sedimentarygeochemistry - universitas...

Chapter 3

Sedimentary Geochemistry

How Sediments are Produced

Knut Bjørlykke

The composition and physical properties of sedimen-tary rocks are to a large extent controlled by chemicalprocesses during weathering, transport and also duringburial (diagenesis). We can not avoid studying chemi-cal processes if we want to understand the physicalproperties of sedimentary rocks. Sediment transportand distribution of sedimentary facies is strongly influ-enced by the sediment composition such as the contentof sand/clay ratio and the clay mineralogy. The primarycomposition is the starting point for the diageneticprocesses during burial.

We will now consider some simple chemical andmineralogical concepts that are relevant to sedimen-tological processes.

Clastic sediments are derived from source rocksthat have been disintegrated by erosion and weather-ing. The source rock may be igneous, metamorphicor sedimentary. The compositions of clastic sedimentsare therefore the product of the rock types within thedrainage basin (provenance), of climate and relief. Thedissolved portion flows out into the sea or lakes, whereit is precipitated as biological or chemical sediments.Weathering and abrasion of the grains continues dur-ing transport and sediments may be deposited anderoded several times before they are finally stored ina sedimentary basin.

After deposition sediments are also being subjectedto mineral dissolution and precipitation of new mine-rals as a part of the diagenetic processes. For the mostpart we are concerned with reactions between minerals

K. Bjørlykke (�)Department of Geosciences, University of Oslo, Oslo, Norwaye-mail: [email protected]

and water at relatively low temperatures. At tempera-tures above 200–250◦C these processes are referredto as metamorphism which is principally similar inthat unstable minerals dissolve and minerals which arethermodynamically more stable at certain temperaturesand pressures precipitate.

At low temperatures, however, unstable mineralsand also amorphous phases may be preserved for along time and there may be many metastable phases.

Many of the reactions associated with the dissolu-tion and precipitation of minerals proceed so slowlythat only after an extremely long period can theyachieve a degree of equilibrium.

Reactions will always be controlled by thermody-namics and will be driven towards more stable phases.The kinetic reaction rate is controlled by temperature.

Silicate reactions are very slow at low temperatureand this makes it very difficult to study them in thelaboratory.

Biological processes often accompany the purelychemical processes, adding to the complexity. Bacteriahave been found to play an important role in both theweathering and precipitation of minerals. Their chiefcontribution is to increase reaction rates, particularlyduring weathering.

In this chapter we shall examine the processesbetween water and sediments from a simple physical-chemical viewpoint. A detailed treatment of sedimentgeochemistry is however beyond the scope of thisbook.

87K. Bjørlykke (ed.), Petroleum Geoscience: From Sedimentary Environments to Rock Physics,DOI 10.1007/978-3-642-02332-3_3, © Springer-Verlag Berlin Heidelberg 2010

88 K. Bjørlykke

Na +

H

H

H

H

H

H H

H

O

O O

O

105°

Fig. 3.1 The strong dipole of water molecules causes them tobe attracted to cations which thereby become hydrated. Smallcations will be most strongly hydrated and less likely to beadsorbed on a clay mineral with a negative charge

Water (H2O) consists of one oxygen atom linked totwo hydrogen atoms, with the H-O-H bonds formingan angle of 105◦ (Fig. 3.1). The distance between theO and the H atoms is 0.96 Å, and between the hydrogenatoms 1.51 Å. Water molecules therefore have a strongdipole with a negative charge on the opposite side fromthe hydrogen atoms (Fig. 3.1). This is why water hasa relatively high boiling point and high viscosity, andwhy it is a good solvent for polar substances. Anotherconsequence of this molecular structure is that waterhas a high surface tension, important for enabling par-ticles and organisms to be transported on its surface.The capillary forces which cause water to be drawn upthrough fine-grained soils are also a result of this highsurface tension.

A number of concepts are particularly useful fordescribing and explaining geochemical processes:

1. Ionic potential2. Redox potential Eh3. pH4. Hydration of ions in water5. Distribution coefficients6. Isotopes

3.1 Ionic Potential

Ionic potential is a term introduced by V.M.Goldschmidt to explain the distribution of elementsin sediments and aqueous systems. It must not be

confused with ionisation potential. Recent authorshave proposed the term “hydropotential” for the con-cept, to avoid confusion.

Ionic potential (I.P.) may be defined as the ratiobetween the charge (valency) Z and the ionic radius R:

IP = Z

R

The ionic potential is an expression of the chargeon the surface of an ion, i.e. its capacity for adsorbingions. Small ions carrying a large charge have a highionic potential while large ions with a small chargehave a low ionic potential (see Fig. 3.2).

Ions with low ionic potential are unable to break thebonds in the water molecule and therefore remain insolution as hydrated cations (e.g. Na+, K+). This meansthat the ion is surrounded by water molecules with theirnegative dipole towards the cation (Fig. 3.1).

This is because the O–H bond is stronger than thebond which the cation forms with oxygen (M–O bond-ing, M = metal); this is particularly true of alkali metalions (Group I) and most alkaline earth elements (GroupII, I.P. <3). Metals with an ionic potential only slightlylower than that required to form M–O bonds, namelyMg2+, Fe2+, Mn2+, Li+ and Na+, will be the moststrongly hydrated. The hydration strongly affects thechemical properties of the ion and its capacity to beadsorbed or enter into the crystal structure of a mineral.Since the ions are surrounded by water molecules, wecan use the expression “hydrated radius” to describethe space occupied by the ion and its water moleculeswithin a crystal structure (Fig. 3.3).

If the M–O bond is approximately equal in strengthto the O–H bond (I.P. 3–12), the metal ion replacesone of the hydrogen atoms to form very low solu-bility compounds of the type M(OH)n (see Fig. 3.2).Examples of these so-called hydroxides that we com-monly encounter in sedimentary rocks as a resultof weathering are Fe(OH)3, Al(OH)3 and Mn(OH)4.These hydroxides have very low solubility.

Ions with high ionic potential (>12) form an M–Obond that is stronger than the H–O bond, giving solubleanion complexes such as SO4

−−, CO3−−, PO4

3− andreleasing both of the H+ ions into solution.

This approach can be used to explain the behaviourfor elements on both sides of the Periodic Table(electropositive and electronegative) which form ionicbonds. The elements in the middle, however, have agreater tendency to form covalent bonds in which the

3 Sedimentary Geochemistry 89

2.0

1.5

1.0

0.5

Soluble cations e.g. Na+

Hydrolysates e.g.Fe(OH)3 .Very lowsolubility

Soluble anioncomplexes e.g.SO4--

Ioni

c ra

dius

Charge (valency)0 1 2 3 4 5 6 7

Cs

Rb

K

Na

Li

Ca

Sr

Ba

Mn ScFe

Be

Mg GaFe

B

Al

ThU

Ti

PSi

C

ZrNb.Ta

N

S

V

R.E

Mn

Fig. 3.2 Ionic radius and charge (valence) for some geochem-ically important elements. Ions with low ionic potential aresoluble as cations (e.g Na+, K+) while ions with intermediateionic potentials will bond with OH− groups and have very low

solubility, forming hydrolysates (e.g Al(OH)3), Fe (OH)3). Highionic potentials make soluble cation complexes like CO−−

3 andSO−−

4 . The ratio between these parameters - the ionic potential -can be used to explain their behaviour in nature

Li

+, r = 0.6 Å R = 3.8 Å

Na

+, r = 0.95 ÅR = 3.6 Å

K

+, r = 1.33 Å R = 3.3 Å

Rb

+, r = 1.48 ÅR = 3.2 Å

Cs

+, r = 1.69 ÅR = 3.2 Å

Mg

++, r = 0.65 ÅR = 4.2 Å

Ca

++, r = 1.0 ÅR = 4.0 Å

Sr

++, r = 1.13 ÅR = 4.0 Å

Ba

++, r = 1.43 ÅR = 3.0 Å

Naked radius (r)Hydrated radius (R)

Fig. 3.3 Ionic radius (inÅngstrom units) of hydratedand non-hydrated (“naked”)ions of alkali metals andalkaline-earth metals. Thesmaller ions have higher ionicpotentials and form strongerbonds with water moleculesso that they become hydrated.This hydration effect isreduced with increasingtemperature

strength of the M-O bond is not merely a function ofthe valency and radius, and the picture becomes farmore complex. The concept of ionic potential is never-theless still useful; we see that during weathering,elements with low ionic potential remain in solutionalong with the anionic complexes of metals and non-metals with high ionic potential. This is reflected inthe composition of seawater. The hydrolysates, onthe other hand, become enriched on land as insolu-ble residues or through weathering (Al3+, Fe3+, Mn4+,Ti4+, etc.). Note also that Fe++ and Mn++ which occurin reducing environments have lower ionic potentialand are much more soluble that Fe3+ and Mn4+.

The most soluble ions remain in the seawateruntil they are precipitated as salt when seawateris concentrated during evaporation. In addition to

the chlorides (i.e. NaCl, KCl), these are mainlysalts of cations with low ionic potential, and ofanions with high ionic potential, e.g. CaSO4

.2H2O,Na2CO3 and carbonates such as CaCO3 (calcite),CaMg(CO3)2 (dolomite) and MgCO3(magnesite).

The principle of ionic hydration and the size of theionic radius are capable of explaining a whole rangeof geochemical phenomena. Among the Group I ele-ments of the Periodic Table, we know that Li+ andNa+ are enriched in seawater. This because the stronghydration prevents absorption on clay minerals whichusually have a negative surface charge. K+, Rb+ andCs+, on the other hand, have larger ionic radii and con-sequently are less strongly hydrated. This leaves themwith a more effective positive surface charge whichfacilitates their adsorption onto clay minerals, etc.

