bedding planes and slope gradient. Additional factors can be water (precipitation, ground water, introduced water) and construction.Most of the mass movements are shallow, i.e. the slip plane is less than 10 m deep.The mass movements take place mostly in winter.7,7 percent of the examined area is endangered to move, 2,6 percent is actively moving.A slope stability map was prepared (fig. 5).The drying-wetting-test, the alternate-wetting-test and the freezethaw- test were used to classify the rock samples concerning their weatherability. The results could be very well correlated (fig. 7,7).The drying-wetting-test is recommended for the Classification of the rocks. It is standardized, cheap and shows the clearest results. Besides, alternate Wetting and drying is the most important factor of weathering in the region in question.The determination of the petrographic-mineralogic and physical properties of the rocks showed, that the weatherability of the rocks is dependent on microtexture, carbonate content, clay mineralogy and water-absorbing capacity. The mean grain size and the consolidation are relatively unimportant.Most of the mass movements occur within the weathering zone. So the weathering behaviour of the rocks was studied in the laboratory. The examinations showed the following: The drying-wetting-test, the alternate-wetting-test and the freezethaw- test were used to classify the rock samples concerning their weatherability. The results could be very well correlated (fig. 7,7).The drying-wetting-test is recommended for the Classification of the rocks. It is standardized, cheap and shows the clearest results. Besides, alternate Wetting and drying is the most important factor of weathering in the region in question.The determination of the petrographic-mineralogic and physical properties of the rocks showed, that the weatherability of the rocks is dependent on microtexture, carbonate content, clay mineralogy and water-absorbing capacity. The mean grain size and the consolidation are relatively unimportant.The investigations showed, that a combination of field and laboratory examinations is a good basis for the evaluation of slope stability.

F. Häfner, W. Fein, U. Held

### Versuche zur Abhängigkeit einfacher bodenmechanischer Parameter bei veränderten Tonmineralgehalten am Odenwälder Buntsandstein

The layers on the boundary Röt/Muschelkalk (lower/middle Trias, Anis) are very strongly susceptible to landslides.Water is passing through the fissured limestones to the substratum of Röt clay-/siltstones, which contain a high rate of active and inactive clay minerals.The frequency of landslides decreases from north to south (200 km) in the same direction as the marinity of the fossil environment. It should be proved whether the content of clay minerals influences the stability of overconsolidated soils.As the content in the northern part (Rhön) is known, two different points of investigations (A and B in fig. 1) were selected. One in the siltstone of Röt, the other in analogue materials in the boundary zone of Perm/Trias (A in fig. 1).The clay minerals mainly consist of illites and kaolinites (fig. 2). Additionally, in some cases montmorillonite and corrensite appear in the Röt-claystone.Grain size distribution of 110 samples (fig. 3) and the diagram of plasticity (fig. 4) of both test series mainly agree with each other, accordingly the angle of internal friction (23 – 25° in Röt material) and the cohesion (0,0025 – 0,035 MN/m2) are close together.A dependence between these parameters and the clay content was not distinguishable.In order to register the influence of active clay minerals a second investigation serie B was started. Different quantities of montmorillonite were now mixed with the basic material (grain size distribution see fig. 5). The specimens of both series were disturbed and homogeneous.Samples with more than 10 percent (–80 percent) montmorillonite content showed — as expected — a distinct dependence on the consistency indices (fig. 6). Samples with > 20 percent montmorillonite belong within the ränge of TA (distinct plastic clays) according to DIN 18 196.The water contents of the liquid limit reach > 100 percent in the end, thereby also the angle of internal friction will be influenced.Resulting from this only the adding of 20 percent and more montmorillonite to prepared samples alters the soil-mechanic parameters so strongly that the stability of slopes is noticeably influenced. Because these differences of clay-mineral contents do not exist in the natural samples taken between north and south of Hesse the environment difference does not influence the instability of the slopes, but the height of slopeä, the water storage, the joints and the exposition will be of more influence for the plastification of the Liegendes and for sliding.

E. Backhaus, F. Schön

### Ingenieurgeologische und bodenmechanische Untersuchungen an Kriechhängen

Investigations of sliding slopes have indicated that the sliding and creeping phenomena occuring in slopes consisting of overconsolidated clay cannot be explained on the basis of shear strength parameters determined in conventional laboratory tests. Thus it follows that when engineering activities take place on such slopes particular attention and improved soil mechanics research methods are required for a reliable prediction of slope stability.The primary object of this investigation was to develop and test such a method. When developing this method procedure special emphasis was placed on geological factors like stratification, softened zones and weathering which influence clay slope stability.Three slopes (located at the western edge of the Schwäbisch Alb; fig.1),on which failure has already taken place in form of sliding or creeping have been investigated, namely:A creeping slope of Keuper marl (Higher Triassic) near Hechingen, a slipped slope (Middle Jurassic) near Thanheim and a creeping Keuper marl slope near Böbingen/Schwäbisch Gmünd. These slopes were mapped and explored using engineering geology methods. Soil samples for laboratory tests were taken from boreholes. Location of the shear bands were determined by inclinomter measurements performed inside boreholes (fig. 7). The groundwater levels within the sliding body were observed in additional shallow boreholes, as the exploration drillings showed a deeper groundwater table which was not significant for the slope stability.Soil mechanics index tests were used for soil Classification and to determine the weathering categories. Specimens obtained from the presumed shear bands were tested in direct shear, ring shear, and triaxial tests. These tests were evaluated with respect to the angle of total shear strength, ϕ’r, and the angle of residual shear strength, ϕ’r. These shear parameters were considered the most suitable ones to model the shear strength of the observed softened bands in the overconsolidated clay.In the backcalculations the sliding mass was assumed to consist of several rigid blocks moving relative to each other and the fixed subsoil along plane slip surfaces (kinematically admissible rigid body failure mechanism, fig. 8). The number, size and dimensions of these blocks were determined on the basis of field measurements and morphological conditions. The values for ϕ’s and ϕ’r obtained by backcalculations were lower than those values measured by shear tests. The difference in the case of direct shear and triaxial tests was of the order of 2 to 5 degrees, while for the ring shear test is was about 2 to 4 degrees. The measured values for ϕ’s varied from 13 to 25 degrees while those of ϕ’r were between 9 and 20 degrees.The following conclusions have been drawn from these investigations: In clay slope stability-analysis geological investigation plays an important role and should be performed in the first stage of the investigation in order to establish an efficient research programme.A reliable prediction of slope stability is very dependent on the location of the shear bands being determined accurately. The location of the shear bands may be found by means of weathering categories, but when slope supporting structures are used the location of the shear bands should be determined by inclinometer measurements (fig. 9).Commonly applied shear strengh parameters as ϕ’, c’ (effective shear Parameters of the overconsolidated clay) and cu (undrained cohesion), determined in conventional tests, are not suitable for slope stability analyses because there is no cohesion from the overconsolidation. The ϕ’s and ϕ’r values also seem to be a little to high and should therefore be used with a higher safety factor. If possible backcalculated values should be used.

