MINERALOGY AND GEOCHEMISTRY OF

QUATERNARY CLAYS IN THE

NUMEDAL AREA, SOUTHERN NORWAY

ELEN ROALDSET

Roaldset, E.: Mineralogy and geochemistry of Quaternary clays in the Numedal area, southern Norway. Norsk Geologisk Tidsskrift, Vol. 52, pp. 335-369. Oslo 1972.

The mineralogy and geochemistry of the ela y fraction ( <2f.!) of the Quater­nary deposits in the Numedal area, southern Norway, have been studied. The Numedal clays contain considerable amounts of illite, but chlorite, vermiculite, mixed layer minerals, quartz, microcline, and acid plagioclase are also usually found. Amphibole is present in most of the samples. Traces of montmorillonite are occasionally found.

During weathering the amount of vermiculite and mixed layered illite­vermiculite increases at the expense of trioctahedral illite and chlorite.

The average chemical composition of the clay fraction ( <2!!) corresponds to that of coarse clays. The clay material is only slightly weathered, and the residual indexes correspond to those of fresh rocks.

Comparison of the mineralogical and chemical data shows the physico­chemical variations during weathering, transport and sedimentation, and the composition of the parent material to be of significant importance to the clay composition.

Elen Roaldset, Institutt for geologi, Universitetet i Oslo, Oslo, Norway.

This paper presents the mineralogy and geochemistry of the clay fraction ( <2JL) of the Quaternary deposits in the Numedal area, southern Norway <Fig. 1).

The Numedal river system is about 300 km long with a drainage area of about.6000 km2• In the lower parts, from Kongsberg to the sea, marine facies are represented. Above the marine limit different sediments such as moraine, glaciofluvial, fluvial, and lacustrine deposits are found.

The investigation is a contribution to the 'Numedal project' (Rosenqvist 1969), the purpose of which is to study the origin and formation of Norwe­gian Quaternary deposits. The present study was undertaken to evaluate the variations in clay mineralogy and geochemistry in relation to parent material, sedimentological and other environmental factors.

Regional geology

The rocks of the Numedal area are of Precambrian, Cambro-Silurian, and Permian ages (Fig. 1).

Precambrian rocks

According to B. E. Løberg 1970 (pers. comm.), the Precambrian rocks .:omprise granites and granitic gneisses, porphyries, tuffs and tuffaceous

336 E. ROALDSET

PRECANBRIAN,

BRECCIA

GRANITE AND GRANITIC GNEISSES

PORPHYRIES AND TUFFITES

QUARTZITE

MICA SCHISTS AND MICA GNEISSES WITH CONGLOMERATE BEDS AND TUFFITE LAYERS

BASIC AND ULTRABASIC BODIES, Dlt<ES AND SILLS

DIORITTIC GNEJSSES AND INTRUSIYES

( AROUND KONGSBERG NIGMATITE )

CANBROSILURIAN,

� SHALE, LIMESTONE AND SANDSTONE

PERNIAN,

CJ LAYAS AND INTRUSIYES

SANPLE TYPE AND LOCATION,

o NOR AlNE

o GLACIOFLUYIAL

'\ Maln diteclion of tlatlot &lrloe

A. FlUVIAL OR LACUSTRINE �·P"""''I"L.,L.JE 67-74(profil)

76 L> ESTUARINE

• MARINE

• BEDROCK

Fig. l. The geology of the Numedal area compiled by B. E. LØberg 1970. Numbers refer to sample numbers. Inset map shows location of Numedal in Norway.

sediments. In the northeastern areas the bedrock consists mainly of mica­

schists, gneisses and quartzites with zones of meta-amphibolite and por­

phyries. The gneisses near Uvdal, Rødberg and Norefjord are of inter­

mediate to basic composition. The pyroxene and amphibole contents are

usually high and the mica content low. However, in the mica-schists and

mica-gneisses the mica content is 10-15 %, while it is less than 2% in the

80-81

83-85

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 337

quartzites and quartz gneisses. The Dagali granite consists of 5-20% micaceous minerals and the Rollag granite of 2-5 %.

Northeast of Kongsberg a north-south breccia separates the rocks of the

Telemark and the Kongsberg-Bamble formations. These rocks were studied by Bugge (1937).

Cambro-Silurian rocks

Low metamorphic Cambro-Silurian sediments are present as erosion rem­

nants in the northwestern areas, and as a 4-5 km wide NE--SW zone across

the valley at Kongsberg. The sediments consist of Cambro-Ordovician schists,

Ordovician and Silurian limestone and Downtonian sandstone (Brøgger & Schetelig 1926). The rocks are contact-metamorphosed by Permian intrusive

bodies. In the north the schists are metamorphosed to phyllite.

Permian rocks

In the area between the sea at Larvik and the Cambro-Silurian zone south of

Kongsberg, the bedrock consists of Permian eruptive rocks of the Oslo

Region. The plutonic rocks are mainly kjelsåsite, larvikite, nordmarkite, and

ekerite of monzonitic to granitic composition (Barth 1945). The correspond­

ing extrusive rocks are basalt and rhomb-porphyries (trachy-andesites). The

lavas are less abundant than the intrusive rocks (Barth 1962).

Quatemary geology

The marine limit at Kongsberg divides the valley system into two morpho­

logic units with marine silt and clay deposits south of Kongsberg, and glacio­

fluvial, fluvial and lacustrine grave! and sand deposits north of the city.

Numedal (north of Kongsberg)

The glacial striations indicate that the direction of movement of the inland glacier was from NW towards SE (Fig. 1). As the ice-sheet melted the directions of movement changed, and were at the latest stage determined by the local topography (Isachsen 1933, Sørlie 1965).

According to Isachsen (1933) the thickest and most extensive moraine

deposits are found in the Dagali and Tunhovd districts. In Nore, Rollag, and Svene the bedrock is often sparsely covered with moraine (Samuelsen 1937). Recent mapping has also revealed very extensive moraine deposits in these

areas.

In the phyllite areas a ground moraine of boulder clay type and an upper

ablation moraine are found (Isachsen 1933). Outside the phyllite areas the

ground moraines occasionally have a high clay content (Reusch 1896, Låg

1948, Sørlie 1965). The Cambro-Silurian phyllite in the northern areas is

believed to have contributed significantly to the clay fraction of these mo­

raines (Isachsen 1933). On the floor of the valley, fluvial and lacustrine silt and sand have filled

338 E. ROALDSET

up ice-eroded basins. A gradual transition from the sandy and gravelly

ablation moraines to the glacio-fluvial deltas is often found. The glacio­

fluvial deltas at Kongsberg are interstratified with thin silt and clay horizons.

A high sensitivity of these clays indicates a marine origin (Foss 1963). The drainage area for the river Numedalslågen is about 6 000 km2 and

the water comes mainly from the northwestern areas. In the Nore-Kongs­

berg district and south of the Cambro-Silurian zone, the distances from the

river to the watershed are about 10-15 km.

Lågendalen (south of Kongsberg)

After the retreat of the inland ice, at the beginning of the isostatic uplift of

the land, Lågendalen was a long and narrow fiord consisting of deep ice-

---.._ eroded basins separated from each other by narrow thresholds. The oldest

marine clays and glacio-fluvial deltas were deposited at this time (Fig. 2A). The highest marine limit (about 18 0m a.s.l.) is found near Kongsberg, where

large glacio-fluvial terraces were built up. The lowest marine limit is found

near Larvik (about 140 m a.s.I.). Glacio-fluvial deposits (deltas) are fre­quently found in side valleys, where some have built up to the marine limit.

Marine clays were deposited sirnultaneously with the build-up of glacio­

fluvial deltas. Glacio-fluvial deltas are therefore frequently found overlying

marine clays. In other words, the clays are 'bottomsets' or 'pro-delta' depos­

its.

Due to the land-rise, there was a continuous lowering of the shore-line in 'Lågendalfjorden'. The shallow-water sediments in the valley were soon lifted

A

bed rock

B

Fig. 2. The post-glacial development in Lågendalen A. Formation of glacio-fluvial deltas. B. Erosion and resedimentation of fluvial deposits.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 339

above sea-level, and immediately eroded. The estuary of Numedalslågen

started moving southwards, and the river as well as its tributaries eroded

the marine clays as soon as these sediments were lifted above the erosional

base (Fig. 2B). At the same time there was a deposition of fluvial sediments

(mainly sand and silt) in front of Numedalslågen (estuarial sediments). As a

consequence of this, well-sorted fluvial sand is found on top of the marine clays.

The fluvial sediments were again lifted above sea-level, and erosion and

redeposition occurred. Consequently fluvial sand and gravel as well as the

clays might have gone through several cycles of erosion and deposition. This

process is going on continuously.

Weathering of bedrock

The rocks of the area have been exposed to glacial erosion, and the post­

glacial weathering is normally of no significance. However, in some cases the

Precambrian and Permian rocks show signs of weathering (Schetelig 1918,

Låg 1945, Roaldset & Rosenqvist 1971). In the larvikite areas, soils formed by in situ weathering are common. During the weathering processes the dark

minerals have been transformed to chlorite (Låg 1945). The weatLering is assumed to be of preglacial (Tertiary) age.

Depositional environments

The river-water in Numedalslågen has a pH-value of about 5, which is in

agreement with results obtained by Johnson et al. (1968) in New Hampshire

under similar climatic conditions and Quaternary geology. They claimed that the acidity of the river-water originates from acid rain-water and from the

development of podsol soil profiles. This could also be the case in the Nume­dal area where iron podsol and humus podsol are the most abundant soil profile types. In south Norway the pH-value of rain-water varies from 4.9 to 6.2 and the pH-value of humus is about 4 (Låg 1968). Clays above the marine limit were presumably transported and deposited in a slightly acid environment, while the marine clays were deposited at a pH-value of about 8.

As a result of the isostatic rise of the land, marine clays were lifted above sea-level. This resulted in a gradual change in the environmental conditions. Originally the pore-water was of the same composition as sea-water. A con­

tinuous leaching of these sediments gradually decreases the salinity, and the

pH-value becomes more neutral. Norwegian marine clays usually have a pH­

value of between 7 and 8 (J. Moum 1969, pers. comm.).