90 K. Bjørlykke

This is demonstrated in nature during weatheringand transport. While similar amounts of potassium andsodium are dissolved during weathering of basementrocks, the potassium concentration in the sea is muchlower (K/Na ratio of only 1:30). This is because K+ ismore effectively removed by adsorption because it isless protected by hydration. The same is true to an evengreater extent for Rb+ and Cs+, which are adsorbedeven more readily. These ions therefore have a rela-tively short residence time in seawater, between beingdelivered by rivers and then removed by accumulatingsediment.

With regard to Group 2 elements, Mg++ for examplewill be more strongly hydrated than Ca++ because it isa smaller ion. As a result, Mg++ has a greater tendencyto remain in solution in seawater. However, despite thefact that the Mg/Ca ratio in seawater is 5, it is calciumcarbonate which is the first to form through chemi-cal and biological precipitation. Dolomite or magnesitedo not precipitate directly from seawater and this is inpart due to the strong hydration of Mg++. Normally, ifwe had naked (unhydrated) ions, MgCO3 and FeCO3

would be more stable than CaCO3 because Mg++ andFe++ have greater ionic potentials and stronger bondingto the CO2−

3 ion. However with increasing tempera-ture the hydration declines because the bonds with thedipole of the water molecules become weaker. Mg++ isthen more likely to be incorporated into the carbonatemineral structures. Therefore during diagenetic pro-cesses at 80–100◦C, magnesium carbonates precipitatemore readily even if the Mg++/Ca++ and Fe++/Ca++

ratios are low. Even if Mg is preferred in the carbon-ate structure and also in the clay minerals, very littlemagnesium is usually available in the deeper parts ofsedimentary basins except in the presence of evaporiteswith Mg salts.

3.2 Redox Potentials (Eh)

Oxidation potential (E) is an expression of the ten-dency of an element to be oxidised, i.e. to give upelectrons so it is left with a more positive charge.This potential can be measured by recording thepotential difference (positive or negative) which ariseswhen an element functions as one electrode in agalvanic element. The other electrode is a standardone, normally hydrogen. The oxidation potential of

the reaction H2 = 2H+ + 2e (electrons) is defined asE0 = 0.0 V at 1 atm and H+ concentration of 1 mol/l at20◦C. Different conventions have been used to assignplus and minus values. In geochemical literature,metals with a higher reducing potential than hydro-gen are assigned negative values, e.g. Na = Na+ +e− = −2.71V, while strongly oxidising elements aregiven a positive sign, e.g. 2F− = F2 + 2e = 2.87V.A list of redox potentials shows which elements willact as oxidising agents, and which will be reducingagents. Reactions which result in a negative oxidationpotential (E) will proceed spontaneously, while thosewhich have positive voltage will be dependent on theaddition of energy from an outside source. We canpredict whether a redox reaction will occur by usingNernst’s Law (see chemistry textbooks).

3.3 pH

The ionisation product for water is[H+

] · [OH−] =

10−14. The concentration of H+ in neutral water willbe 10−7. pH is defined as the negative logarithm of thehydrogen ion concentration, and is therefore 7 for neu-tral water (at 25◦C). However, the ionisation constant(product) varies with temperature, e.g. at 125◦C theionisation constant for water is [H+] · [OH−] = 10−12.In other words, neutral water then has a pH of 6. It isimportant to remember this when considering the pHof hot springs or in deep wells, for example oil wells.

In nature the pH of surface water mostly liesbetween 4 and 9. Rainwater is frequently slightly aciddue to dissolved CO2, which gives an acid reaction:

H2O + CO2 = H2CO3 (carbonic acid)

H2CO3 = H+ + HCO−3 = 2H+ + CO2−

3

Humic acids may give the water in lakes and riversa low pH. Sulphur pollution from burning oil and coalgives SO2, which is oxidised in water to sulphuric acid:

2SO2 + O2 + 2H2O = 2H2SO4

In areas with calcareous rocks or soils this sulphuricacid is immediately neutralised and the water becomesbasic, as is the case across much of Europe. By con-trast, in areas with acidic granitic rocks as in the southof Norway and large areas of Sweden, the rock does


not have sufficient buffer capacity to counteract acidrain or acidic water produced by vegetation (due tohumic acids). Organic material also contains a certainamount of sulphur, and drainage of bogs, or drought,can produce an acidic reaction. This is because H2Sfrom organic material is oxidised to sulphate when thewater table is lowered, allowing oxygen to penetratedeeper in these organic deposits.

The water near the surface of large lakes and thesea can have a high pH because CO2 is consumeddue to high organic production (photosynthesis). If theorganic material decomposes (oxidises) on its way tothe bottom, CO2 is released again, causing the pH todecrease with depth since the solubility of the CO2

increases with the increasing pressure.CO2 is also less soluble in the warm surface water

than in the colder water at greater depth.Seawater is a buffered solution, with a typical pH

close to 8, though this varies somewhat with tempera-ture, pressure and the degree of biological activity.

Eh and pH are important parameters for describingnatural geochemical environments, and the diagramobtained by combining these two parameters is partic-ularly useful.

The lower limit for Eh in natural environments isdefined by the line Eh = −0.059 pH, because other-wise we would have free oxygen, and the upper limitcorresponds to Eh = 1.22 − 0.059 pH, beyond whichfree oxygen would be released from the water. Ifwe also set pH limits at 4 and 9 in natural envi-ronments, we can divide the latter into four maincategories:

1. Oxidising and acidic2. Oxidising and basic3. Reducing and acidic4. Reducing and basic

Variations of pH and Eh are the major factorsinvolved in chemical precipitation mechanisms in sedi-mentary environments where there is not strong evap-oration (evaporite environments).

The solubility of many elements is highest in thereduced state and they are precipitated by oxidation.This is particularly characteristic of iron and man-ganese, whereas others such as uranium and vanadiumare least soluble in the reduced state.

3.3.1 Distribution Coefficients

When a mineral crystallises out of solution, the com-position of the mineral will be a function of the compo-sition of the solution and the temperature and pressure.Trace elements which are incorporated in the mineralstructure are particularly sensitive to variations of thesefactors. With constant temperature and pressure, theconcentration of an element within a mineral which isbeing precipitated, is proportional to its concentrationin the solution. The ratio between the concentration ofan element in the mineral and its concentration in thesolution (water) is called the distribution coefficient.

A number of elements substitute for Ca++ in thecalcite lattice: Mn++, Fe++ and Zn++ have distributioncoefficients (k) < 1. This means that they will be cap-tured, so that the mineral becomes enriched in theseelements relative to the solution.

Mn++/Ca++(mineral) = k · Mn++/Ca++(solution)

k here is about 17, that is to say the manganese con-centration in the calcite is 17 times greater than in thesolution.

At low temperatures (25◦C) Mg++, Sr++, Ba++ andNa+ have distribution coefficients <1. This means thatthe mineral phase will contain proportionately lessof these elements than the aqueous phase. For Sr++,k is about 0.1 (0.05–0.14) in calcite, such that theSr content in calcite is relatively low. The Sr con-tent in aragonite is considerably higher because theSr++ ion, which is larger than the Ca++ ion, is moreeasily accommodated within the lattice. By analysingtrace elements in minerals like calcite we can infersomething about the environment when the mineralsprecipitated. Limestones with a high content of stron-tium may have had much primary aragonite whichwas replaced by calcite. Calcite containing significantamounts of iron must have precipitated under reducingconditions because only Fe++ would be admitted intothe calcite structure.

3.4 Isotopes

A number of elements occur in nature as differ-ent isotopes: the atomic number (protons) is constantbut there are different numbers of neutrons. They

92 K. Bjørlykke

therefore have the same chemical properties althoughtheir masses are slightly different. Isotopes which areradioactive (unstable) break down at a specific ratecharacteristic for the isotope species (the disintegrationconstant). By analysing the reaction products formedin the minerals they can be dated. The 87Rb −87Srand the 40K −40Ar methods are the ones most com-monly used in determining the age of rocks. The ratiosbetween lead isotopes can also be employed becauseof the 235U −207Pb, 238U −206Pb and 238Th −208Pbreactions.

Dating sedimentary rocks is a complicated proce-dure and the results are often difficult to interpret.The main problem is that clastic sediments are com-prised of fragments and minerals which have beeneroded from older rocks and the measured radio-metric age may be strongly influenced by the ageof these source rocks. Separating the newly formed(authigenic) mineral to be dated, can be particularlychallenging.

