The occurrence and behaviour of rare earth and associated … · The occurrence and behaviour of rare earth and associated elements in lateritic regolith profiles in Western Australia

The occurrence and behaviour of rare earth and associated

elements in lateritic regolith profiles in Western Australia

XIN DU

BSc (EnvSc)

This thesis is presented for the degree of Doctor of Philosophy

in Soil Science of the University of Western Australia

School of Earth and Environment

Faculty of Natural and Agricultural Sciences

2012

i

Statement of candidate contribution

This thesis contains published work and/or work prepared for publication, some of

which has been coauthored. The bibliographical details of the work and where it

appears in the thesis are outlined below (with percentage contributions from coauthors

in parentheses).

i. Du, X. (70%), Rate, A.W. (15%), & Gee, M. (15%), 2012. Redistribution and

mobilization of titanium, zirconium and thorium in an intensely weathered lateritic

profile in Western Australia. Chemical Geology 330-331, 101-115.

ii. Du, X. (70%), Rate, A.W. (15%), & Gee, M. (15%), 2012 (In press). Particle size

fractionation and chemical speciation of REE in a lateritic weathering profile in

Western Australia. Explore.

iii. Du, X. (70%), Rate, A.W. (15%), & Gee, M. (15%), 2011. Translocation and

fractionation of rare earth elements within intensely weathered lateritic profiles in

Western Australia. Mineralogical Magazine 75, 784. Oral presentation in the 2011

Goldschmidt Conference, Prague, Czech Republic. available at:

http://goldschmidt.info/2011/abstracts/finalPDFs/784.pdf

iv. Du, X. (70%), Rate, A.W. (15%), & Gee, M. (15%), 2010. Geochemical

mass-balance in intensely weathered soils, Darling Range, Western Australia. In:

Gilkes, R.J. & Prakongkep, N. (Editors). Proceedings of the 19th World Congress

of Soil Science: Soil Solutions for a Changing World. IUSS: Brisbane, Australia.

(Published on DVD; ISBN 9780646537832; available at:

https://events.ccm.com.au/ei/viewpdf.esp?id=126&file=E%3A%5Ceventwin%5Cdocs

%5Cpdf%5Csoil2010Abstract01769%2Epdf)

We hereby declare that the individual authors have granted permission to the candidate

(Xin Du) to use the results presented in these publications.

Student Signature

Coordinating Supervisor Signature


https://events.ccm.com.au/ei/viewpdf.esp?id=126&file=E%3A%5Ceventwin%5Cdocs%5Cpdf%5Csoil2010Abstract01769%2Epdf


iii

ABSTRACT

The nature of the redistribution and fractionation of rare earth elements (REE) during

supergene weathering is not fully understood, especially in lateritic weathering which

is characterised by intense weathering and formation of ferruginous materials.

Therefore, the geochemical characteristics and mode of occurrence of REE were

investigated in four lateritic regolith profiles (GE, MQ I, MQ II and JG) developed on

granitoids with dolerite dykes in Western Australia. The outcomes of this study are

important factors to consider when using REE as tracers for the regolith weathering,

pedogenesis and sedimentation.

A high deficiency of REE relative to parent granitoids is typical of the regolith studied,

especially in the GE and JG profiles, suggesting high mobility of REE. Breakdown of

weathering-susceptible light REE (LREE)-rich minerals, such as allanite and/or

REE-rich fluorocarbonate facilitates depletion of LREE at early stages of weathering.

The released REE either are partially leached away by solutions, or precipitate as

secondary phosphates (e.g. rhabdophane and florencite). These secondary phosphates

play an important role in sequestering REE and hence limiting their further mobility,

especially for LREE. Trace to minor amounts of REE are associated with clay minerals,

Fe oxides/oxyhydroxides and organic ligands, and thus are retained in regolith, as

revealed by the sequential extraction experiment.

Heavy REE (HREE)-rich minerals, such as zircon, are relatively weathering-resistant

and thus HREE hosted by these minerals are not susceptible to inter-horizon transport.

Residual accumulation of weathering-resistant minerals is the main control on the

retention of HREE in intensely weathered regolith. The occurrence of REE is

dominated by mineral phases (the residual species) in intensely weathered lateritic

regolith as revealed by the sequential extraction. Therefore, the abundance, stability

and composition of LREE-rich secondary phosphates and HREE-rich

weathering-resistant minerals control the fractionation of REE in lateritic regolith, both

of which are largely affected by the weathering conditions.

iv

Cerium fractionated from the other REE and showed positive anomalies in the

duricrust of GE (Ce*=6.1) and JG (Ce

*=25.3) profiles. In the JG profile, neoformed

poorly crystalline (hydr)oxide phases enriched in Ce, Zr and Th were observed as a rim

attaching onto: (i) the wall of Al/Fe-rich pores in the duricrust, and (ii) the boundary

layer between Al-rich and Fe-rich rims in iron nodules of the upper ferruginous zone.

This suggests that: (i) Ce fractionates from other REE as hydrous cerianite

(CeO2∙nH2O) and precipitates with Fe oxyhydroxides during oxidative processes;

(ii) Zr and Th mobilize at the sampling scale during lateritization which is attributed to

the breakdown of thorite and REE-rich fluorocarbonates during initial weathering.

In the JG profile, trace concentrations of Yb (0.02-0.12 wt%) were determined in iron

cores and clay layers of iron nodules and minor concentrations of REE were associated

with extracted Fe oxide/oxyhydroxide species. Fine-grained (<10 µm) REE-bearing

phosphates were incorporated in crystalline Fe oxides in the duricrust, or occurred in

the clay layer of iron nodules. These imply that: (i) translocation of REE occurred both

at mineral and profile scales; and (ii) Fe oxides/oxyhydroxides are important in

redistribution and fractionation of REE at advanced stages of lateritization.

High concentrations of REE reside in the silt and clay size fractions, inferring that

formation of secondary minerals and adsorption by clay minerals augmented REE

concentrations in fine particle fractions. The sand size fraction had the lowest

concentrations but the highest mass of REE, indicating the dilution effect of quartz and

the importance of weathering-resistant minerals in retention of REE.

In addition, redistribution of Th into secondary phosphates as a trace component and

strong partitioning into gravel rather than matrix showed translocation of Th both at

mineral and profile scales. Absence of primary sphene crystals and observation of

dissolved ilmenite and rutile reflect the mobility of Ti at the mineral assemblage scale.

Fluctuation of Ti/Zr in the ferruginous zone in contrast to the consistency of Zr/Hf

throughout the JG profile (within the range of parent granitoids) implies that Ti and Zr

fractionate from each other during extreme weathering and advanced lateritization.

v

ACKNOWLEDGEMENTS

This thesis was completed with financial support from the International Research Fees

Scholarship (China Scholarships) and a Top-Up Scholarship provided by the

University of Western Australia (UWA), China Scholarship Council (CSC), and an

analytical funding (In-Kind Student Support) provided by the Association of Applied

Geochemists (AAG).

I would like to gratefully thank my PhD supervisors, A/Prof Andrew W. Rate and

A/Prof Mary Gee in the University of Western Australia for their expert assistance,

guidance, mentoring, and encouragement throughout my candidature. You helped me

with sample collection even in hot summer with flies all-around and helped me without

any complaints in correcting my grammatical mistakes and provided expert

commentary on my thesis. I feel very honoured to have had the opportunity to learn

from and be inspired by such exceptional scientists.

Thank you to our talented analytical chemist Michael Smirk, who provided me with

analytical assistance and practical advice; and I am thankful to Rick Pearce and Tracey

Quinn for their analytical expertise to help me analysing the trace elements in

sequential extractions and particle size fractions; I also want to thank Mr Frank

Nemeth, who helped me with the preparation of polished thin sections and polished

mounts with patience.

I am very grateful to David Adams, Janet Muhling and Peter Duncan from the Centre

for Microscopy, Characterisation and Analysis (CMCA) for their assistance, teaching

and sharing their expertise with me during my microscopy and microprobe analysis.

I would like to acknowledge the Australian Synchrotron for beamtime allocation which

allowed the completion of the Synchrotron X-ray Powder Diffraction (SXRD) and

Synchrotron X-ray Fluorescence Microscopy (SXFM) work in this thesis. I would also

like to thank Dr Justin Kimpton, the beamline scientist of SXRD, Dr David Paterson

and Dr Daryl Howard, the beamline scientists of SXFM and Dr Chris Ryan, the

scientist who developed the Maia detector and Geopixe software. Without your kind

vi

support, this work would not have been completed.

I would also like to thank Prof. Bob Gilkes and Prof. Martin Fey for their great advice

and comments on my research during my candidate.

In addition, thank you to all the lovely UWA Soil Science administration staff, Gail,

Karen and Margaret, for being so helpful and supportive during my study and thank

you to the amazing soil groups of staff and students, Ursula Salmon, Prof. Peng Bo,

Bree Morgan, Talitha Santini, Andrew Lucas, Yongjun Lu and Dr Nattaporn

Prakongkep and Georgina Holbeche, for sharing your extensive knowledge of the

study and for encouraging me when I was down.

And finally, thank you to my sweet and supportive family, Mum, Dad, Mum-in-law,

Dad-in-law and particularly my husband, Yan, who tolerated my temper when I met

research problems and who cooked delicious food to encourage me. Thank you for

your understanding, support and for making me laugh every day. Thank you.

vii

TABLE OF CONTENTS

1 Introduction ........................................................................................... 1

1.1 General introduction ........................................................................................ 1

1.2 Thesis scope ..................................................................................................... 3

1.3 Thesis structure ................................................................................................ 4

2 Literature review ................................................................................... 9

2.1 An introduction to rare earth elements ............................................................ 9

2.2 Chemistry of the REE .................................................................................... 11

2.3 Data presentation of REE .............................................................................. 11

2.4 REE-hosting minerals in granitoids and weathered regolith ......................... 12

2.5 Weathering intensity and geochemistry of REE ............................................ 13

2.5.1 Proxies for weathering intensity and flux change .................................. 13

2.5.2 Redistribution of REE in weathered regolith ......................................... 15

2.5.3 Fractionation of REE during weathering ................................................ 18

2.5.4 Mineral transformation of REE during weathering ................................ 19

2.6 Geochemical pathways of REE during lateritization .................................... 20

2.6.1 Definition of lateritic profiles ................................................................. 20

2.6.2 A typical lateritic profile ......................................................................... 22

2.6.3 Lateritization ........................................................................................... 22

2.6.4 Geochemical behaviour of REE during lateritization ............................. 23

2.6.5 Anomalies of Ce in lateritic regolith ...................................................... 25

2.6.6 Anomalies of Eu in lateritic regolith ...................................................... 26

2.7 Summary ........................................................................................................ 26

3 Description of the study areas ............................................................ 33

3.1 General geology and climate ......................................................................... 33

3.2 Sampling and profile description ................................................................... 33

4 Redistribution of major elements in lateritic profiles during

intensive weathering in Western Australia ............................................. 39

4.1 Abstract .......................................................................................................... 39

4.2 Key words ...................................................................................................... 39

4.3 Introduction ................................................................................................... 40

4.4 Materials and methods ................................................................................... 41

viii

4.4.1 Analytical methods ................................................................................. 41

4.4.2 Weathering intensity-Chemical Index of Alteration (CIA) .................... 44

4.4.3 Mass balance calculation ........................................................................ 44

4.4.4 Statistical analyses .................................................................................. 45

4.5 Results ........................................................................................................... 45

4.5.1 Weathering intensity of parent rocks and the regolith ............................ 45

4.5.2 Mineralogical properties ......................................................................... 52

4.5.3 Mass balance analysis of elemental loss and gain .................................. 53

4.5.4 Depth functions of pedogenic discontinuities ........................................ 59

4.5.5 Grain size distribution of major elements in MQ II profile .................... 60

4.6 Discussion ...................................................................................................... 60

4.6.1 Significant processes during lateritization .............................................. 60

4.6.2 Genesis and sources of Fe redistribution ................................................ 61

4.6.3 Degrees of lateritization ......................................................................... 64

4.6.4 Principal components analysis ............................................................... 64

4.6.5 Mineralogy and element grain size distribution ..................................... 67

4.6.6 Mobility of Ti and Zr .............................................................................. 68

4.7 Summary of the chapter ................................................................................. 69

5 Redistribution and mobilization of Ti, Zr and Th in an intensely

weathered lateritic profile in Western Australia .................................... 73

5.1 Abstract .......................................................................................................... 73

5.2 Key words ...................................................................................................... 74

5.3 Introduction ................................................................................................... 74

5.4 Materials and methods ................................................................................... 76

5.4.1 Analytical methods ................................................................................. 76

5.4.2 Mass balance calculation ........................................................................ 79

5.5 Results ........................................................................................................... 82

5.5.1 Bulk Ti, Zr and Th concentrations in regolith ........................................ 82

5.5.2 Mass balance of Ti and Th ...................................................................... 85

5.5.3 Mineralogical characteristics of Ti, Zr and Th in the JG profile ............ 86

5.5.4 Grain size distribution of Ti, Zr and Th in the lateritic regolith ............. 94

5.5.5 Partition of Ti, Zr and Th into different extraction species .................... 97

5.6 Discussion ...................................................................................................... 97

5.6.1 Mode of occurrence of Zr and Th in the lateritic regolith ...................... 97

5.6.2 Sources of Zr in poorly crystalline phases in duricrust .......................... 98

ix

5.6.3 Partitioning of Th between gravel and matrix ........................................ 99

5.6.4 Mobility of Ti in the JG profile ............................................................ 100

5.6.5 Geological parent mineralogy vs. weathering conditions .................... 101

5.7 Summary of the chapter ............................................................................... 102

6 Distribution and fractionation of REE in intensely weathered

lateritic profiles in Western Australia ................................................... 103

6.1 Abstract ........................................................................................................ 103

6.2 Key word ..................................................................................................... 104

6.3 Introduction ................................................................................................. 104

6.4 Methods and materials ................................................................................. 105

6.4.1 Analytical methods ............................................................................... 105

6.4.2 Calculation methods ............................................................................. 107

6.5 Results ......................................................................................................... 109

6.5.1 Geochemical data of REE ..................................................................... 109

6.5.2 Mineralogy of REE in the parent rock .................................................. 123

6.5.3 Mineralogy of REE in the regolith ....................................................... 125

6.5.4 REE in grain size fractions and chemical extractions of regolith ......... 132

6.6 Discussion .................................................................................................... 135

6.6.1 Evolution of REE-bearing minerals during intense weathering ........... 135

6.6.2 Reason for stronger depletion of LREE over HREE ............................ 137

6.6.3 Fractionation of REE in weathered regolith ......................................... 139

6.6.4 Ce and Eu anomaly ............................................................................... 139

6.6.5 Grain size fractionation and chemical speciation of REE .................... 140


7 Mode of occurrence of REE in an intensely weathered lateritic

profile in Western Australia ................................................................... 143

7.1 Abstract ........................................................................................................ 143

7.2 Key words .................................................................................................... 144

7.3 Introduction ................................................................................................. 144

7.4 Materials and methods ................................................................................. 145



7.5 Results ......................................................................................................... 149

7.5.1 Bulk geochemical data of REE ............................................................. 149

x

7.5.2 Mineralogy of REE in the parent rock ................................................. 161

7.5.3 Mode of occurrence of REE in lateritic regolith .................................. 162

7.6 Discussion .................................................................................................... 176

7.6.1 Geochemical pathways and fractionation of REE ................................ 176

7.6.2 Enrichment mechanism of Ce in ferruginous zone .............................. 177

7.6.3 Effects of Fe oxides/oxyhydroxides on mode of occurrence of REE... 178


8 Particle size fractionation and chemical speciation of REE in a

lateritic profile in Western Australia ..................................................... 181

8.1 Abstract ........................................................................................................ 181

8.2 Key words .................................................................................................... 181

8.3 Introduction ................................................................................................. 182

8.4 Materials and methods ................................................................................. 183



8.5 Results ......................................................................................................... 185

8.5.1 Concentrations of REE in different particle size fractions ................... 185

8.5.2 Mass loading of REE in different particle size fractions ...................... 185

8.5.3 Speciation of REE from sequential extraction ..................................... 186

8.6 Discussion .................................................................................................... 193


9 Conclusion and future work ............................................................. 199

9.1 Conclusion ................................................................................................... 199

9.2 Future work ................................................................................................. 201

9.3 Summary ...................................................................................................... 204

10 References ....................................................................................... 207

11 Appendices ...................................................................................... 229

xi

TABLE LIST

Table 2.1 Summary of REE in common minerals in granitoid rocks ............................. 28

Table 2.2 Concentrations of REE in different types of parent rock ................................ 30

Table 2.3 Concentrations of REE in different horizons of lateritic regolith profiles ...... 31

Table 4.1 Selected physical and chemical properties of matrix fractions (<2mm) of the

profiles studied ......................................................................................................... 43

Table 4.2 Concentrations of major elements in gravel and matrix of four lateritic

profiles...................................................................................................................... 47

Table 5.1 Sequential extraction procedures of trace elements in the lateritic regolith.... 77

Table 5.2 Concentrations of Ti, Zr and Th in grain size fractions of the JG profile ....... 81

Table 5.3 Element concentrations of minerals in Figure 5.4 and Figure 5.5 based on

EPMA in parent meta-granitoids and lateritic regolith in the JG profile ................. 88

Table 5.4 Element concentrations from EPMA in Figure 5.8 (a) and (b) ....................... 92

Table 5.5 Concentrations of Zr, Ti and Th in different sequential extraction species ..... 97

Table 6.1 concentrations of REE in parent rock and lateritic regolith of the GE and MQ

profiles.................................................................................................................... 118

Table 6.2 Concentrations of REE and associated elements from EPMA analyses of

representative minerals in parent granitoids from the GE profile .......................... 129


representative minerals in parent granitoids from the MQ profile ......................... 130


representative minerals in lateritic regolith from the MQ profile .......................... 131

Table 6.5 Concentrations of REE in sequential extractions of representative regolith in

the GE and MQ I profiles ....................................................................................... 134

Table 7.1 Concentrations of REE and derived fractionation parameters in parent

meta-granitoids and lateritic regolith from the JG profile ..................................... 154

Table 7.2 Element concentrations from EPMA analyses of representative minerals in

parent meta-granitoids (Figure 7.6) of the JG profile ............................................ 157

Table 7.3 Element concentrations from EPMA analyses of REE-bearing phosphates in

xii

lateritic regolith (Figure 7.8) of the JG profile....................................................... 158

Table 7.4 Element concentrations from EPMA analyses of weathering-resistant

minerals in lateritic regolith of the JG profile ........................................................ 159

Table 7.5 Element concentrations in Figure 7.11, Figure 7.12 & Figure 7.13 of the

duricrust and iron nodules in the JG profile ........................................................... 174

Table 7.6 REE concentrations of random spots in iron core and clay layer in iron

nodules from the A horizon and upper ferruginous zone of the JG profile ............ 175

Table 8.1 Concentrations of REE in grain size fractions of the JG profile ................... 190

Table 8.2 Concentrations of REE in different chemical extractions of representative

regolith in the JG profile ........................................................................................ 192

xiii

FIGURE LIST

Figure 3.1 Sampling sites (a, labelled as box) and sketches of the profiles sampled (b).37

Figure 3.2 Photographs of regolith from selected horizons of the GE profile. ............... 38

Figure 3.3 Photographs of regolith from selected horizons of the MQ I profile ............ 38

Figure 4.1 Ternary A-CN-K and A-FM-CNK plots of regolith samples from four

lateritic profiles (GE, MQ I, MQ II, JG) based on chemical compositions of matrix

and gravel samples. .................................................................................................. 51

Figure 4.2 Semi-quantitative mineralogical composition of regolith samples and parent

granitoids determined by random powder XRD analysis based on weighted

average of matrix and gravel .................................................................................... 55

Figure 4.3 Mass balance of major elements in regolith samples from four lateritic

profiles, based on weighted average concentrations of major elements in matrix

and gravel at each depth, using Zr as the reference element .................................... 56

Figure 4.4 Depth functions of the molar ratio Na/K and concentration ratio

Al2O3/Fe2O3 for MQ two profiles and concentration ratio (Ti/Zr)/10 for four

profiles...................................................................................................................... 57

Figure 4.5 Major element concentrations in grain size fractions of the regolith samples

from the MQ II profile. ............................................................................................ 58

Figure 4.6 The distribution of Al2O3 vs. SiO2 and Al2O3 vs. Fe2O3 in matrix and gravel

from four lateritic profiles ........................................................................................ 63

Figure 4.7 Schellmann SiO2-Al2O3-Fe2O3 diagrams showing different degrees of

lateritization of weathered regolith from four lateritic profiles ............................... 64

Figure 4.8 Principal component analyses of major elements in regolith samples and

parent granitoids from four lateritic profiles ............................................................ 66

Figure 4.9 Principal component factors of regolith samples and parent granitoids from

four lateritic profiles calculated using major element composition. ........................ 66

Figure 4.10 Calculated τ values of Al, Fe and Si referenced to Zr in matrix and gravel

from four lateritic profiles ........................................................................................ 71

Figure 5.1 Variation of Ti, Zr and Th with depth in the JG profile ................................. 84

Figure 5.2 Variation of Ti/Zr, Zr/Hf and Ti/Th with depth in the JG profile .................. 84

xiv

Figure 5.3 Mass balance calculations of Ti and Th against depth in the JG profile,

based on weighted average concentrations in matrix and gravel, using Zr as the

reference element ..................................................................................................... 85

Figure 5.4 Backscatter electron images of Ti-, Zr- and Th- hosting phases in parent

meta-granitoids of the JG profile ............................................................................. 87

Figure 5.5 Backscatter electron images showing Ti retained as ilmenite and Ti oxides

in the ferruginous mottled zone of the JG profile .................................................... 91

Figure 5.6 Diffraction patterns from SXRD showing evidence for transformation of Ti

from ilmenite and rutile in the ferruginous mottled zone (JG4) to anatase in the

duricrust (JG5) of the JG profile .............................................................................. 92

Figure 5.7 (a) The only partially dissolved zircon grain identified in the duricrust

(circled) and (b) a typical fractured, partially metamict zircon grain in the A

horizon (<1 m depth)................................................................................................ 92

Figure 5.8 Neoformed poorly crystalline Zr-hosting phases associated with Ce on pore

walls around Al/Fe matrix in the duricrust of the JG profile ................................... 93

Figure 5.9 Forms of Th persisting in regolith samples of the JG profile ........................ 95

Figure 5.10 Grain size distribution of Zr, Ti and Th in the JG profile. ........................... 96

Figure 6.1 REE distribution patterns of (a) rocks and regolith samples normalized by

the average chondrite; and (b) regolith samples normalized by the parent granitoid

in lateritic GE profile ............................................................................................. 112

Figure 6.2 REE distribution patterns of (a) rocks and regolith samples normalized by

the average chondrite; and (b) regolith samples normalized by the parent granitoid

in lateritic MQ I profile .......................................................................................... 113

Figure 6.3 REE distribution patterns (a) rocks and regolith samples normalized by the

average chondrite; and (b) regolith samples normalized by the parent granitoid in

lateritic MQ II profile. ............................................................................................ 114

Figure 6.4 normalized ratios (La/Sm)PR (LREE/MREE) and (La/Yb)PR (MREE/HREE)

and CIA of regolith samples against depth in three lateritic profiles ..................... 115

Figure 6.5 SiO2-Al2O3-Fe2O3 ternary plots and associated variation of REE

concentrations and ratios against the S/SAF weathering index for the GE profile.121

Figure 6.6 Mass balance calculations of REE against depth for three lateritic profiles,

based on weighted average concentrations of REE in matrix and gravel, using Zr

xv

as the reference element ......................................................................................... 122

Figure 6.7 Backscatter electron images of REE-bearing accessory minerals in parent

granitoids of the GE profile.................................................................................... 124


granitoids of the MQ profiles ................................................................................. 127