Material

The material studied was collected during the summer months of 1967 and

1968. The sampling was supervised by I. Th. Rosenqvist and P. Jørgensen.

340 E. ROALDSET

The profile samples were taken by means of shovel-drill, while the other samples were collected in road cuts. The samples were kept in unsealed plastic cups until the laboratory investigation started.

The sample localities and the sediment types are shown in Fig. 1. The distribution of sediment types is as follows:

No. of samples

24

2

5

3

2

2

4

43

85

Sediment type

moraine esker glacio-fluvial deltas flu via l river-water Cambro-Silurian shales lacustrine marine

All the moraine samples are from the C horiwns of the podsol profiles.

Description of some samp/e localities

At Hvittingfoss (E 54-59) a quick clay slide occurred in 1960 (Eide 1961). Four of the six samples are of remouldered clay from this slide, while two (E 54-55) are of undisturbed marine clay. The marine clays are overlain by estuarial sand.

All the Aserum samples (E 67-79) were collected near the lake Aserum­vannet. The clay from this locality is an extremely soft quick clay. Grain size analyses show that about 50 % of the sample material has a particle size less than 2.�t· The average water content varies from 50 to 70 %.

River-transported mud and clay deposits are found in the more tranquil parts of Larvikfjorden, and the sand in the central parts. Three samples of river-transported mud from Larvikfjorden are included in this investigation (E 83-85). For samples E 83 and E 84 the salinity of the pore-water was found to be 2.7 % and 3.3% respectively.

Methods

The analytical procedure is shown in Fig. 3.

Material coarser than 7 4!1 was separated by sieving, and the silt and ela y fractions by dispersion and sedimentation. No flocculent agents were em­ployed. During the fractionation processes the material came in contact only with distilled or deionized water, the intention being to avoid more extensive

changes of the clay minerals. However, a few samples <E 77-79, E 83-85)

were pretreated with 35 % �� in order to remove organic matter. When the clay material has been in contact only with deionized or distilled water,

the term 'untreated' is used.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 341

X-RAV OIFFRACTION A NALVSIS

SOOIUM CITRATE TREATED SAMPLE

l U N TREATEO SAMPLE

l ORIED AT GLYCEROL HEATED HEATED 32'/, RE L. SATURATED TO 95°C TO 125°C HUMIOIT V 1/2 HR. 112 HR.

Fig. 3. The analytical procedure.

GRAIN SIZE DISTRIBUTIONS

SAMPLE

l ns:c������ION ___ GRA IN SIZE ANAL VS IS

l CLAV<2�

· OETERMINATION OF ORIE O AT 110°C CATION EXCH ANGE CAPACI T V l

1 os o •c

l DETERMINATION OF /IGNITION tOSS

HEATEO HEATED FUSED WITH 1o ��re ;o ��re LITHIU M TrRABORATE

OETERM INAT ION X-RAV FLUORESCENS O F Na BY NAA ANALVSIS

After removal of material coarser than 2 cm, grain-size analyses were car­

ried out on the moraine samples. The grain-size data were obtained by wet

sieving and the falling drop method (Moum 1965).

MINERAL ANALYSIS

In this investigation the ela y fr action ( <4-t) is studied.

Qualitative and semi-quantitative determinations of mineralogy were per­

formed by X-ray diffraction analysis. A Siemens diffractometer having a

Ni-filtered CuKa radiation and a scanning speed of 1° per minute was used.

The X-ray tube was operated at 720 W (36 KV, 20 rnA).

Tests were performed on untr:eated clay and on the same clay after

removal of iron oxides (Mehra & Jackson 1960). Oriented samples were

prepared by pipetting suspended clay material on to a glass slide (Bradley et al. 1937). Sedimentation and drying took place at room temperature at a relative humidity of 32 %. The X-ray investigation was made by scanning over the area 2-30°, 2E>.

After the X-ray diffraction effects were recorded at room temperature, the

sample was treated with glycerol using a method similar to that of Brunton (1955), and then examined for new diffraction effects.

After the second X-ray examination the sample was heated to 95 °C, 125

°C, 200 °C and 450 °C for 30 minutes andjor 2 hours, air-quenched and

X-rayed again.

Gibbs (1965) indicated that size-segregated mounts are produced by tech­

niques utilizing any form of settling of the <2!-l clay fraction on the mount.

This segregation can be described as a sort of graded bedding where the

large particles concentrate in the lower parts of the mounts and the particle

size gradually decreases upward. Thus, the clay mineral mounting technique

used here results in an enrichment of quartz and feldspar in the lower por-

342 E. ROALDSET

tions, and of the most fine-grained clay material in the upper portions. Gibbs

(1968) assumed that in an 'average' mount of 40 to 100!-t thickness, using

CuKa radiation, 90 % of the X-ray diffraction pattern will originate from

the upper 5-12-jJ.. In the case of mineral segregation on the glass slide, large

deviations may occur. In spite of this, useful additional information concern­

ing the clay material is presumably acquired by semi-quantitative mineral

analyses based on this method.

Mineral identification criteria

The mineral identification criteria were taken from 'The X-ray identification

and crystal structures of clay minerals' edited by Brown (1961), and supple­

mented by methods outlined by Graff-Petersen (1961), Jørgensen (1965) and

Gjems (1967).

By comparing untreated and citrate treated material, certain information

regarding the mineralogy of the mixed layers is obtained.

The diagnostic criteria for the clay minerals are summarized in Fig. 4.

Illite is recognized by a strong first order basal reflection at 10 A which

remains unchanged by thermal treatment (Molloy & Kerr 1961). Jf the in­

tensity ratio I<oot>fl<o0:2) is less than 4, the illite is assumed to be of dioctahe­

dral type (Graff-Petersen 1961). Trioctahedral illite and biotite studied by

Bradley & Grim (1961), have intensity ratios I(Oot)/1<002> in the range of 5-10.

In the present work mixtures of di- and trioctahedral illites are assumed to

have intensity ratios between 5 and 10. Since quartz, chlorite and vermiculite

CLAV MINERAL· IOENTIFICATION

X ·RAV OIFFRACTOGRAM OF ORIENTED AGGRE GATE l FRACIION < 2� l

ØJUED AT 32 •J.REl

HUMIOIT Y

GLYCEROL SATURATED

HEATED TO '50 •c 2 HOURS MINERAL

NO CHANGE NO CHANGE CHLORITE

E XPANOS TO 14.5 A - CO LLAPSES TO 10 A ----VERMICULITE

�X PAN OS TO 17-18 l --+ COLLAPSES TO 10 l MONTMORILLONITE

NO CHANGE - C O LLAPSES TO 10 A ----REGUlARlY I<IXEO lAVE RED SY NME TRICAL PEAK VERMICULITE -ILLITE

NO CHANGE --+NO CHANGE REGULARLY MIXEO LAVERE O SYMME TRI CAL PEAK CHLORITE -IL LI TE

NO CHANGE ....,...COLLAPSES TO 10 l RANO OMLY MIXEO LAYEREO UNSYNMETRICAL PEAK VERNICULITE- Il li TE

'------NO CHANGEI-------RANDOMlV MIXED lAVERED CHlORITE ·Il LITE

E X PAN OS '----+C OLLAPSES TO t0-14A �MIXEO lAVERE O CHLORITE • MONTM ORILLONITE OR CHlORITE- VERMICULITE

'------C O LLAPSES TO 10 l---+MIXEO lAVE RED MONTM ORILLONITE • ILLITE

NO CHANGE --NO CHANGE ------MICA, I<US C O.VITE AND/OR 9\0TITE SVMME TRICAL PEAK

NO CHANGE -+NO CHANGE ------llliTE UNSVMME TRICAL PEAK

Fig. 4. The identification criteria of the clay minerals.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 343

are present in all samples, the (060) reflection cannot be used to differentiate

between the di- and trioctahedral illites. The 1 M and 2 M polymorphs are

identified by reflections in the interval 20°-30°, 28 (Bradley & Grim 1961).

The asymmetrical shape of the illite reflection is interpreted as due to de­gradation of illite, and represents hydrous mica and mixed-layer minerals

with varying amounts of water (Gaudette et al. 1966).

Chlorite is identified by weak first and strong second order reflections at

14 and 7 A respectively, and the mineral is non-expandable. Thermal treat­

ment will cause a small shift in position and an increase in the intensity of the 14 A reflection (Brindley & Ali 1950, Weiss & Rowland 1956). The

mineral is stable up to 450-700 °C.

Substitution of iron for magnesium in the octahedral layer changes the

absolute as well as the relative intensities of the basal reflections (Brindley

& Gillery 1956). The presence of other clay minerals makes it impossible

to determine the composition of the chlorite by the intensity ratios. Clay chlorites with a low thermal stability which could be confused with

vermiculite have been discussed by Brindley (1961). Vermiculite is identified by a strong 14 A reflection shifting to 14.5 A

upon glycerol treatment, and to about 10 A upon heating to 450 °C (Weiss

& Rowland 1956). Vermiculite has a high cation exchange capacity up to

130 mekv./100 g (Brown 1961). With K in the interlayer position the lattice

contracts to about 10 A, and thus acquires the character of biotite CBarshad 1948).

Montmorillonite is identified by the expansion of the mineral to about

17-18 A on treatment with glycerol (MacEwan 1944). Upon heating, the

basal reflection shifts to 1 O A.

Mixed-layer minerals are composed of elementary layers of two or more

types. Mineral structures also occur where the individual components have

a distinct zonal distribution (Shimane & Sudo 1958). When the elementary

layers occur with a fixed periodicity, the term regular mixing is used, while random mixing is used in the opposite case.

Vermiculite-illite, vermiculite-montmorillonite, montmorillonite-illite, chlo­

rite-montmorillonite and vermiculite-chlorite mixed-layer minerals are iden­tified according to the scheme shown in Fig. 4.

In addition, the following minerals were identified: amphibole (8.4 A), gypsum (7.56 A), quartz (4.26 A), anhydrite (3.50 A), potassium feldspar

(3.25 A), low temperature plagioclase (3.18-3.19 A), and calcite (3.03 A)

Table 1. Relative intensities of minerals in the X-ray diffraction diagrams.