G. Gudehus, M. Goldscheider, R. Lippomann

### Ingenieurgeologie und Tonmineralogie vulkanogener Sedimente

Excavation of deep roadcuts in tuffitic sediments of the northern ‘Siebengebirge’ montains, W-Germany, have entailed considerable slope failures. Oligocene tuffs and tuffites of the trachytic Siebengebirge volcanism are deposited on tertiary sand and coal or directly on residual clays of the Devonian basement. Because the sliding planes (inclination 6° – 10°) cannot be explained by the easily determined Standard soil parameters, extensive geotechnical and clay-mineralogical investigations were carried out. The tuffites are of variable grain size depending on their degree of weathering and decomposition. Predominantly, however, they may be classified as fine-sandy, clayey silts.Clay-mineralogical analyses have proved that the tuffites consist of smectites, illites, kaolinites and quartzes, while the thin gliding planes contain over 6 0 percent of highly swellable montmorillonites.The swelling of tuffitic sediments was tested under different conditions in order to establish the relation of swell pressure and deformation. The tests have proved a significant dependence of the swelling potential on following soil parameters: initial water content, liquid limit, anddry density.The shear strength of the material was tested in the ring-shear appa’ratus, considering the shear path and the degree of particle orientation. A new test method was developed which allows to carry out ring-shear tests on normal and overconsolidated samples and constant volume.To eliminate the influence of surcharges the shear stress and the normal stress were standardized by an equivalent normal stress. The tests have proved that the angle of internal friction is a soil-specific parameter and that the cohesion depends upon the void ratio, increasingly with growing overconsolidation. In addition to the ring-shear tests, triaxial cu-tests were carried out and interpreted according to the critical-state-theory.The post-failure behaviour of tuffites can be described as follows: the residual shear strength depends on the overconsolidation rationormally Consolidated samples are deformed plasticallyoverconsolidated samples show discrete shear jointsafter 8 cm shear displacement a residual shear strength of overconsolidated undisturbed samples of only 3 0 percent (8°) of the maximum strength remains; this compares well with calculations based on actual failures.According to the investigations the main reason of slope failure in swellable overconsolidated tuffites is the decline of the shear strength after load relief by roadcutting. Usually, the failure starts at the base and expands progressively into the slope.The orientation of clay minerals supports the decline of shear strength after initial movements.

K.-H. Heitfeld, W. Echle, H. Düllmann, R. Azzam, R. Hasenpatt

E.-W. Raabe

### Bruchmechanismus steifer geklüfteter Tone bei einaxialer Kompression

Recent results in fracture analysis have suggested that the Coulomb-Mohr criterion appears to be best suited for soils, while failure criteria based on microcrack nucleation may be best suited for rocks that have a great number of inherent ‘Griffith-cracks’ in the form of joints or grain boundaries, required by these theories.In this context stiff fissured clays take up an intermediate position. They are substantial soils, but due to their sharp physical discontinuities they exhibit structural attributes of a rock mass.A lot of efforts have been devoted to investigate the strength behavior of stiff fissured clays based on the Coulomb-Mohr criterion, yet the understanding of long-term as well as short-term strength and failure mechanism still remains insufficient.The main objective of the investigations presented was to determine, whether dilatancy mechanics similar to those that are described in the rock mechanics literature may control the strength behaviour of stiff fissured clays in the lower isotropic stress region.In order to obtain the stress-dilatancy relationship of a soil specimen under uniaxial compression, a new method was developed based on optographic trace recording of location points attached to the sample surface (fig. 6).The results of dilatancy measurements offer a progressive Volumetrie strain rate with temporal intensity in various sample regions (fig. 9).Comparing test readings with microfabric studies in scanning electron micrographs and thin section analysis (fig. 11, 12 and 13), dilatancy obviously results from microcrack nucleation.New cracks were found to appear first at large components, such as quartz grains, close to inherent fissures and bending or forking of fissures (fig. 14).At peak stress no faulting was obtained. Collapse of specimen occurs in post-failure region either as longitudinal Splitting or inclined shear-like fracturing (fig. 10).Analytical energy calculations revealed that failure mechanisms obtainedcan be divided into six sequential events i)initiation of microcracksii)stable propagation of microcracksiii)critical energy releaseiv)unstable microcrack propagation including coalescence of separate cracks leading tov)ultimate strengthvi)formation of macroscopic fracture.Thus failure mechanism of stiff fissured clays under uniaxial compression stands in contrast to soil mechanical understanding of shear failure.Taking into account that failure mechanism below the brittle-ductile transition operates in the same manner, new findings for solving longterm strength behaviour may be expected.