Figure 6.9 Backscatter electron images of REE-bearing minerals in regolith of the MQ

profiles.................................................................................................................... 128

Figure 6.10 Concentrations of selected REE (La, Ce, Sm, Dy, and Yb) in grain size

fractions of the MQ II profile. ................................................................................ 133

Figure 6.11 Mass loading of selected REE (La, Ce, Sm, Dy, and Yb) in grain size

fractions of the MQ II profile ................................................................................. 133

Figure 7.1 REE distribution patterns of (a) meta-granitoids and regolith samples

normalized by the average chondrite composition; and (b) regolith samples

normalized by the parent meta-granitoid in the JG profile .................................... 151

Figure 7.2 Normalized ratios (La/Sm)PR (LREE/MREE) and (La/Yb)PR (MREE/HREE)

of regolith samples against depth in the JG profile ................................................ 152

Figure 7.3 Plots of (La/Sm)PR and (La/Yb)PR vs. La for the JG profile, illustrating the

degrees of depletion and fractionation of REE. ..................................................... 152

Figure 7.4 SiO2-Al2O3-Fe2O3 ternary plots and associated variation of REE

concentrations and ratios against the S/SAF weathering index for the JG profile. 156

Figure 7.5 Mass balance calculations of REE against depth in the JG profile, based on

weighted average concentrations of REE in matrix and gravel, using Zr as the

reference element ................................................................................................... 160


meta-granitoids of the JG profile ........................................................................... 164

Figure 7.7 REE distribution patterns of fluorocarbonate and thorite normalized by the

parent meta-granitoids in the JG profile................................................................. 165

Figure 7.8 Backscatter electron images of REE-bearing secondary phosphate minerals

in regolith samples of the JG profile ...................................................................... 166

Figure 7.9 Images of REE-bearing secondary phosphates located in the clay layer of

iron nodules at 1.5m depth in the JG profile .......................................................... 168

xvi

Figure 7.10 Mapping of secondary rhabdophane in iron nodule at 1.5 m depth of the

JG profile................................................................................................................ 169

Figure 7.11 Cerium fractionated from other REE and occurring as a rim along the

Al/Fe-rich pores in the duricrust ............................................................................ 171


boundary between clay and iron layers in iron nodules ......................................... 172

Figure 7.13 Cerium fractionated from other REE and occurring as joint matrix between

two iron cores within one large nodule .................................................................. 173

Figure 7.14 Backscatter electron images of crystalline Fe oxides intergrown with

micron-size Ce-rich secondary phosphates in the duricrust ................................... 174

Figure 8.1 Concentrations of REE in grain size fractions in the JG profile ................. 188

Figure 8.2 Mass loading of REE in grain size fractions in the JG profile .................... 188

Figure 8.3 Distribution of REE percentages in sequential extractions of the

representative regolith of the JG profile ................................................................ 189

Figure 8.4 Normalized ratios of (La/Sm)PR and (La/Yb)PR in particle size fractions and

sequential extractions in the JG profile .................................................................. 196

xvii

APPENDIX LIST

Appendix 11.1 Abbreviation ......................................................................................... 229

Appendix 11.2 ICP-OES analyses of the reference standards determined repeatedly

with samples for each analysis ............................................................................... 232

Appendix 11.3 R script for principal component analysis of major elements .............. 234

Appendix 11.4 Photographs of polished thin sections of iron nodules mounted on

quartz slides from the JG profile ............................................................................ 235

Appendix 11.5 Detailed operation procedure of the sequential extraction method ...... 236

Appendix 11.6 ICP-MS analyses of the reference standards determined repeatedly with

samples for each analysis ....................................................................................... 240

Appendix 11.7 EPMA detection limits of element concentrations in Ti-, Zr- and Th-

bearing minerals in the JG profile .......................................................................... 242

Appendix 11.8 EPMA detection limits of element concentrations in minerals of parent

granitoids and regolith samples from the GE and MQ profiles ............................. 243

Appendix 11.9 Concentrations of REE in grain size fractions of the MQ II profile .... 245

Appendix 11.10 EPMA detection limits of element concentrations in REE-bearing

minerals from the JG profile .................................................................................. 247

1

1 Introduction

1.1 General introduction

Supergene weathering is an important geochemical process for element cycling in

Earth surface environments, involving water/rock interaction and resulting in many

fundamental chemical changes. In turn, the composition of rock and regolith may

provide useful insights into the chemistry and nature of these interactions, including

the mechanisms of element mobility in crustal environments (McLennan, 1989). As an

index for petrological evolution, regolith weathering, pedogenesis and sediment tracing,

the geochemical cycling of rare earth elements (REE) is worthy of further and

sustained research.

It is widely accepted that REE can mobilize, redistribute and fractionate during

supergene weathering (Aide and Pavick, 2002; Aubert et al., 2001; Banfield and

Eggleton, 1989; Braun et al., 1993; Koppi et al., 1996; Laveuf and Cornu, 2009;

Nesbitt, 1979; Tyler, 2004). Although REE have been widely studied, the geochemical

behaviour of REE during weathering cannot be easily generalized because of: (i) wide

variance of REE-bearing minerals and their relative concentrations of REE; (ii)

different accessibility of these minerals to solutions and variance of solution chemistry;

and (iii) location-specific physicochemical and biological factors during weathering

(Bao and Zhao, 2008; Price et al., 1991). Even though, the mobilization and

fractionation of REE during weathering is proposed to be constrained mainly by the

primary REE-bearing minerals and weathering conditions (Aubert et al., 2001; Braun

et al., 1998; Ji et al., 2004; Nesbitt, 1979).

However, the precise sequence of events, behaviour and fractionation of REE during

rock weathering and pedogenesis are not completely understood, especially during

formation of iron nodules/ferruginous duricrust in intensely weathered lateritic profiles

defined by high concentrations of Fe oxides and oxyhydroxides. Although iron oxides

and oxyhydroxides are known to have high surface areas thus rendering them very

efficient sinks for many cations (e.g. Cu, Ti, V and Zn, etc.) and anions (e.g. phosphate

Occurrence and behaviour and rare earth and associated elements in lateritic regolith profiles in Western Australia

2

and silicate) (Singh and Gilkes, 1992), their effects on the translocation and

fractionation of REE in lateritic soils under superficial weathering and lateritization are

still not well understood.

Diverging views exist with regard to the stage at which REE fractionation starts during

the course of weathering and how weathering intensity affects REE fractionation.

Different mobility of heavy REE (HREE) and light REE (LREE) has been widely

reported in lateritic profiles: e.g. higher mobility of HREE relative to LREE (e.g.

Braun et al., 1993; Huang and Gong, 2001; Ma et al., 2007) in contrast to higher

mobility of LREE over HREE (Beyala et al., 2009; Braun et al., 1990; Ndjigui et al.,

2009; Nesbitt and Markovics, 1997).

The geochemical behaviour of REE during extreme weathering has been far less

studied than incipient or moderate weathering in temperate zones. Elements that are

conserved in temperate zones, such as REE, Ti and Zr, may become mobile during

extreme chemical weathering in tropical regions (Braun et al., 1993). Consequently, a

systematic understanding of the mobilization and translocation of REE and the

commonly considered conservative elements Ti, Zr and Th and their redistribution into

different solid phases is fundamental in order to use REE as a tracer for weathering and

sedimentation and Ti, Zr or Th as immobile elements to assess mass flux or volume

change during supergene weathering.

As the product of intense weathering, lateritic regolith represents one of the most

common superficial formations in the tropics, and is commonly diachronous, extending

over tens of millions of years (Dequincey et al., 2006). The genesis of the ferruginous

horizon of laterite, where polyphases are usually involved during weathering and

lateritization, is poorly understood; and the major and trace element behaviour during

its formation is also difficult to define. In Western Australia, extreme weathered

regolith (e.g. bauxitic) has been widely investigated, including the geological,

geographical and morphological information and geochemical characteristics.

However, the investigation of the geochemical behaviour of REE in intensely

weathered regolith profiles in Western Australia has not been as widely investigated,

Chapter One: Introduction

3

though the extreme weathering intensity may further enhance the mobility of the REE

during the lateritization processes.

Accordingly, this study investigates the abundance and mode of occurrence of REE in

the parent rock and lateritic regolith building on previous work by Sadleir and Gilkes

(1976 and the references therein), Brimhall et al. (1988; 1992) and Anand and Paine

(2002), Anand and Butt (2003; 2010) and Anand et al. (1991) etc.. The aim is to

improve our understanding of the geochemical behaviour and fractionation

mechanisms of REE during intense weathering and lateritization. Associated mobility

of Ti, Zr and Th are also examined, particularly in regards to the ferruginous duricrust

and iron nodules formed under extreme weathering and strong lateritization.

1.2 Thesis scope

This thesis investigates the occurrence, behaviour and fractionation of REE and

associated elements in four intensely weathered lateritic profiles developed on variably

metamorphosed granitoids with dolerite dykes in Western Australia.

Specifically, this thesis addresses the following objectives:

i. To understand more fully the environmental behaviour of major elements in

different solid phases and mineralogical and geochemical processes of lateritic

weathering.

ii. To investigate the mass flux change and mode of occurrence and thus to assess

the mobility of Ti, Zr and Th at different scales under intense lateritic weathering.

iii. To quantify the abundance, redistribution and residence of REE in different

regolith horizons in lateritic profiles and to investigate factors affecting the

mobilization and translocation of REE during intense weathering.

iv. To improve the understanding of the mode of occurrence and fractionation

mechanism of REE during lateritization, especially in iron nodules and ferruginous

duricrust, whose effects on the translocation and fractionation of REE under intense

supergene weathering are still not well understood.

v. To study further the REE redistribution in different grain size fractions and


4

chemical speciation in lateritic regolith and to improve understanding of the

translocation and fractionation of REE affected by weathering and particle size sorting

processes.

1.3 Thesis structure

This thesis is composed of nine chapters:

Chapter One (Introduction) provides general background information and introduces

the objectives, scope and structure of this thesis.

Chapter Two (Literature Review) surveys several critical areas of current knowledge

about geochemical behaviour and fractionation of REE during supergene weathering

and lateritization, defines the terminology used in this thesis, and gives a

comprehensive overview of the previous research on the topic.

Chapter Three (Description of the studied areas) describes the geological, geographical

and morphological information of the areas studied. It also describes the regolith

profiles studied (GE, MQ I, MQ II and JG profiles) in terms of horizons, texture,

colour, depth etc. and provides essential details of sampling procedures.

Chapters Four to Eight are five self-contained research papers. Each of these chapters

addresses one primary objective for this study:

Chapter Four (Geochemistry of major elements, objective i) investigates the

mineralogy and geochemistry of protolith and regolith for four lateritic profiles and

quantifies the mass flux change of major elements in different grain size fractions as

well as bulk regolith. This provides important insights into elemental redistribution

into different solid phases and enhances a holistic understanding of the environmental

behaviour of major elements during lateritization.

Following on from this, Chapter Five (mobility of Ti, Zr and Th, objective ii), taking

the JG profile as an example, investigates the mode of occurrence of Ti, Zr and Th in


5

ferruginous materials and assesses the mobility of Ti, Zr and Th at the mineral

assemblage and profile scales by the combined use of electron probe micro-analysis

(EPMA) and synchrotron X-ray powder diffraction (SXRD), together with bulk

geochemical data. The source for Zr and Th in the neoformed phases is proposed to be

from the breakdown of thorite and REE-bearing fluorocarbonates rather than zircon

during the early stage of weathering. Strong partitioning of Th into gravel rather than

matrix reflects redistribution of Th at the profile scale. Fluctuation of Ti/Zr ratios in the

ferruginous zone in contrast to the consistency of Zr/Hf ratios throughout the profile

suggests that Ti and Zr fractionated from each other and partitioned between gravel

and matrix during extreme weathering and advanced lateritization. The results in this

chapter prove that Ti, Zr and Th are mobile at a variety of scales, despite their accepted

use as reference elements for studying element mass flux change. This novel

investigation substantially contributes to the current understanding of geochemical

behaviour and partitioning of Ti, Zr and Th under intense supergene weathering.

Chapter Six (Residence and fractionation of REE, objective iii) investigates the

residence, translocation and fractionation of REE in three (GE, MQ I and MQ II)

moderate to intensely weathered lateritic profiles. It addresses the factors controlling

the mobility and affecting the fractionation of REE during intense weathering. Strong

depletion of REE in the regolith suggests high mobility of REE beyond the profile

scale during extreme weathering. Higher depletion of LREE than HREE in regolith

compared with parent granitoids reflects the fundamental controls on the mobility of

REE by the mineralogy of the parent rock and the subsequent characteristics of the

weathering conditions. This is important information to enable confident use of REE as

tracers of geochemical processes in intensely weathered settings. The specifics of REE

depletion and fractionation help us to understand the geochemical pathways of REE

during supergene weathering, and have the potential to be a strategic clue for

mineralogical exploration.

Continuing along a similar theme, Chapter Seven (Mode of occurrence of REE,

objective iv) investigates the mode of occurrence of REE retention in ferruginous


6

materials in the JG profile developed over granitoids. Mode of occurrence of REE is

determined by the combined use of EPMA and synchrotron X-ray Fluorescence

Microscopy (SXFM). The occurrence of Ce, associated with Zr and Th, as neoformed

poorly crystalline (hydr)oxide phase forming a rim or coating around an Al-Fe matrix

in the pore system of ferruginous duricrust indicates mobilization and fractionation of

Ce during strong lateritization. To my knowledge, there are no other similar published

studies that identify Ce fractionated from other REE and associated with Zr and Th,

concentrated in ferruginous materials in lateritic regolith. Potential geochemical

signatures and fractionation mechanisms of Ce during lateritization are also explored.

The co-occurrence of Ce, Zr and Th in neoformed phases have been discussed in the

previous chapter (Chapter Five), which investigates the mobility of commonly

considered immobile elements Ti, Zr and Th.

In addition to an extensive study of the mineralogical residence of REE in lateritic

regolith of the JG profile, distribution of REE in grain size fractions and chemical

speciation is also examined in Chapter Eight (objective v). The dominance of REE by

mineral phases resistant to weathering identified in Chapter Seven is consistent with

residual forms in sequential extraction, grain size and electron microprobe data, in

intensely weathered regolith. However, REE can also be retained in the regolith as the

adsorbed ions or complexes/ligands of organic matter, clay minerals, and Fe

oxides/oxyhydroxides. Significant concentrations of REE were determined in the water

soluble, exchangeable and adsorbed species, and these are likely to represent

potentially bioavailable forms of REE. The potential for REE bioavailability is critical

for understanding the behaviour of REE in natural environments or under

anthropogenic influences.

Chapter Nine is a final integrated conclusion, which brings together the relevant

findings from each research paper in addressing the thesis objectives, as well as

identifying possible areas for future research.


7

The submission details for each of the chapters are as follows:

(i) Du, X., Rate, A.W., and Gee, M. 2012. Redistribution and mobilization of

titanium, zirconium and thorium in an intensely weathered lateritic profile in

Western Australia. Chemical Geology, 330-331, 101-115.

(ii) Du, X., Rate, A.W., and Gee, M. 2012 (In Press). Particle size fractionation and

chemical speciation of REE in a lateritic weathering profile in Western

Australia. Explore.

(iii) Du, X., Rate, A.W., and Gee, M. 2011. Translocation and fractionation of rare

earth elements within intensely weathered lateritic profiles in Western Australia.

Mineralogical Magazine 75, 784. Oral presentation in the 2011 Goldschmidt

Conference, Prague, Czech Republic; available at:


(iv) Du, X., Rate, A.W., and Gee, M. 2010. Geochemical mass-balance in intensely

weathered soils, Darling Range, Western Australia. In: Gilkes, R.J. &

Prakongkep, N. (Editors). Proceedings of the 19th World Congress of Soil

Science: Soil Solutions for a Changing World. IUSS: Brisbane, Australia.

(Published on DVD; ISBN 9780646537832; available at:

https://events.ccm.com.au/ei/viewpdf.esp?id=126&file=E%3A%5Ceventwin%

5Cdocs%5Cpdf%5Csoil2010Abstract01769%2Epdf)




9

2 Literature review

2.1 An introduction to rare earth elements

The Lanthanides, also called rare earth elements (REE), are members of Group IIIA in

the periodic system and share very similar physical and chemical properties because of

their common outer electron shell configuration. The REE include: lanthanum (La),

cerium (Ce), praseodymium (Pr), neodymium (Nd), promethium (Pm), samarium (Sm),

europium (Eu), gadolinium (Gd), terbium (Tb), dysprosium (Dy), holmium (Ho),

erbium(Er), thulium(Tm), ytterbium (Yb), and lutetium (Lu). A further element yttrium

(Y) is also a member of Group IIIA and has a similar chemistry to that of the REE, so

is sometimes grouped with the REE.

The REE are lithophile elements (Goldschmidt, 1937) that, excepting Pm, invariably

occur together naturally in all types of crustal rocks (Kabata-Pendias and Pendias,

2001). The terrestrial abundance of the REE decrease with increasing atomic weight,

and, in accordance to the Oddo-Harkins rule (Harkins, 1917), REE with even atomic

numbers are more abundant than those with odd atomic numbers (Kabata-Pendias and

Pendias, 2001). Usually the REE are subdivided into the light REE (LREE; from La to

Sm, those with low atomic numbers and masses) and the heavy REE (HREE; from Gd

to Lu, those with higher atomic numbers and masses). Very occasionally the term

middle REE (MREE), is loosely applied to the elements from around Pm to Ho

(Henderson, 1984).

Although similar slight variations in the physical and chemical properties of REE can

result in differences in solubility of compounds, stability of complexes and different

valences, Ce and Eu being the main examples of the latter. These slight differences

result in fractionation of the REE by many different geochemical processes, as a

consequence REE have traditionally been utilized in studies of the evolution of

petrological systems and magmatic processes (Bea and Montero, 1999; Forster, 1998;

Poitrasson et al., 1996; Taylor and McLennan, 1985), and, because of the potential for

differential leaching (Compton et al., 2003; Huang and Gong, 2001) during weathering,


10

REE are also widely used in the study of pedogenesis and weathering processes (Aide

and Smith-Aide, 2003; Laveuf et al., 2008; Lee et al., 2004; Patino et al., 2003; Viers

and Wasserburg, 2004), aqueous systems (Braun et al., 1998; Dupre et al., 1999; Viers

et al., 2000; Viers et al., 1997; Viers and Wasserburg, 2004), and for mineral

exploration (Bierlein et al., 1999; Brugger et al., 2008; Rao et al., 2004; Smith et al.,

2000; Yang et al., 2009).

In order to understand geochemical processes using REE as indices, it is fundamental

to determine REE concentration and characterize REE-bearing minerals. In recent

years, due to advances in instruments such as inductively coupled plasma mass

spectrometry (ICP-MS), laser ablation secondary ionising microprobe and modern

electron microprobes, and greater accessibility to synchrotron techniques and

secondary-ion mass spectrometry ion probe (SIMS), fast and accurate analyses of REE

in bulk samples and minerals with extremely low detection limits, tremendously high

sensitivity, and at micron-scale of spatial resolution can now be achieved (Bidoglio et

al., 1992; Cao et al., 2000; Chen et al., 1993; Schmidt et al., 2007; Williams, 1996),

enhancing studies on REE partitioning during hydrothermal alteration and supergene

weathering (Barrea and Bonzi, 2001; Brugger et al., 2008; Nakai et al., 2001; Schmidt

et al., 2007; Takahashi et al., 2000). For example, micro X-ray fluorescence (µ-XRF)

and micro X-ray absorption near edge structure (µ-XANES) spectroscopy enable

measurement of the distribution and oxidation state of Eu in-situ in scheelite at near

µm-resolution (Brugger et al., 2008); 100 ppm REE diopside glass standards

determined by NanoSIMS have yielded good reproducibility and accuracy and the

spatial resolution can achieve 5-10 µm by this technique (Ito and Messenger, 2009).

Rock weathering, as one of the critical processes in the surficial geochemical cycling

of elements, plays an important role in mobilization and translocation of REE. During

early pedogenesis, the chemical composition of an immature soil will be strongly

controlled by the composition of geological parent materials, whereas the chemical

composition of mature soils reflects the dominant effects of the weathering

environment and processes (Thanachit et al., 2006). These processes mobilize and

Chapter Two: Literature review

11

fractionate REE, and hence REE fractionation patterns can be used as tracers of

pedogenetic processes including dissolution of primary minerals, formation of

secondary minerals, redox processes, transport of material and ion exchange

(Middelburg et al., 1988).

2.2 Chemistry of the REE

Originating in identical external electronic shells (5s, 5p, 6s), the similar chemical and

physical properties of the REE result in their being classed as a geochemically

coherent group. Rare earth elements occur mainly as tri-valent (3+

) ions in nearly all

natural mineral, and only Eu2+

, Sm2+

and Yb2+

are stable in aqueous solutions, while

Ce is also stable as tetra-valent Ce4+

(Cerny et al., 1989). The mobility of REE is

limited because of the very low solubility of their phosphate minerals. Rare earth

elements can form stable complexes with carbonate, fluoride, hydroxide or sulphate

anions in alkaline solutions. The similar radii and oxidation states of the REE allow for

liberal substitution of the REE for each other into various crystal lattices. This

substitution accounts for the characteristic multiple occurrences of REE within a single

mineral (Castor and Hedrick, 2006). However, differences in filling of 4f-orbitals,

from zero electrons (La) to 14 electrons (Lu) results in a regular decrease of the ionic

radius, known as the ‘lanthanide contraction’ (Cerny et al., 1989). The differences in

mass and effective ionic radius lead to differences in typical isomorphic substitutions:

e.g. REE3+

for Cr3+

,V3+

, Fe3+

, Nb3+

, Sc3+

; Ce4+

for Th4+

, U4+

, Zr4+

; and Eu2+

for Ca2+

,

Sr2+

, Pb2+

(Cerny et al., 1989), forming different types of REE-bearing minerals.

2.3 Data presentation of REE

Rare earth element concentrations are usually presented graphically; where the REE

concentrations are normalized to a chosen reference material by dividing the

concentrations of each REE in the sample with the same REE in the reference

materials. The ensuing value is then plotted as the logarithm of the normalized

abundance versus atomic number (referred to as a REE distribution pattern in this

thesis). The reference materials can vary, depending on sample type and the aim of the


12

study, however, the mean carbonaceous chondrite values are the most commonly used

(McDonough and Sun, 1995). Other international reference materials include the North

American shale composite (NASC), an average upper crust composition (UCC), and

Post-Archean Australian Shale (PAAS) (Henderson, 1984; McLennan, 1989). In

addition to these commonly used reference materials, a part of the system under

investigation, e.g. parent geological material or one soil horizon from the investigated

profile (e.g. Laveuf et al., 2008), can also be used to normalize the REE concentrations

in order to show the fractionation and mass flux of REE as regolith evolves. In this

thesis, average chondrite (Anders and Grevesse, 1989) and the parent rock were chosen

as reference materials to normalize the REE concentrations to show the

depletion/accumulation and fractionation of REE after persistent intense weathering.

In addition, fractionation of REE in this thesis refers to the variation of concentration

of one particular REE (e.g. Ce) or a group of REE (e.g. HREE), relative to the other

REE or another group of REE (e.g. LREE). To study REE fractionation in as much

detail as possible, the REE are divided into three groups in this thesis (Henderson,

1984): LREE (from La to Nd), MREE (from Sm to Ho) and HREE (from Gd to Lu).