Montmorillonite Chlorite K- Na-

glycerol Vermi- Il lite Quartz feldspar feldspar

Calcite

expanded culite

17-18 A 14 A 10 A 4.26 A 3.25 A 3.19 A 3.03 A

0.5 l 2 l l 3 2

344 E. ROALDSET

(Brown 1961). The amphiboles may have strong reflections in the 8.3-8.8 A

and the 3.1-3.2 A regions (ASTM 1960).

Semi-quantitative determinations

Semi-quantitative determinations were performed along the lines suggested

by Johns, Grim & Bradley (1954), and modified by Jørgensen (1965). By

testing standards of known composition and by the method of known addi­

tives, it was found that equal quantities of minerals would give the relative

X-ray intensities shown in Table 1.

The relative intensity of the 14-10 A reflection is based on the intensity

of the 14 and 10 A reflections and is arbitrarily put at 1.5.

The integrated intensity (peak area) is estimated by taking the product of

the maximum peak height and the peak width at half maximum height (Nor­

rish & Taylor 1962).

The intensity data obtained were converted to intensity percentages by using the data in Table 1.

DETERMINATION OF CATION EXCHANGE CAPACITY

The cation exchange capacities were determined as suggested by A. M.

Bystrom-Asklund 1971 (pers. comm.). The exchangeable interlayer cations

are replaced with Sr2+ using a SrC12""solution, and the Sr content is deter­

mined quantitively by X-ray fluorescence analysis. The standard curves are

constructed by adding different quantities of SrC03 to powdered clay

material.

DETERMINATION OF IGNITION LOSS

The ignition loss was determined as the weight loss of material dried at 110 °C after heating for 2 hours at 1050 °C.

X-RA Y FLUORESCENCE ANAL YSIS

The major elements Si, Ti, Al, Fe, Mn, Mg, Ca, and K were determined by X-ray fluorescence spectrography employing a Siemens Sequenz Rontgen­spektrometer SR 1 at Institutt for geologi, Universitetet i Oslo. To achieve a homogeneous sample, clay material dried at 110 °C was mixed with

lithium-tetraborate in the ratio 1:4, heated at 1050 °C for 30 minutes,

chilled and ground to a fine powder and before analysis, pressed into a pellet.

The standard curves were based on artificial mixtures of oxides and lithium­

tetraborate. Most of the standard mixtures were kindly placed at my disposal

by P. Jørgensen. However, Fe and Al rich standard samples were specially

made for this investigation.

NEUTRON ACTIVATION ANALYSIS

Instrumental neutron activation analysis was used to determine Na following

the method of Gordon et al. (1968). Samples of 100 mg were irradiated

together with standards in a JEEP-Il reactor (Kjeller, Norway) at a thermal

neutron flux of 1.5 X 1013 neutrons cm-2 sec-1 for 30 minutes.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 345

The standard rocks AGV-1 and G-2 were included in the experiments.

These provided a check on the accuracy of the analytical procedure.

Sodium was determined by the 24Na isotope with a half life time of 15

hours. Duplicate analyses were within 15 %. The counting equipment involved in this work was a Ge(Li) ORTEC 24

cm2 coaxial detector with a resolution of 3.2 KeV (FWHM) for the 1331

Ke V peak of 6°Co. The sample preparation as well as the counting were

carried out at Mineralogisk-geologisk Museum, Universitetet i Oslo.

STATISTICAL ANALYSIS

The mineralogical and chemical data were analysed statistically by the stan­dard IBM programme FACTO edited by W. J. Dixon, UCLA, 1964. The programme is written in FORTRAN and processes average content, correla­

tion coefficients and R-mode factor analysis.

Results and discussion GRAIN-SIZE DISTRIBUTION

The moraine samples were grain-size analysed and the characteristic distribu­

tion curves are shown in Fig. 5. The moraines are normally nonsorted and

without any stratification (exemplified by E 10). However, a few exceptions

can be found. Samples E 8 and E 16 are enriched in a bluish-grey clay.

Moraine samples E 2 and E 21 are relatively enriched in the sand and silt

fractions. The clay content is low and the finest fractions are presumably

washed out. The distribution curves of samples E 8, E 11 and E 15 show

these sediments to have bimodal distribution indicating that the samples

could be mixed sediments.

CLAV 100

.... BO z LIJ u et: 60 LIJ CL. l-:I: 40 C) LIJ �

30

o

SILT SAND GRAVE L

.. ·•····

.·

...;;

l,.. ....... .......... �·g ........ l ,, X;...--

�� " æj� J,/ " o

�

p')/ �� . o"'

.tr'� .... ,

.. ,

... r�/ .... �. / -·� - .-c' /o·

,,c�/ ,; · "

,. _&/ "

_, �-::"·· l ,/ �-- -_;, . .. ··· o· ... .. ·- · - · -

o" /

v>< ILO

6 20 60 10 011 2

GRAIN SIZE

o· /

........... o----o·-·-x--

6 1mm 2

E2 E B t----

E10 t--E 16

4 6 10mm

Fig. 5. Characteristic grain-size distribution curves of moraine samples.

346 E. ROALDSET

30 25

l

30 25

Treated with sodium citrate

20 15

Trealed with

sodium citr ate

' 5

20 15

10

10

7 10

10

20 o

A

5 28

J Fig. 6. X-ray diffracto-

grams of E 53-vermi-

culite rich marine clay

(below) and E 13-chlo-

rite rich morainic clay

(above) illustrating the

different effects of

20 1 Mehra & Jackson's

5 28 (1960) rinsing method.

CLAY MINERAL ALTERA TJON BY SODIUM DITHIONITE- SODIUM CITRATE

TREATMENT

Iron oxide was removed from the sample material by reducing ferric ions by sodium dithionite and complexing the divalent ions with sodium citrate. An

optimum pH for maximum reaction kinetics occurs at approximately pH 7.3. Therefore a sodium bicarbonate buffer was used to hold the pH at the

optimum level. The method was described by Mehra & Jackson (1960), who

found no mineralogical alterations during the rinsing process. This was also

assumed by Gibbs (1967).

However, in a series of clay samples from the Numedal area, the clay

mineralogy was undoubtedly altered by this process. This is indicated by

the X-ray diffractograms and the cation exchange capacities (CEC) of un­

treated and citrate treated material.

The changes in X-ray characteristics accompanying the sodium citrate ex­

traction are illustrated in Fig. 6. For the moraine material and some marine

clays the removal of amorphous coatings and free iron oxides brings out

some X-ray diffraction peaks that are otherwise difficult to detect. For clays

from the central part of the valley system, mostly marine, pronounced

changes in X-ray characteristics accompanied the sodium citrate extraction.

For the latter group changes in the X-ray diffractograms appear as a

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 347

Table 2. The cation exchange capacities.

CEC (meqv/100 g) Clay mineral content (untreated sample) (intensity %)

Sam p le Untreated

Citrate- chlorite vermiculite treated Ill i te and and

sample sample illite-chl. vermic.-chl.

E 2 2.3 5-10 40 5-10 E 32 20 126 40 40 E 42 20 104 25-40 5-10 40 E 44 16 59 40 5-10 5-10 E 45 18 70 40 5-10 10-25 E 53 14 31 10-25 10-25 40 E 62 19 89 25-40 10-25 25-40 E 73 11 62 25-40 25-40 10-25

reduced size of the 14 A peak accompanied by an increase in the 10 A peak and a pronounced 14-10 A background (Fig. 6) while the gypsum and an­

hydrite reflections disappear. For these samples the cation exchange capaci­

ties increased remarkably as a result of the citrate treatment. The results

of the cation exchange capacity determinations are presented in Table 2. The lowest CEC values are obtained for sample E 2 which mainly consists

of chlorite and illite-chlorite mixed layer minerals, and the highest values

are obtained for samples with a high content of vermiculite and illite-vermi­

culite mixed layer minerals. The effects of the sodium-citrate extraction illus­

trated by the X-ray diffractograms were most pronounced for samples rich

in vermiculite and chloritized vermiculite.

A dose relationship between the presence of vermiculitic minerals and the

cation exchange capacities was predicted by Jørgensen (1965).

Based on the above results the mineral alterations are interpreted as fol­

lows: A collapse of the unstable chlorite and vermiculite minerals to 10--11 A.

The minerals have possibly been transformed to illite. A collapse of the vermiculite structure to 12.6 A as Mg from the interlayer

positions is exchanged for Na from the citrate solution. The basal spacing of Na saturated vermiculite is 12.6 A (Barshad 1948).

Gypsum and anhydrite minerals dissolve. R. W. Berry 1969 (pers. comm.) applied Mehra & Jackson's (1960) method

to different sized fractions of a quick clay sample and found that the alter­ation of the material increased as the particle size decreased.

Quigley & Martin (1963) studied chloritized weathering products of glacial

tills and found that citrate-extractable chlorite existed to a depth of 55

inches. Their method differed slightly from Mehra & Jackson's (1960) in that

the temperature during the rinsing process was kept at 100 °C instead of 80

°C. The extractable chlorite breaks down under heat treatment as low as

350 °C. This is in accordance with the results in this work.

348 E. ROALDSET

CLA Y MINERALOG Y

Description of individual X-ray diffractograms

The X-ray curves of some samples have been selected to display the charac­

teristic features of the different mineral groups. The moraine and the marine

clays are divided into three and two groups respectively, while the recent

lacustrine and marine sediments are grouped separately.

In the Figures the terms 'untreated' and 'sodium citrate treated' refer to

the material before glycerol and heat treatment.

The semi-quantitative composition of the clay material is given in Table 3.

3 4 5 7 10 14 Ad

25 20 15 10 50 29

l»'"' humidity

. l E 5 LJ�"'·'

l;' f 4 i 1 110 210 Ad t

So is io i l l 29 10 so

32�o re l.

E 16

Fig. 7. X-ray diffractograms for oriented samples of moraine clays: E 5 Moraine clay with a high illite and chlorite content. E 16 Moraine clay with a high illite content.

(:lumidity

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 349

E 1-3, E 5-9. These samples are from the moraine areas in Numedal and have almost identical X-ray diffractograms (see E 5 in Fig. 7). The clay material is unweathered and gives distinct X-ray diffraction peaks.