V. Feeser

### Anisotropie des Eigenspannungszustandes der wechsellagernden Locker- und Festgesteinsschichten des Frankfurter Raumes

In the Frankfurt area several building projects are carried out, of which the subsoil is influenced to a depth. of 40 m and more. The so caused deformations which lead to Settlements at the surface depend on the stresses existing before construction as well as on additional stresses. The same applies to tunnel structures.As the states of stress depend on depth, the knowledge about the anisotropic State of in-situ stresses is important for exact numeric investigations. The examination of the anisotropic State of in-situ stress was the aim of this research project to which soil mechanical and geological investigations were applied.Field and laboratory investigations were carried out. In boreholes, reaching a depth of 4 0 m, the horizontal earth pressures at rest σh’ were directly measured using a pressuremeter. As the corresponding vertical stresses σv’ = ϒ’.t are known, the coefficient of earth pressure at rest MATH TYPE could be calculated in dependence on the depth t (fig. 2). The resulting dependence of KO = f(t) on depth with values of 1 to 3 near the surface could be confirmed by oedometric tests on specimens from different depths. By these oedometric tests the overconsolidation ratio OCR was determined in dependence on the depth. A well-known empirical relation between OCR and KO allowed to determine KO independently once more and to compare it with the KQ-values obtained from pressuremeter tests (fig. 1, 2, 8). The correspondence between these KO-values was quite satisfactory. Moreover the correspondence between them and the results of geological investigations was satisfactory to the degree which can be expected on such investigations.The area of Frankfurt is appropriate for the program because of a sequence up to 300 m thickness, slightly faulted and consisting of Tertiary clays, silts, sandstones and limestones is exposed, dipping 3° towards NW (fig. 3, 4). Therefore, a maximum vertical load during the geological past of 50 to 200 m according to the respective location within the city can be derived (fig. 10, 11). The directions of the principal stresses are permanently maintained vertical or horizontal, respectively. Some open joints in clays and open bedding planes encountered during excavation of an underground tunnel indicate locally reduced horizontal stresses. In such areas also increased Settlements caused by tunneling are recorded.At Frankfurt the KQ-values are distinctly lower than those determined in the London Clay with KO = 3 − 4 close to the surface. For a comparison clay mineralogical analyses were performed. Characteristic for the Frank furt area is a frequently changing and in the Miocene often high content of swelling clay minerals up to 70 percent related to the grain size 2μm

E. Franke, H. Mader, K. Schetelig, T. Schneewolf

### Vorbelastung und Erdruhedruck eines Kreidetons

P. R. Khera, H. Schulz

### Ingenieurgeologische Untersuchungen an Geschiebemergeln im Hamburger Raum

The ground moraines of the Hamburg area almost exclusively consist of tills. They represent about 40 percent of the soil cover.The strength behaviour of these tills depends on their petrographic structure and consistency. Yet, the different structures of fabric influence the strength to varying degrees according to the valid consistency: with soft to stiff consistency the properties of the grain size fraction 0,06 mm which functions as binding agent in tills, are most importantwith firm consistency an often observed joint structure can be of additional geotechnical significance. Compared with that the marl within the boulder clay is only soil-mechanically active if it occurs as mosaic layers.the determined plasticity indices correspond to the granular skeleton/ binding agent ratio. Because of a high binding agent content the Niendorf and Elster moraine show higher plasticity indices and higher liquid limits.Density and dry density rise with increasing age of the moraines. The corresponding decrease of the porosity is due to the filling of the pore space with finely distributed clay and calcium carbonate. With comparable granular skeleton and increasing lime content porosity decreases (fig. 4).An exponential correlation exists between shear strength and water content: the latter being constant, shear strength will increase according to density resp. age of the marls (fig. 5).The behaviour of cohesive soil under influence of water is described in disintegration tests.The granular skeleton (grain size fraction 0,06 mm) mainly consists of quartz, feldspar, lime and boulder fragments whereas the binding agent consists of swellable clay minerals (smectite), transition – minerals (mixed-layer minerals) and unswellable minerals (kaolinite, chlorite) with a subordinate content of finely distributed calcite, dolomite and quartz.Quantitative estimates of the clay mineralogical composition based on standardized suspensions show about the same content of kaolinite within the different marls of the Hamburg area (Elsterian moraine, Drenthe moraine, Niendorf moraine, Fuhlsbüttel moraine). This is different from the content of smectite which is highest in the Elster marls and thus helps to identify them (fig. 1). In most cases the lime content supports the stratigraphical Classification of marls as well.The influence of the varying petrographic structure of the different marls on their soil mechanical behaviour is confirmed by geotechnical investigations: the determined plasticity indices correspond to the granular skeleton/ binding agent ratio. Because of a high binding agent content the Niendorf and Elster moraine show higher plasticity indices and higher liquid limits.Density and dry density rise with increasing age of the moraines. The corresponding decrease of the porosity is due to the filling of the pore space with finely distributed clay and calcium carbonate. With comparable granular skeleton and increasing lime content porosity decreases (fig. 4).An exponential correlation exists between shear strength and water content: the latter being constant, shear strength will increase according to density resp. age of the marls (fig. 5).The behaviour of cohesive soil under influence of water is described in disintegration tests.As the disintegration rate depends on the initial water content of the samples an apparatus was developed to register continuously the disintegration process (figs. 6 and 7).The disintegration process is determined by the grain skeleton and the kind of cementation (fig. 8). Thus, high lime contents (as they can be found in the Niendorf moraine) have a structure preserving effect because of the increase of fine grains.Summarizing it can be said that different strength behaviour of tills can be explained geologically and described by mineralogical composition and geotechnical parameters.