The normalized values (La/Sm)PR and (La/Yb)PR are used for identifying fractionations

between LREE-MREE and LREE-HREE relative to the parent rock (PR). Both Ce and

Eu anomalies were calculated by the following formula (1) and (2) respectively (N

refers to the relevant REE concentration in the reference material):

Ce*=(Ce/CeN)/[(La/LaN)

0.5×(Pr/PrN)

0.5] (1)

Eu*=(Eu/EuN)/[(Sm/SmN)

0.5×(Gd/GdN)

0.5] (2)

2.4 REE-hosting minerals in granitoids and weathered regolith

Rare earth elements occur in more than 200 minerals distributed across a wide variety

of mineral classes (Cerny et al., 1989; Henderson, 1984; Kanazawa and Kamitani,

2006). It is not possible to cover all of these minerals here; rather, discussion is limited

to REE-bearing minerals that are important and widespread during weathering of

granitoids and pedogenesis in regolith developed from such rocks. For a detailed list of

REE-bearing minerals refer to Henderson (1984) or Cerny (1989).


13

The major minerals in granitoids commonly contain negligible contents of REE, e.g.

mica and orthopyroxene. Quartz does not contain REE (Compton et al., 2003; Laveuf

and Cornu, 2009), whereas feldspar may contain negligible to moderate amounts and

commonly has a positive Eu anomaly (Bea, 1996; Condie et al., 1995; Laveuf and

Cornu, 2009). Amphibole and clinopyroxene have appreciable REE concentrations but

low Th and U (Bea, 1996). Garnet is not only a very efficient concentrator of the REE

but is preferentially enriched in HREE (Henderson, 1984), in a similar manner, epidote

is preferentially enriched in HREE and has a moderate to strong positive Eu anomaly

(Table 2.1).

Accessory minerals generally play important roles in the distribution of REE because

of their typically high partition coefficients for REE, indicating that these accessory

minerals act as sites of concentration for REE (Henderson, 1984). This is particularly

important in granitoids, which are abundant in REE and often contain a wide range of

accessory minerals. If these accessory minerals are particularly susceptible to

weathering, then the whole-rock REE pattern may be dramatically changed. Therefore,

the abundance and stability of these REE-bearing minerals are fundamental controls on

the subsequent mobilization and translocation of REE during weathering.

2.5 Weathering intensity and geochemistry of REE

2.5.1 Proxies for weathering intensity and flux change

Weathering modifies rocks and sediments at or near the Earth’s surface by a

combination of physical and chemical processes (Taylor and Shirtliff, 2003). To

quantitatively evaluate the chemical weathering intensity and pedogenesis, different

indices based on whole rock geochemical analyses have been proposed for evaluating

pedogenic processes and the degrees of alteration. In this thesis, three types of proxies

have been used to characterize element mass flux and weathering intensity of regolith:

(i) Chemical Index of Alteration (CIA) (Nesbitt and Young, 1982); (ii) Trace element

concentration ratio; (iii) Mass balance calculation (Brimhall et al., 1991).


14

Based on the molar proportions of four major elements, the CIA is calculated

according to the formula (3) below (Nesbitt and Young, 1982):

CIA=100×Al2O3/(Al2O3+CaO*+Na2O+K2O) (molar basis) (3)

Where CaO* is CaO associated with the silicate fraction of samples (excludes

carbonates). It assumes that alkali metal elements are leached out from

aluminosilicates, such as feldspar and mica in ionic form, while Al2O3 is residual

during weathering and forms phyllosilicate clay and oxyhydroxide minerals. This

index has been used widely to identify multiple parent materials in soil profiles and

provide background weathering information to improve our understanding of

elemental mobility during weathering.

Trace element concentration ratios are favoured both as means of assessing element

partitioning and flux change during pedogenesis and weathering processes (Sheldon

and Tabor, 2009). The trace element ratios e.g. Ti/Zr, Ti/Th, Zr/Th, Zr/Nb, Ho/Dy,

Lu/Hf, Sm/Nd are considered to be relatively stable during initial and moderate

weathering, but sensitive during advanced stages of weathering (Condie et al., 1995;

Fernández-Caliani and Cantano, 2010; Sheldon and Tabor, 2009) and thus can be used

to evaluate element partitioning during intense weathering.

The mass balance calculation is a commonly used method to assess gains and losses of

various elements in soils, and is extensively used to understand pedogenesis and

investigate weathering intensity (Brimhall et al., 1991). It assumes that an immobile

element (e.g., Zr) behaves conservatively and can be used to calculate the mass flux

change of the other mobile elements (τ) and correct their concentrations for volumetric

strain (ε) during weathering and pedogenesis. The formula is given in Equation (4):

1))((,

,

,

,

, pj

wj

wi

pi

C

C

C

C

ji

(4)

In Equations (4), C represents concentration; subscript i identifies the immobile

element, j identifies the element of interest, w identifies weathered material and p

identifies parent rock. If τi,j = 0, the element has behaved conservatively; if τi,j = −1, the

element j is completely depleted during weathering; positive τi,j values signify absolute

enrichment.


15

This method of mass balance calculation is based on the assumption of a genetic

relationship between regolith and underlying material, and a reference element that is

relatively conservative during weathering in the regolith profile. The most frequently

used immobile elements are Zr, Hf, Sc, Ti, Nb, Ta and Th; however, in intensely

weathered lateritic profiles, it is difficult to define the ‘immobile element’. Although Ti,

Zr and Th have been considered to be the least mobile elements and thus used as

immobile reference elements in previous studies of laterite (e.g. Braun et al., 1993;

Brown et al., 2003; Ji et al., 2004), the mobility of these elements during intense

weathering is still under debate (e.g. Kahmann et al., 2008; Kurtz et al., 2000;

McLennan, 1995; Sheldon and Tabor, 2009).

2.5.2 Redistribution of REE in weathered regolith

The nature of REE redistribution during supergene weathering associated with

mineralogical reactions is not fully understood (McLennan, 1989). This is an important

avenue of research not only because more complete understanding of REE

geochemistry during weathering would provide important additional constraints on the

understanding of the weathering process; but also because documentation of REE

release and subsequent migration could provide insight on the geochemical behaviour

of REE during sedimentation processes and in aqueous geochemistry (McLennan,

1989).

Many field-based studies have demonstrated that REE can mobilize, redistribute and

fractionate during supergene weathering (e.g. Aide and Pavick, 2002; Aubert et al.,

2001; Banfield and Eggleton, 1989; Braun et al., 1993; Koppi et al., 1996; Laveuf and

Cornu, 2009; Nesbitt, 1979; Tyler, 2004), however, there is little agreement regarding

the overall magnitude or potential for fractionation among the REE. In some cases,

there is a failure to clearly address the distinction between small scale REE mobility

associated with mineral reactions and the larger scale transport of REE into or out of

the system under consideration (McLennan, 1989). Although REE have been studied

widely, the geochemical behaviour of REE during weathering cannot be easily


16

generalized because of: (i) the wide variance of both REE-bearing minerals and their

concentrations of REE; (ii) the different accessibility of these minerals to solutions and

the variance of solution chemistry; and (iii) the location-specific physicochemical and

biological factors during weathering (Bao and Zhao, 2008; Price et al., 1991). Despite

these uncertainties, the redistribution, mobilization and fractionation of REE during

weathering are proposed to be constrained mainly by the primary REE-bearing

minerals and the weathering conditions (Aubert et al., 2001; Braun et al., 1998; Ji et al.,

2004; Nesbitt, 1979).

The abundance of REE in regolith is usually dependent on the abundance of REE in

parent materials (e.g. Braun et al., 1990; Macfarlane et al., 1994), sites of concentration

in mineral phases (e.g. Braun and Pagel, 1994; Braun et al., 1993), the relative stability

of the mineral phases with respect to fluids (groundwater, penetrating soil water, etc.,

e.g. Braun et al., 1993; Lottermoser, 1990), weathering conditions (e.g. Ji et al., 2004;

Walter et al., 1995) and the types of soil (e.g. Hu et al., 2006; Mourier et al., 2008).

Concentrations of REE in the most common parent rock types are listed in the Table

2.2. Commonly, granitoids have higher REE contents than intermediate and mafic

rocks, while calcareous and ultramafic rocks have the lowest REE contents (Ding et al.,

2002). However, although developed from similar parent materials, regolith which

have undergone different weathering and pedogenic conditions, may also vary

significantly in their REE abundance (Ndjigui et al., 2009).

Generally, REE contents increase with depth (Nesbitt, 1979; Tyler, 2004) or with

decreasing weathering intensity (Ohlander et al., 1996; Taunton et al., 2000a). The

upper part of weathered regolith profiles is usually depleted in REE compared with

accumulation in the bottom part (Braun et al., 1993). Depletion of REE in the upper

part of profiles may reflect acidic conditions which cause REE mobilization, whilst

REE-retaining alkaline conditions dominate in the lower part (Nesbitt, 1979).

The main reason for these different REE redistribution patterns in weathered profiles is

thought to relate to weathering conditions. The mobility of REE is a complex function

of ionic radius and charge in solution, pH, Eh, water flux, history and weathering state


17

of the soil, redox conditions, water table, amount and type of inorganic and organic

ligands, microorganisms, and the nature of secondary and intermediate minerals

formed under different weathering conditions (Aubert et al., 2001; Braun et al., 1998;

Ji et al., 2004; Ma et al., 2004; Price et al., 1991; Taunton et al., 2000a; Taunton et al.,

2000b). The high degree of variability of this wide range of weathering conditions may

lead to variable abundance and translocation of REE in regolith. In addition,

volumetric change during weathering of the parent materials should also be considered

when interpreting REE mobility, thus, relative mass flux change based on mass

balance calculations has an advantage given the volumetric change. Lateral

redistribution of REE in landscapes as a component of catchment scale pedogenetic

processes should also be considered (Sommer et al., 2001; Sommer et al., 2000)

because REE are likely to be subject to horizontal redistribution by groundwater.

In addition, data on the nature and stability of REE complexes at low temperature are

of critical importance to the geochemical exploration for REE in supergene

environments (Wood, 1990). The REE can be mobilized in solution by forming stable

complexes such as carbonate, fluoride, phosphate and oxalate. In neutral and alkaline

solutions (7≤pH≤9), carbonate complexes dominate REE speciation thus enhancing

REE mobilization (Johannesson et al., 1996); conversely, phosphate tends to removal

REE from solution (Johannesson et al., 1995). The complex Ln(CO3)2- (Ln denotes

REE) is strongly enriched in HREE over LREE whereas LnCO3+ (stable in seawater) is

the opposite (Cantrell and Byrne, 1987). At pH 6.5-9.5, LnCO3

+ predominates whereas

at pH ≥9.5 the Ln(CO3)2- complex is favoured (Wood, 1990). When the pH is 2-6.5,

REE occur mainly as simple ions and sulphate complexes (Wood, 1990). However,

lack of systematic experimental data for: (i) all REE across a wide range of pH, (ii) the

presence of multi-complex phases, (iii) the limited investigations of the complexation

of Ce4+

and Eu2+

and stability constants for REE phosphate complexes, and (iv)

conflicting data about the thermodynamics of REE hydrolysis (Wood, 1990), all

restrain further understanding of the mobilization of REE during water/rock interaction

under natural supergene weathering conditions.


18

2.5.3 Fractionation of REE during weathering

Although fractionation of REE may occur during weathering processes (Koppi et al.,

1996), exactly how weathering intensity exerts effects on REE fractionation, and the

precise sequence of events and fate of REE during weathering is not fully understood.

This information is necessary in order to use REE as tracers for supergene weathering,

pedogenesis and sedimentation processes.

Differences in weathering rates and the formation of more element-specific secondary

minerals (Tyler, 2004) during weathering may also result in REE fractionation in

weathering profiles. However, diverging views exist regarding at which stage the

movement and differentiation of REE start during rock weathering and soil formation

(Zhang et al., 2007). Banfield and Eggleton (1989), Price et al. (1991) and Sharma and

Rajamani (2000) illustrated that REE contents change dramatically during the initial

stages of weathering, while Middelburg et al. (1988) and Duzgoren-Aydin and Aydin

(2009) proposed that the migration and differentiation of REE occurs at advanced

stages of weathering. In early and intermediate weathering, mineral abundances may

control REE abundances (Banfield and Eggleton, 1989; Nesbitt, 1979) but for

advanced weathering in laterite, the relatively greater mobility of the HREE appears to

be more significant (Braun et al., 1993; Braun et al., 1990; Brown et al., 2003).

The lower mobility of LREE compared to HREE commonly results in a significant

relative enrichment of LREE and depletion of HREE in weathering products after

extensive weathering (Braun et al., 1993; Braun et al., 1990; Compton et al., 2003;

Koppi et al., 1996; Nesbitt, 1979), however higher mobility of LREE over HREE have

also been reported in intensely weathered environments (e.g. Beyala et al., 2009; Braun

et al., 1990; Ndjigui et al., 2009; Nesbitt and Markovics, 1997). Light REE are known

to be hydrolysed more easily than HREE, whereas HREE preferentially form more

stable inorganic complexes than LREE, particularly with carbonate, fluoride,

hydroxide or sulphate anions in alkaline solutions (Åström and Corin, 2003) and are

more likely to desorb from clay minerals than LREE. This may explain why HREE are

more prone to mobilization than LREE and translocation to the lower parts of regolith


19

profiles (Aubert et al., 2004; Aubert et al., 2001; Cantrell and Byrne, 1987; Ma et al.,

2007). However, this process is believed to be controlled by pH and the type of REE

complexation (Cantrell and Byrne, 1987; Johannesson et al., 1995; Johannesson et al.,

1996; Wood, 1990).

The relationship between weathering intensity and the mobilization and fractionation

of REE is also unclear. Some previous studies have shown that there is no significant

correlation between the degree of REE fractionation and any of the following;

chemical weathering intensity (by CIA) (Caspari et al., 2006), physical weathering

intensity (by particle size fraction index) (Caspari et al., 2006) and rock weathering

and soil formation (Minarik et al., 1998; Zhang et al., 2007). However, the degree of

inter-horizon transport of REE has been proposed to have great potential to become an

index of weathering intensity (Aide and Christine-Aide, 2012). Then this poses the

question that what is the association between the mobility and fractionation of REE

and the weathering intensity? What factors control REE fractionation during

weathering? And what mechanisms are involved during initial and advanced

weathering? An important research issue, the mobilization and fractionation

mechanisms of REE during weathering, is therefore discussed in detail in this thesis.

2.5.4 Mineral transformation of REE during weathering

The transformation of REE minerals during weathering processes depends on their

susceptibility and the weathering conditions. Rare earth elements-bearing minerals can

be subdivided into three groups according to weathering susceptibility: (i) strongly

resistant to weathering e.g. xenotime and zircon; (ii) moderately resistant to weathering,

e.g. monazite; and (iii) weakly resistant to weathering, e.g. allanite and the

fluorocarbonates (bastnäsite and parisite) (Bao and Zhao, 2008). During initial and

moderate stages of weathering, the weathering-susceptible minerals: feldspar, biotite,

allanite, epidote, apatite etc., preferentially dissolve in low pH solutions. The REE

released during this process may: (i) be lost from the weathering profile via transport in

solution (Condie et al., 1995), (ii) form secondary minerals (Braun and Pagel, 1994),

(iii) be incorporated into or adsorbed onto clay minerals (Vos et al., 2006), and (iv) be


20

adsorbed onto Fe/Mn oxyhydroxides (Aide and Smith-Aide, 2003; Aide et al., 1999).

Simultaneously, weathering-resistant heavy minerals such as zircon and rutile are

retained as residuum in weathered regolith, and the REE (especially HREE) included

in these heavy minerals are not expected to be highly mobilized during pedogenesis,

except under intense or extreme weathering. Under extreme weathering conditions,

both secondary and weathering-resistant heavy minerals may partially or completely

alter (Braun et al., 1993; Taunton et al., 2000a; Taunton et al., 2000b). In this process

quartz, being relatively resistant to weathering, acts as a diluent in the regolith (Hardy

and Cornu, 2006). Incorporation or reformation of secondary florencite, rhabdophane

and/or churchite (Braun and Pagel, 1994) is believed to be the main pathway for LREE

retention in weathered regolith (Nedachi et al., 2005).

2.6 Geochemical pathways of REE during lateritization

2.6.1 Definition of lateritic profiles

The study of regolith spans many disciplines of the Earth Sciences (Anand and Paine,

2002), and thus many definitions are confusing and may lead to misunderstanding.

Therefore, it is necessary to define terms from the perspective of regolith geochemistry.

In this thesis the term ‘laterite’ refers to Fe-rich weathering profiles which have

undergone intense supergene weathering (Anand and Butt, 2000; Anand and Butt,

2003; Anand and Butt, 2010; Anand and Paine, 2002); and the other key terminology

for deeply weathered profiles used in this thesis is summarised below, following the

definitions published by Anand and Paine (2002).

A typical laterite profile commonly includes saprolite, mottled clay zone, ferruginous

zone and surface soil. Saprolite refers to nearly isovolumetrically weathered bedrock

retaining the fabric and structure of the parent rock, pseudomorphically replacing the

primary minerals. The mottled zone has macroscopic segregations of subdominant

colour that differ from the surrounding matrix and mottled clay zone is dominantly

composed by secondary clay minerals. The ferruginous zone is composed

predominantly of secondary oxides and oxyhydroxides of Fe (goethite, hematite,


21

maghemite), hydroxides of aluminum (e.g. gibbsite, boehmite) and kaolinite, with or

without quartz. The upper lateritic profile consists of ferruginous mottled zone,

ferruginous duricrust and loose gravel. Ferruginous mottled zone has a goethite rich

halo with sharp or diffuse boundaries, whereas ferruginous duricrust is a cemented

hard layer composed of various Al-Fe secondary segregations that originated from

underlying parent rock. Distinct from the ferruginous duricrust, ferricrete is a product

of cementation and conglomeration of surficial sands and gravel by Fe oxides, where

no genetic relationship between the Fe and the underlying mottled and saprolite zones

is inferred (Anand and Butt, 2010). In this thesis, ‘duricrust’ is used in short for

‘ferruginous duricrust’ since the ferruginous duricrust is the only type of duricrust

present in the lateritic profiles studied. Regolith is used as a collective term for the

weathered and transported materials covering fresh rock, which have been formed by

various geochemical processes e.g. weathering, erosion, transport and/or deposition of

older material.

Laterite represents one of the most common superficial formations in the tropics,

covering approximately 30% of the continents (Dequincey et al., 2002), the formation

of which can extend of 10’s of Ma (Dequincey et al., 2006). Lateritic profiles are

commonly thought to have formed in tropical climates with relatively high

temperatures and seasonal rainfall. However, laterite can also form in wet, cool to cold

climates given sufficient time (Gozzard, 2007). In contrast to common pedogenesis,

during lateritic weathering, lateritic regolith is intensely depleted in base cations and

enriched in iron either in some layers, or throughout the profile, commonly forming

ferruginous zones. In Western Australia, seasonally high rainfall and alternating

arid/humid weathering conditions may have further enhanced the lateritization process

and resulted in a widespread distribution of lateritic profiles. Therefore, studying of

abundance and redistribution of REE in lateritic regolith is significant for

understanding the behaviour and mode of occurrence of REE under intense weathering

conditions and advanced lateritization.


22

2.6.2 A typical lateritic profile

Zoning in lateritic profiles occurs at variable depths (Bourman, 1993), but a typical

lateritic profile would have: fresh bedrock, saprock, saprolite, a mottled clay zone and

a ferruginous zone (Anand and Paine, 2002). In real situations, however, one or more

zones may be missing from the profile. The saprock and saprolite form the lower part

of the regolith and retain the primary mineral constituents of the bedrock. At this depth,

weathering has been less intense and is nearly isovolumetric, whereas the mottled clay

and ferruginous zones comprising the upper part of the profile have been subjected to

stronger, non-isovolumetric weathering, leaching, cementation and soil-forming

processes, and probably precipitation and erosion (Anand and Butt, 2010; Anand and

Paine, 2002). Thin, depleted (commonly quartz-rich) topsoil may often be present

above the ferruginous zone. In a ferruginous zone, nodules and pisoliths may be

present and they are distinguished by their morphology: nodules are irregular, with

re-entrant surfaces, whereas pisoliths are ellipsoidal or spherical. As the sphericity of

nodules increases they merge with pisoliths (Anand and Butt, 2010) and nodules can

also be formed from cementation of one or more pisoliths. In this study, both nodules

and pisoliths are presented in the Jarrahdale regolith profile (JG) studied, and ‘iron

nodules’ is used to denote both types to simplify the terminology.

2.6.3 Lateritization

The geochemical processes for formation of lateritic profiles/landscape are collectively

called lateritization. Intense weathering is an important process during lateritization,

leading to disaggregation, breakdown of original silicate minerals, dissolution of

primary minerals, and leaching of base cations. At the onset of weathering, any

carbonates are dissolved, sulfides are oxidised, easily weathered Fe-Mg silicates are

hydrolysed, and then most of the readily weathered minerals such as feldspar alter to

kaolinite. Resistant minerals, such as quartz and zircon, remain relatively unaltered.

The intense leaching of base cations and formation of secondary clay minerals leads to

the formation of mottled zone (sometimes known as the ‘pallid zone’ if depleted in Fe).

Secondary clay mineral formation by chemical weathering of the primary minerals is


23

termed ‘kaolinization’ and is one of the fundamental processes during lateritization.

With more intense weathering, silicates are increasingly leached out of the profile, and

more secondary oxides, Fe oxyhydroxides, Al hydroxides, and kaolinite are formed.

Continuous dissolution, precipitation, cementation, and erosion results in formation of

the ferruginous zone (Anand and Paine, 2002); note that enrichment of Fe oxides and

oxyhydroxides during lateritic weathering is referred as ferruginization. This process

leads to accumulation of crystalline Fe oxyhydroxides (e.g. goethite), Fe oxides (e.g.

hematite and maghemite), Al hydroxides (e.g. gibbsite), and Al oxyhydroxides

(e.g. boehmite), in the intensely weathered regolith. Though the geochemistry and

genesis of lateritization have been widely investigated (Bourman, 1993; Brimhall et al.,

1991; Schellmann, 1994), the translocation and fractionation of REE during

lateritization in various secondary mineral phases is still not fully understood (Feng,

2011; Ma et al., 2007).

2.6.4 Geochemical behaviour of REE during lateritization

Quantitative understanding of the nature of the migration-fixation and fractionation

mechanisms of REE caused by sorption of clay or Fe oxyhydroxide is currently

inadequate. Such information is important for interpreting the behaviour of REE as

potential clues tracing the processes of lateritization and weathering.

Mottled clay formed during kaolinization is believed to act as a potential reservoir of

REE in weathered lateritic profiles because of adsorption of REE onto the clay surface

(Laveuf and Cornu, 2009). This is an important secondary REE enrichment process

(Bao and Zhao, 2008; Ohlander et al., 1996) during weathering, however, the main

enrichment of REE takes place during precipitation of secondary REE bearing

minerals (Braun and Pagel, 1994; Braun et al., 1993). Adsorption of REE by clay is

controlled by the nature of the clay minerals, pH, ionic strength, the presence of

additional ligands such as carbonate or organic complexes, surface coverage, and

effects specific to the individual REE (Coppin et al., 2002; Fendorf and Fendorf, 1996;

Koeppenkastrop and Decarlo, 1992; Koeppenkastrop and Decarlo, 1993; Laveuf and


24

Cornu, 2009; Piasecki and Sverjensky, 2008; Takahashi et al., 1999). As well as these

controls, differences in clay mineralogy can affect fractionation of REE (Coppin et al.,

2002; Laveuf and Cornu, 2009), potentially explaining the apparently contradictory

signatures of REE adsorbed by clay minerals. Usually, REE adsorption increases with

increasing pH (Coppin et al., 2002), and in high ionic strength solutions HREE are

more sorbed than LREE (Coppin et al., 2002). Clay minerals have a strong affinity for

all REE except Ce (Duzgoren-Aydin and Aydin, 2009). As the most important clay

mineral in lateritic regolith, kaolinite has considerably variable REE concentrations

(Laveuf and Cornu, 2009) and the fractionation of REE by kaolinite sorption is still not

fully understood. The concentrations of REE in different horizons of lateritic regolith

profiles are listed in the Table 2.3.