The glycerol treatment did not change the 14 A reflection, but a small peak in the 18-14 A area might appear. Upon heating the 10 A reflection increased, that at 14 A decreased and the one at 18-14 A disappeared. Thus, according to the diagnostic criteria, the clay material consists mainly of illite and chlorite, with smaller amounts of vermiculite and mixed layered chlorite-illite, illite-vermiculite and chlorite-vermiculite, amphibole, quartz, microcline and acid plagioclase. Small amounts of montmorillonite are found in some samples. The illite intensity ratios I(oo1)/I(o02) = 3-7 are suggestive

of a mixture of di- and trioctahedral illite, the greater part dioctahedral. The X-ray reflections between 20° and 30°, 2E>, show the 2M polymorph to be present.

E 4 and E 16. The samples (see E 16 in Fig. 7) are from the moraine areas. The clay mineral content is remarkably low in sample E 4 and unusually high in E 16. In spite of this difference the X-ray diffractograms are very similar.

After glycerol treatment reflections in the 18-14 A area appear. Upon heating the 10 A reflection increases, the 14 A reflection decreases and the 18-14 A reflection disappears.

The samples contain illite, and small amounts of chlorite, vermiculite, montmorillonite and mixed layered illite-chlorite, amphibole, quartz, micro­cline and acid plagioclase. The illites are of dioctahedral as well as of tri­octahedral type and have 1M structure.

E 10-15, E 17-23, E 34, E 36-37, E 39-41, E 65-66. Most of these samples are from the central parts of the valley system or from the Cambro-Silurian areas near Kongsberg. The group consists of moraine and fluvial sediments, recent river mud and river water (see E 22 and E 37 in Fig. 8).

The most characteristic feature is the presence of considerable amounts of vermiculite and randomly mixed-layer illite-vermiculite, illite-chlorite and chlorite-vermiculite. Other minerals present are illite, chlorite, traces of montmorillonite, amphibole, quartz, microcline and acid plagioclase. The illite intensity ratios I(ooi)/l(o02) = 4-6 are suggestive of a mixture of di- and trioctahedral illites. The 2M polymorph is present.

Reflections due to gypsum and anhydrite are found for the samples E 34 and E 41. The sulphate minerals are believed to originate from Cambro­Silurian host rocks.

E 24-26, E 30-33, E 42-45, E 53, E 60-64. The material of these samples was deposited under marine conditions in the central parts of the 'Lågendal­

fjorden' (see E 33 in Fig. 9). The X-ray diffractograms show great similari­ties with the above group of vermiculite rich clays. All samples have a high

350 E. ROALDSET

3

30

3

30

25

E 22

25

4 5

20

4 5

20

7 10

15 10

7 10

15 10

20 Ad

se 2a

320fo rei. humidity

20 Ad

32"fo rel. humidity

Glycerol

200° c

450°C

Fig. 8. X-ray diffractograms for oriented samples of fluvial and moraine clays. E 22 Citrate-treated river mud with a high illite and vermiculite content.

E 37 Moraine clay with a high content of illite and vermiculite.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 351

E 59

l 25

4 l

'

i o

l 20

5 l

l 15

7

7 l

l 10

10 1'01

1p l 5 28

Untreated 32'/. rel. humidity

Citrate tr.,aled 32'/. rel. humidily

Untreated +

glycerol

450° c

20 Ad

�·l 29

32.,-o rel. wmidity

\J Glycerol

. �

450"C

Fig. 9. X-ray diffractograrns for oriented sarnples of marine clays: E 33 Marine clay with a high content of illite and verrniculite. E 59 Marine clay with a high content of illite and chlorite. Notice that the gypsurn reflection (7.56 A) disappears upon heating.

352 E. ROALDSET

content of illite, vermiculite and mixed-layer illite-vermiculite. Chlorite

minerals are of secondary importance. The minerals identified are: illite,

chlorite, vermiculite, mixed-layer illite-vermiculite, illite-chlorite and chlo­

rite-vermiculite, amphibole, quartz, microcline and acid plagioclase. Traces

of montmorillonite are occasionally found. The illite intensity ratios I(oot)/ 1(002) = 6-10 indicate the presence of considerable amounts of trioctahedral

illite. The 2M polymorph is identified.

In all samples the clay mineral assemblages are altered by the sodium­

dithionite sodium-citrate treatment (Figs. 6 and 8).

E 27-29, E 54-59, E 67-74, E 76. The clays in this group also were depo-

4 l

i o 5 l

E 78

30 io

7 l

i o

}j

1 i

10

� 28

v�;;����·

450° c

2,0 Ad � 28

450" c

Fig. 10. X-ray diffractograms of recent lacustrine (E 78) and recent marine mud (E 83).

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 353

sited in a marine environment (see E 59 in Fig. 9). Here the most common

feature is a high illite and chlorite content. The sodium-citrate treatment

causes no noticeable changes in the X-ray diffractograms. It is noteworthy

that the X-ray reflection of gypsum (7.56 A) disappears by the sodium

citrate and heat treatments. The mineralogical variations among samples

from different localities and among samples from various depths within a

profile are of little significance.

The clay fraction is made up of the following minerals: illite, chlorite,

small amounts of vermiculite, mixed-Iayer illite-vermiculite, amphibole,

quartz, microcline and acid plagioclase. The Asrum samples (E 67-74, E 76)

also contain mixed-layer vermiculite-chlorite, and the Hvittingfoss samples

(E 54-59) contain gypsum. The illite intensity ratios I(oot)/1<002) = 5-10 are

suggestive of mainly tri- but also dioctahedral illites. The 1M polymorph is common, but the 2M type is also identified.

E 77-79, E 83-85. The sample material consists of recent Iacustrine and

marine mud (see E 78 andE 83 in Fig. 10) with a high content of organic

matter. Compared to the illite-chlorite group of marine clays these samples

are relatively enriched in vermiculite and mixed-Iayer illite-vermiculite. The illites are of di- and trioctahedral types.

The mutual resemblances are greater among the samples of Recent lacu­

strine mud than for the corresponding marine samples (Tab le 3 ). Both

groups are presumed to originate from material transported down the river

Numedalslågen. The illite content of the marine mud seems to increase as

the salinity of the pore-water increases (see pp. 5, 6 and Table 3).

The X-ray diffractograms of samples E 80 andE 81 are similar to those of marine mud. The clays were deposited in Larvikfjorden and then Iifted

above sea leve! by the isostatic uplift of the land.

Semi-quantitative mineralogy

Table 3 shows the semi-quantitative composition of the different samples. The mixed-layer minerals are usually randomly interstratified and therefore a

further division into regular and randomly mixed-Iayer minerals is not applied. Both groups are referred to as mixed-Iayer minerals.

All samples contain illite, which in moraine clays amounts to 10-40%

and in marine clays to 40-60 % of their composition. The illites are of di­as well as of trioctahedral type, the trioctahedral being the most common

type in the marine clays.

In addition to illite, all samples contain chlorite (3-40 %), vermiculite (up

to 40 % ), various amounts of mixed-layer minerals, quartz (2-22 % ), micro­

cline (2-22 %), and acid plagioclase (1-20 %). Most of the non-marine

samples contain amphibole (normally below 10 %). Traces of montmoril­

lonite are found in some samples. The montmorillonite minerals are always

of less importance. Mixed-Iayer illite-montmorillonite and chlorite-mont­

morillonite are found only in the following samples: E 2, E 7, E 39 andE 40.

354 E. ROALDSET

Table 3. Mineralogical composition of Numedal clays .

E l

E 2

E 3

E 4

E

E 6

E 7

E B

E 9

E 10

Ell

E 12

El3

E 14

E 15

E 16

E 17

E lB

E 19

E20

E 21

E 22

E 23

E 25

E 26

E 27

E2B

E 29

E 30

E 31

E 32

E33

E 34

:l ... .... .... H

XX

X

XX

XX

XX

XXX

XXX

XX

XX

XX

XX

XXX

XX

XX

XX

xxxx

XX

XX

XX

XX

X

X

X

xxxx

xxxx

xxxx

xxxx

xxxx

XXX

XXX

xxxx

xxxx

XXX

E 35 xxxx

E 36

E37

XX

XX

-E 38 xxxx

E 39

E 40

E41

XX

XX

XXX

XXX

XX

XX

X

XX

XX

XX

X

XX

XX

XX

X

XX

XXX

XX

XX

X

X

XX

XX

XX

X

X

XX

XX

XX

XX

XX

X

X

XX

XX

X

.. ... ... �

� .. ...

X

X

X

X

X

X

X

X

X

X

X

X

X

X

XX

X X

XXX

X

XXX

XXX

X

X

X

XXX

XXX

XX

XXX

XX

XX

XX

X

X

XX

small. but observable amounts

X • - lO"

XX • 10 - 25 f.

:l ... g .... :;::

j " o :0:

X

X

.. ... ... � :;l " l

:l ... .... .... H

X

XX

X

X

XX

XX

XX

XX

XX

XX

XX

XX

X

XX

XXX

X

XX

XX

XXX

X

XX

l

il .. � :: ...... �r-t! ...... HO

X

X

XX

X

X

X

X

X

X

X

XX

X

X

XX

XX

XX

XX

X

XXX

XX

X

XXX

XXX

XX

XX

X

XX

X

X

.. ...

1 ... .. .... ... .. �; ...... "'" 0 ..

X

X

X

X

X

X

X

X

X

X

.. .... o .D

l

XX

X

X

X

X

X

X

X

XX

X

X

X

X

X

XX

X

JIICC - 25 - 40 "

..

j X

X

X

XX

XX

X

X

X

X

X

X

X

X

X

X

XX

XX

XX

X

X

xxxx • more than 40 "

:! ... .... o o .. " ... :0:

XX

XX

X

XX

XX

X

X

XX

XX

XX

XX

XX

XX

X

XX

XX

XX

XX

XX

XX

X

XX

XX

X

X

XX

X

XX

XX

XX

XX

XX

X

.. " .. <l

j .... "'

XX

X

XX

XX

X

X

X

X

X

X X

XX

X

X

X

X

X

X

X

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 355

E 42

E4}

E 44

E 45

E 47

E 48

E 49

E 50

E 51

E 52

E 53

E 54

E 55

E 56

E 57

E58

E 59

E 60

E 61

E 62

E 63

E 64

E 65

E 66

E 67

E 68

E 69

E 70

E7l

E 72

B73

E 74

E 75

E 76

E 77

E78

E 79

EBO

E 81

E 82

E 83

E 84

E 85

" .. ... .... .... ...