Axel Baermann, Karl Wüstenhagen

### Geotechnik der Moränen

The geotechnical properties of moraines are generally considered to be more or less uniform. Frequent observations at building sites, however, have proved the contrary.The petrographic composition of glacial sediments is extremely variable. All rock- and earth-material eroded and transported by ice is indicated as moraines. Distinction is made between moraines of distant and local origin. Moraines of distant origin in Northern Germany consist of rock and soil from Scandinavia and the Baltic. In basal tills such material evolved by lodgement or melt-out. Sole tills contain the total sheared and compressed local rock material. The composition changes after short distances.On the other hand, the petrographic composition of ground moraines is only subordinately influenced by local material and therefore much more uniform over great distances. Till of the Drenthe-Glacial from Juetland and Schleswig-Kolstein can easily be mistaken for a ground moraine of Lower Saxony and the Netherlands of the same period.The youngest, sandy tili of the Warthe-Glacial compares with the Drenthe- moraines, but is found only at the lower Elbe, in NE-Lower Saxony and in Schleswig-Holstein. The tili of the Niendorf-Formation (Middle Saale Glacial) is rieh in clay, lime and flint. The darkgrey Elstermoraine is characterized by low lime-content and high stiffness. It is found in deep excavations only.A main target of all mappings for foundation engineering purposes is to distinguish the different morphologic (drumlins etc.), genetic and stratigraphic types of moraines as to their influences on project and design. Thickness, facies and other relevant geotechnical parameters can be presented according to the selected scale. The usual soil mechanical parameters require refinements and additions in order to describe the ränge of properties of the ground moraines.Critical experiences have been made very often in foundation engineering with the uncalculable occurrance of foundlings and erratic blocks of different rock types. The dark tili of the Elster Glacial is often highly Consolidated and causes problems during excavation or anchoring measures. The Niendorf tili is stable in slopes but difficult to excavate. The tili of the Drenthe-, Fuhlsbuettel- (Warthe-) and Weichsel Periods have proved to be critical due to their varying stiffness. Shear failures in the Drenthe tili have caused considerable damage, and the Fuhlsbuettel tili as well as the Pudasjärvi tili in Finland are known for their pseudothixotropic behaviour (pudding moraine).The reasons for these variable geotechnical properties are manifold. First of all, the petrographic and geochemical parameters are of great importance. Furthermore, the degree of weathering, the diagenetic and the glacial consolidation have to be mentioned. Experience has shown that also the stratigraphic position of the tills is of considerable importance for the engineering geological assessment. Concluding studies are in progress.

Friedrich Grube

### Sedimentpetrographie und Geochemie der Moränen

Chemical and physical investigations of stratigraphically different tills in North Germany were carried out, in order to obtain the lithological parameters which could have an influence on geotechnical parameters and shear wave velocities within glacial tills.Aim and ränge of this study resulted from the objectives of the research projects “Geotechnics of moraines in North Germany” directed by E.F. GRUBE, “Lithological reconnaissance of near-surface sediments by compression and shear waves” directed by R. MEISSNER and “Geotechnics of moraines in North Germany” directed by K. WÜSTENHAGEN. This study can therefore not be considered as an independent research project.The results of these investigations show, that the stratigraphically different tills of the Weichselian, middle and older Saalian glaciations are characterized by specific grain size distributions and carbonate contents. The tili of the Drenthe-stadium shows a grain size distribution, which is not favourable for compaction, and a low carbonate content. In the tili of the Niendorf-formation an increased silt fraction with a corresponding low sand content is to be found. Due to this grain size distribution and a mean carbonate content of about 25 percent a decrease of porosity and an increase of shear wave velocity can be expected, depending on the preconsolidation load by glacial ice.Sufficient data are not available for the definition of a typical grain size distribution pattern of the Weichselian tili. This tili shows a grain size distribution nearly corresponding to the “Fuller”- curve. It can therefore be regarded as favourable for compaction. Nevertheless a lower firmness of the Weichselian tili can be observed in the outcrop despite the high carbonate content of 20 percent, when compared with the tili of the Niendorf-formation. This might be due to the lower preconsolidation load by ice pressure.Weathering processes led to mobilisation, migration and deposition of iron, manganese, aluminium and calcium resulting in a secondary density increase in a depth of 3 – 5 m below the surface of the tills at Emmerlev Kliff and Brodtener Ufer.

S. Christensen

### Verformungsverhalten und Strukturfestigkeit norddeutscher Geschiebemergel

Results of studies on north German till indicate, that its strength is not a consequence of glacial pre-consolidation as was previously assumed. Furthermore, the often discussed compaction of grain structure via newlyformed minerals during various weathering phases, though obviously present, appears to be inadequate to account for the observed mechanical resistance. Arguments from soil mechanics and engineering geology favouring these critical observations are presented in the article.During the present study a new method was developed to investigate the behaviour of deformation characteristics of till. The method allows calculation of the deformation of external surface wall as well as the distribution of partial deformations within the sample related to raster elements. The results obtained deviate significantly from those of previous investigations on stress-deformation relationships within rocks: 1)Till exhibits dilatancy, manifested by an increase in volume or a negative, i.e. the onset of external wall deformation directed against the normal strain occurs just before failure strength.2)A typical distribution pattern of the partial deformation is seen both within the samples and in the ground, which is dependent on the materials and the deformation strategy. This deformation- and densitydistribution is called here the shear structure. Next to the shear plane a loosening of the grain structure could be observed, which, strictly speaking, may be termed dilatancy s. str. The resulting negative deformations are removed in two mutual zones of compression following into a zone of extension. In other words, deformations are conducted, only to a very small extent, towards the external surface wall of the sample. The remaining, proportionally larger part of the sample, which is also subjected to stress, behaves rather intact; it participates only subordinately in the deformation of the external wall.To simulate the deformation sequences under such complex conditions a new computer supported method was developed. The method takes into account not only the material characteristics of the sample (e.g. width and number of shear zones) but also its stress-dependent characteristics.This complicated deformation behaviour is limited not only to till. It was also observed in other loose sediments and rocks. The reason appears to lie in the balancing of deformation within natural dispersion-systems. For approximate quantification, a model plane was constructed, which consists of the sum of contact planes between the individual grains. In its size and geometry, this plane represents the mechanically active internal surface of the sample. On this model plane, the external stress on the system is monitored and partly resolved into two shear components. The strength (friction and cohesion) of the contact planes determines the extent of supporting (below critical) or almost no resistance (above critical) parts of the inner surface.The possibilities of further experimental and theoretical work for better understanding the mechanisms of shear deformation on the basis of a few phenomenological models is discussed.