The fact that isotopes have different masses causesfractionation to take place through both chemical andbiological processes. The simplest example is water,H2O, which contains two oxygen isotopes and twohydrogen isotopes. The oxygen isotopes are fraction-ated through evaporation, with more H16

2 O evaporatingthan H18

2 O. This is because the 18O isotope has greatermass and a phase change from fluid to vapour thereforerequires more energy. H16

2 O has higher vapour pressurethan H18

2 O. This is the reason why rainwater and icecontain less 18O than seawater.

Isotope fractionation is a function of temperature,however, and is much more effective with evaporationat low temperatures than at high ones. The explana-tion for this is that at high temperatures the energyof the molecules are so great that the difference inmass between 18O and 16O is of less consequence. Atlow temperatures the isotopic separation evaporationis much more selective so that the water evaporated ismore enriched in 16O. When water vapour condensesto rainwater, molecules with 18O are most stable. Rainand snow becomes enriched in the heavier isotope(18O), so that the water vapour remaining in the airbecomes more enriched in 16O. Most of the evapora-tion takes place at low latitudes and the water vapour inthe air has a progressively lower 18O-content towardshigher latitudes as the air cools and it rains. The con-centration of oxygen isotopes is expressed in relationto a standard:

δ18O =(

18O/16Osample/18O/16Ostd − 1

)· 1000

This standard may be the average compositionof seawater, called SMOW (Standard Mean OceanWater). Another commonly used standard is PDB (PeeDee Belemnite), which is the composition of calcitein a Cretaceous belemnite. The calcite (CaCO3) wasprecipitated in the sea and its composition was inequilibrium with the seawater at normal temperatures(15–20◦C). There is more 18O in calcite than in thewater (positive fractionation), but with higher temper-atures the less effective fractionation of oxygen lowersthe δ18O values. The relationship between the twostandards is:

δ18OSMOW = 1.031 · δ18OPDB + 30.8

PDB values are preferred for carbonate mineralswhile the SMOW scale is mainly used for water sam-ples and silicate minerals.

Hydrogen has two stable isotopes, 1H and 2H (deu-terium), and an unstable one, 3H (tritium), which hasa half-life of 12 years. The hydrogen isotopes areeven more strongly fractionated than oxygen isotopesduring evaporation. Water molecules with deuterium(heavy water) have lower vapour pressure that watermoles with hydrogen.

In meteoric water there is a linear relation betweenthe deuterium/hydrogen ratio (D/H) and the δ18O.

The isotopic composition of seawater has variedthrough geological time, though not so much duringthe last 200–300 million years. During glacial periods,seawater acquires more positive δ18O values becausethe water bound as ice has more negative δ18O values.Rainwater (meteoric water) has normal δ18O valuesfrom –2 to –15. The values become more negativetowards higher latitudes, and near the poles one canmeasure δ18O values of about –50 and δD (2H) valuesclose to –350 (see Fig. 3.4). Minerals that form in sea-water show decreased 18O/16O ratios with increasedambient temperature during formation. The δ18O/16Oratio in carbonate secreting marine organisms, forexample, is thus a function of both temperature andsalinity. The seawater changes its δ18O values byaround 1–1.5‰. Isotopes can thus provide importantproxy evidence for palaeoclimate studies.

Cold freshwater gives strongly negative δ18O val-ues, whereas evaporites are enriched in 18O isotopes


–50

δ1800/00

δD0 /0

0

–40 –30 –20 –10 0 +10

–300

–200

–100

0 Precipitationat low latitudes

Precipitationat high latitudes

Evaporationin closedbasins

Fig. 3.4 Ratio between the isotopic composition of seawaterand freshwater. Evaporites will deviate from the mixing linebetween these endmembers

(positive δ18O). Shallow marine carbonates that arediagenically modified by freshwater, give lower δ18Ovalues than marine carbonates deposited in deeperwater.

Stable oxygen isotope analyses were first usedby Urey, in 1951, to demonstrate past temperaturechanges in seawater. By taking samples through across-section of a belemnite it was possible to regis-ter annual variations in seawater temperature from 150million years ago (Fig. 3.5).

The precipitation of newly formed (authigenic) min-erals gives an oxygen isotope composition which isa function of the composition of the porewater inwhich the mineral is precipitated, and the tempera-ture. If the porewater isotope composition is known,the temperature (T) can be calculated, and vice versa.

The calcite precipitation formula is:

T = 16.9 − 4.38(18Ocarb −18 Owater)+

0.1(18Ocarb −18Owater)2

Here the values for calcite are given in PDB and forwater in SMOW. We see that if the δ18O value for cal-cite is 0 (PDB) and seawater has 0 (SMOW), the tem-perature is 16.9◦C, which may have been a typical seatemperature when the standards were precipitated.).

The above formula can be expressed graphically,enabling the temperature to be read off a curve asa function of the isotopic composition of the calcite,

which is the assumed composition of the porewaterduring precipitation (Fig. 3.6). Similar calculations canbe done for other precipitated minerals, for example forquartz using the δ18O fractionation as a function of thetemperature for quartz.

Carbon has two stable isotopes (12C − 98.9% and13C − 1.1%). During photosynthesis a greater propor-tion of 12CO2 than 13CO2 forms organic compounds,because 12CO2 has a smaller mass. Organic material istherefore enriched in 12C relative to atmospheric CO2

and HCO−3 in seawater. The isotopic composition of

carbon is expressed as δ13C values:

δ13C = [13C/12C(sample)/13C/12C(std) − 1] · 1000

All samples are compared against a standard ofmarine calcite, the PDB belemnite, which by defini-tion has δ13C = 0‰ PDB. The isotopic compositionof dissolved carbon (CO2) has been relatively constantduring the last 300–400 million years, but limestonescan nevertheless be dated and correlated using differ-ences due to variation in the composition of seawater.Towards the end of the Precambrian the compositionof seawater seems to have been more variable, andthere this type of correlation is particularly valuablesince there are no fossils. In large massive limestonesthe isotope composition does not change significantlyduring diagenesis, because the volume is so great.Atmospheric CO2 has δ13C = –7‰. Land plants havean average δ13C value of –24 (–15 to –30‰), andmarine organisms have a similar range of values.Freshwater containing CO2 released by the breakdownof organic matter, and groundwater filtered through asoil profile, will take up CO2 with negative δ13C valuesfrom roots and organic material.

Bacterial fermentation of organic material(2CH2O = CH4 + CO2) forms gas (methane) whichis very strongly enriched in 12C (δ13C = –55 to –90‰)and CO2 which is positive (δ13C = +15).

Thermal breakdown (thermal decarboxylation) oforganic matter produces δ13C values of –10 to –25.

The strontium isotope ratio (87Sr/86Sr) in seawa-ter has varied considerably through geological time(Fig. 3.7). This is because there are two radically dif-ferent sources of strontium in seawater. Continentalweathering supplies much 87Sr to the sea since graniticrocks contains relatively high concentrations of rubid-ium which can decay to 87Sr. Dissolution of basalt at

94 K. Bjørlykke

202.0016

018

19

t°c

18

17

16

15

0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 cm

Radius

Fig. 3.5 Analyses of oxygen isotopes in a Jurassic belemnite from the centre to the outermost layer. Colder water during the winteris recorded by lower 18O/16O ratios. We can see that the belemnite lived for 4.5 years and died in the spring. (From Urey et al. 1951)

the mid-oceanic ridges will supply strontium with arelatively low 87Sr/86Sr ratio because basalt containsa little potassium and also rubidium.

When there is rapid seafloor spreading a greatdeal of water passes through the mid-oceanic ridges,so that the seawater receives much Sr with a low87Sr/86S ratio. During such periods, for example in the

Jurassic – Cretaceous, the creation of new warm sea-floor will lead to a transgression onto the continents.This reduces the gradients and hence transportingcapacity of rivers, limiting the supply of clastic mate-rial to the ocean.

Since the Jurassic, the 87Sr/86Sr ratio has risenalmost continuously, and by analysing marine calcitic


2

δ180 Calcite (PDB)

–4 –6 –10

20

40

80

120

Late CalciteCement ( type II )

Tem

pera

ture

o C

–12 –14–8–200

100

60

140

160

Late CalciteCement (type I)

Measured range in calcites ( type I & II )

0

– 4–2

2

4

6

8

δ18 O H 2

O (SMOW) =

Fig. 3.6 Relation between the isotopic composition of porewater and carbonate cement, as a function of temperature (from Saigaland Bjørlykke 1987)

0.709

0.708

0.707

0 100 200 300 400 500 600 m.yr.

Cenozoic Cret. Jur. Trias. Perm.. Carbon. Dev. Sil. Ord. Cambr.

Pre-cambr.

87Sr/86Sr

High rates ofseafloorspreading

87Sr from land(weathering anderosion)

Fig. 3.7 87Sr/86Sr ratio inseawater from the Cambrianto present. Based onMacArthur et al. (2001). Thisratio reflects the relativecontribution form weatheringof continental rocks with highcontents of 87Sr and exchangewith basaltic rocks with lowcontents of 87Sr on theoceanic spreading ridges

fossils such as foraminifera, one can obtain ratheraccurate age determinations. This applies particularlyto the Tertiary period, when the rise in the 87Sr/86Srratio was particularly rapid (Fig. 3.7).

The isotopic composition of clastic sediments canalso be used for stratigraphical correlation. Then it canbe more useful to employ isotopes which do not gointo solution and react with water, but retain the orig-inal age of the rocks from which they were eroded. Inthe North Sea and on Haltenbanken the ratio betweenthe rare earth elements samarium and neodymium

(147Sm/143Nd) was used to correlate reservoir rocksboth in-field and regionally.

3.5 Clay Minerals

A number of minerals are referred to as clay min-erals because they predominantly occur in the finestgrain-size fraction (clay fraction) of sediments andsedimentary rocks. However, this is not an accurate

96 K. Bjørlykke

definition, because the clay fraction contains manyother minerals than those we call clay minerals, andbecause the clay minerals themselves are often largerthan 4 μm (0.004 mm). By “clay minerals” we usuallymean sheet silicates which consist chiefly of oxygen,silicon, aluminium, magnesium, iron and water (H2O,OH–). Clay minerals in sedimentary basins are partlyderived from sheet silicate minerals occurring in meta-morphic and eruptive rocks (e.g. biotite, muscovite andchlorite), but during weathering and transport theseclastic minerals are typically altered from their initialcomposition in the parent rock.