XXX

xxxx

JCCCX

xxxx

XXX

xxxx

xxxx

=

xxxx

xxxx

XX

xxxx

T.XXX

xxxx

xxxx

xxxx

xxxx

.XX

xxxx

XXX

xxxx

XX

xxxx

XXX

XXX

XXX

XX

XX

XXX

XXX

XXX

XXX

XXX

xxxx

XXX

XX

XXX

XXX

XXX

=

XX

X

X

X

X

X

XX

X

X

X

X

X

XX

XX

XX

XX

XX

XX

X

X

X

X

X

X

X

X

X

XX

XX

XX

XX

XX

XX

X

X

X

X

XX

X

XX

X

XX

X

X

XX

XX

XX

XX

XX

XXX

XX

XXX

XX

XX

XX

XX

XX

XXX

XXX

X

X

X

X

X

X

XX

XX

XX

X

X

XX

XXX

.. .. ... g .... .... ... ..

! �

X

small, but obeervable amounta

X • -.10"

XX • 10 - 25 "

XX

X

X

X

X

X

X

X

X

X

X

X

X

X

XX

XX

X

X

XX

X

XX

XX

X

X

X

X

""

X

X

XX

XX

XX

XX

XX

XX

XX

XX

XX

XX

X

XX

XX

X

XX

XX

XX

XX

X

X

XX

XX

" .. .... ...... .... ... .,

�! o:

X

X

X

X

X

X

X

X

X

X

.. .... .&

j

X

X

X

X

X

XX

XX

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

XXX •25-40"

X

X

X

XX

X

XX

X

X

X

X

XX

XX

XX

X

X

X

XX

X

XX

xxxx • more than 40 "

� .... g .. " iii!

XX

X

XX

X

X

XX

XX

X

XX

X

XX

XX

XX

XX

XX

XX

X

X

X

X

"

XX

XX

XX

X

XX

XX

X

X

X

X

XX

X

XX

"

XX

: • ... "

j ;;:

XX

X

X

XX

X

X

X

XX

X

X

X

X

"

X

"

XX

XX

XX

X

"

"

X

X

"

X

356 E. ROALDSET

Vermiculite-montmorillonite mixed-layer minerals are not found in any of

the samples. The montmorillonite minerals are more frequent in the moraine

than in the marine samples. This is consistent with results obtained by

Gjems (1967), who generally found a low montmorillonite content in soil

samples from the Precambrian and Permian areas of southwestern Norway. Calcite is present in Cambro-Silurian shales (E 35, E 38) and in marine

clays from Kaupang (E 80, E 81). Small amounts of gypsum and anhydrite

are found in samples E 34 and E 41, both from the Cambro-Silurian areas, in the Hvittingfoss samples (E 54-59), and in marine mud (E 85).

Within each of the different marine profiles the relative proportions of

each mineral vary down the profile. The mineral changes appear as changes in intensity of the reflections due to illite, vermiculite and vermiculitic mixed­

layer minerals (Fig. 11).

The mineralogical data obtained are in accordance with earlier results

(Rosenqvist 1942, Collini 1950, Soveri 1956, Jørgensen 1965, Gjems 1967).

Summary and discussion of X -ray data The clay-mineral assemblages may be expected to differ substantially for the

following reasons:

Variations in composition of primary rock. Variations in physico-chemical conditions of the weathering, transport and

sedimentation processes.

Post sedimentary changes. According to Låg (1948) a great deal of the moramtc soils in Norway originate almost exclusively from the underlying bedrock. Låg (1948) also found a relationship between the morainic and the other sedimentary soils, as the latter may have been washed out of the morainic soils.

On the basis of the data for the Numedal clays, there seems to be no absolute relation between the composition of the parent material and that of the clay material. A high illite content of the clays is presumably con­sistent with an influence from Cambro-Silurian parent rocks. The physico­

chemical variations during transport and sedimentation, however, are as­

sumed to be of significant importance.

Taking the sedimentological environments into consideration it is clear

that clays transported down Numedalslågen and deposited in the central

part of the valley system, have a high content of vermiculite and mixed­

layer illite-vermiculite and chlorite-vermiculite. Undisturbed moraine clays

and little-transported clays supplied from the sides of the valley normally

have a high illite and chlorite and a low vermiculite content.

An evaluation of the semi-quantitative data (Table 3) shows the amount of vermiculite and mixed-layer illite-vermiculite to increase as the amount of

trioctahedral illite, chlorite and interstratified chlorite decreases. The rela­

tionship is illustrated in Fig. 11.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 357

PROFILE DATA MINERALOGY GEOCHEMISTRY DEPTH (metrtsl

CLAY MINERAI..OGY

Ou • r1,_10 •lt4 stOO"J.)

- IM Chlorite

OTHER HINERALS !'-of totalsample}

O 114-lolllite•chtorite and illite-vermiculite

Gllii] 11olllite

---- PARAMETER VALUE -----

Fig. 11. Mineralogical and geochemical variations within the Aserum profile (E 67-74) plotted vs. profile depth.

al= 'mg= MgO

MgO + F�03 (total) + MnO

fm = (MnO + MgO + Fe203 (total)) X 100 _

k = K20

X ' K20 + MgO

CaO X 100 (Na20 + K20) X 100 c = ---'--'--- alk = _;_.......::. __ ....::..._..:._ __

X '

X

The chemical milieu prevailing in the water of Numedalslågen and in the

� horizon of podsol profiles, with pH-values generally about 4 and 5, sug­

gests that the formation of vermiculite from biotite and chlorite may have

occurred here. During the weathering, transport and sedimentation phases intermediate between illite, vermiculite and chlorite groups may be formed.

In laboratory experiments it has been shown that biotite, particularly phlogopite, is altered to vermiculite on substituting the interlayer K+ by

Mg++ and vice versa (Barshad 1948). In biotites the rate of K+ release is influenced by the oxidation of Fe++ in the octahedral sheet of the structure (Reichenbach & Rich 1969). Basset (1959) proposed that the removal of K+

from one interlayer by hydrated cations may induce stronger bonding of K+ in the next interlayer.

In podsol profiles biotite minerals may be converted to mixed-layer

biotite-vermiculite (Johnson et al. 1968), or be completely transformed via

hydrous mica to vermiculite (Walker 1947, 1949, 1950). Trioctahedral mica

is suggested to be unstable in the acid surface soil horizons (Mitchell 1955).

The formation of vermiculite from chlorite with a mixed-layer mineral as

an intermediate stage in this development was described by Jørgensen

(1965). The process was assumed to be an exchange between Mg from the

brucite layer and protons in exchangeable positions.

358 E. ROALDSET

Tab le 4. Chemical composition of Numedal clays (wt. % ).

B l

11 2

B 3

E 4

l

E 6

B 7

E 8

E 9

11 10

E11

E 12

El4

B 15

E 16

E 17

E 18

E 19

B20

E 21

126

E 27

E28

E 29

B 30

E 31

l! 32

B"

BH

B 35

E 36

E'7

l! :58

E 39

B41

B 42

B 43

E44

B 45

E 47

E 48

E 49

B 50

E 51

B 52

B 53

B 54

B 55

E 56

B 57

B58

E 59

49.6 0.91

53.4 t o.92

45.1 0.92

52.0 o.11

56.2 o. 74

52.6 0.68

50.0 0.83

49.9 0.93

49.8 1.09

52.0 o.66

51.0 0.69

54.5 0.77

38.4 o.n

48.5 0.85

51.5 0.79

49.2 0.76

58.3 o. 72

48.4 1.07

48.0 1.37

51.8 0.91

47.2 0.80

47.5 0.95

51.1 0.94

51.7 0.77

52.0 o. 77

51.5 0.78

51.8 0.87

50.5 0.94

50.1 1.10

49.5

55.2

53.1

56.5

0.78

0.87

0.95

1.01

56.2 0.80

42.2 0.82

51.7 0.85

51.0 0.87

53.0 o.86

55.4 1.02

52.2 0.92

51.8 0.97

50.0 0.86

51.5 0.85

52.5 0.84

52.0 0.87

48.7 0.95

51.7 0.80

55.2 0.96

52.0 0.89

53.2 0.92

53.7 0.87

53.0 0.90

16.8 13.4

15.8 12.0

18.2 9.8

19.9 11.5

15.7 9.3

16.5 10.0

18.0 11.9

18.8 12.3

17.5 15.4

17.0 8.8

15.5 9.5

17.6 9.6

18.7 11.8

22.2 11.2

14.9 12.8

17.1 10.9

18.9 7.4

17.0 11.8

19.8 9.4

12.5 7.9

21.5 12.7

18.9 14.3

15.7 12.0

15.4 11.6

15.9 11.6

16.1 11.9

15.6 11.7

15.9 13.3

17.3 7.1

14.6

19.7

15.4

18.2

4.3

8.9

11.1

2.7

19.6 7.5

14.0 17.6

15.1 11.7

15.2 12.3

15.9 11.8

15.0 14i4

16.2 12.7

14.9 13.6

16.1 13.5

15.0 12.9

15.8 11.5

15.6 12.3

16.4 15.1

14.4 q.9

15.1 11.2

15.4 11.9

15.2 12.4

14.9 10.1

15.1 13.o

o.26 '· 78 3.82 1.97

0.15 3.62 3.36 2.14

o. 21 2.81 4.65 1.85

0.35 2.10 3.65 2.12

0.12 l. 96 3.12 3.13

o.13 3.42 3.56 1.:58

0.23 2.81 3. 76 2.14

0.19 3.68 o. 75 2.17

0.20 4.31 1.10 1.43

0.22 3.14 2.86 2.20

o.14 8.6o 3.94 2.81

0.16 3.82 4.10 2.19

0.11 1.15 1.61 1.13

0.13 2.:58 2. 75 2.30

0.19 4.00 4.45 2.67

0.18 3. 94 4.45 2.66

0.11 3.40 3.11 2.50

0.:58 3.84 3.82 2.31

0.12 2. 70 1.60 1.62

0.10 2.40 l. 78 2.30

0.16 4.30 3.95 1.20

0.16 4.85 0.55 2.25

0.16 5.28 1.62 2.20

0.17 5.07 4.18 1.95

o.l8 4.15 3.95 1.58

0.17 4.57 4.15 1.42

0.15 4.10 1.56 2.11

0.18 5.15 1.57 2.00

0.07 2.60 3.46 2.35

0.05 2.21 11.90 0.59

o.�5 2.68 1.53 1.89

o.n 3.12 1.57 1.64

o.o3 2.40 o.55 o.46

0.14 3.00 2. 95 l. 78

0.06 2. 52 2.10 1.11

0.17 4.25 1.60 2.06

0.20 4.45 2.48 2."