Antonin Paluska

### Untersuchungen über typische Verteilungsmuster von Trennflächen und deren Ausbildung unter ingenieurgeologischen Gesichtspunkten

Geological and engineering-geological investigations have been performed In the medium Siegen-Formation in the Ahr-river-valley since 1978. The objective was to establish relations between the geological structure of the rock mass and engineering-geological parameters.During a first study the occuring discontinuity planes could be genetically classified under lithological aspects into 9 groups. Starting with the appraised typical features of discontinuity planes and the numerically described relations between lithology, deformation and weathering of the rock-mass as well as the structural parameters, it was possible to establish regional criteria for the definition of homogeneous areas.Slopes in the Ahr-river-valley resemble the system of discontinuities whereas the different planes predominate according to their local characteristics. Also the size and the Variation of the individual rockfragments depend largely upon the geological parameters of the- respective homogeneous areas.In a second step of the research programme the geometry of the discontinuity planes within the different units and their mechanical properties have been evaluated. This study comprised a great number of roughness measurements both in the field and in the laboratory. First time, a new terrestrical-fotogrammetric method could be applied with sufficient accuracy.The measurements of the shear strength along discontinuity planes were performed according to BARTON (1973), a method which allows the determination of shear strength parameters of planes without coatings by simple laboratory and field tests. Basic relations between the diagenetic roughness of discontinuity planes and the relevant shear parameters could be established. The tilt- and pulltests of BARTON have been supplemented by frame-shear-tests. Considerable differences were found, obviouslycaused by the higher stress-level of the frame-shear-test, which distinctly inflicts the roughness of the planes. The restricted validity of the BARTON-formula was confirmed. Therefore, a new relation has been derived, considering the normal stress, which is better adapted at least to the paleozoic rocks of the Rhenish Mountains.A counter calculation of a rock slide in the research area based upon the probabilistic safety concept has produced fair evidence of the correctness of the established parameters.

K.-H. Heitfeld, K.-H. Hesse

### Verteilungsfunktion der Kluftkörpergröße von verwittertem oberkarbonischem Sandstein

Rock blocks are formed by natural processes, i.e. tectonics and weathering, and by man-made processes. These processes develop statistically as the rock blocks are “polydisperse”, i.e. of different size and of different shape. The distribution of the rock block size is of principal interest in engineering practice, e.g. for the stability of slopes, in mining and other underground excavations, for the fragmentation of rock for building purposes, as for instance embankment fills and also in ore dressing.Two questions arise: firstly as to the methodical assessment of the distribution of the rock block size, and secondly as to the dependency of this distribution on lithology, tectonics, and weathering.The analysis of rock block size is to be carried out on rock faces of both man-made and natural exposure, possibly also on rock cores. One principal problem arises from the fact that only sections of the actual rock blocks can be viewed, depending on the orientation of jointing, the Position of the blocks, and the orientation of the rock face. This has also been named the “tomato salad view”. Therefore, a method has been developed to derive a realistic distribution of the rock block size from simple geometric sections of the rock blocks which are easiliy reproducible. The “line analysis” after ROSIVAL is adapted as measuring method. The chord lengths, i.e. the distances between the boundaries of the rock blocks, i.e. between the intersections of the joints with the rock face, are measured along grid lines the directions of which depend on the orientation of the joint system.The field investigation was carried out on a rock face exposing a weathered sandstone sequence of coal bearing Upper Carboniferous. The face was about 20 metres in height.The distribution of the chord length measured was approximated by different Statistical functions, e.g. RRS, expontential, normal and log — normal distributions. The best fit was obtained by the RRS distribution developed by ROSIN, RAMMLER and SPERLING. More recently this distribution is also called the two-parametric Weibull distribution.In the next phase the investigation is directed towards the statial change of the RRS distribution with changes of lithology, tectonics, and weathering as observed within the exposure. For this purpose theparameters and the characteristic values of the RRS distribution are studied. One of the two parameters, and also both mean and median, are of almost the same magnitude and vary synonymously with respect to space. They react, if tectonics remains unchanged, on changes of both weathering and lithology. They tend to decrease with increasing weathering and decreasing grain size, i.e. joint spacing. Most interesting is the reaction of the second parameter, the exponent of the RRS distribution. This parameter is barely affected by lithology provided the tectonics remain unchanged. However, it reacts significantly on changes in weathering observed towards the ground surface.As an indicator of the Utility of the distribution of the rock block size an equi-distant joint spacing was developed based on integral geometry. A comparison of the mean chord length in RRS distribution with this joint spacing indicates that the mean of the chord length is always less than this mean joint spacing.Therefore, a function had to be developed which would allow to derive the distribution of the actual rock block size from the distribution of the chord length. This was accomplished based on the well-known mathematical relationship between the distribution of the chord length of spheres with their actual diameters. A respective Solution was provided by CAHN and FULLMAN. If the RRS distribution and its derivative are inserted into the Cahn-Fullman function a new distribution is obtained which contains the two parameters of the RRS distribution. For the new distribution equations were developed for mean, variance, and mode. Depending on the spatial configuration of lithology, tectonics, and weathering within the exposure mean and mode of the new function describing the distribution of the rock block size change analogously with the mean of the RRS distribution of the chord lengths.Striking agreement was observed in a comparison between the equi-distant joint spacing and the modified mean and mode of the new function describing the distribution of rock block size. This agreement proves that the proposed analytical approach yields the realistic distribution of the rock block size.

Falk Beyer, Klaus W. John

### Kriechverhalten gefüllter Gesteinstrennflächen

Creep deformations of a rock mass occur predominantly along its discontinuities, preferably on those filled with gouge material. Since so far only little Information has been available on the creep behaviour of filled rock discontinuities a detailed experimental study has been conducted to investigate the critical parameters and to quantify their influence.Creep tests on filled rock discontinuities were performed on sandwichtype samples in modified direct shear frames at different boundary conditions. Factors of influence investigated were type of filier material, shear stress, normal stress, and size of the specimen. Smooth saw-cut surfaces were used as rock discontinuities since these proved to represent boundary conditions which are similar even to those at rough rock surfaces, where the material in the roughness troughs is not involved in the creep movement.The results clearly show for the primary creep phase an exponential dependence of the creep velocity on time, whereby the exponential exponent m is equal to -1 for lower shear stresses, but it increases towards zero for higher shear stresses. The initial creep velocity is directly proportional to the thickness of the filier layer and related to the clay content of the filier by an exponential function. The exponent m was found to be dependent on the clay content, the layer thickness, and the shear stress level related to the residual shear strength. The parameters of these empirical functions were derived from the experimental data.None of the creep tests showed the existence of a secondary stationary creep phase but an immediate transition from the primary to the tertiary, accelerating creep phase was observed. For the higher shear stress levels investigated an exponential decrease of the time of transition with increasing clay content was found. The clay content of the filier, i.e. the content of particles smaller than 2 μm, thus represents the critical material parameter which determines the primary and tertiary creep behavior of a filled discontinuity.Other potential factors of influence, such as normal stress and size of the discontinuity, are automatically taken into consideration if, aspracticed in this study, the creep data are given as a function of the relative shear stress, i.e. the ratio of actual shear stress to residual shear strength. In all tests the consistency index of the filier was kept constant and close to the lower bound of values observed in critical slopes with sliding on filled discontinuities.Potential reasons for the direct dependence of the creep behaviour in the primary and tertiary phase on the clay content of the filier are discussed on the basis of a simplified physical model.