Mica (muscovite and biotite) lose some potassiumwhich is replaced by water (H2O, H3O+) to form illite(hydro mica). Clay minerals are also formed throughweathering reactions, for example by the breakdown offeldspar and mica. Clay minerals which are formed bythe breakdown of other minerals within the sediment,are called authigenic.

Sheet silicates have a structure consisting of sheetsof alternating layers of SiO4 tetrahedra and octahe-dra. In the tetrahedral layers, silicon or aluminiumatoms are surrounded by four oxygen atoms. In theoctahedral layers the cation is surrounded by six oxy-gen or hydroxyl ions. Both bi and trivalent ions canact as cations in the octahedral layer. In sheet sili-cates with trivalent ions (e.g. Al3+) only two of thethree positions in the octahedral layer are occupied,and such minerals are therefore called dioctahedral.With bivalent ions (Mg++, Fe++) all three positionsmust be filled to achieve a balance between the posi-tive and negative charges, so these minerals are calledtrioctohedral.

The main method of identifying clay minerals isX-ray diffraction (XRD), by which the thickness of thesheet silicates is determined using X-rays which arediffracted according to Bragg’s Law: nλ = 2d sin ϕ.Here λ is the wavelength of the X-ray, ϕ the angle ofincidence and d the thickness of the reflecting silicatelayers; d is thus a function of angle ϕ.

Sheet silicates may also be identified by meansof differential thermal analysis (DTA), which recordscharacteristic exothermal or endothermal reactions.

Figure 3.8 shows the structure of some of the mainclay minerals. Illite consists of sheets with two layersof tetrahedra and one of octohedra, bonded togetherby potassium. This ionic bonding is relatively weakso the mineral cleaves easily along this plane. Thebonds within the tetrahedral and octahedral layers are

more covalent and stronger. The potassium content inmica corresponding to the formula of mica is about9% K2O, while illite has a greater or lesser deficit ofpotassium. Smectite (montmorillonite) has the samestructure except that most of the potassium is replacedby water (H3O+), other cations or organic compounds(e.g. glycol). There are strong indications that smectiteconsists of small particles of 10 Å, plus water.

Illite is most likely comprised of several layersof these small 10 Å particles stacked on top of oneanother.

In an atmosphere of glycol vapour, smectite willswell from 14 to 17 Å, while illite is unable to expandbecause there are numerous layers bonded togetherwith K+ or other cations, for example NH+. Smectitehas a very high ion-exchange capacity and to someextent can exchange ions in the octahedral layer. Thestability of smectite declines in aqueous solutions withhigh K+/H+

(Na+/H+

)ratio and with increasing tem-

perature, and it converts to illite. Vermiculite has astructure reminiscent of the smectites, and also under-goes ion exchange and thus charge deficit in the tetra-hedral layer, so that the bonding between each layeris too strong for much swelling to occur. Vermiculitesare mostly trioctahedral, containing mostly Mg or Fein the octahedral layer.

Glauconite is a green mineral which forms on theseabed. It is a potassium and iron bearing silicatesomewhat similar to illite and contains both di - andtrivalent iron. It is therefore formed right on the redoxboundary, and during periods with little or no clasticsedimentation this can result in relatively pure beds ofglauconite.

Kaolinite consists just of a tetrahedral layer and anoctahedral layer and is very stable at low temperatures.There are no positions in the structure where exchangecan precede easily, which gives kaolinite a much lowerion exchange capacity than smectite. At higher tem-peratures kaolinite becomes unstable and will convertto illite if K-feldspar or other sources of potassium areavailable (at 130◦C) or pyrophyllite (Al3Si4O10(OH)2)at higher temperatures.

Kaolinite is part of the kaolin mineral group, whichincludes dickite which tends to form at slightly highertemperatures (100◦C).

Chlorite is a mineral which consists of two tetra-hedral layers and two octahedral layers, totalling14 Å. The octahedral layer is filled with Mg++andFe++. Magnesium-rich chlorites are typical of high


{

{

{

{

Tetrahedron Octahedron

Si4+, Al3+

O– – ,OH–O– – ,OH–

Tetrahedrallayer (SI,Al) (O, OH)4

Octahedral Al3+/Mg++, Fe++

layer

Al3+,Fe3+,Fe++,Mg++

K Al2 AlSi3O10 (OH)2

Muscovite(Dioctahedral)

K (Mg, Fe) AlSi3 O10 (OH)2

Biotite(Trioctahedral)

Illite (K1

– x) (Al, Mg, Fe)2–3 AlSi3 O10(OH)2

Tetrahedral layer (SI,Al) (O, OH)4

Octahedral layer (Al, Mg)

n. H3O+ (Exchangeable cations x)

(Glycol)

Smectite × (Al2–x, Mgx) Si4O10(OH)2

(Montmorillonite)

Tetrahedral layer (Si3, Al) (O, OH)4

Octahedral layer (Al1)

Kaolinite – Al2SiO2O5 (OH)4

Tetrahedral layer (Si3, Al)

Octahedral layer (Mg, Fe, Al)

Octahedral layer (Me, Fe, Al)(Biotite layer)

Chlorite – (Mg, Fe, Al)6 (Si,Al)4 O10 (OH)8

14 Å

7 Å

14–17 Å

10 Å

K+

Fig. 3.8 Simplified illustration of the main groups of clay min-erals. Their physical properties can be explained by their crystalstructure. The chemical bonds between SiO4−

4 and O2− in thetetrahedral structure are very strong. In the octahedral layers the

bonds are weaker because Mg++ is surrounded (co-ordinated)with 6 oxygen. In the illite (mica) structure potassium is co-ordinated with 12 oxygen molecules, resulting in weak bondsthat produce a strong cleavage

temperature metamorphic rocks, while iron-rich onesmay form authigenically in sediments near the seaflooror at shallow depth.

Chamosite is an iron-rich chlorite mineral whichoften forms close to the sediment surface in reduc-ing conditions. Together with siderite, chamosite is

98 K. Bjørlykke

an important mineral in sedimentary ironstones which,especially in England, earlier were used as iron ore.

Clay minerals have a number of properties whichdistinguish them from most other minerals. Because oftheir very large specific surface area they have a greatcapacity for adsorbing ions, which is increased by thefact that clay minerals have negatively charged edgesdue to broken bonds. In water with a low electrolytecontent, clay minerals will therefore repel each other. Ifcations are added, clay minerals will then accumulatea layer of positive ions (a double layer), and repulsionbetween negatively charged clay minerals declines asthe strength of the electrolyte increases. Van der Waal’sforces will therefore cause flocculation more easily insaltwater, where repulsion due to the negative charge isreduced. This is why clays transported by rivers floc-culate into larger particles which sink more rapidly tothe seafloor.

A colloid solution of clay in water is called a sol,which can be regarded as a Newtonian fluid. Floccu-lated clay is called a gel and has thixotropic properties.This means that the shear strength decreases withincreasing deformation so that it changes from a gelto a sol which consists of dispersed colloidal particlesin water (hydrosol). After a time without deforma-tion, it regains its strength. This is typical for smectiticclays.

Sediments which increase their volume when theyare deformed, are called dilatant. When the originalpacking of the grains is destroyed, the new packingmay be less effective and the volume then increases.Walking on a beach the deformation of the sand causesincreased porosity so that water is sucked into the sandat the surface.

Norwegian clays deposited in the sea when theice sheet retreated about 10,000 years ago, consist ofcrushed rock fragments and have approximately thesame composition as the parent rock. During deposi-tion these clays acquired a markedly porous structure,with the minerals stacked like a card house. Saltwaterhelped to hold this structure together because it neu-tralised the negative charges. When the clays are ele-vated above the sea by isostatic recovery of the land,they are exposed to meteoric water. Even if the clayhas a low permeability, freshwater will slowly seepthrough it and remove the saline water. This reducesthe strength of the clay’s structure and hence its sta-bility and the clay becomes “quick”. This is causedby overpressure and weak effective stress between

the grains and hence little friction. The card housestructure may then collapse, releasing the interstitialwater to produce a low viscosity clay slurry that willflow even down very gentle gradients. The additionof salt (NaCl or KCl) binds the clay particles so thatthe shear strength is increased, and this is employedwhen the ground has to be stabilised for buildings andconstruction.

3.6 Weathering

The composition of clastic rocks in sedimentarysequences depends to a large extent on the supply ofsediments from source areas undergoing weatheringand erosion. The physical properties of sedimentaryrocks are controlled by the primary sediment compo-sition and changes during burial (diagenesis). We shallhere briefly look at the processes producing sediments.

Mechanical weathering is the physical breakdownof rocks into smaller pieces which can then be trans-ported as clastic sediment.

Chemical weathering involves the dissolution ofminerals and rocks and precipitation of new mineralswhich are more stable at low temperatures and highwater contents. Parts of the parent rock will then becarried away in aqueous solution by the groundwaterand rivers into the ocean.

Erosion is the combined result of the disintegrationof rock and the removal of the products.

As we shall see, biological processes are not onlyimportant in connection with chemical weatheringbut also with respect to mechanical weathering. Bothmechanical and chemical weathering, which are essen-tially land surface processes, are due to the chemicalinstability of rocks that were formed or modified underother conditions (at greater depths, higher pressures ortemperatures, or in different chemical environments).They are no longer stable when exposed to the atmo-sphere, water and biological activity.

3.6.1 Mechanical Weathering

Igneous and metamorphic rocks which have beenformed at many km depth at higher temperatures andpressures are not stable when exposed at the surface.