The upper ferruginous zones in weathered lateritic profiles are commonly depleted in

all REE except Ce, although it has been reported that Fe oxides have high surface areas

rendering them very efficient sinks for heavy metals (Nedel et al., 2010; Singh and

Gilkes, 1992). Iron oxides are known to contain REE, but the concentration does not

correlate with Fe content (Laveuf and Cornu, 2009). It is rare for REE to substitute for

Fe in the lattice of Fe oxides at ambient temperatures and pressures; however,

transformation from ferrihydrite to goethite with Lu3+

and Eu3+

substitution and

incorporation was reported at 70 ºC and pH 13 (Dardenne et al., 2003).

Scavenging of REE by Fe oxides and oxyhydroxides is believed to be mainly affected

by surface complexing, which is strongly pH-dependent (Bau, 1999; Marmier et al.,

1999; Marmier and Fromage, 1999; Piasecki and Sverjensky, 2008), hence sorption of

REE usually increases with increasing pH within a range of 5-7 (Marmier et al., 1999;

Marmier et al., 1997; Marmier and Fromage, 1999; Piasecki and Sverjensky, 2008).

Surface complexing and pH do not affect Y as much as REE (Bau, 1999).

Fractionation of REE by Fe oxyhydroxides shows MREE enrichment at pH>5 in low

salinity solutions when other strong complex ligands are absent (Bau, 1999), indicating

La, Gd, Y and possibly Lu preferentially remain in the solution rather than being

surface-complexed onto Fe oxyhydroxides (Bau, 1999).


25

However, the fractionation between LREE, MREE and HREE in Fe oxides is subject

to debate (Laveuf and Cornu, 2009) and varied fractionation with enrichment of LREE

(Koeppenkastrop and Decarlo, 1993), MREE (Bau, 1999; Land et al., 1999) or HREE

(Elderfield and Greaves, 1981; Marker and Deoliveira, 1994) have been observed. The

differences in REE fractionation induced by Fe oxides probably arise from the

presence of various proportions of different types of Fe oxides and the presence of

other complex ligands (Laveuf and Cornu, 2009). For example, in solutions at

pH 4.0-7.1 with carbonate present, REE sorption by amorphous Fe oxyhydroxides

initially increases with increasing carbonate concentration and then decreases; this

effect was more pronounced for HREE than LREE (Quinn et al., 2006). In addition,

REE are fractionated during adsorption by Fe oxyhydroxides when humate complexes

are present, resulting in MREE enrichment (pH 5.2) rather than the non-preferential

adsorption by Fe oxyhydroxides or humate complexes (Davranche et al., 2004).

Generally, REE contents in amorphous Fe oxyhydroxides are higher than in crystalline

Fe oxides (Land et al., 1999; Laveuf and Cornu, 2009), and operationally defined

amorphous and crystalline Fe phases displaying enrichment of MREE were reported

by Land et al., (1999).

2.6.5 Anomalies of Ce in lateritic regolith

Cerium is one of the ‘unusual’ REE because it can occur in nature as Ce3+

like the

majority of lanthanides or as Ce4+

in oxidizing conditions. Cerium also has very low

elemental mobility, due mainly to the stability and low solubility of CePO4 and CeO2.

In lateritic profiles, topsoil and clay zones usually have either slight positive or no

apparent Ce anomalies, ferruginous zones commonly have a positive Ce anomaly, and

saprolite may show a negative or positive Ce anomaly (Angelica and Dacosta, 1993;

Braun et al., 1993; Braun et al., 1990; Braun et al., 1998; Ndjigui et al., 2009). In the

ferruginous zone, Fe acts as a redox sensitive element, and strong redox mediated

associations between the oxyhydroxide phases are expected within weathered profiles.

During the primary redox change in lateritic weathering, soluble Ce3+

is released by


26

chemical weathering from REE-bearing minerals and is oxidized to Ce4+

where pH

ranges from 2.5 to 7.5, and Eh from -0.4 to 1.1V; precipitation as CeO2 in the

ferruginous zones (Angelica and Dacosta, 1993; Braun et al., 1990; Braun et al., 1998;

Takahashi et al., 2000) is most likely in lateritic profiles, however, adsorption by

Fe/Mn oxyhydroxides also has the potential to produce a Ce enrichment (Ndjigui et al.,

2008; Nedel et al., 2010; Ohta et al., 2009; Quinn et al., 2006). The variability of

observed Ce anomalies in saprolite may relate to a change in redox condition induced

by fluctuation of groundwater or movement of the weathering front (Braun et al., 1990;

Ndjigui et al., 2009).

2.6.6 Anomalies of Eu in lateritic regolith

Although Eu anomalies are variable in lateritic weathered profiles (Table 2.3), they are

less studied than Ce anomalies. Negative Eu anomalies in the saprolite zone, or even

throughout the profile (e.g. Braun et al., 1993; Braun et al., 1998), may be caused by

breakdown of plagioclase (Panahi et al., 2000), sphene and allanite (Condie et al.,

1995), or the tetrad effect1 (Feng, 2011). Although the tetrad effect in REE patterns

has been reported widely in different geological samples (e.g. Liu and Zhang, 2005;

Monecke et al., 2002; Takahashi et al., 2002), it is still under debate (McLennan, 1994).

Positive Eu anomalies also exist in lateritic regolith (e.g. Braun et al., 1998; Ndjigui et

al., 2009), and they may relate to the type of parent rock and the redox conditions

during weathering (Ndjigui et al., 2009).

2.7 Summary

Although the geochemical behaviour of REE in supergene settings has been widely

investigated since the 1980s, compared with the studies of REE under high

temperature and high pressure settings it is far less studied, especially during intense

lateritic weathering. Many issues remain unresolved or not fully understood, such as

1tetrad effect: a split of chondrite-normalized REE patterns into four rounded segments which either are convex or

concave and formed M-shaped and W-shaped lanthanide pattern Masuda, A., Kawakami, O., Dohmoto, Y., Takenaka, T., 1987. Lanthanide tetrad effects in nature: two mutually opposite types, W and M. Geochemical Journal 21(3), 119-124.


27

migration-fixation mechanisms and mode of occurrence of REE in lateritic regolith,

the impact of Fe oxyhydroxides on translocation and fractionation of REE during

lateritization, the distribution of REE into different particle size solid phases and the

influence of weathering intensity on mobilization and fractionation of REE etc.. In

addition, many results are controversial and have not been interpreted unambiguously

yet; for example, the preferential mobilization of LREE or HREE during weathering,

and at which stage of weathering mobilization and fractionation of REE starts.

Therefore, the study of the geochemical behaviour and fractionation of REE under

intensely lateritic weathering is important and worthy of further research. This thesis

will improve the understanding of the mode of occurrence, fractionation mechanism

and geochemical behaviour of REE during weathering and lateritization in supergene

settings.

28

Table 2.1 Summary of REE in common minerals in granitoid rocks

Mineral* Formula ΣREE LaN/YbN Ce

* Eu

*

primary minerals

Quartz SiO2 None

Plagioclase (Na, Ca)(Si, Al)4O8 3.0-143 ppm 17.0-1184 0.71-1.16 0.43-32.9

K-feldspar e.g. KAlSi3O8 1.3-43.2 ppm 6.6-242 0.74-1.01 0.41-33.8

Biotite K(Mg,Fe2+

)3[AlSi3O10(OH,F)2] 0.03-1.75 ppm 15.2-17.3 0.11-1.12 3.51-6.82

Muscovite KAl2[AlSi3O10(OH,F)2] 0.08-3.78 ppm 0.96-1.38 0.40-1.04 0-10.7

Amphibole Ca2(Mg,Fe2+

)5Si8O22(OH)2 22.6-203 ppm 1.3-4.1 1.08-1.31 0.42-1.16

Clinopyroxene e.g. (Mn2+

, Mg)2Si2O6 3.6-24.8 ppm 0.6-2.8 0.97-1.15 0.91-1.24

Orthopyroxene e.g. (Mg, Fe2+

)3Al4BeSi3O16 0.05-0.14 ppm 1.36

Garnet e.g. Ca3Al2Si3O12 46.8-268 ppm 0-0.8 0-1.07 0-0.8

Cordierite Mg2Al4Si5O18 0.2-5.2 ppm 1.3-4.7 0.12-1.36 0.78-0.98

Tourmaline e.g. NaAl3Al6(BO3)3(Si6O18)(OH)4 0.7-25.1 ppm 0.8-89.3 0.26-1.71 0-0.42

Minerals with REE non as essential structural ion

Zircon ZrSiO4 76.3ppm-1.28% 0-0.12 0.14-1.20 0-0.88

Apatite Ca5(PO4)3(OH,F,Cl) 0.13%-1.29% 0.10-84.0 0.14-71.1 0.03-1.16

Epidote Ca2(Fe3+

, Al)3(SiO4)3(OH) 25.5-271 ppm 5.8-116 0.27-1.03 1.37-8.31

Sphene(titanite) CaTi(SiO4)(O,OH,F) 0.29%-2.73% 2.9-8.3 0.23-1.04 0.55-1.08

Th-orthosilicate ThSiO4 0.5%-23.0% 0-6.7 5.75 0-0.29

Uraninite UO2 0.75%-0.99% 0.05-0.12 0.41-0.68 0-0.29

Fluorite CaF2 up to 14.1% Ce

Rutile TiO2 tens of ppm

29

Mineral* Formula ΣREE LaN/YbN Ce

* Eu

*

Minerals with REE as essential structural ions

Monazite (La,Ce,Nd)PO4 1.62%-52.7% 5.9-526 0.35-0.87 0-0.73

Allanite (REE,Ca)2(Al,Fe3+

)3(SiO4)3(OH) 13.8%-22.8% 0-327 0.11-1.61 0-0.17

Xenotime (HREE)PO4 0.8%-17.4% 0-0.17 0.13-0.99

Fluocerite (Ce,La)F3 up to 65% predominant Ce

Bastnasite (Ce,La)(CO3)F over 52% predominant LREE

Synchysite (Ce,La) Ca(CO3)2F predominant LREE

Parisite (Ce,La)2Ca(CO3)3F2 predominant LREE

Cerianite CeO2 predominant Ce

Zirkelite (Zr,Ca,Ti,Fe,Mg,REE,U,Th)3O5 ~1.9% HREE predominant HREE

Florencite CeAl3(PO4)2(OH)6 up to 26% predominant LREE

Rhabdophane (Ce,La)PO4∙H2O up to 52% predominant LREE

*There are more than 200 REE-bearing minerals; this table only includes the most common and most important minerals in granitoid rocks.

The data were mainly re-calculated and summarized based on the study by Bea (1996), and some missing data were referred to (Henderson, 1984);

Ce*=(Ce/CeCN)/[(La/LaCN)

0.5×(Pr/PrCN)

0.5];Eu

*=(Eu/EuCN)/[(Sm/SmCN)

0.5×(Gd/GdCN)

0.5]; CN refers to the average chondrite (Anders and Grevesse, 1989).

30

Table 2.2 Concentrations of REE in different types of parent rock

Parent

Rock Location

Element concentrations (ppm) Reference

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu ΣREE

Granite France 16.3 37.5 4.66 17.3 4.14 0.35 3.09 0.49 2.45 0.37 0.82 0.12 0.69 0.09 88.4 (Aubert et al., 2001)

Granite Canada 47.7 94 9.73 33.3 4.88 0.93 3.12 0.42 1.97 0.35 1.08 0.17 1.12 0.17 199 (Panahi et al., 2000)

Syenite Cameroon 69.5 140 65.8 10.7 2.93 6.54 2.66 1.13 0.9 0.13 300 (Braun et al., 1993)

Gneiss Cameroon 11.5 23.0 2.61 9.96 1.90 1.19 1.72 0.33 2.46 0.65 1.98 0.35 2.59 0.39 60.7 (Braun et al., 1998)

Serpentinite Cameroon 0.14 0.34 0.05 0.27 0.07 0.01 0.09 0.02 0.11 0.02 0.08 0.02 0.10 0.02 1.33 (Ndjigui et al., 2009)

Dolomite China 2.78 1.91 0.73 4.09 1.61 0.62 2.75 0.43 1.58 0.21 0.43 0.07 0.39 0.06 17.7 (Ji et al., 2004)

Carbonatite Australia 292 621 287 56.5 16.1 34.5 5.40 4.15 5.55 0.62 1323 (Lottermoser, 1990)

Basalt China 16.6 30.7 3.73 17.2 4.44 1.48 4.91 0.75 4.12 0.79 1.99 0.25 1.53 0.22 88.7 (Ma et al., 2007)

Granodiorite Australia 25.0 57.8 25.4 6.02 1.42 5.73 0.85 1.01 2.89 0.48 127 (Nesbitt, 1979)

Aries

kimberlite Australia 268 433 9.94 2.09 1.04 0.15 714 (Singh and Cornelius, 2006)

chlorite

schists Cameroon 29.9 62.3 7.23 27.6 5.56 1.07 5.03 0.83 5.28 1.10 3.38 0.49 3.24 0.47 154 (Beyala et al., 2009)

PAAS1 38 80 8.9 32 5.60 1.10 4.7 0.77 4.4 1.0 2.9 0.50 2.8 0.50 184 (Nance and Taylor, 1976)

NASC2 32 73 7.9 33 5.7 1.24 5.2 0.85 5.8 1.04 3.4 0.50 3.1 0.48 174 (Haskin and Paster, 1979)

UCC3 30 64 7.1 26 4.5 0.88 3.8 0.64 3.5 0.80 2.3 0.33 2.2 0.32 146 (McLennan et al., 1980)

1PAAS: Post-Arcbean average Australian shale;

2NASC: North American shale composite (post-Archean);

3UCC: Post-Archean upper continental crust.

31

Table 2.3 Concentrations of REE in different horizons of lateritic regolith profiles

PR1 Type PR saprolite saprolite Mottled clay Ferruginous zone

2 Duricrust

3 Reference

lower upper matrix nodule matrix gravel

Serpentinite ΣREE4 1.33 270 271 162 140 (Ndjigui et al., 2009)

(La/Sm)PR 2.12 1.67 2.92 2.49

(La/Yb)PR 0.79 0.49 0.80 0.60

5Ce

* 0.02 1.73 1.57 3.29

Eu* 2.28 2.37 2.26 2.32

Serpentinite ΣREE 1.33 105 110 437 240-742 170 82.1 (Ndjigui et al., 2009)

(La/Sm)PR 2.50 1.42 1.96 1.11-1.36 2.22 1.54

(La/Yb)PR 0.71 0.31 0.44 0.26-0.29 0.47 0.32

Ce* 0.46 0.68 11.3 9.76-24.0 2.46 1.26

Eu* 2.09 2.21 2.3 2.29-2.41 2.14 2.19

Granodioritic

gneiss ΣREE 39.4 21.3-47.4 104-483 (Tripathi and Rajamani, 2007)

(La/Sm)PR 0.58-0.86 0.58-1.04

(La/Yb)PR 0.09-0.16 0.13-0.32

6Ce

* 0.85-1.63 1.21-4.16

Eu* 0.56-0.77 0.30-0.79

Chlorite ΣREE 153.4 194.0 131.1 32.72-100.1(w)

7 (Beyala et al., 2009)

(La/Sm)PR

1.98 1.16 1.32-1.51

(La/Yb)PR

1.38 0.78 0.65-1.04

32

PR1 Type PR saprolite saprolite Mottled clay Ferruginous zone

2 Duricrust

3 Reference

lower upper matrix nodule matrix gravel

Syenite ΣREE 300.3 475.8 445.2 247.7-3035 253.1 193.3(w) 301.5(w) (Braun et al., 1993)

(La/Sm)PR

1.27 1.27 0.94-1.61 1.51 1.46 1.51

(La/Yb)PR

2.37 2.55 1.10-2.26 5.17 3.98 5.98

6Ce

* 1.07 1.28 0.98-19.7 1.09 1.14 1.07

Eu* 0.95 0.90 0.57-0.93 0.84 0.78 0.84

Granodiorite ΣREE 118.4 122.0 100.2-255.0 90.8-165.3(w) 446.5(w) (Dequincey et al., 2002)

(La/Sm)PR

0.91 0.57-1.12 1.28-1.33 1.27(w)

(La/Yb)PR

1.00 0.41-1.24 0.79-0.86 0.91

Ce* 0.94 0.21-1.95 0.73-1.05 6.17

Eu* 1.03 0.86-1.02 0.82-0.91 0.78

1PR: parent rock;

2Ferruginous zone is defined by the dominant composition of secondary Fe oxides and oxyhydroxides, Al hydroxides and kaolinite; and the

3duricrust refers to a hard cemented

layer consisting secondary segregations. In this thesis all ferruginous zones consist of duricrust; however, in the references listed in the table, duricrust has been separated out to

emphasize the concentration variations of REE in different zones.

4ΣREE is the total concentrations of REE;

5Ce

*=(Ce/CePR)/[(La/LaPR)

0.5×(Pr/PrPR)

0.5];Eu

*=(Eu/EuPR)/[(Sm/SmPR)

0.5×(Gd/GdPR)

0.5];

6Data of Pr is missing, so Nd is used when calculating Ce

*;

7w refers to whole rock analysis.

33

3 Description of the study areas

3.1 General geology and climate

The study areas were located in the south-western part of Western Australia and lie

within the Darling Range, slightly east of the Darling Fault and Perth Basin (Figure

3.1). The geological history of the Darling Range can be traced back at least 2600

million years and possibly even further (Gozzard, 2007). Since Paleogene, deep and

intense weathering of exposed rocks of the Darling Plateaus, resulted in a widespread

cover of lateritic materials, and this weathering has continued until geologically recent

times (Gozzard, 2007). This area is part of the vast Yilgarn Craton - an ancient region

of varied rock types that occupies much of the south-western part of Western Australia

(Anand et al., 2006; Anand and Paine, 2002; Gozzard, 2007). Large volumes of

granitoids intruded the metamorphic rocks and other rocks of the Yilgarn Craton

between 2700 and 2600 million years ago and dolerite dykes intrude the granitoids

during development of the Darling Fault in the Mesoproterozoic and Neoproterozoic

(Gozzard, 2007).

The area currently has a Mediterranean climate, with a cool wet season from May to

September and a warm, dry season from November to March, with transition periods

in April and October. Average annual rainfall was ca. 1239.5 mm in the Dwellingup

(Darling Range) from 1934 to 2011 and rainfall mostly occurs in winter (Bureau of

Meterology, 2012).

The vegetation of the areas studied shows marked regional changes based largely on

climate, with local variations of geology, soils, topography and drainage. The Darling

Range in the high-rainfall area currently has open eucalypt forests of jarrah

(Eucalyptus marginata) and marri (Eucalyptus calophylla) (Anand and Paine, 2002).

3.2 Sampling and profile description

The profiles studied are Fe-rich lateritic weathered profiles near outcrops of granitoids


34

and dolerites. The dolerites are sub-vertical intrusions that cross cut the metamorphic

fabric of the granitoids and thus called dolerite dykes here. Samples of outcropping

fresh granitoid and dolerite were sampled ca. 5 kg separately at each study site.

Different zones of regolith in each profile were identified based on the physical

properties (e.g. texture, colour, coarse fragment content) and chemical properties (e.g.

mineralogy, Al/Fe concentrations, and organic matter contents). One to two ten-cm

blocks of bulk undisturbed regolith samples per horizon were collected at different

depth and sealed in plastic bags or boxes for transport to the laboratory.

Great Eastern Profile: The first intensely weathered lateritic profile investigated was

adjacent to Great Eastern Highway, Western Australia (31°22'30.95"S, 118°41'27"E),

with very good bedrock (granitoid/dolerite dyke) exposure (Figure 3.1&Figure 3.2).

Sample identities from this location are prefixed with GE. Regolith samples from

different horizons, the parent granitoid and the intrusive dolerite dyke were collected

on 3rd

April, 2009. One outcropped coarse grained granitoid sample (GEPR1B) was

crossed by a late-stage sub-horizontal pegmatite vein (GEPR1A), thus these two

subunits were analysed separately. The profile was ca.12 m deep, including saprolite,

mottled clay, ferruginous duricrust, and A horizon regolith. The saprolite formed from

weathered bedrock with horizons above showing progressive loss of rock fabric

upwards as porosity and the proportion of clay increases. The mottled clay zone was a

pale white kaolinite-rich zone, ca. 3 m thick, with distinct upper boundary with the

ferruginous zone. The ferruginous zone was composed of loose lateritic ferruginous

materials with ferruginous gravel at ca. 7 m depth (GE5), and the cemented

ferruginous duricrust which was a dark red, dense, and hard layer without gravel at ca.

3.5 m depth (GE6).

Mountain Quarry Profiles: The second and third profiles studied were located in

Mountain Quarry (31°54'54" S, 116°3'44" E), on the southern slope of Greenmount

Hill, Western Australia (Figure 3.1). Sample identities are prefixed with MQ. The

second profile (MQ I profile, Figure 3.3) was 3.6 m deep; samples of regolith from

different horizons based on colour and texture were collected on 29th

, May, 2009. The

Chapter Three: Description of the study areas

35

third profile (MQ II profile) was ca. 2 m deep, 10 m away from the second profile,

developed overlying granitoid with stonelines preserved from quartz veins which

imply in-situ weathering below the sub-horizontal component of the stone line at 0.6 m

depth (samples MQ10 to MQ13). Samples of outcropping granitoid and dolerite and

the regolith were also collected at the MQ sampling site.

Jarrahdale profile: The fourth profile was located at the Jarrahdale Railway cutting

(32°17'46"S, 116°5'40"E) at an average elevation of 270 m above sea level in the

Darling Range, 80 km south-east of Perth, Western Australia (Figure 3.1). Sample

identities were prefixed with JG. The lateritic JG profile was ca. 12 m deep overlies

metamorphic basement consisting of granitoid intruded by a dark-coloured dolerite

dyke. The location of intrusive dykes in the granitoids can be mapped from the

overlying duricrust and the profiles developed on granitoid and dolerite are distinctly

different (Gozzard, 2007). The JG regolith is developed overlying granitoids and

divided into seven zones: parent granitoid, saprolite, mottled clay, ferruginous mottled

zone, ferruginous duricrust, upper ferruginous zone and the A horizon. The mottled

clay is pale white kaolinite-rich, consisting of a lower zone (JG2) at 8.6 m depth and an

upper zone (JG3) at 6.5 m depth. The ferruginous duricrust (3 m depth) is gibbsite and

goethite rich. The upper ferruginous zone (JG6, 1.5 m depth) is rich of red iron nodules.

In contrast, the horizon A regolith (JG7-10, <1 m depth) is gravely sandy soil rich of

dark brown to black loose nodules. The sampling of the profile was conducted on the

6th

August 2009. Each zone was identified based on different properties (e.g. texture,

colour, coarse fragment content), sampled at the depth given above in a

ca. 10×10×10 cm cube, put into a sealed plastic box, transported to the laboratory and

air dried. Photographs of each horizon were not taken and thus are not presented in this

thesis.

A number of studies have investigated the geological, morphological and geochemical

characteristics of lateritic bauxite regolith in the Darling Range, and at Jarrahdale the

regolith is widely accepted to have undergone in-situ intense weathering and

lateritization (Anand and Butt, 2010; Anand et al., 1991; Anand and Paine, 2002;


36

Brimhall et al., 1992; Brimhall et al., 1994; Gozzard, 2007; Kew and Gilkes, 2007;

Sadleir and Gilkes, 1976). Note that both fresh and weathered euhedral zircons from

lateritic bauxite profiles at Jarrahdale have been dated at ca. 2650 Ma (Brimhall et al.,

1994) indicating upward lithological continuity of the parent meta-granitoid through

the bauxite profile. Rounded zircon, ilmenite, and rutile have been found in the

surficial meter-depth regolith and are proposed to have been transported by wind and

predominantly from a different, much younger source (ca. 700-1150 Ma) (Brimhall et

al., 1992; Brimhall et al., 1994; Brimhall et al., 1988).