0.19 4.70 1.88 2.34

0.18 4.65 l. 70 2.55

0.18 4.38 1.:58 2.31

0.20 4.06 1.37 2.64

0.16 4.28 1.38 2.12

0.15 4.09 1.66 2.20

0.14 4.15 1.64 2.09

0.14 4.22 1.53 2."

0.19 4.:58 0.99 1.56

0.21 4.66 2.85 2.32

0.19 4. 92 2. 58 2.40

0.17 5.35 2.55 1.93

0.17 5.16 2.78 2.22

0.18 4.71 2.93' 2.37

0.17 5.25 2.96 2.37

5.00

4.62

5.11

5.20

5.10

5.07

5.30

5.65

5.63

4.10

5.56

5.67

2.00

3.74

5.30

5.12

4.55

4.35

3.90

2.69

5.50

5.83

5.72

5.45

5.05

5.15

5.53

5.52

3. 74

Loaa 1p1UOil

5.30

4.07

12.60

4.45

3.92

6."

5.05

6.08

5.58

6.00

3.39

2.4}

25.80

9.50

3.75

7.10

3.26

9.20

12.40

17.50

5.77

3.42

4.77

4.10

4.60

5.20

·5.00

5.85

13.40

5.13 11.70

4.55 7.50

4.97 7.90

7.35 9.65

4.10 5.37

4.19 16.20

5.64 4.55

5.70 4.:58

5.70 3.50

6.30 4.00

5.69 5.00

5. 96 4.45

5.70 4.05

5. 72 3.80

5.60 3. 76

5.n 4.14

5.48 5.70

5.68 3.60

5.19 2.55

5.20 3.71

5.24 3.81'

4.77 4.50

5.17 3.71

100.84

100.28

101.30

lol.96

99.24

99.67

100.02

100.65

102.04

96.93

101.13

100.85

101.43

103.55

100.35

101.41

102.25

102.17

100.91

100.08

103.08

96.71

99.49

100.39

99.78

100.94

96.42

100.91

101.24

100.71

102.97

99.88

96.82

101.64

100.80

97.62

96.91

99.87

105.20

100.96

99.95

96.15

97.87

96.02

96.86

99.47

96.14

100.29

99.10

101.19

99.03

101.65

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 359

E 60

E 61

E b2

B 63

E 64

B 65

B 66

E 67

E68

E 69

E70

E71

E 72

E�

E 74

l! 75

E 76

E78

E80

E Bl

E 83

53.0 0.98 15.2 B.2 0,14 4.77 1,86 2,19 5.00

53.8 0.95 15�7 11.4 0,12 4.35 1.95 2.38 5.02

47.7 0.89 15.0 11.0 0.16 4.04 1.83 1.80 4,60

51.5 0,94 15.7 12.3 0.17 4.40 1,70 2.08 5.05

52,9 1.05 19.3 1o.o o.n 3.90 1.90 1.92 5.oo

49.6 1,07 15.7 14.0 0,12 5.15 2,52 1.59 3. 75

51.3 0.97 18.3 11.7 0.16 4.25 2 00 2.16 3.96

52.4 0.96 16.7 1}.2 0.14 4.30 1,62 1.79 5.18

55.0 0.92 15.8 11,2 0.15 3.82 1.98 2.32 4.97

63.8 0,86 13.7 7.2 0.10 2,76 2.35 2.76 4.44

62.2 0,97 12.6 7.7 0,10 2.53 2.53 2.66 4.23

59.2 0,85 14.2 9.1 0.13 3.52 2.26 2.57 4.99

49.9 0,90 15.2 B.6 0,16 5.09 l. 72 2.52 5,62

52.5 0.90 16.0 11.2 0,17 4.56 1.95 2.34 5.32

53.5 0,89 15.7 11.1 0.15 4.38 1.95 2.39 5.41

50.8 0.85 19.6 9.8 0,14 4_.25 1,84 2.23 5.16

51.2 o.BB 16.5 12.1 0,15 4.90 1,76 2,02 5.57

51.5 0,92 17.6 12.2 0.26 3.22 1.75 1.66 4.07

55.1 0,67 15.1 9.6 0.07 3.20 0.87 2.06 5.00

53.4 0,84 15,5 10,5 0.14 4.71 3,21 1,80 4,95

58.0 1.04 15.4 8,3 0.09 3.10 1.61 2.44 4.15

Loes StM 1gn1t1on

4.25 100.59

4. 75 100,42

10,00 97.02

5.05 98.89

4.42 100,52

5.50 99.00

5.60 100,40

4.25 100.54

4.80 100.96

2.94 100.86

4.62 100.14

3.58 100.40

5.95 100,66

4. 70 99.64

4.75 100.22

5.80 100.47

4.95 100.03

5.35 98.53

8,60 100.27

5.55 100,60

6.00 100.13

Walker (1961) assumed the vermiculite minerals always to be of secondary

origin formed by the alteration of rock forming mica, pyroxene, amphibole, chlorite and feldspar.

The stability of chlorite minerals is elucidated by the stability relations in

the system involving pure Mg-chlorite, phlogopite, kaolinite, K-mica and

K-feldspar (Garrels & Christ 1965). An interesting aspect of the diagram is

that pure Mg-chlorite is unstable at pH-values below 7. This confirms the impression gained from the X-ray diffractograms, namely that the chlorite minerals are unstable in acid surface water and in weakly acid river water.

The dioctahedral illites are always less influenced by chemical weathering

than the trioctahedral illites and chlorites (Droste 1956, Battacharya 1962, Jørgensen 1965).

The remarkably high illite content of the marine clays may be noted. The environmental conditions during the river transport (pH 5) are different from those prevailing during sedimentation in seawater (pH 8). This means that

the mineral paragenesis relatively stable in the river is not necessarily stable

in the sea. The environmental changes may result in mineralogical alterations

which here are expressed as an increasing illite content. During the weather­

ing processes the potassium ions have been partially removed from between

the unit layers of the illite minerals. The illites so 'degraded' are then carried

to the sea, where they rapidly pick up potassium again (Yoder & Eugster

1955). In recent marine mud from Larvikfjorden the illite content increases

with the distance from the river outlet.

360 E. ROALDSET

Table 5. Average chemical composition of Numedal clays.

Weight Standard deviation

% (%)

Si02 51.8 3.92

Ti02 0.89 0.12

Al203 16.5 1.94

Fe203 11.3 2.08

Mn O 0.16 0.05

MgO 3.98 1.07

Ca O 2.36 1.04

Na20 2.10 0.41

K20 5.00 0.75

Loss ignition 5.92 3.52

CHEMICAL ANALYSES

The chemical analyses are given in Table 4, and mean values and standard

deviations in Table 5. The data refer to material dried at 110 °C.

Recalculations of the chemical analyses

To facilitate comparison of the profile samples the weight per cent of each

was recalculated to Niggli parameters.

During the weathering processes the oxides Al203, "K20, Si�, and Ti02

are concentrated, while MgO, CaO, Na20 and FeO are removed. Normally

an increase in the Al203 content is accompanied by an increase in the K,.20 content. In his presentation of analyses of Caledonian sediments Vogt (1927) introduced the following ratio:

V= mol (MgO + CaO + Na20)

According to Vogt, this ratio indicated the degree of residual character of

the sediment and was therefore applicable as a maturity index or a residual index.

Geochemical variations

When comparing the geochemical data for the clay fraction of moraine and

marine material, small variations are found. The moraine clays normally

have slightly higher Ti02, Al20a, Fe.203, MnO and K20 contents and

slightly lower Si02, MgO, CaO and Na20 contents than the marine clays.

Magnesium is the only element notably enriched in marine clays. The

average chemical composition of the ela y fraction ( <2!!) of the Numedal

sediments (Table 5) corresponds to that for coarse clays given by Pettijohn

(1957).

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 361

Silicon. The Si02 content varies roughly antipathetically with the clay

mineral content and ranges from 38 to 64 wt. %. Most of the samples have a Si(h content about 52 %. A low Si(h value is presumably correlated with

a high content of chlorite and vermiculite minerals. Titanum. The Ti(h content varies between 0.66 and 1.37 wt. %, whiJ::h

according to Pettijohn (1957) is higher than normally found in clays. Gold­schmidt (1954) has suggested that only small amounts of Ti are chemically or physically bound to the clay minerals, but that the Ti of hydrolysates is

probably present as very finely crystalline Ti(h (or Ti(h-hydrate) deposited along with the tiny flakes of the clay minerals. Vinogradov & Ronov (1956)

found a strong positive correlation between Ti02 and Al203 in clays from the Russian Platform. Analyses frequently show a high Ti content in biotites, where it may replace Si in tetrahedral, or Fe, Mg in octahedral sites (Deer, Howie & Zussman 1966). Thus, a high Ti content in clays may be related

to the amount of trioctahedral illite and the decomposition of biotite is a likely Ti source.

Magnesium. A high MgO content in marine samples seems to be con­sistent with the content of trioctahedral illite and vermiculite, and thus in

turn due to the composition of the parent material. It may also be considered that the Mg content of marine clays is increased by diagenetic uptake of Mg++ from the seawater. However, because of the similarities between the Mg-values for lacustrine and marine mud (samples E 78 andE 83) the latter effect is believed to be of less significance.