K.-D. Höwing, H. K. Kutter

### Vergleichende Gefügeuntersuchungen für den Tunnelbau im Buntsandstein

A geological exploration including special mapping, drilling and fabric analyses was worked out to prepare a tunnel project in the Lower Triasrsie sandstones of NE Hesse in Germany. These results were compared with a fabric selective inter prettation of aerial fotographs that had been carried out following a method proposed bei REUL (1977). In general the lineation fabric in the aerial photographs fits within the fabrics of fractures and joints. To the contrary the interpretation of larger fractures in the aerial photographs shows considerable deviations to the results from field measurements. On the other hand strain rates derived from aerial photographs are well comparable to those resulting from field interpretations.

C. Munk, H. Prinz, D. H. Zankl

### Seismische Bestimmung der bodendynamischen Kennwerte von oberflächennahen Schichten in Erdbebengebieten der Niederrheinischen Bucht

During the period from 1981 to 1983 seismic down-hole measurements were carried out at 36 selected sites within the Lower Rhine Basin and its vicinity in order to determine systematically the propagation velocities and the amplitude attenuation of compressional waves (P-waves) and shear waves (S-waves) in soft soil deposits of Quarternary and Tertiary age. For comparison, additional velocity measurements were made in Mesozoic and Paleozoic sedimentary rocks.The seismic P-waves were produced by vertical hammer blows to an iron plate lying on the ground surface. For the generation of horizontally polarized S-waves a pendulum construction with a moving iron mass of some 70 kg’ was used. The mass blows horizontally against a wooden plank which is fixed to the ground. The seismic signals were detected in the borehole at various depths with triaxial geophones and digitally recorded on magnetic tape by means of an 8-channel Lennartz PCM-unit. A vertical stacking was performed in order to improve the signal/noise ratio. The data processing was made with a CBM-4032 microcomputer system.From seismic propagation velocities and amplitude attenuation curves the elastic and non-elastic properties (dynamic shear modulus, Young’s modulus, bulk modulu’s, Poisson1s. ratio, Q value, non-elastic damping ratio) of the soil layers investigated were calculated. A Statistical evaluation of these data enabled the authors to deduce generalized relationships between dynamic soil properties and soil type, depth, geological age, and overconsolidation.Based on these results local soil amplification curves for earthquake Vibration in the seismic active regions of the Lower Rhine Basin could be calculated with the frequency domain method of the ‘Steady State Harmonie Oscillations’ (ROESSET 1970). Amplification spectra for selected sites demonstrate that the local earthquake Vibration is strongly affected by the thickness and the dynamic properties of the near-surface-soil deposits.

L. Ahorner, M. Budny

### Einflüsse lithologischer Parameter oberflächennaher Sedimente auf Kompressions- und Scherwellengeschwindigkeiten

Research was concentrated on the measurement of compressional (P) and shear-waves (S) in sands and boulder clay. Results may be summarized as follows.S-velocities (Vs) in the upper hundred meters of non-consolidated material are generally three to five times smaller than P-velocities (Vp). In saturated sands and clay extreme values of VP /V s up to 15 have been observed (fig. 4b). The records with horizontally polarized shear-waves (SH) are not seriously disturbed by other wave types. No converted waves are created by SH-waves at horizontal boundaries (fig. 2). No interference takes place between Love-waves and SH-reflections because Lovewaves propagate with a similar velocity as the first arrivals do. Because of their shorter wavelength SH-waves have a better resolution than Pwaves. Correlation of abserved velocities Vp and Vs to lithologic and mechanical parameters (fig. 7) indicate that (i)Vp and Vs in boulder clay are influenced by its porosity, degree P s of Saturation and the properties of grain boundary contact, which is a function of various parameters like clay-calcite content, size, shape, pressure etc.(ii)Porosity in boulder clay decreases with increasing pressure and increasing content of CaCO3. Increasing percentage of clay results in an increase of porosity. Vp and Vs decrease with increasing porosity in boulder clay as shown in fig. 7d.(iii)The influence of the State of Saturation is most different for Vs and Vp: while both values are about constant up to high values of Saturation, Vs shows an abrupt increase at about 99 % Saturation while Vs shows a small decrease which is caused by changes in density and shear modulus (fig. 7a,b). For boulder clay the decrease in Vs may be attributed to the volume change of the water absorbing clay minerals which reduce the shear modulus and show the strength of the material (fig. 7b).(iv)The effective pressure (lithologic-pressure minus pore-pressure) has a strong influence on the grain-boundary contacts (fig. 3). The Vg values in dry and saturated sands as well as in boulder clay follow a relationship like Vp -values follow a similar relationship in dry and partially saturated sands while for Vp in saturated sands and boulder clay a P rather constant value is found for the low pressure conditions of near-surface layers.(v)Boulder clay shows a decrease in Vp for an increase of the clay content. This is related to the different elastic properties of the grain boundary contacts of the clay minerals and those of the coarser material. An increasing content in CaCO3, on the other hand, is correlatable with an increase in Vp and Vs . This correlation is possibly due to the cementation and stabilization of the grain-boundary matrix (see figs. 7e and 7f).(vi)A correlation seems to exist between the Vg-values and the resistance against penetration, as shown in the regression lines of fig. 6. Further research on this subject is in progress.From the above mentioned items the relationships between Vp, Vs and the pressure and Saturation (especially in sands) are the most solid and general ones. Future investigations have to show whether the other correlations have a local or also a more general character.