37

Figure 3.1 Sampling sites (a, labelled as box) and sketches of the profiles sampled (b). On the map (a) the dashed line labelled Darling Fault represents

the western margin of the Darling Range. In the sketch of regolith profile (b) ‘m’ denotes matrix and ‘g’ denotes gravel.

38

GE1, 12.5 m depth, saprolite GE3, 10 m depth, mottled clay GE5, 7 m depth, lower ferruginous

zone

Figure 3.2 Photographs of regolith from selected horizons of the GE profile.

MQ1, 3.6 m depth, C horizon MQ2, 3.3 m depth, lower B horizon MQ4, 2.2 m depth, upper B horizon

Figure 3.3 Photographs of regolith from selected horizons of the MQ I profile

39

4 Redistribution of major elements in lateritic profiles during

intensive weathering in Western Australia

4.1 Abstract

In order to understand the geochemical behaviour of major elements in different solid

phases in laterite and to investigate geochemical pathways of lateritic weathering, the

redistribution of major elements in matrix (<2 mm) and gravel (>2 mm) in four

intensely weathered lateritic profiles (GE, MQ I, MQ II and JG) in Western Australia

was investigated.

The GE and JG regolith samples were highly weathered with chemical indices of

alteration ca. 99%, nearly complete loss of Na, Ca and Si, and enrichment of Fe in the

ferruginous zone. In the GE and JG profiles, Fe mainly occurred as goethite, hematite

and maghemite, while Al mainly occurred as secondary clay minerals (kaolinite) and

gibbsite in ferruginous zone; gravel was more enriched in Al and Fe but more depleted

in Si than the matrix, which was consistent with gravel having higher weathering

intensity and degree of lateritization than the matrix.

The regolith samples from both MQ profiles were less weathered than the GE and JG

profiles, lack of gibbsite and hematite, and showed weak lateritization. The chemical

indices of alteration ranged from 55% to 92% in both MQ profiles and gravel had

higher concentrations of Si than matrix. The presence of a pedogenic discontinuity in

both MQ profiles was identified from the molar ratio Na/K, concentration ratios

Al2O3/Fe2O3 and Ti/Zr, implying that mass movement had occurred in the upper part of

both profiles during weathering.

Significant depletion of base cations and Si, coupled with enrichment of Fe and Al,

reveal that intense leaching of cations, kaolinization, desilication and ferruginization

took place in lateritic regolith during weathering and lateritization.

4.2 Key words

Major elements; laterite; weathering; mass balance; Western Australia;


40

4.3 Introduction

Chemical weathering is one of the critical processes in the geochemical cycling of

elements and translocation of elements from crust to sediments. During early

pedogenesis, the chemical composition of a soil will be strongly controlled by the

composition of geological parent materials, though this influence diminishes in

importance with time (Schaetzl and Anderson, 2005; Thanachit et al., 2006). The

development of a soil reflects the weathering processes associated with the dynamic

environment in which it has formed. The mobilization and redistribution of elements

during weathering follows various pathways as elements behave differently during

various pedogenic processes, including: dissolution of primary minerals, formation of

secondary minerals, redox processes, transport of material and ion exchange

(Middelburg et al., 1988).

Lateritic regolith represents one of the most common superficial formations in the

tropics, and is commonly diachronous, extending over tens of millions of years

(Dequincey et al., 2006). In contrast to common pedogenesis, during lateritic

weathering, regolith is intensely weathered and enriched in Fe, either in some layers or

throughout the profile, commonly forming a hard cap of ferruginous duricrust or

ferruginous gravel. Though many studies have been conducted on lateritic regolith

profiles (e.g. Beauvais, 1999; Brimhall et al., 1991; Brown et al., 2003; Costa, 1997;

Dequincey et al., 2002; Dequincey et al., 2006; Fernández-Caliani and Cantano, 2010),

the mobilization and redistribution of major elements into different grain size fractions

of lateritic regolith during intense weathering are not yet fully understood. A holistic

understanding of elemental behaviour during weathering and lateritization processes

cannot be achieved solely by determination of total elemental concentrations in bulk

regolith. It is essential to determine the relative elemental concentrations in different

solid phases as well, since partitioning of elements into matrix (<2 mm) or gravel

(>2 mm) may reflect the weathering history and weathering processes involved. The

analysis of the geochemical and mineralogical features of lateritic regolith, including

matrix and gravel, has the potential to improve our understanding of weathering and

lateritization (Beauvais, 1999).

A number of studies have been investigated the geological, geographical, morphological

and geochemical characteristics of lateritic bauxite regolith in the Darling Range

(Anand and Butt, 2010; Anand et al., 1991; Anand and Paine, 2002; Brimhall et al.,

Chapter Four: Redistribution of major elements in lateritic profiles during intensive weathering in Western Australia

41

1992; Brimhall et al., 1988; Kew and Gilkes, 2007; Sadleir and Gilkes, 1976).

Accordingly, this study investigates the mobilization and redistribution of major

elements into different solid phases from lateritic regolith building on previous work.

The objective of this study is to investigate the key geochemical and mineralogical

pathways of lateritic weathering and the relative partitioning of major elements into

matrix and gravel in four intensely weathered lateritic profiles in Western Australia.

This information will be helpful to understand the genesis of ferruginous materials

during lateritization.

4.4 Materials and methods

4.4.1 Analytical methods

In this study, chemical compositions of matrix (<2 mm, represented by suffix ‘m’) and

gravel (>2 mm, represented by suffix ‘g’) were analysed separately. Exceptions were

the duricrust in the GE profile (GE6), a very hard cemented material without

corresponding loose matrix or iron nodules, and the saprolite (JG1) and mottled clay

(JG2&3) in the JG profile, both soft pale materials without gravel. These four samples

were crushed and/or ground to ≤ 200 µm and oven dried at 105 °C overnight prior to

chemical analysis.

Pre-treatment of regolith samples included hand-picking roots/rhizomes and sieving

through 2 mm plastic mesh for separation of gravel (>2 mm) from the matrix (<2 mm)

and weighing each subsample separately. The matrix fraction (<2 mm) was used to

determine the pH, total carbon and particle size distribution (Table 4.1). Soil pH was

determined potentiometrically at 23 °C in the supernatant in a 1:5 suspension of soil:

deionised water and 1:5 suspension of soil: 0.01 M CaCl2 (Rayment and Higginson,

1992). Total carbon was determined by Elementar (Vario Macro, Hanau, Germany).

Subsamples of matrix and gravel were ground to ≤ 200 µm and oven dried at 105 °C

overnight. The bulk raw regolith matrix (< 2 mm) from MQ II profile was separated

further into three size fractions without crushing or grinding: clay (<2µm), silt (2-20 µm)

and sand (> 20 µm) by the sedimentation and wet sieving method (Day, 1965) in order

to understand the behaviour of major elements in different particle size fractions. The

particle size fraction limit recommended by the International Society of Soil Science

(ISSS) has been adopted in Australia (Marshall, 1947; Marshall, 2003; Prescott et al.,

1934). Different particle size fractions were rinsed with MilliQ water three times, oven


42

dried at 105 °C overnight and ground to ≤ 200 µm.

Fusion beads for elemental analyses were made by mixing 0.1 g (to an accuracy of

0.1 mg) of finely ground sample or reference material with 0.7 g 12:22 Norrish flux

(lithium metaborate:lithium tetraborate=12:22) and heating in a muffle furnace at

1050 °C for 40 minutes. Duplicate fusion beads were made on 10% of samples to check

preparation errors. After cooling, the fusion beads were dissolved in 100 mL of 10%

analytical grade HCl. The major elements were determined by inductively coupled

plasma-optical emission spectroscopy (ICP-OES, Perkin-Elmer Optima 7300DV) at the

University of Western Australia. Certified international standard materials, including

stream sediment reference standards STSD-2, STSD-4 (Canada Centre for Mineral and

Energy Technology, CANMET), an in-house standard reference and 12 blanks were

prepared in the same way as the samples and analysed together with samples to check

the accuracy and precision. The variation between the tested and expected values of the

standards was within 5% (Appendix 11.2). The concentrations of major elements in

matrix and gravel are listed in Table 4.2.

Primary minerals in the weathering products were identified by means of random

powder X-ray diffraction (XRD) from 4 to 70 2 using CuKα radiation and a Philips

PW 1830 diffractometer with a diffracted beam graphite crystal monochromator, after

grinding to <63 µm and homogenisation. The proportion of mineral phases (not include

amorphous or poorly-crystalline phases) were identified semi-quantitatively using the

software Traces (GBC Scientific Equipment). All primary mineral phases were

identified manually and cross-checked with the dataset of the International Centre for

Diffraction Data (ICDD). The main-peak area of each primary mineral was measured,

and the mineral proportion was calculated using the main-peak areas of each primary

mineral divided by the sum of main-peak areas of all primary minerals identified by the

Traces software. The clay fraction (<2 µm) was separated by dispersion and

sedimentation, basally oriented on ceramic slides, air-dried, and then scanned at

1° 2/min from 1 to 30° 2/min. For clay mineral identification, the oriented aggregates

were treated with ethylene glycol. Mineral names were abbreviated according to

Whitney and Evans (2010).


43

Table 4.1 Selected physical and chemical properties of matrix fractions (<2mm) of the

profiles studied

Sample1 Depth

2 pH

3 pH

4 clay% silt% sand% TC

5 Description

6

(m) (H2O) (CaCl2) <2µm 2-20µm >20µm

GE

GEA1 0.09 6.05 5.09 1.9 3.2 94.9 4.66 A Horizon (m&g)

GEA2 0.12 6.48 5.03 1.5 2.8 95.7 1.46 A Horizon (m&g)

GEA3 0.23 6.70 5.33 1.9 2.9 95.2 0.80 A Horizon (m&g)

GE6 3.5 6.31 6.12 2.3 7.0 90.7 0.38 Duricrust (g)

GE5 7.0 6.62 5.96 19.3 2.3 78.4 0.21 Lower ferruginous zone

(m&g) GE4 8.4 6.45 5.75 18.9 9.7 71.5 0.04 Mottled clay (m&g)

GE3 10.0 6.79 5.52 22.6 8.1 69.3 0.03 Mottled clay (m&g)

GE2 11.4 6.56 5.26 22.6 7.7 69.7 0.03 Mottled clay (m&g)

GE1 12.5 6.42 4.77 24.7 12.6 62.7 0.07 Saprolite (m&g)

MQ I

MQ9 0.2 6.24 5.19 19.8 6.9 73.2 2.06 A Horizon (m&g)

MQ8 0.5 6.37 5.13 16.1 0.8 83.1 1.17 A Horizon (m&g)

MQ7 0.7 6.27 5.02 35.7 1.9 62.4 0.39 A Horizon (m&g)

MQ6 0.9 5.94 4.91 68.8 0.7 30.5 0.46 B Horizon (m&g)

MQ5 1.1 5.81 5.05 71.9 0.7 27.4 0.39 B Horizon (m&g)

MQ4 2.2 5.47 5.02 25.2 1.2 73.6 0.14 B Horizon (m&g)

MQ3 2.8 5.77 4.60 22.8 2.7 74.5 0.08 B Horizon (m&g)

MQ2 3.3 5.51 4.49 16.5 1.7 81.8 0.06 B Horizon (m&g)

MQ1 3.6 5.88 5.56 0.82 2.1 97.0 0.03 C Horizon (m&g)

MQ II

MQ15 0.08 5.88 4.88 18.4 10.3 71.3 2.04 A Horizon (m&g)

MQ14 0.25 6.15 5.16 30.5 1.7 67.8 0.28 A Horizon (m&g)

MQ13 0.6 5.48 4.83 50.3 0.8 48.9 0.32 A/B Horizon (m&g)

MQ12 1.1 5.45 4.75 25.6 2.0 72.5 0.11 B Horizon (m&g)

MQ11 1.6 5.62 4.62 23.3 2.4 74.3 0.09 B Horizon (m&g)

MQ10 2.0 5.63 4.65 18.0 0.8 81.2 0.07 C Horizon (m&g)

JG

JG7 0.02 5.50 4.83 8.4 6.8 84.8 6.06 A Horizon (m&g)

JG8 0.15 5.59 5.06 7.0 6.5 86.4 2.19 A Horizon (m&g)

JG9 0.3 5.54 5.13 5.8 6.3 87.9 1.06 A Horizon (m&g)

JG10 0.4 5.47 5.06 6.8 6.5 86.7 0.73 A Horizon (m&g)

JG6 1.5 5.60 5.26 4.3 2.5 93.2 0.47 Upper ferruginous zone

(m&g) JG5 3.0 5.08 4.71 6.5 3.5 90.0 0.30 Duricrust (m&g)

JG4 5.0 4.90 4.39 9.0 7.5 83.5 0.20 Ferruginous mottled zone

(m&g) JG3 6.5 4.55 3.96 28.3 7.3 64.4 0.08 Mottled clay (m)

JG2 8.6 3.76 3.45 29.6 8.5 61.8 0.17 Mottled clay (m)

JG1 10.0 3.34 3.15 27.0 15.4 57.7 0.27 Saprolite (m)

1The first two letters of the sample codes identify each profile: Mountain Quarry (MQ) and Great Eastern

Highway (GE) and Jarrahdale granitoid profile (JG). 2depth refers below surface (m);

3pH was determined at 23 °C in a 1:5 suspension of soil: deionised water;

4pH was determined at 23 °C in a 1:5 suspension of soil: CaCl2 solution;

5TC refers to total carbon, determined by vario Macro Elementar Analyser;

6‘m’ denotes matrix and ‘g’ denotes gravel.


44

4.4.2 Weathering intensity-Chemical Index of Alteration (CIA)

To evaluate the intensity of chemical weathering quantitatively, the elemental

concentrations were converted into oxide concentrations and Chemical Index of

Alteration (CIA) (Nesbitt and Young, 1982) were calculated. The CIA calculates loss of

mobile elements relative to less mobile elements in bulk samples, providing a single

parameter estimate of the intensity of chemical weathering. The formula (Nesbitt and

Young, 1982) is:



carbonates). In this study, all regolith have low pH conditions, and no carbonates were

observed by scanning electron microscopy (SEM), and thus all Ca was assumed to be

associated with the silicates.

4.4.3 Mass balance calculation

To quantify net element fluxes from pedogenic weathering, a geochemical mass balance

calculation was used (Brimhall et al., 1991). The formula for normalized concentration

(τi,j) in Equation (2) assumes that an immobile element (e.g. Zr) behaves conservatively

and can be used to correct mobile element concentrations for volumetric strain (ε)

during weathering and pedogenesis.

1))((,

,

,

,

, pj

wj

wi

pi

C

C

C

C

ji

(2)

In Equation (2), C represents concentration, i represents the immobile element, j

represents the element of interest, w represents weathered material and p identifies

parent rock. If τi,j = 0, the element j has behaved conservatively at the sampling scale; if

τi,j = −1, the element j has been depleted completely during weathering; positive τi,j

values signify absolute enrichment.

Equation (2) provides a tool for estimating elemental loss or gain for a profile; however,

mass balance equations have two critical assumptions: a genetic relationship between

regolith and the underlying rock and a fully conserved reference element. In this chapter,

the element Zr was used as a reference element for two main reasons: (i) its existence in

the weathering-resistant, very low-solubility host minerals zircon; (ii) its relatively high

concentrations compared with other high field strength elements, resulting in robust

estimates of mass balance. The mobility of Ti, Zr and Th is discussed in detail in the

next chapter, taking the JG profile as an example.


45

4.4.4 Statistical analyses

The converted major oxide composition data were transformed with a centered log-ratio

(CLR) method using CoDaPack software (Reimann et al., 2008) and then subjected to

principal component analysis using R software (R Development Core Team, 2011) to

assess trends in chemical compositions of regolith samples in a more comprehensive

manner and investigate the principal geochemical processes during lateritization. The R

script is listed in Appendix 11.3.

4.5 Results

4.5.1 Weathering intensity of parent rocks and the regolith

The CIA values for parent rock and regolith matrix and gravel are presented in Table

4.2. In the GE profile, the CIA of granitoids was ca. 52%, higher than the dolerite

(36%). Most GE regolith had CIA values above 90%, reflecting extreme weathering

conditions, except the A horizon matrix (GEA1m, 0.1 m depth, CIA=82%). The

duricrust (GE6, 6 m depth) and the upper mottled clay matrix (GE3m, 10 m depth) had

the highest CIA (> 99%). In addition, the gravel CIA was higher than the matrix in the

A horizon regolith, but lower than the matrix from the lower ferruginous zone (3.5 m

depth) to saprolite (12.5 m depth).

In the MQ I profile, the granitoids had similar CIA (ca. 52%) to the granitoids from the

GE profile. The regolith had lower CIA values and thus was less weathered than the GE

regolith. The CIA in MQ I regolith increased from 63% in gravel and 69% in matrix of

the A horizon (0.2 m depth) to ca. 92% in both gravel and matrix at 1.1 m depth (MQ5),

then decreased to 53% in gravel and 56% in matrix of the C horizon (3.6 m depth).

In the MQ II profile, the CIA increased from 58% in gravel and 71% in matrix of the A

horizon to ca. 83% in both gravel and matrix of the B horizon (0.6 m depth), and then

decreased to 55% in gravel and 63% in matrix of the C horizon (2.0 m depth). The CIA

of gravel was less than or equal to CIA in the corresponding matrix in both MQ profiles,

similar to the subsurface regolith of the GE profile.

In the JG profile, however, the CIA of gravel and matrix in the ferruginous zone

(1.5-6.5 m depth) were similar, ranging from 95% to 99%. In the A horizon, the CIA of

gravel (ca. 99%) was higher than matrix (92%-99%), similar to the A horizon regolith

from the GE profile.


46

Two kinds of outcropping rock were sampled at these sites, loosely classified as either

granitic or doleritic. The most obvious distinction between dolerite and granitoid is that

dolerite has less Si and Zr but more Ti, Fe, Mg, Ca, Mn and P. These variations in

chemical composition were expected to be expressed in regolith (Kew and Gilkes,

2007); the chemical compositions of regolith from the four profiles showed similarity to

the granitoid rather than the dolerite, with Ti and Zr residually enriched (more evidence

from rare earth elements in Chapter Six and Seven). Using A-CN-K and A-CNK-FM

ternary plots, weathering trends in major element composition were identified (Figure

4.1). In the GE and JG profiles, most regolith samples were highly weathered, and the

plots cluster closely at A which reflects the presence of phyllosilicate clay minerals as

the main mineralogical components. In contrast, MQ regolith samples plot separately

from each other, and the weathering trend shown by the regolith chemical compositions

indicates that the regolith was weathered and developed from the granitoid rather than

the dolerite. In ternary plots, the regolith from 2.0 m to 0.6 m depth of the MQ II profile

followed the weathering trend, reflecting in-situ weathering, which is consistent with

the overlying stoneline (Chapter Three). A similar weathering trend was also shown in

the MQ I profile, indicating decreasing Ca and Na and increasing Al and Fe because of

breakdown of primary silicates (mostly feldspar) and formation of secondary clay

minerals and iron oxides. The CIA values of the A horizon regolith in both MQ profiles

were lower than the B and C horizon regolith and did not follow the weathering trends;

this may reflect biogeochemical cycling of Na, K and Ca, or erosion and transportation

of the A horizon regolith as suggested by the stoneline. In the A-CNK-FM plot (Figure

4.1), the MQ regolith showed the depletion of alkaline elements and accumulation of Al

and Fe with increasing weathering intensity.

47

Table 4.2 Concentrations of major elements in gravel and matrix of four lateritic profiles

Sample Depth Proportion1 CIA Al Ca Fe K Mg Na Si S Ti P Mn Zr

Unit m % % % % % % % % % % ppm ppm ppm

d.l.2 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.1 0.1 0.1

Profile GE

Rock

GEBPRA 54.3 7.56 1.01 1.01 2.94 0.14 2.52 34.9 0.00 0.08 89.0 111 106

GEBPRB 53.0 7.51 1.18 1.11 2.86 0.14 2.64 35.4 0.01 0.08 107 129 102

GEPR1A 50.0 5.72 0.42 0.13 2.68 0.01 2.82 38.6 0.02 0.02 11.2 29.5 41.9

GEPR1B 51.3 9.58 2.60 1.55 1.31 0.34 4.02 31.2 0.01 0.16 318 230 183

GEPR2 52.5 7.46 1.14 1.08 2.65 0.15 2.89 34.6 0.01 0.08 131 230 102 3GEPR3 35.9 7.06 8.14 8.35 0.17 4.47 1.28 22.5 0.07 0.61 350 1577 49.0