When uplifted and unloaded the rocks expand,mostly in the vertical direction, producing horizontalfractures (sheeting) parallel to the land surface. Thisis because as the rocks near the surface are unloadedby the reduction in overlying rock, they can expandvertically but not horizontally. In this way the ver-tical stresses become less than the horizontal ones,and joints develop normal to the lowest stresses. Wesee this most clearly in granites, which are homo-geneous, while expansion in metamorphic and sedi-mentary rocks occurs along bedding surfaces, alongtectonically weak zones with crushing, or along frac-tures that were formed at great depth. Joints openedby stress release in turn provide pathways for ground-water to circulate, increasing the surface area of rockexposed to chemical weathering.

In areas that experience freeze-thaw cycles, frostweathering becomes very important. When waterfreezes in cracks in rock, it expands by 9% and cangenerate very high stresses, further widening cracksnear the surface. The surfaces of exposed rocks arealso subjected to daily temperature fluctuations whichcause greater expansion of the outer layers relative tothe rest of the rock. Desert regions in particular expe-rience very wide daily temperature ranges, though theimportance of this process for mechanical weatheringhas been questioned. The roots of plants and moss canalso contribute to mechanical weathering as they growinto fractures, take up water and expand.

3.6.2 Biological Weathering

Rocks are a source of nutrients for plants, and plantsare capable of dissolving and breaking down the majorrock-forming minerals. Moss, which consists of algaeand fungi living in symbiosis, produces organic com-pounds that can slowly dissolve silicate minerals. Evenin the earliest stages of weathering, we see that fungushyphae penetrate into microscopic cracks.

Plant roots produce CO2 which helps to lowerthe pH and dissolve minerals such as feldspar andmica, thus freeing an important plant nutrient, potas-sium. Plants also produce humic acids, which likewisestrongly influence the solubility of silicate minerals,and also affect the stability of clay minerals. The pro-duction of humic acids is perhaps the major factorinfluencing the rate of weathering. In heavily vegetated

areas, such as rain forest in the tropics, the weatheringrate is exceptionally high because so much humic acidis produced.

Bacteria and fungi, which are found in almostall soil types, are active in breaking down minerals.Animals also contribute to weathering, and certainmarine organisms such as mussels are able to bore intosolid rock (see Chap. 8). Microbiology has become akey area of research in the quest to understand howminerals are dissolved and precipitated.

3.6.3 Chemical Weathering

There is no sharp demarcation between biological andchemical weathering, because we find biological activ-ity in almost all soils and rocks near the surface.The chemical environment in water at the surface ofthe earth is very much affected by local biologicalactivity, and in most cases it is biological processesthat cause weathering to continue after rainwater hasbeen neutralised through reaction with minerals. Wewill therefore use the term “weathering” here for bothchemical and biological processes.

3.6.4 Weathering Profiles (Soil Profiles)

Both chemical and biological weathering are to a largeextent controlled by climate. The crucial factor is theratio between precipitation and evaporation in an area.In areas where precipitation far exceeds evaporation,podsol profiles develop in which there is a net trans-port of ions down through the soil profile as mineralsare dissolved. In other words, we get weathering dueto the fact that rainwater is slightly acidic (on accountof its CO2 and H2SO4 content) and contains oxygen.Rainwater is initially undersaturated with respect to allminerals. Some minerals are only very slightly soluble,others more soluble in this slightly acidic, oxidisingwater. Dissolved ions are transported down to the watertable, but ferrous iron liberated from iron-bearing min-erals will be oxidised and precipitated as ferric iron(Fe(OH)3). Vegetation at the top of the soil profileproduces CO2 from roots and organic compounds, par-ticularly humic acid, which will increase the solubilityof silicate minerals. Similarly, aluminium derived from

100 K. Bjørlykke

Rainfall

Podsol profile Layer ALeaching (grey)Precipitationof Fe(OH)3 (red)Layer BCapillary water

Water table

Phreatic (ground) water Layer C

Podsol soilprofile.Downwardstransport.

Brown-Earthprofile

Little leaching.Precipitation of carbonates(caliche)

Water table

Phreatic (ground) water Layer C

Evaporation

Rainfall

Evaporitic soil

Precipitation of salts andcarbonates

Water tablephreatic (ground) water Layer C

Evaporationand runoff

Fig. 3.9 Simplifiedrepresentation of soil profilesas a function of rainfall(precipitation) andevaporation

a solution of feldspar and mica, for example, pre-cipitates as Al(OH)3 but is less noticeable becausealuminium hydroxide is white. The uppermost partof the soil profile, where dissolution due to under-saturated rainwater and organic acids dominates, iscalled the A-horizon. Some of the dissolved salts andparticularly iron hydroxide is precipitated in the B-horizon below (Fig. 3.9). These may develop into alayer of solid rock (hard-pan) cemented with iron andaluminium oxides and hydroxides.

Where precipitation is approximately equal to eva-poration, there is less leaching within the soil profile.At a certain depth (about 0.5–1 m) carbonate willbe precipitated and form an indurated layer (calcrete)which may be eroded and from conglomerates.

The organic content is greater in the B-horizonwhich is brown due to less oxidation of organic mat-ter, hence the term brown-earth profiles. If evaporationis greater than precipitation there will be a net upwardtransport of porewater, causing dissolved salts from thegroundwater to be precipitated high in the soil profile.

3.6.5 What Factors Control WeatheringRate and Products?

Because weathering is the most important sediment-producing process, we are interested in understanding

how the rate of weathering is related to rock type, pre-cipitation, temperature, vegetation, relief etc. We alsotry to establish correlations between weathering prod-ucts, particularly clay minerals, and these factors. Bystudying sediments from older geological periods, wecan learn something about weathering conditions atthose times. Weathering products will also bear thestamp of the rocks undergoing weathering. The stabil-ity of a mineral during weathering is largely a functionof the strength of the bonds holding the cations in thecrystal lattice. Potassium (K+) in mica is held by weakbonds (low ionic potential) which are responsible forthe pronounced cleavage. In biotite, the Mg++ and Fe++

in the octahedral layer will also be weakly bonded.During weathering cations like K+, Na+, Ca++, Mg++

and Fe++can be attacked by protons (H+) which willreplace them and send them into solution. Chain sil-icates like hornblendes and pyroxenes will also berelatively unstable and rapidly weather. In feldsparsthe alkali ions are dissolved so that the whole mineraldisintegrates. Stability is lowest in calcium-rich plagio-clase, while pure albite (sodium feldspar), orthoclaseand microcline (potassium feldspar) are more stable.The breakdown of these silicate minerals will pri-marily liberate alkali cations. Silicon and aluminiumhave very low solubility and form new silicate min-erals, largely clay minerals, though some silicic acid(H4SiO4) goes into solution.


1. 2K(Na)Al2 AlSi3O10(OH)2 (muscovite) + 2H+

+ 3H2O = 3Al2Si2O5(OH)4 (kaolinite)+ 2K+(Na+)

2. 2K(Mg, Fe)3AlSi3O10(OH)2 (biotite)+ 12H+ + 2e + O2 =Al2Si2O5(OH)4 (kaolinite) + 4SiO2 (in solution) +Fe2O3 + 4 Mg++ + 6H2O + 2K+

3. 2K(Na) AlSi3O8 (feldspar) + 2H+ + 9H2O =Al2Si2O5(OH)4 (kaolinite) + 4H4SiO4 + K+(Na+)

We see that potassium has been replaced by hydro-gen ions in the new silicate minerals. The same appliesto sodium in albite. The equations show that the reac-tions are driven to the right by low K+/H+ and Na+/H+

ratios.The degree of weathering depends on how under-

saturated the water is with respect to the mineralscomprising the rock, and on the volume of water flow-ing through the rock. If the reaction products in thesolution, K+, Na+ and silica (H4SiO4), are not removedby water flow, the reactions will cease. This is whyweathering is always found to begin along crackswhere water can penetrate (Fig. 3.10). The verticaland horizontal joints that often develop in response topressure release when previously deeply buried rockis exposed at the land surface provide the initial path-ways. As the weathering process spreads outward fromjoints, blocks of unweathered rock are gradually iso-lated. They have rounded corners and may becomeentirely round (spheroidal weathering) (Fig. 3.11a,b,c).In desert areas, where there is little rainfall, weatheringproceeds much more slowly. Illite and montmorillonitemay be formed under higher K+/H+ and Na+/H+ ratiosthan kaolinite, and they are frequently formed wherethere is less water percolation and the removal ofpotassium or sodium is slower.

There is often also a high silica content in thewater in desert areas due to frequent silica algae

(diatom) blooms and because silica is concentrated bywater evaporation. This helps enhance the stability ofsmectite.

Granites subjected to weathering over a very longperiod often develop a special topography. Fracturesand fault zones weather fastest and form valleys wherethe groundwater collects, which further acceleratesthe weathering. The more massive granitic areas willstand out in the terrain, and because precipitation runsswiftly down into the depressions between the ele-vated portions, the topographic difference will becomemore and more pronounced. Granites surrounded bysedimentary rocks will, because of their high contentof feldspar, normally weather faster than the sedimentswhich contain more quartz and other stable minerals.Particularly if the sediments consist of quartzites andshales, the granite will form a depression in the terrain.The weathering products from granites will normallybe quartz grains which form sand grains the same sizeas the quartz crystals in the granite, and clay con-sisting of kaolinite, and possibly also some illite andsmectite formed from feldspars and micas. We have atthe outset a bimodal grain-size distribution with sandand clay, but very little silt.