Manganese. The MnO-content varies from 0.06 to 0.38 wt. %. According to Goldschmidt (1954) the ionic potential of divalent manganese is low and manganese is therefore vulnerable to leaching. In hydrolysate sediments the manganese content depends very much on conditions of sedimentation. Under reducing conditions manganese is selectively removed from the hydro­

lysate material, while under oxidizing conditions a selective manganese fixa­tion takes place (Goldschmidt 1954). The process of selective solution under reducing conditions will also affect iron, but manganese will be preferentially dissolved. Berry & Jørgensen (1971) apart from this found the MnO-content to increase in the finer colloidal fractions relative to other size fractions.

For the Numedal clays it is clear that a high vermiculite and mixed-layer illite-vermiculite content is coherent with a low MnO content. These clays have presumably been under weakly reducing conditions.

A plot of MnO content vs. Vogt's residual index (Fig. 12) shows the relationship between the manganese content and the environmental condi­tions. During the weathering processes the manganese is dissolved.

The similarities between recent lacustrine and marine mud do not include the MnO content which is high in the oxidizing lacustrine milieu, while low in the reducing marine milieu.

To a large extent the chemical data (Table 4) are supported by the semi­quantitative X-ray diffraction data (Table 3). This is illustrated by the Aserum profile (E 67-74) in Fig. 11. The marine clays at Aserum are leached by

362 E. ROALDSET

2.0

><

w o

� ...J

1.5

..: :l o

<Il w 0:

<Il 1.0

1-

<!> o >

0.5

0.05

o

o

o

0.10 0.15 0.20 0.25 0.30 0.35

WEIGHT '!. MnO

Fig. 12. Vogt's residual index plotted vs. MnO content, contours 1-3-5 plots per l % area.

penetrating surface water and by artesian ground water (Andersland et al.

1966). A sandy horizon at 1-1.5 m depth drains the infiltration into Aserum­vannet. The Aserum samples mainly consist of illite, chlorite and mixed-layer illite-vermiculite and illite-chlorite.

Characteristic variations are found within the different groups of marine clays.

In the illite-chlorite group high Fe and K contents and low Na and Ca

contents are related to the amount of illite. This confirms the supposition

that trioctahedral illites are usually present. A relation between the amount

of Al and K and the amount of mixed-layer minerals may be found.

In the illite-vermiculite group high Al and alkali contents are related to

the amount of illite, while the Fe and Mg contents are associated with the

chlorite and vermiculite content. An increasing amount of mixed-layer

minerals is accompanied by an increase in K and a decrease in Na and Ca

contents.

The formation of mixed-layer minerals at the expense of 14 A and 10 A

minerals is normally attended by an increase in Al and K and a leaching of

Mg and Ca.

Similar mineralogical and chemical alterations to those found within the

marine profiles, have presumably influenced the composition of other

Numedal clays.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 363

50

AMPHIBOLE e

BIOTITE e

K20 + No20 +Co O

MICROCLINE e 50

Average of 160 Pleistoc�neo ctays from Scandinavia

e ILLITE

e MUSCOVITE

e MONTMORILLONITE

Fig. 13. Contoured triangular diagram showing the distribution of (F€:203 + FeO +

MgO) - (Al203) - CK20 + Na20 + CaO) in all analysed samples, contours 1-2-4-5 plots per 1 % area. (The mineral analyses are obtained from Deer, Howie & Zussman 1966; and the average of 160 Pleistocene Scandinavian clays, bulk samples, from Englund & Jørgensen 1972).

Geochemical variations - weathering

The chemical data, recalculated to mol percentages of basic oxides, plot

within the central part of a (Fe.20s + FeO + MgO) - (AhOa) - (�O +

N�O + CaO) triangle (Fig. 13) covering only 13 % of its total area. The moraine clays show larger variations in composition than the marine clays, which mainly plot in the black area. Compared to the average composition of 160 Pleistocene clays (bulk samples) from Scandinavia given by Englund & Jørgensen (1972), the Numedal clays are enriched in F�03 (total Fe) and MgO. The mineral compositions plotted in Fig. 13 are obtained from Deer et al. (1966). The diagram indicates that the Numedal clays have a higher

contribution from biotite than from muscovite. This is in harmony with the X-ray diffractograms which in many cases show the presence of con­

siderable amounts of trioctahedral illite. Each of the analysed clays plots in the triangular diagram within the area given by microcline, muscovite, chlor­ite, biotite and amphibole. The Fe-rich clay minerals are believed to originate from biotite granites and amphibolites.

A consistent geochemical variation is found for the MgOfALl03 and the

K20fAW3 ratios between marine and the non-marine clays. Higher amounts

364 E. ROALDSET

80

70

60

.., o � <( ...... o Cl 50 � � o E

40

30

20

D

D

•• l • l

.. ,' • l

.. ,' . •

• l l

• l l

• • .l •

• • / • • • • l

l • • • • .o. l •

• l •• •• l

l • l • 1 D l

DD D [Jf

l D 1 D

l l D

i •' J• ,'o

•

D l DD ,' D

D

l l

D D 1 l

l D D

l

l l

l l

l

l

l l

D D e OGVGIOCARIS SHALE

D e ALUM SHALE

D

• Marin" clays

D Not marin" clays

• Cambro- silurian shales

30

Fig. 14. MgO/Al20s vs. K20/Al203 plots for the Numedal clays. Notice the higher MgO and K20 content in marine samples.

of MgO and K20 are consistent with deposition in a marine environment

(Fig. 14). Mg++ and K+ are assumed to have been removed from the sea

water by adsorption on the degraded clay minerals. The Numedal clays have

similar ratios Mg0/Ak03 to the middle Ordovician and Silurian shales of

the Oslo Region (Bjørlykke 1965). This strengthens the assumption that the

Cambro-Silurian shales have contributed to the Pleistocene clays.

Vogt's residual indices (Fig. 15) for the Numedal clays are similar to

those of fresh eruptive rocks, and show the clays to be almost unweathered.

The marine clays constitute a relatively homogeneous group. The highest as

well as the lowest residual indices are found among the moraine clays. The

marine conditions are expected to have stopped the weathering processes.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 365

10 � MARINE CLAVS

>

0 OTHER CLAYS ( moraimt, glacio-fluvial, fluvial,limnic)

u z Ill ::1 5 Cl Ill 0:: LI..

o 0.5 1.0

� :t m rrrlhilli D 0.5 1.0 1.5 2.0

15

> 10 u z Ill ::1 Cl Ill a:: LI.. 5

o 0.5

ALL SAMPLES

VOGT'S RESIDUAL INDEX Fig. 15. Histogram of Vogt's residual index.

,o 2.5

2.5

For the Numedal clays a dose relation is found between the mineralogical

and chemical composition and the degree of weathering. The amount of

expandable minerals increases as the weathering increases.

Summary of chemical analyses The composition of the Numedal clays corresponds to that of coarse clays

given by Pettijohn (1957). However, the Ti and Mg contents are a little

higher than normally found for clay sediments. The mutual similarities are

greatest among the marine clays which in all diagrams plot within a more

limited area than the non-marine clays. The clay material has residual

indices similar to those of fresh eruptive rocks. The mineralogical data

indicate that most of the clays have been slightly affected by weathering

processes. Marine conditions have stopped the weathering processes.

366 E. ROALDSET

The geochemical data suggest a strong influence from basaltic Precam­

brian andfor Cambro-Silurian parent rocks for the clay fraction of the

Quatemary deposits.

STATISTICAL ANALYSIS

The mineralogical and chemical data are treated statistically by correlation

and factor analyses using the following 17 variables and 73 samples.

The amount (intensity percentage) of the 14 A, 14-10 A minerals (chlorite,

vermiculite and mixed-layer minerals), illite, amphibole, quartz, K-feldspar

and plagioclase.

The amount (weight percentage) of Si02, TiO.z, Al20s, Fe<zOs, MnO, MgO,

CaO, Na20, K20 and the ignition loss.

During the interpretation of the data low correlation coefficients are set

aside.

Each of the minerals is negatively correlated to the illite content. As the

amounts of quartz, feldspar and amphibole increase the amount of clay

minerals decreases. This generalization is not due to chemical alteration, but

depends only upon the grain-size distribution within the <2:!-l fraction. The

content of illite and chlorite increases with decreasing particle size (Berry

& Jørgensen 1971). A strong negative correlation, however, is found between illite and mixed­

Iayer minerals while a small positive value characterizes the relation between

the 14 A and the 14-10 A minerals. These relations confirm the assumption

of an increase in the content of vermiculite and mixed-layer illite-vermiculite

minerals mainly at the expense of trioctahedral illite and to a lesser extent of

chlorite. The 14 A and 8.4 A reflections are inversely related. Thus, the 14 A

minerals are assumed to be derived preferentially from the amphibole

minerals (Brindley 1961, Tomita et al. 1970). An increasing illite content Ieads to an increase in the Fez03 (Fe total),

MgO and K20 contents and a smaller ignition loss. This confirms the

presence of considerable amounts of trioctahedral illite. The amphibole, K­

feldspar and plagioclase minerals are negatively related to the FezOs and

MgO contents and positively to the Na20 and CaO contents. The Ti02

content seems to be connected mainly to the chlorite minerals.

Acknowledgements. - This work is a part of a thesis submitted for the cand. real. degree at the University of Oslo in 1970. The work was suggested and supervised by Professor l. Th. Rosenqvist to whom I am deeply indebted. Docent Per JØrgensen is gratefully thanked for valuable discussions and advice, and cand. real. Arild O. Brun­felt for advising the neutron activation determinations, and Dr. W. L. Griffin for reading the manuscript and correcting the English language.

October 1971

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 367

REFERENCES

Andersland, O. B., Bjerrum, L. & Rolfsen, E. N. 1966: Undrained behaviour of quick clays under load tests at Aserum. Internal report F. 229. Norges geotekn. inst. Oslo.

ASTM, 1960: American Society for Testing Materials. Index to the X-ray powder data file. Publication 48 - l.

Barshad, L. 1948: Vermiculite and its relation to biotite as revealed by phase exchange reactions, X-ray analyses, differential thermal curves and water content. Am. Miner­alogist 33, 655-678.

Barth, T. F. W. 1945: Studies on the igneous rock complex of the Oslo Region, Il. Systematic petrography of the plutonic rocks. Skr. Norske Vidensk.-Akad. Oslo, Mat.-naturv. Kl. no. 9.