Regolith

GEA1g 0.20 98.2 27.0 0.07 26.4 0.08 0.03 0.31 3.30 0.04 0.93 250 363 330

GEA1m 0.09 0.80 82.2 2.18 0.11 1.51 0.34 0.03 0.08 35.9 1.68 0.27 61.7 87.8 181

GEA2g 0.54 97.5 26.2 0.18 22.7 0.17 0.04 0.27 4.09 0.04 0.76 157 290 314

GEA2m 0.12 0.46 94.5 3.12 0.03 2.21 0.15 0.02 0.03 36.9 0.80 0.33 28.2 65.9 214

GEA3g 0.56 98.4 24.0 0.02 25.5 0.09 0.03 0.25 3.59 0.04 0.99 151 371 316

GEA3m 0.23 0.44 96.0 2.50 0.02 1.36 0.10 0.02 0.01 38.3 0.42 0.30 87.3 45.0 187

GE6 3.5 W2 99.6 18.0 0.00 20.9 0.09 0.12 0.07 6.20 0.00 0.29 105 92.0 300

GE5g 0.35 92.3 14.0 0.09 27.8 1.14 0.18 0.22 7.46 0.12 0.23 162 186 313

GE5m 7.0 0.65 97.6 17.4 0.01 8.38 0.61 0.13 0.04 14.7 0.00 0.36 63.5 55.6 201

GE4g 0.06 90.7 7.30 0.02 1.72 0.70 0.11 0.21 36.1 0.01 0.16 48.8 55.9 158

GE4m 8.4 0.94 93.8 6.93 0.01 0.75 0.65 0.10 0.02 35.2 0.00 0.17 10.9 37.3 161

GE3g 0.25 98.1 17.5 0.03 1.62 0.11 0.03 0.18 22.5 0.02 0.13 10.7 26.6 127

GE3m 10.0 0.75 99.1 11.6 0.01 1.02 0.14 0.03 0.04 29.0 0.02 0.12 25.6 17.3 116

GE2g 0.08 95.3 6.70 0.05 1.02 0.11 0.02 0.16 37.1 0.01 0.14 20.3 28.9 140

GE2m 11.4 0.92 97.4 6.28 0.01 0.52 0.15 0.05 0.04 35.8 0.09 0.17 10.0 17.9 154

48


Unit m % % % % % % % % % % ppm ppm ppm

d.l.2 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.1 0.1 0.1

GE1g 0.06 94.1 10.2 0.01 2.82 0.75 0.08 0.09 30.5 0.02 0.14 82.4 48.9 191

GE1m 12.5 0.94 93.5 8.15 0.01 0.76 0.78 0.08 0.02 33.6 0.01 0.18

48.8 31.4 207

Profile MQ

Rock

MQPR1 53.2 7.35 0.72 1.30 3.37 0.41 2.70 33.5 0.03 0.14 230 181 160

MQPR2 51.0 7.55 1.32 1.14 2.92 0.27 2.96 31.0 0.02 0.15 233 154 160 4MQPR3 36.8

7.02 7.55 9.59 0.32 3.77 1.41 21.9 0.09 0.89

465 1705 73.8

Regolith

MQ I

MQ9g 0.21 62.9 7.15 0.49 2.14 2.67 0.16 1.47 33.4 0.00 0.21 121 225 168

MQ9m 0.20 0.79 69.4 7.84 0.45 2.42 2.31 0.19 1.07 28.7 0.06 0.28 183 295 198

MQ8g 0.42 61.5 8.21 0.54 2.64 3.37 0.20 1.77 31.5 0.00 0.21 131 187 158

MQ8m 0.50 0.58 71.1 8.34 0.39 2.64 2.29 0.19 1.10 28.1 0.07 0.30 164 232 186

MQ7g 0.29 77.1 10.3 0.20 4.22 2.23 0.17 1.06 29.7 0.00 0.34 84.0 82.8 172

MQ7m 0.70 0.71 77.1 9.58 0.21 3.58 2.09 0.16 0.95 27.4 0.06 0.34 43.5 70.1 172

MQ6g 0.27 84.8 13.3 0.45 6.19 1.43 0.21 0.68 24.1 0.00 0.45 118 79.8 139

MQ6m 0.90 0.73 89.9 13.0 0.10 5.96 1.11 0.21 0.47 20.9 0.08 0.45 81.2 49.1 130

MQ5g 0.30 92.4 13.4 0.06 6.73 0.83 0.19 0.38 22.8 0.00 0.45 81.4 71.6 124

MQ5m 1.1 0.70 91.9 12.9 0.07 5.94 0.94 0.19 0.34 19.3 0.06 0.42 68.3 40.0 112

MQ4g 0.15 64.9 4.15 0.03 1.14 2.19 0.04 0.59 38.3 0.00 0.09 45.7 32.2 115

MQ4m 2.2 0.85 73.6 8.75 0.07 2.24 2.66 0.09 1.02 29.5 0.07 0.24 42.9 24.8 147

MQ3g 0.15 63.1 6.71 0.07 2.02 3.29 0.10 1.33 34.3 0.00 0.11 8.8 34.7 133

MQ3m 2.8 0.85 68.5 8.83 0.12 2.58 2.82 0.14 1.66 29.7 0.06 0.20 24.2 29.9 154

MQ2g 0.39 55.4 7.01 0.20 1.13 2.78 0.09 2.95 34.7 0.00 0.08 10.0 34.7 135

49

Sample Depth Proportion

1 CIA Al Ca Fe K Mg Na Si S Ti P Mn Zr

Unit m % % % % % % % % % % ppm ppm ppm

d.l.2 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.1 0.1 0.1

MQ2m 3.3 0.61 62.4 7.08 0.20 1.51 2.06 0.11 2.18 30.3 0.05 0.16 21.3 31.4 149

MQ1g 0.41 52.9 7.93 0.78 1.20 3.41 0.16 3.11 33.0 0.01 0.10 55.4 88.3 131

MQ1m 3.6 0.59 56.0 6.77 0.72 1.36 2.47 0.19 2.26 32.5 0.06 0.16 53.0 97.8 163

MQ II

MQ15g 0.42 58.3 7.85 0.66 2.12 3.08 0.20 2.23 32.2 0.00 0.20 104 191 168

MQ15m 0.08 0.58 71.0 8.49 0.44 2.66 2.36 0.30 1.06 30.5 0.04 0.29 248 227 150

MQ14g 0.20 60.1 8.83 0.92 2.35 2.93 0.15 2.23 31.7 0.01 0.21 39.8 107 185

MQ14m 0.25 0.80 73.7 8.41 0.20 2.90 2.30 0.13 0.98 30.8 0.03 0.29 72.4 49.5 193

MQ13g 0.26 83.2 11.4 0.15 4.60 1.51 0.16 0.90 27.0 0.00 0.35 43.4 56.9 150

MQ13m 0.60 0.74 82.7 10.6 0.08 4.09 1.79 0.15 0.75 26.2 0.06 0.33 56.7 33.1 160

MQ12g 0.16 70.9 7.39 0.07 2.87 1.84 0.12 1.43 34.4 0.00 0.18 16.5 41.8 148

MQ12m 1.1 0.84 73.9 9.10 0.09 3.27 2.00 0.14 1.46 27.6 0.06 0.23 17.7 22.2 145

MQ11g 0.26 57.7 6.85 0.18 1.40 3.21 0.08 2.19 35.3 0.00 0.10 13.2 33.2 105

MQ11m 1.6 0.74 67.4 8.20 0.13 2.30 1.91 0.13 2.11 28.9 0.06 0.18 6.6 20.8 133

MQ10g 0.26 54.9 5.81 0.21 1.06 2.81 0.05 2.16 37.3 0.00 0.10 10.0 29.4 108

MQ10m 2.0 0.74 62.5 7.89 0.14 1.68 2.13 0.09 2.61 30.2 0.06 0.17 43.7 15.0 146

Profile JG

Rock

JGPR1 46.8 7.79 1.57 1.41 3.45 0.35 3.72 33.5 0.02 0.13 307 225 160

JGPR2 47.2 7.61 1.51 1.53 3.17 0.33 3.66 32.2 0.02 0.13 252 218 159

Regolith

JG7g 0.24 99.2 23.6 0.03 20.4 0.18 0.03 0.03 6.70 0.03 0.74 184 205 348

JG7m 0.02 0.76 99.0 15.6 0.01 5.25 0.18 0.03 0.01 19.3 0.01 0.74 121 79.1 472

JG8g 0.60 98.7 24.2 0.06 24.2 0.19 0.02 0.08 4.44 0.03 0.89 243 203 354

JG8m 0.15 0.40 92.9 8.52 0.16 2.82 0.54 0.04 0.06 28.8 0.01 0.65 135 305 425

50


Unit m % % % % % % % % % % ppm ppm ppm

d.l.2 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.1 0.1 0.1

JG9g 0.51 99.3 27.3 0.05 22.2 0.12 0.02 0.03 4.05 0.04 0.77 210 183 354

JG9m 0.3 0.49 92.7 9.04 0.09 3.03 0.65 0.03 0.12 31.0 0.01 0.69 102 209 507

JG10g 0.69 99.3 26.2 0.04 21.5 0.12 0.03 0.05 4.74 0.05 0.67 239 163 346

JG10m 0.4 0.31 92.2 9.35 0.11 3.32 0.78 0.04 0.10 30.5 0.01 0.67 87.2 108 490

JG6g 0.82 99.4 22.9 0.03 23.4 0.08 0.02 0.04 2.04 0.02 0.99 132 168 349

JG6m 1.5 0.18 98.6 21.4 0.00 5.52 0.41 0.04 0.01 13.3 0.02 0.38 80.2 57.5 445

JG5g 0.47 98.6 21.8 0.01 1.79 0.41 0.04 0.01 13.7 0.02 0.33 20.6 34.1 349

JG5m 3.0 0.53 98.1 13.2 0.01 2.49 0.31 0.03 0.02 25.6 0.01 0.44 60.7 50.5 291

JG4g 0.40 95.4 7.09 0.01 0.86 0.46 0.06 0.01 35.4 0.01 0.37 1.6 39.4 292

JG4m 5.0 0.60 95.3 8.05 0.01 1.18 0.54 0.05 0.01 32.2 0.00 0.52 29.0 64.6 482

JG3 6.5 W 94.1 7.50 0.01 0.78 0.63 0.06 0.02 34.2 0.02 0.38 9.0 37.1 341

JG2 8.6 W 86.2 9.73 0.00 0.85 2.18 0.05 0.04 31.4 0.04 0.17 0.8 22.9 164

JG1 10.0 W 64.7 8.35 0.27 0.84 3.93 0.06 1.26 33.6 0.03 0.10 2.1 59.6 105

4Batch 1

RSD5 0.2-0.6 0.0 0.0-0.3 0.0-0.1 0.0 0.0 0.7-1.1 0.0-0.2 0.0 10.1-20.4 3.3-16.3 8.6-8.9

Batch 2

RSD 0.1-0.2 0.0-0.4 0.0-0.1 0.0-0.5 0.0 0.0-0.7 0.0-0.9 0.0 0.0-0.1 0.9-30.2 2.5-17.5 2.6-9.3

Batch 3

RSD 0.0-0.2 0.0-0.3 0.0-0.2 0.0 0.0 0.0-0.3 0.0-0.2 0.0 0.0 7.4-31.4 0.4-8.9 0.9-10.4

1Proportion

refers to the weight proportion of gravel (g) or matrix (m) in percentage; ‘W

2’ refers to this sample without corresponding gravel/matrix and thus determined by

whole-rock analysis; 2d.l. refers to detection limit.

3GEPR3 and

4 MQPR3 are dolerite, whereas other rock samples are granitoid.

4Batch refers to each time determination of element concentrations based on the profile;

5RSD is the range of relative standard deviations (precision) of the duplicates/triplicates analysed by ICP-OES.

51

Figure 4.1 Ternary A-CN-K and A-FM-CNK plots of regolith samples from four lateritic profiles (GE, MQ I, MQ II, JG) based on chemical

compositions of matrix and gravel samples. Dashed line with arrow indicates weathering trend for MQ profiles. (a) and (e) GE profile; (b) and (f) JG

profile; (c) and (g) MQ I profile; (d) and (h) MQ II profile. Triangles represent granitoids and circles represent dolerite, whilst squares indicate regolith

gravel and diamonds indicate matrix samples.


52

4.5.2 Mineralogical properties

In the GE profile, the parent granitoid (GEPR2) was characteristically identified by

quartz (ca. 67%), muscovite (ca. 4%), K-feldspars (ca. 8%) and albitic plagioclase

(ca. 21%) (Figure 4.2a). The saprock (GEBPRa&b) was composed of similar minerals

to the parent granitoid but in different proportions: quartz (71%-74%), muscovite

(ca. 4%), K-feldspars (ca. 7%) and albitic plagioclase (14%-17%). From saprolite

upwards, anorthitic plagioclase and K-feldspar had been weathered and altered to

kaolinite and gibbsite, whereas quartz and muscovite remained as residual components.

The proportion of kaolinite increased gradually upward from the saprolite (12.5 m depth)

to upper mottled clay (8.4 m depth), and then decreased in the lower ferruginous zone

(7.0 m depth). The concentration of gibbsite increased from the saprolite until duricrust

(3.5 m depth), except in the lower ferruginous zone. Muscovite was absent in the

mottled clay (8.4-11.4 m depth) but occurred in the ferruginous zone, reflecting its

moderate resistance to weathering. Quartz was residually enriched in the saprolite due

to breakdown of feldspar, but was less abundant in the mottled clay and ferruginous

zones due to further desilication and formation of Al hydroxides and Fe oxyhydroxides

under intense weathering. Goethite and hematite were present in the ferruginous zone;

however, significant amounts of hematite were only identified in the duricrust.

In the MQ profiles, the mineralogy followed similar patterns (Figure 4.2b & c):

kaolinite dominated the clay mineral assemblage and the proportion increased gradually

with decreasing depth. In contrast, plagioclase and muscovite decreased upward until

muscovite was completely absent at 1.1 m depth in the MQ I profile; muscovite then

increased again above 1.1 m depth. K-feldspar was also present its lowest proportion

(ca. 4%) at 1.1 m depth in the MQ I profile.

In the JG profile, the mineral characteristics were more complex than the other three

profiles (Figure 4.2d). In the saprolite (10 m depth) and mottled clay (6.5-8.6 m depth),

all plagioclase had been replaced by kaolinite and gibbsite. In the ferruginous zone and

the A horizon, further weathering and degradation resulted in the replacement of

kaolinite by gibbsite and formation of Fe oxyhydroxides. The regolith from the

ferruginous zone and the A horizon was made up of loose matrix with abundant iron

nodules; the iron nodules contained more Al (gibbsite) and Fe oxyhydroxides (goethite)

but less Si (quartz) than the matrix. The loose matrix was composed of heterometric,

sub-rounded to angular quartz grains of low sphericity with lower amounts of


53

K-feldspar, kaolinite, gibbsite and boehmite. Iron nodules ranged from 1-3 cm in

diameter and were composites of hematite, maghemite, quartz, gibbsite and boehmite

mineral assemblage. The iron nodules from the ferruginous zone were red and

concentrically zoned, with a core of hematite surrounded by a goethite rim (Fe-rich

layer) and an Al-rich layer, similar to the studies of Anand (2002; 2010) and

Fernandez-Caliani and Cantano (2010). The iron nodules from the A horizon, however,

were non-concentrically zoned, clay matrix cemented with Fe oxides (mainly hematite

and maghemite) without layers, containing less gibbsite but more quartz and the fabric

and texture changed to dark brown/black due to greater organic matter contents

(Appendix 11.4). Quartz grains in the iron nodules appeared strongly corroded and

spongiform, showing dissolution features, similar to those described by Abreu (1990)

and Fernandez-Caliani and Cantano (2010).

4.5.3 Mass balance analysis of elemental loss and gain

Mass balances of major elements were calculated at each sampling depth, based on

weighted average concentrations of major elements in matrix and gravel samples, using

Zr as the reference element. Depth profiles of absolute elemental loss or gain are plotted

in Figure 4.3.

In the GE profile (Figure 4.3a), Al was the least depleted element and enriched in

mottled clay (τ(Zr, Al) = 0.5) at ca. 10 m depth. In the ferruginous zone and the A horizon,

τ(Zr, Fe) ranged from 1.1-5.6, representing Fe enrichment, especially in the duricrust

(τ(Zr, Fe) = 5.6). Silicon was depleted most in the duricrust (τ(Zr, Si) = −0.9) and least in the

mottled clay (τ(Zr, Si) ca. −0.3). The elements Na and Ca were significantly depleted

throughout the profile, with τ(Zr, Na)< −0.98 and τ(Zr, Ca)< −0.95; K and Mg were less

depleted than Na and Ca, with τ(Zr, K) from −0.99 to −0.85 and τ(Zr, Mg) from −0.94 to

−0.60.

In the MQ I profile (Figure 4.3b), τ(Zr, Na) and τ(Zr, Ca) first decreased upwards until 1.1 m

depth, and then increased until the surface soils, consistent with the changes in

plagioclase content. Silicon was slightly enriched in the B and C horizons (τ(Zr, Si)

ranging from 0.01 to 0.1) but depleted in the A and B horizons (τ(Zr, Si) ranging from

−0.1 to −0.2). Aluminium and Fe were enriched throughout the profile with the highest

τ(Zr, Al) = 1.4 and τ(Zr, Fe) = 6.0 at 1.1 m depth.

In the MQ II profile (Figure 4.3c), similar as the MQ I profile, depletion of Na and K


54

first increased upwards to 0.6 m depth and then decreased. Aluminium was enriched in

most regolith (except τ(Zr, Al) = −0.05 at 0.25 m depth) and Fe was enriched in all

regolith with τ(Zr, Fe) ranging from 0.5 to 2.5. The changes in mass balance of major

elements at 1.1 m depth in the MQ I profile, and at 0.6 m depth in the MQ II profile,

were consistent with the previous assumption that, above 1.1 m depth in the MQ I

profile and 0.6 m depth in the MQ II profile, transported materials were present.

In the JG profile (Figure 4.3d), Na and Ca showed near-complete loss (τ(Zr, Na) and τ(Zr, Ca)

ca. −0.99) throughout the profile except the saprolite (τ(Zr, Na) = −0.5 and τ(Zr, Ca) = −0.7).

Silicon was enriched in the saprolite (τ(Zr, Si) = 0.6) but depleted upwards until the upper

ferruginous zone (τ(Zr, Si) = −0.95). Iron was depleted in the saprolite (τ(Zr, Fe) = −0.1),

mottled clay (τ(Zr, Fe) ca. −0.7), even the duricrust (τ(Zr, Fe) = −0.3) but strongly enriched

in the upper ferruginous zone (τ(Zr, Fe) = 5.2) and the A horizon (τ(Zr, Fe) average 2.7).

Based on the study of Brimhall et al. (1991), Zr concentrations in the A horizon (<1 m

depth) have been enhanced by aeolian input, and thus τ(Zr) in the A horizon would be

higher without external input, suggesting that calculated Fe enrichments in the A

horizon were conservative estimates.


55

Figure 4.2 Semi-quantitative mineralogical composition of regolith samples and parent

granitoids determined by random powder XRD analysis based on weighted average of

matrix and gravel, for: (a) the GE profile; (b) the MQ I profile; (c) the MQ II profile,

and; (d) the JG profile; (Qz: quartz; Ms: Muscovite; Kfs: Potassium feldspar; Fsp:

plagioclase feldspar; kln: kaolinite and halloysite group; Gbs: gibbsite; Gth: goethite;

Hem: hematite; Mgh: maghemite; Boe: boehmite).

56

Figure 4.3 Mass balance of major elements in regolith samples from four lateritic profiles, based on weighted average concentrations of major elements

in matrix and gravel at each depth, using Zr as the reference element (Brimhall et al., 1991): (a) the GE profile; (b) the MQ I profile; (c) the MQ II

profile; (d) the JG profile (the vertical dashed line refers to τ(Zr) = 1.0 without depletion or accumulation relative to the parent granitoid; trend of Ca

was similar to Na, and Mg to K, so not plotted above).

57

Figure 4.4 Depth functions of the molar ratio Na/K and concentration ratio Al2O3/Fe2O3 for MQ two profiles and concentration ratio (Ti/Zr)/10 for four

profiles illustrating the pedogenic discontinuity (shown by the boxed area) at 1.1 m depth in MQ I profile and 0.6 m depth in MQ II profile.

58

Figure 4.5 Major element concentrations in grain size fractions of the regolith samples from the MQ II profile.


59

4.5.4 Depth functions of pedogenic discontinuities

Pedogenic discontinuities represent zones of change in physical and chemical

properties primarily originating from processes during soil development, rather than

the parent lithology, which is more commonly referred as lithological discontinuities

(Schaetzl and Anderson, 2005; Tsai and Sang, 2000). In both MQ profiles, the

distribution of major elements in A horizon regolith is not correlated to the weathering

trend (Figure 4.1). In addition, a stone line observed within the MQ II profile may

indicate either the presence of an erosional episode at/near a discontinuity (Parsons and

Herriman, 1966; Ruhe, 1958) or a mass movement with biogenic involvement

(Johnson, 1990; Johnson and Balek, 1991; Lichte, 2000). Either of these reasons

indicates the presence of pedogenic discontinuity in both MQ profiles. Therefore,

depth functions for various pedogenic parameters, including molar ratio Na/K,

concentration ratios Al2O3/Fe2O3 and (Ti/Zr)/10 were prepared for the MQ I and MQ II

profiles (Figure 4.4).

In the MQ profiles, abrupt changes of Na/K, Al2O3/Fe2O3 and (Ti/Zr)/10 occurred at

1.1 m depth of the MQ I profile and at 0.6 m depth of the MQ II profile, including

gravel and matrix. The (Ti/Zr)/10 first increased upwards and then decreased until

surface soil, showing great variations (variation up to > 70% in MQ I profile and > 50%

in MQ II profile relative to the parent granitoid). However, below 1.1 m depth of the

MQ I profile and at 0.6 m depth of the MQ II profile, (Ti/Zr)/10 was relatively

consistent. The depth of abrupt change coincided with the presence of stone line in the

MQ II profile, implying that a mass movement (i.e. erosion and re-deposition with or

without biological recycling) had occurred during the weathering history of these two

profiles.

In the GE and JG profiles, (Ti/Zr)/10 was relatively consistent in the saprolite and

mottled clay zone, close to the parent granitoid (variation < 30%), including both

gravel and matrix; then Ti and Zr in gravel and matrix started to fractionate in the

ferruginous zone, and significantly differentiated from each other in the A horizon. The

variation of (Ti/Zr)/10 in the ferruginous zone showed that Ti and Zr partitioned


60

between gravel and matrix at the advanced stage of weathering and lateritization and

further details will be discussed in the next chapter using the JG profile as an example.

4.5.5 Grain size distribution of major elements in MQ II profile

The distribution of major elements into different grain size fractions from the MQ II

profile (Figure 4.5) showed that Al, Fe, Mg, Ti and Na were enriched in the clay and

silt fractions. Potassium and Si, however, were enriched in the sand and gravel

fractions. The silt fraction had the highest Zr concentrations. In the A horizon, gravel

had the highest concentrations of Ca; however, in the B and C horizons, sand was the

main host for Ca.

4.6 Discussion

4.6.1 Significant processes during lateritization

Based on the geochemical and mineralogical data, the regolith from these four lateritic

profiles has experienced moderate to extreme weathering. The mass balance

calculations (Figure 4.3) reflect the substantial loss of alkaline and earth-alkaline

elements in the saprolite of the GE and JG profiles, in agreement with the depletion of

plagioclase (Figure 4.2). Compared with Na and Ca, K and Mg were less depleted and

corresponded to the residual occurrence of muscovite and potassium feldspar. Both

alkalis and Si continued to be lost as weathering intensified. Kaolinite was further

altered into gibbsite and Fe was enriched in the ferruginous zones as goethite and

hematite (Figure 4.2). Similar trends have been reported by Anand and Paine (2002).

At pH above 5, Si (even from quartz) has a higher solubility than Al- and Fe- oxides

(Breemen and Buurman, 1998), and thus would be preferentially removed from the

system. The near-complete loss of Na and Ca and significant enrichment of Al and Fe

in the ferruginous zone indicate that the regolith has undergone an advanced stage of

lateritization. Only residual or partially weathered quartz and some

weathering-resistant minerals remained in the ferruginous regolith.

Both MQ profiles were less weathered than GE and JG profiles (Figure 4.1), and the


61

chemical composition of regolith showed intermediate depletion of Si, Na and K and

enrichment of Al and Fe (Figure 4.3). The plots of SiO2 vs. Al2O3 and Fe2O3 vs. Al2O3

(Figure 4.6) clearly show that (i) the regolith was weathered in-situ from granitoids; (ii)

silicon consistently decreased whereas Al and Fe increased with increasing weathering

intensity. Aluminium was residually enriched by dissolution of feldspar and mica and

the formation of kaolinite at the expense of primary minerals (kaolinization) in the B

horizon of both MQ profiles and in the mottled clay zone of the GE and JG profiles

(Figure 4.2). Idealized weathering reactions for dissolution of silicates, e.g. feldspar

and muscovite, and for formation of kaolinite at the early stages of weathering are

presented below:

CaAlSi3O8+H2O+2H+⇌Al2Si2O5(OH)4+Ca

2+ (loss of Ca)

2KAl3Si3O10(OH)2+3H2O+2H+⇌3Al2Si2O5(OH)4+2K

+ (loss of K)

4.6.2 Genesis and sources of Fe redistribution

The nature of formation of iron oxides is generally more dependent on the

environmental conditions at the time of formation than on the particular structures of

the primary mineral from which the Fe was released (Anand and Paine, 2002). Under

anaerobic conditions, Fe can mobilize and redistribute over a range of spatial scales.

This mobilization can be vertical or lateral, at a horizon, or even a landscape, scale.

In the GE and JG profiles, residual accumulation of Fe is likely to be the result of

continuous cyclic dissolution-precipitation processes with redox changes occurring as

a result of changes in regolith water regime (Anand and Butt, 2010; Anand and Paine,

2002; Tripathi and Rajamani, 2007). Redistribution of Fe in the duricrust may be

related to a capillary effect triggered by seasonal fluctuation of the water table (Braun

et al., 1993; Ndjigui et al., 2009). In addition, in the JG profile, the location of intrusive

dykes in the granitoids can be mapped from the overlying duricrust and the profiles

developed on dolerite and granitoid are distinctly different; both of which suggest that

duricrust has formed in-situ (Anand and Paine, 2002; Sadleir and Gilkes, 1976). The

regolith from the GE profile was also formed residually from weathering and

pedogenic processes; the complete profile components (including saprolite, mottled


62

clay and ferruginous zone) occur without any discontinuity, erosion or missing zones

and no apparent enrichment of any trace elements derived from external sources in

subsurface regolith (regolith <1 m depth given aeolian input) was observed. Therefore,

Fe enrichment in the GE profile is more likely a result of vertical residual

accumulation rather than lateral movement.