Basic rocks (e.g. gabbro) will weather far morerapidly than granite because basic plagioclase (Ca-feldspar), pyroxenes and hornblende are very unstableand dissolve faster than silica-rich (acid) minerals.During progressive weathering sodium, potassium,magnesisum and calcium will be removed by thegroundwater but some potassium may be adsorbed onclay minerals. The weathering residue will be enrichedin elements with low solubility such as Ti, Al, Siand Mn (Fig. 3.12). In a normal, oxidising weatheringenvironment, all the iron will precipitate out again asiron oxide (Fe(OH)3) while the magnesium will tendto remain in solution. In areas with high rainfall theconcentration of ions like K+, Na+ and silica will be

Weathering = Rock + H2O + H+ Weathering product + Ions in solution

Exfoliation cracks

Granite

Rainwater

Fig. 3.10 Weathering of granites along extensional fractures

102 K. Bjørlykke

A B

C

Fig. 3.11 (a, b) Weathered granite (North of Kampala, Uganda) showing different stages of weathering. (c) Concentric – spheroidalweathering in a basic intrusive rock, Weathering is much faster in the absence of quartz

diluted and kaolinite will precipitate. Where porewa-ter circulation is slower we may get a higher build-upof Mg++, Ca++ and silica concentrations in the water,so that smectite (montmorillonite) or chlorite preci-pitates. Smectite requires porewater with a relativelyhigh silica concentration (Fig. 3.13) and is thereforeoften found in sediments derived from volcanic rocksthat contain glass or soluble silicate minerals. Biogenicsources of silica (diatoms, radiolaria) will also increase

the silica concentration in porewater because amor-phous silica is much more soluble than quartz.In desert environments evaporation of water afterrainfalls will concentrate silica in the porewater andmake smectite stable. Figure 3.12 shows analyses ofrocks at various stages of transformation due to weath-ering. Weathering proceeds particularly rapidly inamphibolites: Na+, Mg++ and Ca++ are quickly leachedout, while A13+, Fe3+ and Ti4+ become enriched. K+ is


%200

180

160

140

120

100908070605040302010

J1 J2 J3 J4 J5 J6 J7CaOMgONa2O

K2O

SiO2

Al2O3

Fe2O3

TiO2

Fig. 3.12 Chemical analyses of changes in the chemical com-position of samples representing progressive weathering of anamphibolite compared to an unweathered sample. The stable

elements (Ti, Fe and Al) are enriched while there is a strongdepletion of Na, Mg and Ca. Potassium is depleted but isadsorbed on clay minerals in the soil

8.0

6.0

0.0

4.0

2.0

–4.0 –3.0 –2.0 –1.0 0.0

Quartzsaturation

Amorphoussilica saturation

Log aSiO2 (aq)

LogaK+/aH+

Illite

Gibbsite Kaolinite

Smectite

Microcline

Mixedlayerclay

Fig. 3.13 Activity diagram showing the stability of some min-erals as a function of silica and the K+/H+ ratio (after Aagaardand Helgeson 1982). Weathering reactions are characterised by

reduction in both these two parameters towards the lower left ofthe diagram

104 K. Bjørlykke

however to large extent adsorbed on clay minerals andmuch more Na+ than K+ is therefore supplied to theoceans with the rivers.

After the alkali cations have been dissolved out ofthe silicates and kaolinite has been formed, extremelyslow leaching of quartz commences. When the concen-tration of silica in the porewater is sufficiently low (seeFig. 3.13), kaolinite will be unstable and be replacedby gibbsite Al(OH)3 or (A12O3•3H2O). Since gibbsitecannot form as long as the porewater is in equilibriumwith quartz, all the quartz must have dissolved first orbecome encapsulated (e.g. in a layer of iron oxides).Clearly, gibbsite will form far more rapidly during theweathering of basic rocks than of granites, since theinitial silica content is considerably lower. It takes avery long time to dissolve all the quartz in a granite.The solubility of quartz at surface temperatures andpH 7–8 is about 5 ppm, increasing at higher pH val-ues. Alkaline (basic) water can therefore increase thesolution rate of quartz. The end product of the weath-ering process is laterite, which consists of gibbsite andiron oxides or hydroxides. Under atmospheric (oxidis-ing) conditions with a neutral pH, aluminum hydroxideand iron oxides may for practical purposes be regardedas insoluble.

At low pH values, e.g. under the influence of humicacids in humid tropical climates, aluminum is moresoluble than iron and selective leaching of A13+ willproduce iron-rich laterites. Aluminum hydroxide canbe dissolved at low (acidic) or high (basic) pH val-ues and may reprecipitate as aluminum oxide, whichhas a lower iron content and is thus a higher gradealuminium source (bauxite).

3.7 Distribution of Clay Mineralsand other Authigenic Mineralsas a Function of Erosionand Weathering

3.7.1 What Determines the Type of ClayMinerals We Find in Sedimentsand Sedimentary Rocks?

When rocks are subjected to erosion and weather-ing, clastic minerals are broken down and perhapssomewhat altered, with respect to the minerals in the

parent rocks. We can also get new minerals formedin the source rock itself, and precipitation of newminerals through weathering.

The more rapidly erosion and transport take placecompared to the rate of weathering, the closer the com-position of sediments is to the source rock. Glacialsediments represent one end of the scale in terms ofsediment composition. Because glaciations are char-acterised by very high rates of erosion and low tem-peratures, chemical weathering will be very weak, andQuaternary sediments – including clays – will havea composition which essentially represents the aver-age of the rocks which have been eroded. Sedimentsdeposited in fault-controlled basins (e.g. rift basins)have a short transport distance between the site oferosion and deposition and the sediments have lit-tle time to weather on the way. Clastic chlorite andbiotite break down relatively rapidly during weather-ing and are therefore likely to be preserved in additionto feldspar in rift basins. These unstable minerals aregood indicators of rapid erosion and/or cold climates,since otherwise they are unlikely to survive.

An analysis of the distribution of clay minerals inmodern sediments shows that we find clastic chlo-rite almost exclusively in high latitudes, except aroundislands of basic volcanic rocks (e.g. basalts). Clasticchlorites from metamorphic rocks or altered basicrocks are more magnesium-rich than authigenic chlo-rites, which are iron-rich (chamositic). In temperateareas with moderate to high precipitation, weatheringproceeds relatively rapidly. In desert areas weatheringproceeds very slowly because all weathering reactionsrequire water. The type of weathering and the distri-bution of clay minerals are clearly related to latitude(Fig. 3.14).

Smectite and illite are typical of deserts becausethe weathering is much slower when water is nearlyabsent. When feldspar and other unstable minerals arealtered (weathered) gradually in a dry climate, alkaliions and alkaline earth ions such as K+, Na+, Ca2+ andMg2+ will not be removed rapidly enough due to littlepercolation of fresh rainwater.

Rainwater will cause some dissolution of mineralsand, during the periods of drying, the silica con-centration may be higher so that smectite becomesstable.

As a result we will usually get illite or smectiteformed as authigenic minerals because they are stablein the presence of high K+/H+ ratios in the porewater.


2700mm

Evaporation

Precipitation

Temperature°C252015105

Chlorite

Illite and smectite

LateriteKaolinite

Smectite (montmorillonite)

24002100190015001200900600300

Fig. 3.14 Simplified diagram showing the distribution of weathering and common clay minerals as a function of latitude andrainfall. Cold areas and deserts are characterised by little weathering and more mechanical erosion

Smectite (montmorillonite) is thus a common claymineral in desert areas, and its ability to swell whenwet renders sediments very plastic during floods. Thisexpansion of smectite also lowers its permeabilityand may be the reason why water can flow overthe surface for a long time before sinking into theground. In addition, capillary forces will prevent rapidpercolation of water through dry soil. On the oceanfloor near desert regions we find that illite and smec-tite are typical minerals, brought there by aeoliantransport.

In tropical areas where precipitation is relativelyhigh, the rate of weathering will be very rapid. Thisis not only because weathering processes acceleratewith temperature, but also because vegetation produceslarge amounts of organic acids (humic acids) whichare very effective in breaking down silicate minerals.Microbiological organisms such as fungi and bacteriaalso help in the breakdown process by producing CO2

which forms carbonic acid, H2CO3.Gibbsite (A12O3•3H2O) and iron oxides (haematite,

goethite, Fe2O3•3H2O) are constituents of the lat-erite which we find only in tropical areas with rapidweathering and slow erosion. Whereas iron oxides arealso found at higher latitudes, gibbsite occurs almostexclusively in humid tropical areas.

Laterisation is a very slow process and takesmillions of years, even in tropical regions withrapid weathering. It is therefore primarily in tropical

areas that we find bauxite for the aluminum indus-try. Iron-rich laterites may have an iron content ofover 50%, and in some areas (e.g. India) have beenexploited as iron ore. Laterite forms a very hardcement-like crust over the weathering profile and isalso virtually devoid of nutrients, so crop cultivation isimpossible. In East Africa (especially Uganda), how-ever, erosion has incised through a layer of Tertiary lat-erites. While the laterite cover remains on flat elevatedsurfaces, fresher, more fertile, rocks and weatheringmaterial is exposed in the valley sides. The vegetationin some tropical areas is more abundant, even if thesoils are very poor in nutrients, because the vegetationrecycles those nutrients which are available. If the veg-etation is removed and organic material is no longerproduced, oxidation and the absence of humic acids(increased pH) will lead to precipitation of oxides andhydroxides which make the soil hard and uncultivable.