Barth, T. F. W. 1962: Theoretical Petrology. Second edition. John Wiley & Sons, New York.

Basset, W. A. 1959: The origin of the vermiculite deposit at Libby, Montana. Am. Mineralogist 44, 282-299.

Berry, R. W. & Jørgensen, P. 1971: Grain size, mineralogy and chemistry of a quick clay sample from the Ullensaker slide, Norway. Eng. Geol. 5, 73-84.

Bhattacharya, N. 1962: Weathering of glacial tills in Indiana. l. Clay minerals. Bull. Geol. Soc. Am. 73, 1007-1020.

BjØrlykke, K. 1965: The middle Ordovician of the Oslo Region, Norway. 20. The geochemistry and mineralogy of some shales from the Oslo Region. Norsk geol. tidsskr. 45, 435-456.

Bradley, W. F. & Grim, R. E. 1961: Mica clay minerals, pp. 208-241 in G. Brown (Ed.): The X-ray Identification and Crystal Structures of Clay Minerals. Mineralogical Society, London.

Bradley, W. G., Grim, R. E. & Clark, G. L. 1937: A study of the behavior of mont­morillonite upon wetting. Zeitschr. f. Kristallographie 97, 216-222.

Brindley, G. W. 1961: Chlorite minerals, pp. 242-296 in G. Brown (Ed.): The X-ray Identification and Crystal Structures of Clay Minerals. Mineralogical Society, London.

Brindley, G. W. & Ali, F. Z. 1950: X-ray study of thermal transformation in some magnesium chlorite minerals. Acta Cryst. 3, 25-30.

Brindley, G. W. & Gillery, H. F. 1956: X-ray identification of chlorite species. Am. Mineralogist 41, 169-186.

Brown, G. (Ed.) 1%1: The X-ray Identification and Crystal Structures of Clay Minerals, Mineralogical Society, London.

Brunton, G. 1955: Vapor pressure glycolation of oriented clay minerals. Am. Miner­alogist 40, 124-126.

BrØgger, W. C. & Schetelig, J. 1926: Kristianiafeltet. Rektangelkartene Kongsberg og Tønsberg med Larvik. Norges geo/. undersøkelse.

Bugge, A. 1937: Geologisk kart Flesberg og Eiker. Norges geo/. undersøkelse 143, 1-63.

Collini, B. 1950: Om våra kvartara lerors mineralogiska sammansiittning. Gelo. foren. Stockholm Forh. 72, 192-206.

Deer, W. A., Howie, R. A. & Zussmann, J. 1966: An Introduction to the Rock Forming Minerals, Longmans, Green & Co. Ltd., London.

Droste, J. B. 1966: Alteration of clay minerals by weathering in Wisconsin till. Bull. Geo/. Soc. Am. 67, 911-918.

Eide, O. 1961: Undersøkelser av rasskråning ved Hvittingfoss. Rapport O. 978. Norges geotekn. inst. Oslo.

Englund, J. O. & Jørgensen, P. 1972: Chemical classification system for argillaceous sediments and factors affecting their composition. Geol. foren. Stockholm Forh. 94.

Foss, I. 1963: Grunnundersøkelser for prosjektert ny kontorbygning ved Kongsberg våpenfabrikk, Kongsberg, Rapporter 63/24. Norges geotekn. inst. Oslo.

Garrels, R. M. & Christ, C. L. 1965: Solutions, Minerals and Equilibria. Harper & Row, New York.

Gaudette, H. E., Eades, J. L. & Grim, R. E. 1966: The nature of illite. Clays Clay Minerals 13, 33-48.

Gibbs, R. J. 1965: Error due to segregation in quantitative clay mineral X-ray diffrac­tion mounting techniques. Am. Mineralogist 50, 741-751.

368 E. ROALDSET

Gibbs, R. J. 1967: Quantitative X-ray diffraction analysis using clay mineral standards extracted from samples to be analysed. Clay Minerals 7, 79-90.

Gibbs, R. J. 1968: Clay mineral mounting techniques for X-ray diffraction analysis: A discussion. Am. Mineralogist 53, 242-244.

Gjems, O. 1967: Studies on clay minerals and clay-mineral formation in soil profiles in Scandinavia. Medd. Det norske Skogforsøksvesen 21, 303-415.

Goldschmidt, V. M. 1954: Geochemistry. Clarendon Press, Oxford. Gordon, G. E., Randle, G., Goles, G. G., Corliss, J. B., Benson, M. H. & Oxley, S. S.

1968: Instrumental activation analysis of standard rocks with high resolution X-ray detectors. Geochim. et Cosmochim. Acta 32, 369-398.

Graff-Petersen, P. 1961: Lermineralogien i de limniske jurasedimenter på Bornholm. Thesis. University of Copenhagen.

Isachsen, F. 1933: lsavsmeltningen og de kvartærgeologiske forutsetningene for be­byggelsen i Numedals og Hallingdals øverste bygder. Norsk geogr. tidsskr. 14.

Johns, W. D., Grim, R. E. & Bradley, W. F. 1954: Quantitative estimations of clay minerals by diffraction methods. Jour. Sedimentary Petrology 24, 242-251.

Johnson, N. M., Likens, G. E., Bormann, F. H. & Pierce, R. S. 1968: Rate of chemical weathering of silicate minerals in New Hampshire. Geochim. et Cosmochim. Acta 32, 531-545.

JØrgensen, P. 1965: Mineralogical composition and weathering of some late Pleistocene marine clays from the Kongsvinger area, southern Norway. Geo[. foren. Stockholm Forh. 87, 62-83.

Låg, J. 1945: Weathering of syenite in Kjose, Vestfold. Norsk geo[. tidsskr. 25, 216-224.

Låg, J. 1948: Undersøkelser over opphavsmaterialet for Østlandets morenedekker. Medd. Det norske Skogforsøksvesen 10, 1-223.

Låg, J. 1968: Relationship between the chemical composition of the precipitation and the contents of exchangeable ions in the humus layer of natura! soils. Acta Agr. Scand. 18, 148-152.

MacEwan, D. M. C. 1944: ldentification of the montmorillonite group of minerals by X-rays. Narure 154, 577.

Mehra, O. P. & Jackson, M. L. 1960: Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. Clays Clay Minerals. 7th Con/., 317-327. Pergamon Press, London.

Mitchell, W. A. 1955: A review of the mineralogy of Scottish soil clays. Jour. Soil. Sei. 6, 94-98.

Molloy, M. W. & Kerr, P. F. 1961: Diffraction patterns of A.P.I. reference clay minerals. Am. Mineralogist 46, 583-605.

Moum, J. 1965: Falling drop used for grain-size analysis of finegrained materials. Sedimentology 5, 343-347.

Norrish, K. & Taylor, R. M. 1962: Quantitative analysis by X-ray diffraction. Clay Miner. Bull. 5, 98-109.

Pettijohn, F. J. 1957: Sedimentary Rocks. Harper & Row, New York. Quigley, R. M. & Martin, R. T. 1963: Chloritized weathering products of a New

England glacial till. Clays Clay Minerals, 10th Conf., 107-116. Pergamon Press, New York.

Reichenbach, H. Graf von & Rich, C. l. 1969: Potassium release from muscovite as influenced by particle size. Clays Clay Minerals 17, 23-29.

Reusch, H. 1896: Geologiske iagttagelser fra Telemarken, Indre Hardanger, Numedal og Hallingdal. Skr. Videm·k. Selsk. Kra. 2, 1-102.

Roaldset, E. & Rosenqvist, l. Th. 1971: Adsorbed rare earth elements as a due to the origin of some glacial clays. Bull. Groupe franc. Argiles, 191-194.

Rosenqvist, l. Th. 1942: Om norske leirers petrografi. Medd. fra VegdirektØren 2, 24-30.

Rosenqvist, I. Th. 1969: Numedalsprosjektet - en presentasjon. Norsk geo[. tidsskr. 49, 330-332.

Samuelsen, A. 1937: Eikerbladets og Flesbergbladets kvartærgeologi. Norges geol. undersøkelse 143, 63-109.

Schetelig, J. 1918: Natur og fjeldgrund. - From Lorens Berg: Sande herred. En bygdebok. Kra.

Shimane, H. & Sudo, T. 1958: A chlorite mineral associated with vermiculite. Clay

Miner. Bull. 3, 297-301.

MINERALOGY AND GEOCHEMISTRY OF QUATERNARY CLAYS 369

Soveri, U. 1956: The mineralogical composition of argillaceous sediments of Finland. Ann. Acad. Scient. Fenn. Ill. Geologica-Geographica 48, 1-32.

SØrlie, K. A. 1965: Kvartærformer i Jondalen og søndre del av Numedal. Unpublished cand. real. thesis at the University of Oslo.

Tomita, K., Tateyama, H. & Oba, N. 1970: Formation of vermiculite and kaolin mineral from hornblende. Jour. Japan. Assoc. Min. Pet. Econ. Geol. 64, 64-71.

Vinogradov, A. P. & Ronov, A. B. 1956: Composition of the sedimentary rocks of the Russian platform in relation to the history of its tectonic movements. Geochemistry 6, 533-559.

Vogt, T. 1927: Sulitjelmafeltets geologi og petrografi. Norges geo!. undersøkelse 121. Walker, G. F. 1947: The mineralogy of some Aberdeenshire soil clays. Clay Miner.

Bull. l, 5-8. Walker, G. F. 1949: The decomposition of biotite in the soil. Mineralog. Mag. 28,

693-703. Walker, G. F. 1950: Trioctahedral minerals in the soil-clays of North-East Scotland.

Mineralog. Mag. 29, 72-84. Walker, G. F. 1961: Vermiculite minerals, pp. 297-324 in G. Brown (Ed.): The X-ray

Identification and Crystal Structures of Clay Minerals. Mineralogical Society, London. Weiss, E. D. & Rowland, R. A. 1956: Effect of heat on vermiculite and vermiculite­

chlorite. Am. Mineralogist 41, 899-914. Yoder, H. S. & Eugster, H. P. 1955: Synthetic and natural muscovites. Geochim. et

Cosmochim. Acta 8, 225-280.