In the MQ profiles, transition zones (at 1.1 m depth in MQ I profile and 0.6 m depth in

MQ II profile) showed the lowest Al2O3/Fe2O3 ratio. It is possible that the Fe released

from primary minerals was partially oxidized by oxygen in penetrating

rainwater/atmosphere and, as a result, was precipitated in the upper part of the regolith

during weathering. With continued weathering and the influence of soil creep or

colluviation, and alternative wetting and drying of the soil due to seasonality or

longer-term changes between humic and arid climates, translocation and redistribution

of Fe was facilitated.


63

Figure 4.6 The distribution of Al2O3 vs. SiO2 and Al2O3 vs. Fe2O3 in matrix and gravel

from four lateritic profiles (A_g: gravel of A horizon; A_m: matrix of A horizon;

regolith_g: gravel of subsurface regolith; regolith_m: matrix of subsurface regolith; in

(g) and (h), dolerite had Fe2O3 concentration 13.7 wt%, so not showed).


64

4.6.3 Degrees of lateritization

To classify the degree of lateritization and demonstrate the effects of weathering

intensities on lateritic regolith, ternary SiO2-Al2O3-Fe2O3 diagrams (Schellmann, 1981)

were plotted (Figure 4.7). The progression from fresh parent granitoids to loss of Si

and relative enrichment of Fe and Al involves various degrees of weathering intensity.

Al2O3 and Fe2O3 only separate under the extreme weathering (strong lateritization).

The GE and JG profile have apparently undergone strong lateritization, whereas both

MQ profiles are still within the ‘weak lateritization’ status, but have both undergone

intense kaolinization.

Figure 4.7 Schellmann SiO2-Al2O3-Fe2O3 diagrams showing different degrees of

lateritization of weathered regolith from four lateritic profiles: (a) the GE profile; (b)

the MQ profiles; (c) the JG profile; regolith include matrix and gravel.

4.6.4 Principal components analysis

In order to systematically analyse the data, chemical compositions of all regolith from

four profiles are standardised and subjected to principal component analyses. Two

principal components extracting 76.0% of variance in major element data are

recognized and plotted in Figure 4.8 and the factor score of each regolith sample is

plotted in Figure 4.9.

Principal Component 1 was controlled by base cations (positive loadings) and

conservative elements such as Ti, Zr, Al and Fe (negative loadings). Parent granitoids

were enriched in base cations (ellipse a in Figure 4.9), with moderate weathered MQ

regolith having some base cation depletion (ellipse b). In contrast, GE and JG regolith


65

was characterized by more residually accumulated Zr, Ti, Al and Fe (ellipse c, d, e and

f). This reflects the relative mass flux depletion of base cations in weathered regolith

and relative conservation of Al and Fe in ferruginized regolith during lateritic

weathering. Principal Component 2 was mainly affected by Si, P and Mn. The regolith

of mottled clay and ferruginous mottled zones from the GE and JG profiles contained

high concentrations of Si and separated from the other regolith, indicating initial

residual enrichment of Si due to depletion of base cations without desilication. In

ferruginous zone of the GE and JG profiles, this Si enrichment was weakened by

desilication and formation of gibbsite. The ferruginous surface gravel from the GE and

JG profiles had lower concentrations of Si but higher concentrations of P and Mn than

corresponding matrix, and thus, separated from the matrix and the subsurface regolith.

The C horizon regolith from the MQ I profile and the saprolite from the JG profile (see

arrows) were close to parent granitoids, reflecting: (i) incipient weathering conditions;

(ii) weak depletion of base cations; and (iii) the genetic relationship with the granitoid.

In addition, different types of granitoids were slightly differentiated from each other;

e.g., the coarse grained granite was separated from the late-stage sub-horizontal

pegmatite vein.


66

Figure 4.8 Principal component analyses of major elements in regolith samples and

parent granitoids from four lateritic profiles. Compositional data were transformed

using centered log-ratios.

Figure 4.9 Principal component factors of regolith samples and parent granitoids from

four lateritic profiles calculated using major element composition.


67

4.6.5 Mineralogy and element grain size distribution

The distribution pattern of major elements into different subgroups can be interpreted

by regolith mineral characteristics. In the GE and JG profiles, gravel was more

enriched in Al and Fe but more depleted in Si, whereas matrix was more enriched in Si

but more depleted in Al and Fe (Figure 4.10). This corresponds to the weathering

intensity and mineral composition of gravel and matrix. Both profiles are more

intensely weathered; cyclic reducing and oxidizing conditions likely lead to mottle

formation, with Fe oxides and oxyhydroxides migrating into clay matrix or voids.

Repeated dissolution and cementation results in the formation of gravel, which are

dominated by gibbsite, goethite and hematite. The occurrence of maghemite in

near-surface regolith in the JG profiles may be induced by a combination of heat (bush

fire) to dehydroxylate goethite and organic matter (Anand and Gilkes, 1987; Anand

and Paine, 2002; Perrier et al., 2006). However, in both MQ profiles, Si in gravel was

higher than in matrix, but Al and Fe varied with the depth. The MQ gravel containing

higher Si likely results from less weathering intensity and relatively fast drainage. The

absence of gibbsite in both MQ profiles reveals that kaolinite did not further alter into

gibbsite and no apparent partitioning of Al occurred between matrix and gravel in MQ

regolith profiles. Variation of Fe with depth correlates to the weathering intensity of

matrix and gravel samples. Therefore, the element distribution between matrix and

gravel is a reflection of the mineralogical composition, which is more dependent on the

weathering environment and weathering intensity than the parent lithology.

In addition, the enrichment of Al, Fe, Mg, Ti and Na in silt and clay fractions indicates

that Al released from feldspar, muscovite, etc., in the parent granitoids formed clay

minerals and sesquioxides in the regolith. Iron released from altered magnetite and

other iron-bearing phases in the parent granitoids produced goethite in the regolith.

Due to different charge/ionic radius and hosting minerals, the distribution patterns of

Na and K, Mg and Ca were not the same. These elements inherited from the parent

granitoids, separate from each other during weathering and pedogenic processes

depending on the mineral weathering rate and weathering conditions. The differences

of charge/ionic radius also affect the ability of ion exchange of these elements onto


68

clay minerals, Al and Fe oxides/oxyhydroxides and organic matter.

In addition, Si was residually enriched in the sand fraction as quartz, and Zr was

residually enriched in the silt fraction as zircon. Enrichment of Ti in the fine fractions

indicated the physical, and possibly chemical, weathering of ilmenite and rutile or

newly formed anatase (Anand et al., 1991; Anand and Paine, 2002). The geochemical

distribution trends of major elements in different grain size fractions are therefore

mainly controlled by the physical characteristics and chemical stabilities of hosting

minerals inherited from the parent rock and secondary formed minerals in the regolith,

which in themselves are constrained by the weathering conditions.

4.6.6 Mobility of Ti and Zr

Although Ti and Zr are commonly considered to be and most frequently used as the

least immobile elements in weathered profiles (Beyala et al., 2009; Braun et al., 1993;

Brimhall et al., 1991; Nesbitt and Markovics, 1997; Taboada et al., 2006b),

mobilization of Zr induced by eroded zircon and redistribution of Ti during extreme

weathering have been discussed (Anand and Paine, 2002; Anand et al., 2010; Braun et

al., 1993). Titanium occurs in rocks mainly as rutile, ilmenite and sphene, or in the

structure of silicates such as micas, amphiboles and pyroxenes. The susceptibility of

silicate minerals to weathering results in release of some Ti in the early stage of

weathering of igneous rocks and continuing release as weathering proceeds (Anand

and Paine, 2002). Zirconium occurs in rocks largely as zircon, which is very resistant

to weathering and hence Zr is not considered to be transported under low temperature

and low pressure conditions (Henderson, 1984; Linnen et al., 2005; Vos et al., 2006).

In intensely weathered profiles (e.g., laterite), it is difficult to identify the most

appropriate immobile element(s) not subject to dissolution and physical translocation

given different scales.

In the GE and JG profiles, a variation of (Ti/Zr)/10 from the ferruginous zone upwards

was observed, illustrated that either or both Zr or Ti were relatively mobile under

extreme weathering conditions. In the Jarrahdale bauxitic lateritic profile, both rutile


69

and zircon have been introduced by aeolian input, inducing higher concentrations of Ti

and Zr in the surface soils (<1 m depth) (Brimhall et al., 1988; Brimhall et al., 1991;

Foo, 1999; McLennan, 1995). The euhedral grains of zircon, observed by scanning

electron microscopy in the regolith of four profiles showed zircon’s stability. However,

eroded ilmenite and rutile, and poorly crystalline Ce, Zr and Th (hydr)oxides, were

observed in the ferruginous zone of the JG profile, suggesting mobility of both Ti and

Zr at the sampling scale during the advanced stages of weathering and lateritization.

The mobilization and redistribution of Ti, Zr and Th will be investigated further in the

next chapter taking the JG profile an example.

4.7 Summary of the chapter

In this chapter, the bulk geochemistry of major elements in matrix and gravel from

four intensely weathered lateritic profiles (GE, MQ I, MQ II and JG) was investigated.

The regolith from the GE and JG profiles had undergone intense weathering and strong

lateritization with the CIA above 99% in the ferruginous zone. Both MQ profiles were

less weathered than the GE and JG profiles with only weak lateritization.

In the GE and JG profiles, gravel was more enriched in Al and Fe but more depleted in

Si than matrix. In both MQ profiles, however, gravel had higher Si than the associated

matrix with varied concentrations of Al and Fe. In the ferruginous gravel of the GE and

JG profiles, Fe was mainly enriched as goethite, hematite and maghemite, while Al

was mainly enriched as secondary clay minerals and gibbsite. In both MQ profiles,

however, lower weathering intensities and relatively fast drainage resulted in the

absence of gibbsite and hematitie and low concentrations of goethite.

Using Zr as the reference element, Ca and Na were near-completely depleted and large

loss of K, Mg and Si also occurred in the GE and JG profiles, indicating breakdown of

plagioclase and incongruent dissolution of potassium feldspar, muscovite and quartz at

advanced stages of weathering. The increasing proportion of gibbsite in the duricrust

and ferruginous gravel of the GE and JG profiles revealed that kaolinite was further

altered into gibbsite and cemented with iron oxides with weathering intensifying.


70

Taking the MQ II profile as an example, Al, Fe, Mg, Ti and Na were enriched in clay

and silt fractions, K and Si was enriched in sand and gravel fractions and Zr was

enriched in the silt fractions as zircon. The redistribution of elements into different

grain size fractions is mainly controlled by the physical characteristics and chemical

stabilities of hosting minerals inherited from the parent rock and newly formed

secondary minerals in the regolith. Intense leaching of cations, kaolinization,

desilication and ferruginization were identified as significant processes during

lateritization using principal component analysis, and these mechanisms were further

substantiated by the geochemical mass balance calculations and mineralogical

analyses.

71

Figure 4.10 Calculated τ values of Al, Fe and Si referenced to Zr in matrix and gravel from four lateritic profiles (the dashed line indicates τ(Zr) = 0,

without enrichment or depletion).

73

5 Redistribution and mobilization of Ti, Zr and Th in an intensely

weathered lateritic profile in Western Australia

5.1 Abstract

The mobility of titanium, zirconium and thorium, elements commonly considered

insoluble during supergene weathering, is still not well understood, especially in

intensely weathered regolith. Thus, an intensely weathered lateritic profile (JG)

developed on meta-granitoids in Jarrahdale, Western Australia, was investigated. The

mobility of Ti, Zr and Th has been assessed at both mineral assemblage and profile

scale and the mode of occurrence has been investigated through the combined use of

geochemical data from bulk regolith, particle size fractions and sequential extractions,

with in-situ data determined by electron probe micro-analyzer and synchrotron X-ray

powder diffraction.

Neoformed poorly crystalline phases containing trace to minor amounts of Zr, Ce and

Th unassociated with silicates or phosphates were identified on the walls of Al/Fe-rich

pores in the ferruginous duricrust. This implies that some mobilization and

redistribution of Zr and Th occurs within a sample scale. Breakdown of primary thorite

and rare earth element rich fluorocarbonates is thought to be the source for Zr and Th in

the neoformed phases rather than zircon. Thus, the mineral hosts of Zr, Ti and Th in the

parent rock and their relative susceptibility to weathering are the fundamental controls

on subsequent mobility during initial weathering. Trace amounts of Th in secondary

phases, such as rhabdophane and florencite, shows translocation of Th at the mineral

scale; whilst strong partitioning of Th into gravel rather than matrix reflects

redistribution of Th at the profile scale. The absence of primary sphene from the

regolith and the presence partially dissolved ilmenite and rutile grains in the ferruginous

mottled zone suggest mobilization and translocation of Ti at a mineral assemblage scale.

Furthermore, the fluctuation of Ti/Zr in the ferruginous zone is in contrast to the

consistency of Zr/Hf throughout the profile in general (within the range of parent

meta-granitoid). This suggests that Ti and Zr fractionate from each other and partition


74

between gravel and matrix during extreme weathering and advanced lateritization. This

study demonstrates that Ti, Zr and Th are mobile at a variety of scales, an important

consideration that is often overlooked when calculating element mass flux in intensely

weathered regolith.

5.2 Key words

Zirconium; titanium; thorium; laterite; weathering; regolith;

5.3 Introduction

The mobility of titanium, zirconium and thorium during fluid/rock interaction has been

discussed extensively (Bednar et al., 2004; Cornu et al., 1999; Duvallet et al., 1999;

Kurtz et al., 2000; Langmuir and Herman, 1980; Melfi et al., 1996; Taboada et al.,

2006a; Taboada et al., 2006b). It is accepted that these so-called ‘immobile’ elements

can be highly mobile during hydrothermal alteration (Jiang, 2000; Rubin et al., 1993;

Vanbaalen, 1993); however, their mobility during supergene weathering is still not well

understood. Despite this uncertainty, Ti, Zr and Th are generally considered to be

immobile and are often used as reference elements to evaluate the mass flux of other

elements (Braun et al., 1993; Brimhall et al., 1991; Gouveia et al., 1993). The

assumption of immobility is because: (i) the main host minerals, ilmenite, rutile and

anatase for Ti, zircon for Zr, and most minerals hosting Th are resistant to weathering

(Henderson, 1984; Linnen et al., 2005; Taboada et al., 2006a; Vos et al., 2006); and (ii)

the solubility of Ti, Zr and Th is very low in the absence of strong complexing ligands

(Brookins, 1988; Kabata-Pendias and Pendias, 2001; Linnen et al., 2005; Taboada et al.,

2006a; Tilley and Eggleton, 2005).

The susceptibility of primary minerals during weathering is the main control on the

subsequent mobility and redistribution of Ti, Zr and Th. Titanium has been known to

become mobile during weathering due to: (i) the alteration from sphene/ilmenite to

secondary rutile and anatase at the mineral assemblage scale (Anand and Paine, 2002;

Schroeder et al., 2002), or (ii) as a dissolved element or organo-metallic compounds at

Chapter Five: Redistribution and mobilization of Ti, Zr and Th in an intensely weathered lateritic profile in WA

75

centimetre or even profile scale under tropical weathering conditions (Cornu et al.,

1999). It has also been noted that Zr can be released by the breakdown of amphibole

during weathering, mobilized and sorbed as colloidal (hydr)oxides (ZrO2∙nH2O) to the

surface of goethite in the pores and granules of the gibbsite-rich matrix in bauxitic

profiles (Duvallet et al., 1999; Melfi et al., 1996). Amphibole, however, hosts only trace

concentrations of Zr, the main Zr-hosts in igneous rocks are accessory minerals such as

zircon. Despite the acknowledged longevity of zircon in supergene environments, it is

not totally stable, for examples: (i) corroded zircon and partially dissolved

radiation-damaged (metamict) zircon have been found in extremely weathered lateritic

profiles (Braun et al., 1993; Delattre et al., 2007), and (ii) dissolution of zircon can take

place in systems with a pH below 3 and either a naturally high chloride concentration

(Colin et al., 1993), or where organic matter is present (Hodson, 2002). Thorium usually

occurs in rocks and regolith as: (i) a trace constituent in phosphate, oxide and silicate

minerals, e.g. monazite, cerianite, allanite; (ii) in the rare minerals thorite (ThSiO4) or

thorianite (ThO2), or (iii) it is sorbed onto clays and other soil colloids (Langmuir and

Herman, 1980). Thorium, as well as Ti and Zr, is insensitive to redox change and

mainly occurs in its tetravalent form in natural environmental systems (Buettner and

Valentine, 2012; Langmuir and Herman, 1980). Most Th-hosting minerals are highly

resistant to weathering; hence Th has long been considered a very insoluble and

immobile element in natural systems (Langmuir and Herman, 1980). However,

mobilization and redistribution of Th in lateritic soils where dissolution of thorite, and

transformation and precipitation of thorianite, is further enhanced by the presence of

organic matter has been reported, but Th was still considered the least mobile element at

the profile scale (Braun and Pagel, 1994; Braun et al., 1993; Braun et al., 1998). In

addition, breakdown of primary Th-hosting minerals and adsorption of Th onto clays,

oxyhydroxides and organic matter increases the likelihood of mobilization of Th in

supergene environments (Cromieres et al., 1998; Langmuir and Herman, 1980; Reiller

et al., 2002; Seco et al., 2009; Zhang et al., 2006).

The objective of this study is to investigate the mode of occurrence and mobility of Ti,

Zr and Th in intensely weathered lateritic regolith. Therefore, an intensely weathered


76

lateritic profile (JG) developed on meta-granitoids in Jarrahdale, Western Australia, was

investigated by integrating the geochemistry of bulk regolith samples, particle size

fractions and sequential extractions with in-situ data from electron probe micro-analyzer

(EPMA) and synchrotron X-ray powder diffraction (SXRD) techniques.



This study was performed on a lateritic profile (JG) developed over meta-granitoid

rocks in Jarrahdale, Western Australia. Regolith samples were separated into two

subsample groups by sieving: gravel (>2 mm, represented by suffix ‘g’) and matrix

(<2 mm, represented by suffix ‘m’), with the exception of saprolite (JG1) and mottled

clay (JG2&3), which had only matrix without gravel. Following this division, the

regolith matrix was further separated, using sedimentation and wet sieving methods

(Day, 1965), into the following three size fractions as recommended by the International

Society of Soil Science (ISSS; Marshall, 1947; Marshall, 2003; Prescott et al., 1934):

clay (<2 µm), silt (2-20 µm) and sand (>20 µm). These particle size fractions were

rinsed with deionised water three times, oven dried at 105 °C overnight (together with

matrix and gravel subsamples), and then ground to ≤ 200 µm prior to fusion. In all

chemical procedures high-purity water (≥18 MΩ.cm, Millipore Milli-Q system),

analytical-grade reagents and acid-washed containers were used.

The chemical species and association behaviour of Ti, Zr and Th in matrix from

saprolite (JG1m), the upper part of the mottled clay (JG3m), and ferruginous duricrust

(JG5m) were assessed by a sequential extraction method to determine the element

partitioning during intense weathering and formation of duricrust. An in-house

laboratory reference soil material was prepared together with selected samples. Regolith

trace elements were operationally defined into five species (modified from Hall et al.,

1996): (i) water soluble, adsorbed, exchangeable and carbonate bound (WAE); (ii)

organic matter and sulphide bound (Org); (iii) amorphous Fe-Mn oxyhydroxides bound

(FeAm); (iv) crystalline Fe-Mn oxide bound (FeCry); and (v) residual species (Res).


77

Carbonates are unlikely to be present in the regolith being studied here (due to the low

pH). Sulphides are also scarce in the lateritic regolith, therefore it is assumed that

species Org is mainly hosted by organic matter complexes. A brief summary of the

method used is shown in Table 5.1 and the detailed extraction method and chemical

preparation are listed in the Appendix 11.5. The residual samples and reference

materials were rinsed with deionised water three times, oven dried at 105 °C overnight

and ground to ≤200 µm prior to fusion in order to determine element concentrations.

Table 5.1 Sequential extraction procedures of trace elements in the lateritic regolith

Step Speciation Reagent

i water soluble, adsorbed, and

exchangeable (WAE)

To 1 g of sample, add 20 mL of 1.0 M CH3COONa

(adjust to pH 5 with CH3COOH) at room temperature.

(25°C), shake 6 h, centrifuge for 15min at 3000 rpm;

rinse with 5 mL MilliQ H2O twice, mark 30 mL; repeat.

ii organic matter bound (Org) Add 40 mL 0.1M Na4P2O7 at room temperature (25°C),

shake 1h, centrifuge; repeat; rinse with 5 mL MilliQ H2O

twice, mark 50 mL; repeat.

iii amorphous Fe oxyhydroxide

bound (FeAm)

Add 20 mL 0.25M NH2OH∙HCl in 0.25M HCl, vortex,

water bath at 60°C for 2h, centrifuge; rinse with 5 mL

MilliQ H2O twice, mark 30 mL; repeat.

iv crystalline Fe oxide bound

(FeCry)

Add 30 mL 1.0M NH2OH∙HCl in 25% CH3COOH,

vortex, water bath at 90 °C for 3 h, centrifuge; rinse with

10 mL 25% CH3COOH twice, mark 50 mL; repeat.

v residue (Res) MilliQ water wash residue three times, oven dries at

60 °C. Fuse with 12:22 Norrish flux (Lithium

metaborate/ Lithium tetraborate), dilute with 100 mL

10% HCl.

The fusion beads were made by mixing 0.1 g finely ground sample or reference material

(weighing accuracy 0.1 mg) with 0.7 g 12:22 Norrish flux (lithium metaborate:lithium

tetraborate) and heated in a muffle furnace at 1050 °C for 40 minutes. Duplicate fusion

beads were also made for 10% of the samples to check reproducibility. After cooling,

the fusion beads were dissolved in 100 mL of 10% HCl.


78

Total concentrations of Ti in the fusion beads and extracted solutions were determined

by inductively coupled plasma-optical emission spectroscopy (ICP-OES) whereas

concentrations of Zr and Th were determined by inductively coupled plasma-mass

spectroscopy (ICP-MS); both analyses took place at Genalysis Laboratory Services of

Intertek Commodities in Maddington, Western Australia. Certified international

standard materials, including stream sediment reference material STSD-2, STSD-4

(Canada Centre for Mineral and Energy Technology, CANMET) and an in-house

standard soil material were prepared in the same way as the samples and analysed

together with the samples to check accuracy and precision. The variation between tested

values and certified values was within 10% of the certified values (Appendix 11.2 &

Appendix 11.6). The total concentrations of Ti, Zr and Th in different grain size

fractions are listed in Table 5.2.

Texture, morphology, and phase composition of individual grains were determined

using polished thin sections of air dried and resin impregnated regolith and outcrop

samples. These polished thin sections were examined using a JEOL JSM-6400 scanning

electron microscope (SEM) with a Link analytic energy dispersive spectrometer (EDS),

utilizing both secondary electron (SE) and back-scattering electron (BSE) imaging at

15kV accelerating voltage with a 3 nA beam current. The chemical composition of

selected representative mineral grains was analysed using a JEOL 8530 field emission

electron probe micro-analyzer (EPMA) at 20 kV accelerating voltage and 5 nA beam

current. Software Probe for EPMA from Probe Software Inc. was used for setting up

and analysing the data. Standard references for microprobe calibration were synthetic

glass 612 from National Institute of Standards and Technology (NIST), in-house

standard synthetic rare earth elements (REE) phosphates, rutile, zircon and thorite;

standard Brazil monazite was also analysed with samples for cross checking. All

microscopy analyses were conducted at the Centre for Microscopy, Characterisation and

Analysis (CMCA), University of Western Australia. The detection limits of EPMA of

element concentrations in minerals are listed in the Appendix 11.7.