Volcanic ash consisting of glass and unstable vol-canic mineral assemblages may alter to smectite onland or on the seafloor. In deep sea sediments zeoliteslike phillipsite are common.

Areas with volcanic rocks, particularly amorphousmaterial (volcanic glass), will often form zeolites.These require a high concentration of both silica andalkaline ions in the water, which is the situation whenglass dissolves. Zeolites, particularly phillipsite, areformed authigenically on the Pacific Ocean bed and arealso found in lakes (e.g. in East Africa).

106 K. Bjørlykke

In summary we can say that the factors which deter-mine which types of clay minerals are “produced” inthe various areas are:

1. The rocks which are eroded/weathered (sourcerocks)

2. Rate of erosion3. Temperature4. Precipitation5. Vegetation6. Permeability of source rocks and sediments (perco-

lation of water).

Typical distribution of various minerals:

1. Chlorite and biotite – high latitudes (cold climate)– rapid erosion.

2. Kaolinite – humid temperate and humid tropicalregions – good drainage.

3. Smectite (montmorillonite) – low precipitation orpoor drainage. Typical of desert environments, butalso formed in impervious, e.g. basaltic, rocks inmore humid environments. Typically formed fromvolcanic rocks.

4. Gibbsite – tropical humid climate – long weatheringperiod.

5. Zeolites – formed in areas with volcanic materialand restricted porewater circulation. Require a highconcentration of silica and alkali ions.

3.8 Geochemical Processes in the Ocean

The ocean can be regarded as a reservoir of chemicalsdissolved in water. It looks as though the composi-tion of seawater has not altered radically throughoutthe geological ages from the early Palaeozoic until thepresent day, although there have certainly been somevariations.

The supply of elements to ocean water from riversand by water circulation at the spreading ridges mustbe balanced by a removal of the same elements fromthe ocean water (Fig. 3.15). The annual additionof salts dissolved in river water is about 2 × 109

tonnes/year. The figure was probably less in the geo-logical past because vegetation was sparse or absent.The development of land plants that produce humic

acids, which in turn produced more rapid weathering,has probably increased the supply of salts (sinceDevonian times). This trend was sometimes slowed byperiods with higher sea level that caused widespreadtransgressions and converted huge tracts of coastalland areas into continental shelf (e.g. during the UpperCretaceous) reducing the weathering and the supply ofsalts and nutrients to the ocean.

Some ions like potassium (K+) are adsorbed to clayminerals supplied by rivers (B in Fig. 3.15). Sodium(Na+) is so strongly hydrated that it has a tendencyto remain in solution, while potassium will be far lesshydrated and can be more easily adsorbed onto clayminerals and rapidly removed from seawater.

Most of the elements which the rivers bring tothe ocean are precipitated by organic processes.Organisms can build their own internal chemical envi-ronment, and use their energy to precipitate mineralswhich are not normally stable in seawater. Carbonate-secreting organisms, e.g. foraminifera, molluscs etc.,will precipitate aragonite or calcite even when thewater is cold and undersaturated with respect to theseminerals. Diatoms are so effective in precipitating sil-ica (amorphous silicon dioxide, SiO2), that in mostplaces the seawater near the surface in the photiczone (photosynthetic zone) is much more depletedwith respect to silica than to quartz. Organisms, whenthey die, will in most cases start to break downby oxidation of organic matter and by dissolutionof the mineral skeletons. In reducing environmentsmuch of the organic matter will however be pre-served. In shallow tropical waters, like on a carbonatebank, the seawater may be saturated with respectto carbonate (calcite), but carbonate also accumu-lates in cold water like the Barents Sea becausethe rate of dissolution is slower than the rate ofprecipitation.

The more efficient organisms are at building skele-tons despite undersaturation, the more rapidly they willdissolve. Diatoms, for example, dissolve to the extentof 99–99.9% before they have sunk to the seabed.Only a very small proportion is therefore preserved insediments.

Photosynthesising organisms in the surface wateruse up CO2 and produce oxygen and organic matter:

CO2 + H2O + nutrients = CH2O + O2


Addition ofdissolved ions

WeatheringA

B

Adsorption onclay minerals

Uptake of organisms in the ocean

Dissolution oforganisms inthe water columnand on theseafloor

Circulation of seawater throughbasalt in spreadingridge

C

Geochemical processes in the ocean

F

E = C – D A = B + E + F + G + H

H

Precipitation of poorlysoluble salts duringevaporation, e. g. chloridesulphates, and carbonates

GGrowth of authigenic minerals onthe seafloor, particular by zeolites

D

E

Deposition of organicmatter in sediments

Fig. 3.15 The chemical composition of the seawater remainsnearly constant over geologic time. The supply of ions in solu-tion from rivers and from spreading ridges must therefore be

equal to the removal of dissolved components by precipitationof minerals and adsorption on clay minerals. The most solublecomponents (Na, Cl, KCl) are only removed by evaporation

This helps to keep the pH high so that carbonatesare stable or dissolve slowly. At depth in the absenceof sunlight, respiration and oxidation prevail, releas-ing CO2 and thus lowering the pH and increasing thesolubility of carbonate.

The depth at which the solubility of carbonateincreases relatively rapidly is called the lysocline. Thedepth where the rate of solution is greater than therate of carbonate sedimentation is called the carbonatecompensation depth (CCD). Near the equator the CCDmay be at 4–5 km, becoming shallower at higher lati-tudes. Carbonate can therefore still accumulate on theseafloor even when the water column is undersaturated,if the rate of supply exceeds the rate of dissolution.This is analagous to snow accumulating when the tem-perature is above freezing so long as it falls quickerthan it melts This solution of organic material liber-ates nutrients which can then be returned to the surfacethrough upward flow. In this manner they are repeat-edly recycled. The annual biological production in theocean is therefore many times greater than the sup-ply of nutrients from the land. That fraction of theorganic production which has been removed from theocean by preservation in seabed sediments must be

replaced with nutrients, mostly from land. They can-not be returned to the ocean before the sediments areelevated and subjected to erosion and weathering. Theamount of organic matter which is deposited in sedi-ments is a function of the rate of production minus therate of solution.

The growth of authigenic (newly formed) mineralson the seabed (C in Fig. 3.15) is also an important pro-cess in removing elements from seawater. The mostsignificant are the zeolites, which can develop wheresediments on the seabed have a high silicate or alu-minium content, particularly from volcanic material(glass). They may remove Na+, K+ and Ca2+ fromseawater, but growth can also proceed very much at theexpense of elements already present in the sediment(e.g. in the Pacific Ocean). This applies particularly tophillipsite, heulandite, clinoptilite and analcite.

Apart from this there is little direct chemical pre-cipitation from seawater with normal salinity. This isbecause biological precipitation is more efficient inmany cases and prevents the build-up of sufficient con-centrations of the elements needed for chemical pre-cipitation. Sulphate-reducing bacteria are active in theuppermost few centimetres of the sediment, however,

108 K. Bjørlykke

removing sulphur from seawater in the form of sul-phate and reducing it to sulphides which are thenprecipitated (e.g. iron sulphides, FeS and FeS2).

In areas that are almost cut off from the open ocean,where evaporation exceeds the freshwater rainfall andsupply by rivers, we find that even very soluble saltsare being precipitated. Under these conditions, thereis enrichment of elements which are otherwise pre-cipitated only to a limited degree through biologicalor chemical processes, such as Na, Cl, S, Mg andtrace elements such as B and Br. The amount of saltthus precipitated in evaporites has probably varied verymarkedly throughout the geological ages.

Biological and chemical precipitation, as describedabove, are not sufficient alone to account for the geo-chemical balance of the ocean. Important geochemicalreactions are taking place in the spreading ridges in theoceans (F in Fig. 3.15). Heat from the basalt that isflowing up along the spreading ridge, drives convectioncells which cause ocean water to flow through thebasalts and up along the ridge. The seawater reactswith hot basalt (basic rock melt) and disolves mineralscontaining iron and other metals. Since seawater con-tains sulphur (as S04

−), this leads to precipitation ofsulphides, for example iron sulphides and copper sul-phides. When hot water flows up pipe-like chimneysin the ocean floor near spreading ridges, it mixeswith the seawater and again sulphides are precipi-tated. When water is oxidised, iron oxides and man-ganese oxides are precipitated around the spreadingridges.

3.8.1 Residence Periods for DifferentElements in the Ocean

How long does an element spend in the sea after beingdelivered by rivers, before it is chemically or biologi-cally precipitated? This residence period is an expres-sion of how rapidly an element is removed comparedto its concentration in the ocean. For example, sodiumhas the longest period of residence (about 200 mil-lion years) because little sodium is removed throughbiological or chemical precipitation except in evapor-ite basins. The residence time for potassium is about 1million years because it is more rapidly adsorbed ontoclay minerals and thereby removed. Rare earths haveperiods of only a few 100 years.

3.9 Circulation of Water in the Oceans

Ocean currents are driven by:

1. The rotation of the earth (Coriolis effect)2. Tidal forces3. Differences in water density due to variations in salt

content and temperature4. Wind forces due to atmospheric circulation.

Ocean currents are extremely important for redis-tributing heat from low latitudes to higher latitudes.This circulation is strongly dependent on the topogra-phy of the ocean floor and the distribution of oceanand continents. Bottom currents in the ocean basinsare very different from those at the surface, and oftenflow in opposite directions. While warm surface waterflows from the equator to the poles, cold surface watersinks at the poles and flows along the ocean floor tothe equator. Both currents are deflected by the Corioliseffect, towards the right in the northern hemisphere andtowards the left in the southern hemisphere. While sur-face currents (like the Gulf Stream) will be deflectedeastwards, deeper currents will be deflected towardsthe west of the oceans (e.g. the Atlantic Ocean). Thesedeep-sea currents which follow the depth contours,may be strong enough to transport silt and fine sand,and the resultant deposits are called contourites.