S. Hübner, R. Meißner, H. Stümpel

### Erkundung oberflächennaher Strukturen und Eigenschaften mit dem elektromagnetischen Reflexionsverfahren

In the course of this project a great number of electromagnetic profilings, soundings and measurements in boreholes and in the laboratory were carried out. Most of the investigations were done in soils some also in solid rock.In the initial phase of the work only a small part of our techniques were available. Therefore major developments of highfrequency devices for the laboratory and field measurements were necessary.The numerous results of the elctromagnetic measurements can be summarized with certain simplifications in one important aspect. Essentially, the results depend on the water content of the material and the dissolved ions. As the electromagnetic properties strongly depend on the water content and the dissolved ions, the electromagnetic and the d.c. resistivity of soils and solid rock are related with each other. The properties of the samples were measured in the laboratory in the frequency ränge from 10 Hz to 100 MHz. With few exceptions, the resistivity of the samples at 10 Hz is equal to their d.c. resistivity with sufficient accuracy.

F. Thyssen

### Erkundung oberflächennaher Strukturen mit reflexionsseismischen Verfahren

For the determination of velocity-depth-relations döwn to approximately 50 m results of seismic reflection measurements are presented. The recordings were carried out by self-constructed computer-controlled equipment. The influence of the uppermost part (cm) of the weathered material on the seismograms for the horizontal profiles is discussed in one case. In the field records longitudinal wave reflection arrivals can only be correctly correlated with each other after digital data processing. Addi tional vertical seismic sections from borehole measurements would be necessary.

M. Degutsch, F. Thyssen, M. Holdorf

### Untersuchungen an Tonschiefer im Hinblick auf dessen Eignung als Dammschüttmaterial

Previously, the mechanical properties of slate embankments as well as their permeability and susceptibility to erosion could only be reliably determined by large scale test fills. In order to obtain more precise predictions, petrographic investigations, ultra-sonic measurements and Los-Angeles-Tests were carried out on rock samples and their results were compared with measurements and experiences from existing dams.Rocks up to a quartz content of about 75 percent appear as slates. All rocks of the series slate/quartz yield a very well graded granular fill and such rock fills are easily compacted (γd =21 - 23 kN/m3) in general. Sandstones, quartzites and solid slates in which mica and clay minerals enclose the quartz and feldspar grains in wave like manner, form favourable embankment materials with high shear resistance and rapidly occurring and small deformations. Furthermore, in such slates smooth and continuous cleavage planes do not exist. The quartz content alone is not a suitable parameter for the evaluation of the rock.Severe disintegration of the rock and the tendency to long term Settlements are encountered with platy slates having cleavage planes intersecting the whole piece of rock and which consist of perfectly orientated phyllosilicates, and in cases where the proportion of swelling minerals is >5 percent, frequently combined with a haematite content > 3 – 5 percent (red slates).Ultra-sonic measurements using P-waves beyond 45 kHz yield wave velocities in rock samples of 4 – 6.5 km/s. The anisotropy $$\frac{{{\text{V}}_{{\text{max}}} }}{{{\text{V}}_{{\text{min}}} }}\; = \;\frac{{{\text{v}}\quad {\text{s}}}}{{{\text{v}}\quad {\text{s}}}}$$ was identified as a very important parameter.Favourable embankment material yields an anisotropy of the wave velocity not more than 1.5. Higher values indicate distinct disintegration of the rock particles and longterm settelements resulting from this. Modified Los Angeles-Tests using 10 rock pieces each of about 1 kg and 100/500 cycles (drum rotations) are also very instructive. Rock materials yielding a content of fines (< 2 mm) exceeding 25 percent after the test, have a tendency to continuing deformations.The permeability of slate embankments varies between 10–4 and 10–8 m/s, depending on the kind of rock. Due to the well-graded nature of slate fills, the susceptibility to piping is low, except for fills having a silt/fine sand content of 10 percent and consisting of mica.Comparing the test results on slates with the investigations on the compaction, settlement and Saturation behaviour of granite fills (BRAUNS et al. 1980; KAST and BRAUNS 1981) it can be concluded that both rocks represent suitable embankment materials in general. Both kinds of rock yield a satisfying shear resistance and deformation behavior of the embankments provided sufficient compaction energy and an Optimum water content are applied during construction (BRETH 1972; SCHADE 1984).

K. Schetelig, R. Sellner, F. Dannenberg

### Einfluß von Verwitterungsgrad und Trennflächengefüge auf den maschinentechnischen Lösevorgang an überkonsolidierten Tonund Mergelsteinen

The structural strength of claystones and marlstones during excavation with hydraulic excavators depends on several factors, which include the degree of weathering (fig. 12a) as expressed by point-load index (fig. 12d), the average joints spacing throughout the rock body (figs. 12c and 18), and the orientation of the joint sets (figs. 9a and b). Due to the complexity of the fracturing (and consequent shearing) processes which occur during excavation of the rock, a number of investigations was conducted in the field rather than in the laboratory. In order to determine the forces applied during excavation, the oil pressure (figs. 4 and 8) and length of the piston-stroke (figs. 4 and 7) of the hydraulic cylinders of the excavator were measured. These values were then interpreted in terms of the three above mentioned geotechnical parameters and statistically analysed (figs. 13 to 16). This provided the basis for a Classification scheme for the economic feasibility of excavation as related to the degree of weathering and the spacing of vertical joint sets (fig. 18).

H. Gruhn

### Festigkeitsuntersuchungen an Gesteinen und an Gesteinsverbänden mit dem Prallhammer nach Schmidt

The objective of this research program was to verify whether the rebound hammer SCHMIDT system is appropriate to furnish index-figures for the valuation of rock strength in situ.For many years the rebound hammer has been used for testing and Controlling concrete constructions. After a suitable method for field tests had been worked out — the regulations according to DIN 1048 part 2 and 4 had proved to be unsatisfactory in field work — an easy portable coredrilling machine was constructed in order to receive adequate samples for comparative compressive strength tests. It appears that there is a good correlation between the rebound values gained in field work and the results of the uniaxial compressive strength tests. The test result received by rebound measurements is called “Prellhärte” (rebound hardness).Further examinations were extended to cores. Please note that the cores always are fixed identically for the rebound tests.Following preliminary tests a portable clamping device was developed. The examinations performed in that way principally provided the same interdependences between rock strength and rebound values as the testing in situ did.Subsequently, it was tested whether a loss in rock strength resulting from weathering of rock formations can be identified by rebound hammer tests.Measurement profiles in the Devonian beds of the “Rheinische Schiefergebirge” in the “German Buntsandstein” and in the “Rotliegende” were taken and interpreted. Five measurement points marked in a distance of 20 cm each are combined to one measurement value (1m total).It turned out that a continuous reduction of the rebound values can be observed from the unweathered to the completely weathered parts in slight-ly jointed, more or less homogeneous rock formations. Even in those cases where the reduction of rock strength leads to an increase in decomposition, the intensity of this process is reflected in the index-figures.Concerning future proceedings there is the possibility to use these index-figures to classify differently weathered rocks (in Germany DIN 18 300).