The vertical circulation of seawater is highly sensi-tive to variations in temperature and salt concentration.In periods with glaciation at the poles, the temperaturegradient in the surface water flowing from the equatorto the poles is far greater than in non-glacial peri-ods (such as the Mesozoic). Oxygen-rich cold waternow flowing down from the polar regions into theocean basins is important for maintaining oxidisingconditions in the deep ocean basins.

Animals and bacteria use oxygen from seawatercontinuously (for respiration), and oxidation of deadorganic material also requires oxygen. If we did nothave this downward flow of cold surface water, thewater in the ocean basins would be reducing. Someocean basins are isolated from this circulation, and wemay then have a more permanent layering of waterbased on temperature and salt content. Warm surfacewater with low density can flow over heavier, colderbasal water without the water masses mixing to anymajor extent. The boundary between warm and coldwater masses is called a thermocline. If the density


difference is largely due to salt content, we call theboundary a halocline. A pycnocline is the boundarybetween two water masses with different densities,without a specified cause. Lakes in temperate andcold regions have good circulation. This is due towater attaining maximum density at 4◦C, below whichtemperature density inversion causes turnover.

The addition of freshwater to a basin (e.g. BalticSea, Black Sea) also leads to stratification of the waterdue to salt concentration because brackish water flowson top of marine water with higher salinity. Evaporitebasins produce water which is heavy due to its highsalt content and therefore forms a layer which flowsalong the bottom. If a salinity stratification becomesestablished, it will weaken or destroy the circulationand lead to reducing conditions in the bottom layer. Ifthe density contrast due to salt concentration is greaterthan that due to temperature, this will impede or pre-vent the downward flow of cold, oxygen-rich surfacewater (Fig. 3.16).

There are indications that in previous geologicalperiods (e.g. the Cretaceous) the bottom water in the

ocean basins was warm, salty water (about 15◦C) com-pared with the present situation with cold basal water(2–3◦) and a normal salinity. A higher average temper-ature in the oceans leads to reduced CO2 solubility anda deeper carbonate compensation depth (CCD). Thevolume of water welling up from deeper water layersto the surface corresponds to the amount of downflow.If we have basal water with high salinity, the reduceddensity contrast results in less downward flow and con-sequently less upwelling, so less nutrients are added tothe surface water.

We have seen that a number of different processescontrol the geochemical equilibrium of the ocean.There must also be an equilibrium between the addi-tion and removal of chemical components for theocean water composition to remain relatively constant.The conditions in which this equilibrium was main-tained, however, have varied through geological time.During the first part of the Earth’s history, up to about2.5 billion years ago, the atmosphere was reducing.Most geochemical processes acted very differentlythen from the way they do now. Weathering was less

Evaporation

............................................................

Anoxic

Pycnocline

Low salinity Organic production

Freshwater

Addition ofnutrients

Positive water balance

Black Sea model

Threshold

Negative water balance

Mediterranean and Red Sea model

> Supply of freshwater

Threshold

Oxic Heaviersaltwater

Fig. 3.16 Circulation of water in basins with a surplus of fresh-water input compared to evaporation (Black Sea model). Thisresults in poor vertical circulation and anoxic (reducing) bot-tom waters. The Mediterranean and Red Sea model represents

excess evaporation compared to freshwater input. The surfacewater will then have the highest salinity and density and sink tothe bottom. This increases the vertical circulation and helps tomaintain oxic conditions

110 K. Bjørlykke

efficient because of the low oxygen concentration inthe atmosphere and limited biological weathering. Thesupply of ions to the ocean via rivers was conse-quently less. On the other hand, seafloor spreadingwas most probably faster with more seawater circu-lated through the spreading ridges. Isotope studies(87Sr/86Sr) of seawater in early Precambrian rocksindicate that at that time the composition of seawaterwas more strongly controlled by circulation throughthe basalt on the spreading ridges. We can say thatthe chemical composition of the ocean was buffered bymaterial from the spreading ridge, i.e. mantle material(Veizer 1982).

3.10 Clastic Sedimentationin the Oceans

Clastic sediments are produced chiefly on the con-tinents and are brought to ocean areas through flu-vial or aeolian transport. Island arcs associated withvolcanism may produce large amounts of sedimentcompared to their area because they are tectonicallyactive, which leads to elevation and accelerated ero-sion. Volcanic rocks are for the most part basic andweather quickly, forming large quantities of sedimentsaround volcanic island groups, while fine-grained vol-canic ash becomes spread over wide areas. Submarinevolcanism may also produce some sediment, for exam-ple along the Mid-Atlantic Ridge, but this is verylimited.

The main supply of clastic sediment is fed into theocean through deltas, then transported along the coastand down the continental slope to the abyssal plains.Around Antarctica there is a significant amount ofdeposition of clastic, glacial sediments. In areas in themiddle of the Atlantic Ocean, far from land, the rateof sedimentation is as low as 1–10 mm/1,000 years.

The Atlantic receives a relatively large supply of clasticsediment, in particular from seven major rivers: theSt. Lawrence, Mississippi, Orinoco, Amazon, Congo,Niger and Rhine. Exceptionally high sedimentationrates characterise the Gulf of Mexico, where rapiddeposition of thick sequences from the Mississippidelta has prevailed since Mesozoic times.

The South American and African continents drainmainly into the Atlantic. The water divide between theAtlantic and the Pacific Oceans lies far to the westin South America, and that with the Indian Ocean inAfrica is far to the east (Fig. 7.16).

The Pacific Ocean is surrounded by a belt of vol-canic regions and island arcs. There are relativelyfew rivers that carry large amounts of clastic sedi-ment directly into the Pacific Ocean, in contrast tothe Atlantic. Sediment which is eroded, for exampleon the Asian continent, is deposited in shallow marineareas (marginal seas) such as the Yellow Sea and ChinaSea. The sediments are cut off from further transportby the island arc running from Japan and southwards.The Pacific Ocean is therefore dominated by volcanicsediments.

Volcanic sedimentation takes the form of volcanicdust and glass, which may be transported aerially overlong distances. After sedimentation, volcanic glass willturn into palagonite, an amorphous compound formedby hydration of basaltic tuff. Palagonite may then befurther converted into montmorillonite or zeolite min-erals. The zeolite phillipsite is very widely found in thePacific, but is scarce in the other oceans. Pumice is alsoa volcanic product, and may drift floating over greatdistances. The eruption of volcanoes in the PacificOcean area in historic times has shown that large erup-tions produce 109–1010 tonnes of ash, and much thesame amount of pumice and agglomerates.

Submarine volcanism, by contrast, produces verylittle ash to form sediment. The lava which flows outonto the seabed will solidify as an insulating crust on

Table 3.1 Review of the ratio between mechanical and chemical denudation of the different continents (After Garrelsand Machenzie 1971)

ContinentAnnual chemical denudation(tonnes/km)

Annual mechanical denudation(tonnes/km)

Ratio mechanical/chemicaldenudation

North America 33 86 2.6South America 28 56 2.0Asia 32 310 9.7Africa 24 17 0.7Europe 42 27 0.65Australia 2 27 10.0


contact with the water (often forming pillow lava), sothat little volcanic matter goes into suspension.

Weathering and erosion processes are responsi-ble for the entire volume of sediment which can bedeposited in sedimentary basins. Material added byrivers takes the form of clastic and dissolved matter.The ratio between the quantities of these two forms ofsediment addition is a function of precipitation, tem-perature and relief. Dry areas, like Australia, producemainly clastic material, while the African continentproduces mainly dissolved material because of theintensive weathering in some parts of the continent(Table 3.1).

Further Reading

Garrels, R.M. and Machenzie, F.T. 1971. Evolution ofSedimentary Rocks. W.W. Norton & Co Inc., New York, NY,397 pp.

Kenneth 1982. Marine Geology. Prentice Hall. EnglewoodCliffs. 813 pp.

Chamley, H. 1989. Clay Sedimentology. Springer, New York,623 pp.

Chester, R. 1990. Marine Geochemistry. Unwin Hyman,London, 698 pp.

Eslinger, E. and Pevear, D. 1988. Clay Minerals forPetroleum Geologists and Engineers. SEPM ShortCourse 22.

Garrels, R.M. and Christ, C.L. 1965. Solutions, Minerals andEquilibria. Harper and Row, New York, 450 pp.

Manahan, S.E. 1993. Fundamentals of EnvironmentalChemistry. Lewis Publ., Chelsea, MI, 844 pp.

Saigal, G.C. and Bjørlykke, K. 1987. Carbonate cements inclastic reservoir rocks from offshore Norway – Relation-ships between isotopic composition, textural developmentand burial depth. In: Marshall, J.D. (ed.), Diagenesisof Sedimentary Sequences. Geological Society SpecialPublication 36, 313–324.

Veizer, J. 1982. Mantle buffering and the early Oceans.Naturvissenshaffen 69, 173–188.

Velde, B. 1995. Origin and Mineralogy of Clays. Springer,Berlin, 334 pp.

Weaver, C.E. 1989. Clays, muds and shales. Developments inSedimentology 44, 819 pp.

sedimentarygeochemistry - universitas...

Documents