E. Krauter, H. Woszidlo, H. Büdinger

### Einfluß und Wirkungsweise von CaO bei der Belastung von Aggregaten

The structure stability of clay soils is often weak because of unfavourable interactions between soil water and soil matrix. Loading of soils with a high clay content is accompanied by the development of positive pore water pressures, which further reduce structural stability. Hence, stabilizing measures of clay material must prevent the appearance of positive pore water pressures by changing the interaction between soil water and matrix.Stability of clay aggregates from a pseudogley-pelosol soil from Lias (52% < 2 μm) and from pure clays (bentonite, kaolinite) was investigated in confined compression tests. The surfaces of the aggregates were treated with CaO in amounts of 5% related to dry soil. Treated and untreated aggregates were stored at a water suction of p = − 7 cm WS and free air-entry for 1 day to 30 weeks. The compression tests were carried out in measuring compaction and pore water pressure in relation to load and time after CaO-treatment.As a result of Urning, appearance of positive water pressures and compaction were reduced. In some cases, even negative water pressures occured. The stabilizing effect of liming first increased with time after CaO-treatment and then decreased again. Maximum stabilization (minimum compaction) coincided with minimum pore water pressures. The relation between pore water pressures and stability was more distinct with samples from pseudogley-pelosol soils than from pure clays. The stability effect itself depended on the material under investigation and on the conditions after liming (water and air supply). Maximum stabilization occured before CaCO3-formation was completed.In these experiments, no new minerals except CaCO3 were formed. Therefore, the stabilizing effect of liming clayey materials in this case is based on mechanisms connected with properties of CaCO3, respectively its formation. These mechanisms are: Hydration of CaO: the formation of Ca(OH)2 is water consuming and dries up the aggregates. As a result, water pressures are lowered and stability is improved.Particle-size-effect of the CaC03-crust on the edges of the aggregates: this is a more mechanical effect which prevents a direct contact of clay particles with their enveloping water molecules and enables the water to drain between the CaCO3-particles under conditions of loading. This is another way to reduce the development of positive water pressures.Hence, both mechanisms change the unfavourable interactions of soil water and soil matrix under loading and stabilize the soil structure.

H. Bohne, K. H. Hartge, H. Graf von Reichenbach

### Stabilitätsrelevante Mineralreaktionen in chemisch behandeltem bindigen Baugrund

Soil stabilization was used already in ancient times in road and building eonstruetion as well. Its fundamental rules have been developed in the past decades and are still worked out in our days by the method of “trial and error”. Only few attempts have been made to learn the chemical mechanisms of the soil — Stabilisator interaction.In order to get an idea of the chemical reactions, seven natural clays from significant engineering projects in Bavaria and Mali/Africa as well as three monomineralic clay samples were selected for the investigations. To characterize the interaction between the clays and the added stabilizers (portland cement, hydrated lime, orthophosphoric acid, water glass and bitumen) within a 6 months’ period (resp. 26 months) the present study examines the reactivity of the clay — stabilizer — water -systems in terms of (a) electrical conductivity and zeta potential, (b) X-ray analyses of the original material and the reaction products as well as electron microscopical investigations, (c) Charge distribution and intracrystalline swelling of the affected layer Silicates.Measuring the electric conductivity we found different curves in the important clay - lime and clay - portland cement systems. The measured zeta potentials showed that the reactions were strongly influenced by surface phenomena. The X-ray diffractograms of the stabilized clay systems showed weak lines for new mineral phases in the initial State of the reaction. These peaks gradnally strengthened and after three weeks reflections at 14 — 12 Å and at 3.07 - 3.0 Å indicated that gel – like varieties of CSH-phases had formed. The most distinguished peaks however appeared in the ränge of 8.0 - 7.0 Å. These lines can now be attributed to hydrated calcium - aluminates like C4AH13. These new phases resulted in the rapid formation of calcium hydroxid and the dissolution of sheet silicates, as can be shown by conductivity and chemical measurements. The products formed after short periods of aging are largely gelatinous. and amorphous, however, with further curing, poorly ordered crystallites of CSH and CAH phases are formed.Because of the strong bonding between the new phases and the clay particles no chance was given to separate the reaction products for further investigations. Finally, we succeeded by means of a direct measuring method as it is already known from colloid chemistry: complete exchangeof the calcium interlayer cations of the stabilized clay minerals by n-alkylammonium ions provided the best method for the characterization of changes in the interlayer space. It could be found that the natural bentonite has mean layer Charge of -0.32 eq/Si4O10. No other method gives such detailled information of the Charge distribution of layer Silicates. Applying these investigations on stabilized clay systems we got the following results: portland cement as well as lime will change the Charge in a congruent wav. The charqe distribution remains homogeneous during the reaction time, but there is a shifting to a higher total Charge density. Using water - glass, orthophosphoric acid or bitumen as a stabilizing agent, only small changes are caused and the Charge distribtuions become inhomogeneous. The reasons can be of different origins: redox reactions in the basic solutions may Charge changes within the octahedral and/or tetrahedral sheets of the structure,disintegration of layer crystals and reconstruction of clay crystals with layers of different Charge may be the other reason.The results of the study as well as the methods used during the investigations may be of great importance for the interpretation of a number of geotechnical phenomena like swelling, creeping and sloping. It is obvious that the colloid chemical properties of clay — Stabilisator or clay — water systems are invluencing the mechanical behaviour of any soils.

E. E. Kohler
Weitere Informationen