In addition, selected regolith samples from the ferruginous mottled zone and duricrust


79

were analysed by synchrotron X-ray powder diffraction (SXRD). For these samples, the

matrix fraction was first treated using dispersion and sedimentation to remove the clay

fraction (≤2 µm), and wet sieving to remove coarse sand (>500 µm). The resulting

fraction (2-500 µm) was cleaned in deionised water, oven dried at 50 °C, passed under a

magnet to remove magnetite and added to lithium heteropolytungstate (LST) heavy

liquid to separate the heavy mineral fraction. The heavy mineral fraction was then finely

ground and mounted in capillaries prior to analysis by SXRD at a beam energy of

15.02064 KeV (yielding a wavelength of 0.82616 Å) over an angular range of 4-60° 2θ,

to provide for adequate dispersion/resolution and high peak/background in order to

identify minor constituents. SXRD samples were scanned twice with a 0.5° 2θ-step

difference. SXRD patterns were acquired on beamline 10BM1 (Powder Diffraction) at

the Australian Synchrotron in Melbourne, Australia.

5.4.2 Mass balance calculation

To quantify the net element fluxes from pedogenic weathering, a geochemical mass

balance calculation was used (Brimhall et al., 1991). The formula for normalized

concentration (τi,j) in Equation (1) assumes that an immobile element (e.g., Zr, Th)

behaves conservatively and can be used to correct mobile element concentrations for

volumetric strain during weathering and pedogenesis.

1))((,

,

,

,

, pj

wj

wi

pi

C

C

C

C

ji

(1)

For an immobile element i (τi,j = 0), the volume strain ε can be calculated from the ratios

of density data and the concentration of element i in the regolith and primary rocks:

1)/)(/( ,,, wipiwpwi CC

(2)

In Equations 1 and 2, C represents concentration, ρ represents density, i represents the

immobile element, j represents the element of interest, w represents weathered material

and p identifies parent rock. If τi,j = 0, the element j has behaved conservatively; if τi,j

= −1, the element j is completely depleted during weathering, and positive τi,j values

signify absolute enrichment. If wi, = 0, there is no volume change; wi, > 0 indicates

dilation and wi, < 0 means contraction (collapse).


80

Although Equation 1 provides a tool for estimating element loss or gain for a profile,


regolith and underlying rock and a fully conserved reference element. Therefore, a

conservative immobile reference element is a prerequisite for mass flux calculation.


81

Table 5.2 Concentrations of Ti, Zr and Th in grain size fractions of the JG profile

Element concentrations

depth Ti Zr Th Hf

Unit m wt% ppm ppm ppm

Detection limit 0.01 0.2 0.1 0.1

Methods Fusion/ICP-OES Fusion/ICP-MS Fusion/ICP-MS Fusion/ICP-MS

Upper ferruginous zone 1.5

JG6sand 0.71 246 40.8 7.5

JG6silt 1.12 788 42.6 25.5

JG6clay 0.42 9.6 41.3 0.6

JG6matrix 0.38 445 37.6 12.8

JG6gravel 0.99 349 94.0 9.8

Duricrust 3.0

JG5sand 0.37 182 41.7 5.9

JG5silt 0.97 652 61.3 19.9

JG5clay 0.62 16.9 50.2 0.8

JG5matrix 0.44 291 45.0 8.1

JG5gravel 0.33 349 196 11.0

Ferruginous mottled zone 5.0

JG4sand 0.24 179 23.7 5.8

JG4silt 1.39 1376 66.8 42.4

JG4clay 0.87 35.7 49.1 2.1

JG4matrix 0.52 482 35.6 12.6

JG4gravel 0.37 292 119 7.6

Upper Mottled clay 6.5

JG3sand 0.24 383 36.1 13.3

JG3silt 0.93 914 127 30.4

JG3clay 0.56 171 79.0 5.8

JG3matrix 0.38 341 48.0 10.6

Lower Mottled clay 8.6

JG2sand 0.39 199 153 7.6

JG2silt 0.54 718 287 25.7

JG2clay 0.31 234 220 11.6

JG2matrix 0.17 164 167 4.9

Saprolite 10.0

JG1sand 0.12 128 15.9 4.3

JG1silt 0.23 208 45.1 7.1

JG1clay 0.09 75.6 58.5 2.1

JG1matrix 0.10 105 31.0 2.8

parent meta-granitoids >11.0

JGPR1 0.13 160 16.4 4.12

JGPR2 0.13 159 18.5 5.45


82

5.5 Results

5.5.1 Bulk Ti, Zr and Th concentrations in regolith

5.5.1.1 Abundance of Ti, Zr and Th in the parent rock and regolith

All regolith except saprolite had a higher Ti concentration than the parent

meta-granitoid (0.13 wt%; Figure 5.1a). Furthermore, in the A horizon (0-0.4 m depth)

Ti was more abundant in gravel (0.67-0.89 wt%) than in matrix (0.65-0.74 wt%),

whereas in the ferruginous zone (1.5-5 m depth) Ti was more concentrated in matrix

than gravel, excepting the upper ferruginous zone (JG6, 1.5 m depth). The concentration

of Zr varied significantly in the profile (Figure 5.1b), 105 ppm in the saprolite (10 m

depth), 164-341 ppm in the mottled clay (6.5-8.6 m depth), 291-482 ppm in the

ferruginous zone and 346-506 ppm in the A horizon. With the exception of the saprolite

(105 ppm), the regolith samples were enriched in Zr compared with the parent

meta-granitoid (159 ppm). The concentrations of Zr in the A horizon were higher than

in the regolith below, and the concentrations in matrix were higher than in gravel. Both

Ti and Zr concentrations in matrix generally increased upwards, but Th was extremely

enriched in the lower mottled clay matrix (167 ppm) at 8.6 m depth (Figure 5.1c). This

extreme enrichment of Th is not thought to be an analytical error, as analyses of the

particle size fractions from the lower mottled clay showed similar enrichments in all

fractions (sand, silt and clay). Relative to the average concentration in the parent

meta-granitoid (17 ppm), Th was significantly enriched in ferruginous gravel (up to

196 ppm in duricrust) in contrast to weak enrichment in ferruginous matrix (average

26 ppm), implying strong partitioning of Th into gravel during weathering and

lateritization.

5.5.1.2 Variation of ratios of Ti, Zr and Th with depth

During intense weathering and lateritization processes it is difficult to define an

‘immobile’ element, and thus it is instructive to examine element ratios for potential

immobile elements. The (Ti/Zr)/10 value (Figure 5.2a) varies little from saprolite (1.0)

to ferruginous mottled zone (1.1 in matrix and 1.3 in gravel) and remains close to that of


83

the meta-parent granitoid (ca. 0.8). In the duricrust and upper ferruginous zone,

however, values in both gravel (1.0-2.8) and matrix (0.9-1.5) are more variable,

suggesting Ti and Zr fractionate from each other and are partitioned between gravel and

matrix at advanced stages of lateritization. This fractionation is most apparent in the

upper ferruginous zone. Compared with (Ti/Zr)/10, (Zr/Hf)/10 remains largely constant

and is within the range of parent meta-granitoids (2.9-3.9) throughout the profile

(Figure 5.2b). In contrast, (Ti/Th)/100 is more variable from the saprolite to horizon A

regolith, especially in gravel from the upper part of the profile (Figure 5.2c). The

constant ratios of Ti/Zr in the lower part of the profile and Zr/Hf throughout the profile

suggest that either these elements have undergone a relative mass flux change at a

similar rate, or they may remain residual during weathering at the investigated scale. It

is difficult to envisage similar rates of element mass flux under persistent intense

supergene weathering, and thus it is more likely that Ti, Zr and Hf are effectively

conservative during initial and moderate weathering. Therefore, as Ti/Zr and Zr/Hf

appear less affected by external processes than Ti/Th, Ti/Zr and Zr/Hf may be more

suitable discrimination ratios for moderate weathering. The fractionation between Ti

and Zr in the ferruginous zone suggests that Ti and Zr partition between gravel and

matrix that subject to extreme weathering and strong lateritization. Given the constant

ratio of Zr/Hf consistent with the parent meta-granitoid and the relatively higher

concentrations than Hf, resulting in robust estimates of mass balance, Zr is used as the

reference element. Similar mass balance calculations have been reported in previous

studies in the lateritic bauxitic profiles in Jarrahdale (Brimhall et al., 1992; Brimhall et

al., 1994).


84

Figure 5.1 Variation of Ti, Zr and Th with depth in the JG profile (gravel, ‘g’ is

represented by a triangle and matrix, ‘m’ by a circle in this figure and Figure 5.2). Note

that Th is strongly partitioned into gravel and the increase in Th concentration in the

lower mottled clay matrix does not correlate with any similar spike (positive or negative)

in the Zr or Ti.

Figure 5.2 Variation of Ti/Zr, Zr/Hf and Ti/Th with depth in the JG profile. In order to

aid comparison the ratios have been divided by either 10 or 100 as indicated in the

graphs. Note that Zr/Hf remains within the range defined by the parental meta-granitoid

(bounded by dashed lines; the Ti/Zr range is so small it is represented by one line at this

scale).


85

5.5.2 Mass balance of Ti and Th

Using Zr as the reference element, Ti and Th were enriched throughout the regolith

profile (Figure 5.3). Values of τ(Zr,Ti) slightly increased from 0.24 in the saprolite to 0.56

in the ferruginous duricrust, and then sharply increased to 2.1 in the upper ferruginous

zone, and averaged 1.3 in the A horizon. Compared with Ti, τ(Zr,Th) showed that Th was

significantly enriched in the duricrust and extremely enriched in the lower mottled clay.

In the ferruginous duricrust (3 m depth), ε(Ti) = −0.5, ε(Zr) = −0.2 and ε(Th) = −0.7, all

suggesting regolith collapse. Collapse can be inferred from the increase in

concentrations of the immobile elements (Ti, Zr and Th) because the loss of mobile

elements is not exactly compensated by an inversely proportional decrease in bulk

density during intense weathering and lateritization (Brimhall et al., 1992).

Figure 5.3 Mass balance calculations of Ti and Th against depth in the JG profile, based

on weighted average concentrations in matrix and gravel, using Zr as the reference

element. As τ(Zr,Ti) and τ(Zr,Th) are both above 0 throughout the profile, this implies mass

flux increase of Ti and Th relative to Zr.


86

5.5.3 Mineralogical characteristics of Ti, Zr and Th in the JG profile

5.5.3.1 Occurrence of Ti, Zr and Th in parent meta-granitoids

Representative accessory mineral data from the parent meta-granitoid are presented in

Table 5.3. Titanium was predominantly partitioned into ilmenite (FeTiO3, 28-32 wt%,

Figure 5.4a & b) and sphene (CaTiSiO5, also known as titanite, 20-22 wt%, Figure 5.4b).

Zircon grains (ZrSiO4, Figure 5.4c) had high concentrations of Zr (ca. 45-47 wt%), a

minor amount of Th (ca. 0.1 wt%) and varied concentrations of REE (up to 0.3 wt%

total REE). Thorite (ThSiO4, Figure 5.4d) is the main host for Th (18.9-32.9 wt%), and

also contained 7.7-13.7 wt% Zr, 0.6-0.9 wt% Ti and 3.5-5.2 wt% total REE.

In addition to their main host minerals, significant concentrations of Ti, Zr and Th also

occurred in many widely disseminated accessory minerals, for example, REE-rich

fluorocarbonates (Figure 5.4e & f) contained varied concentrations of Ti

(0.02-0.09 wt%), Zr (up to 0.27 wt%) and Th (0.6-6.4 wt%); magnetite 0.03-0.05 wt%

Ti and up to 0.02 wt% Th; and allanite ca. 0.02 wt% Ti and ca. 0.08 wt% Th.

Another Zr-hosting mineral, probably zirkelite, was observed as a string some hundreds

of microns long and one micron wide associated with quartz in the parent

meta-granitoid (Figure 5.4g & h). As this mineral size range is below the spatial

resolution of the electron microprobe, the chemical data could be separated from the Si

interference originating from the quartz.

87

(a) (b) (c) (d)

(e) (f) (g) (h)

Figure 5.4 Backscatter electron images of Ti-, Zr- and Th- hosting phases in parent meta-granitoids of the JG profile: (a) ilmenite surrounded by apatite;

(b) ilmenite intergrown with sphene; (c) zoned euhedral zircon crystal; (d) thorite crystal rich in Zr (a fracture resulted from electron beam impact); (e)

and (f) REE-bearing fluorocarbonates containing Zr and Th; (g) and (h) probable zirkelite ‘string’ associated with quartz. Compositional analyses of

minerals in (b), (c), (d), (e) and (f) are listed in Table 5.3 (Ap: apatite; Fsp: feldspar; Ilm: ilmenite; Py: pyrite; Qz: quartz; Spn: sphene; Zrk: zirkelite).


88

Table 5.3 Element concentrations of minerals in Figure 5.4 and Figure 5.5 based on

EPMA in parent meta-granitoids and lateritic regolith in the JG profile

Mineral

Sam. No.1 1 2 3 4 5 6 7 8 9

Zrn Thr Fc Fc Ilm Spn Ilm Ilm Rt

El*.(wt%)

2 Fig 5.4(c) Fig 5.4(d) Fig 5.4(e) Fig 5.4(f) Fig 5.4(b) Fig 5.4(b) Fig 5.5(b) Fig 5.5(c) Fig 5.5(d)

Si 15.2 11.3 0.77 5.61 b.d. 13.9 b.d. b.d. 0.25

Zr 45.4 9.79 b.d. 0.27 b.d. b.d. b.d. b.d. 0.38

Ti 0.04 0.76 0.02 0.09 32.2 22.2 33.1 32.1 51.0

Pb 0.07 0.46 b.d. 0.08 b.d. b.d. b.d. b.d. 0.02

Th 0.10 32.9 2.06 2.31 b.d. b.d. 0.01 b.d. 0.14

U 0.22 7.00 0.09 0.23 0.02 b.d. b.d. 0.17 0.14

Al 0.03 0.35 0.48 1.26 b.d. 1.13 b.d. 0.01 2.68

Y 0.13 1.17 0.04 0.08 b.d. b.d. b.d. b.d. b.d.

La b.d. b.d. 20.0 16.2 b.d. b.d. b.d. b.d. b.d.

Ce b.d. 1.83 30.6 23.4 b.d. b.d. b.d. b.d. 0.05

Pr b.d. 0.26 2.51 1.90 b.d. b.d. b.d. b.d. b.d.

Nd 0.11 1.13 5.42 4.33 b.d. b.d. b.d. b.d. b.d.

Sm b.d. 0.45 0.54 0.46 b.d. b.d. b.d. b.d. b.d.

Eu b.d. 0.05 0.19 0.14 b.d. b.d. b.d. b.d. b.d.

Gd 0.02 0.39 0.22 0.15 b.d. b.d. b.d. b.d. b.d.

Dy 0.03 0.17 b.d. b.d. b.d. b.d. 1.04 b.d. b.d.

Yb 0.13 0.25 0.06 0.04 0.06 b.d. b.d. 0.07 0.02

Lu 0.03 0.04 b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Fe 0.64 0.93 1.40 2.90 34.1 0.42 27.9 33.4 3.91

Mg b.d. 0.16 0.36 0.32 0.06 b.d. 0.03 0.07 0.01

Ca 0.08 0.32 4.90 5.26 0.20 19.1 b.d. b.d. 0.04

Sr 0.39 0.01 b.d. b.d. b.d. 0.08 b.d. b.d. b.d.

K 0.01 b.d. b.d. 0.04 b.d. b.d. b.d. b.d. 0.01

P b.d. 0.50 0.02 0.02 b.d. 0.01 b.d. b.d. 0.03

S b.d. 0.03 0.14 0.11 b.d. b.d. b.d. b.d. 0.08

As 0.01 b.d. b.d. b.d. 0.01 b.d. 0.01 0.01 b.d.

F b.d. 1.19 7.09 10.6 b.d. 0.50 b.d. b.d. b.d.

O 33.7 24.4 11.6 14.9 31.5 39.2 30.3 31.3 38.2

total 96.3 95.9 88.5 90.7 98.2 96.6 92.4 97.1 96.9

1No.: Each mineral analysed is allocated with a number; the number allocated is consistent throughout

the thesis. 2Tb, Ho, Er, Tm and Na are below the detection limit (b.d.).

Zrn-zircon; Thr-thorite; Fc-fluorocarbonate; Ilm-ilmenite; Spn-sphene; Rt-rutile.


89

5.5.3.2 Residence of Ti, Zr and Th in lateritic regolith

In the regolith, Ti was mainly hosted by ilmenite, rutile and anatase (Figure 5.5 &

Figure 5.6) and primary sphene was not observed. Ilmenite in the ferruginous mottled

samples (Figure 5.5a & b) contained 30-37 wt% Ti and rutile/anatase (Figure 5.5c, d, e

& f) with ca. 51 wt% Ti. Trace concentrations of Ti were incorporated into zircon

(average 0.12 wt%) and secondary phosphate minerals e.g. rhabdophane (ca. 0.01 wt%)

and florencite (ca. 0.05 wt%), and sorbed onto, or co-precipitated with, Al/Fe oxides

(0.04-0.25 wt%).

Generally, the minerals ilmenite, rutile and anatase are resistant to weathering; however,

partially dissolved ilmenite and rutile (Figure 5.6) was identified in the ferruginous

mottled zone. Powder diffraction patterns (SXRD; Figure 5.6) showed that ilmenite and

rutile, although present in the ferruginous mottled zone (JG4, 5 m depth; Chemical

Index of Alteration (CIA)=95%) are not seen in the ferruginous duricrust (JG5, 3 m

depth; CIA=98%); rather, anatase is present. Evidently there is a change in the mineral

location of Ti between the ferruginous mottled zone and the ferruginous duricrust.

In the regolith the only mineral with a high Zr concentration identified by SEM and

EPMA was zircon. No corroded zircon was observed in the regolith profile, except for

two fractured grains, both of which were half partially dissolved and half crystalline

(Figure 5.7): one grain was in the ferruginous duricrust (Figure 5.7a) and the other in

the A horizon (Figure 5.7b). The apparent dissolution may result from physical

weathering (due to metamict areas) rather than chemical dissolution, because half of the

mineral retained its crystalline structure. In addition to zircon, Zr occurs in submicron

poorly crystalline phases associated with Ce, forming a rim and coating around Al/Fe

matrix in the pore system of the ferruginous duricrust (Figure 5.8). Quantitative analysis

by microprobe revealed that the Zr does not always exist with Si as zircon (ZrSiO4), and

the low sum of oxides <100% in many cases most likely reflects a hydrous state (Table

5.4).

In the ferruginous zone, REE-rich fluorocarbonates were absent and most primary


90

thorite was also weathered. The REE and Th released during weathering partially

precipitated as secondary REE-bearing phosphates such as rhabdophane and florencite

(Figure 5.9a & b). The concentration of Th in these secondary phosphates varied from

0.08 wt% in florencite up to 9.8 wt% in rhabdophane. Trace concentrations of Th were

also hosted by zircon (up to 0.56 wt%, Figure 5.9c), ilmenite (0.01-0.05 wt%) and

anatase (0.05-0.14 wt%). Thorite (ThSiO4) was rare in the regolith; only one

micron-sized grain was observed in the ferruginous duricrust (Figure 5.9d), with a

significant concentration of Th (ca. 45 wt%) and a minor amount of REE

(ca. 0.41 wt%). In addition, up to 5 wt% Th was determined in neoformed poorly

crystalline phases in ferruginous duricrust (Table 5.4).


91

(a) (b)

(c) (d)

(e) (f)

Figure 5.5 Backscatter electron images showing Ti retained as ilmenite and Ti oxides in

the ferruginous mottled zone of the JG profile: (a) and (b) slightly fractured and

decomposed ilmenite; (c) decomposed ilmenite surrounded by Ti oxides; (d), (e) and (f)

are decomposed Ti oxides. Note that (b) is backscatter image from the JEOL

microprobe 8530 and the remainders are backscatter images from the SEM JEOL 6400.

The compositional analyses of minerals (b), (c) and (d) are listed in Table 5.3.


92

Figure 5.6 Diffraction patterns from SXRD showing evidence for transformation of Ti

from ilmenite and rutile in the ferruginous mottled zone (JG4) to anatase in the duricrust

(JG5) of the JG profile (peaks are only labelled for Ant: anatase; Ilm: ilmenite; Rt: rutile.

Off scale peak at 3.342 d-spacing is quartz which was not totally removed by the

separation procedure).

(a) (b)

Figure 5.7 (a) The only partially dissolved zircon grain identified in the duricrust

(circled) and (b) a typical fractured, partially metamict zircon grain in the A horizon

(<1 m depth).

Table 5.4 Element concentrations from EPMA in Figure 5.8 (a) and (b)

Sam Element concentrations* (wt%)

Si Zr Ti Pb Th U Al Ce Gd Fe S F O Total

(a) 0.15 6.04 0.20 0.07 5.15 0.03 2.24 9.14 0.19 20.7 0.05 0.05 12.7 56.8

(b) 0.29 1.07 0.07 0.03 0.60 0.02 12.7 1.44 0.02 32.7 0.20 0.07 22.1 71.4

*Y, La, Pr, Nd, Sm, Eu, Tb, Dy, Ho, Er, Tm, Yb, Lu, Mg, Sr, Na, K, P and As were below the detection

limit of the microprobe; the analysis spot was located at the brightest areas of the Ce-mapping

corresponding with the highest concentrations of Ce.

15

20

25

30

35

3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8

d spacing

Co

un

ts (

tho

us

an

d)

10

15

20

25

Ant

Ilm

RtJG5

JG4


93

(a)

(b)

Figure 5.8 Neoformed poorly crystalline Zr-hosting phases associated with Ce on pore

walls around Al/Fe matrix in the duricrust of the JG profile. This co-occurrence of Zr

and Ce is unassociated with silicates or phosphates and thus is most likely (hydr)oxides

(Table 5.4): (a) edge of a ca. 2 mm nodule cemented with clay matrix in the duricrust;

(b) quartz surrounded by and cemented with Al/Fe matrix in the pore system of

duricrust; (CP is the backscatter image).


94

5.5.4 Grain size distribution of Ti, Zr and Th in the lateritic regolith

In order to exclude the externally sourced Ti and Zr in the A horizon, regolith samples

from 1.5 m to 10 m depth were used to investigate the distribution of Ti, Zr and Th in

different grain size fractions. The results are shown graphically in Figure 5.10 and the

data are presented in Table 5.2.

The silt fraction had the highest concentration of Ti in all regolith samples (Figure

5.10a). In the ferruginous zone, Ti concentration in the clay increased with depth from

1.5 m to 5 m, and then decreased from the mottled clay (6.5 m depth) to the saprolite

(10 m depth).

In all regolith samples except the lower mottled clay zone (8.6 m depth), the silt fraction

had the highest concentration of Zr, whereas the clay fraction contained the lowest

concentration of Zr (Figure 5.10b). In the lower mottled clay zone, silt still had the

highest concentration of Zr and sand had the lowest.

In the ferruginous zone (1.5-5 m depth) gravel had the highest concentrations of Th

whereas the silt fraction contained the highest concentration of Th in the mottled clay

(6.5-8.6 m depth). In the saprolite (10 m depth), the concentration of Th in clay was

slightly higher than both sand and silt (Figure 5.10c).

95

(a) (b)

(c) (d)

Figure 5.9 Forms of Th persisting in regolith samples of the JG profile with/as: (a) secondary REE-bearing mineral rhabdophane; (b) secondary

REE-bearing mineral florencite; (c) substituted with minor concentration in zircon; (d) micron-size grain of thorium orthosilicate mineral (ThSiO4); the

EDS spectra plots in the column on the right correspond to the minerals circled in the SEM backscatter images on the left.

96

Figure 5.10 Grain size distribution of Zr, Ti and Th in the JG profile.


97

5.5.5 Partition of Ti, Zr and Th into different extraction species

Sequential extraction showed that Ti predominantly occurred in the Res, with only trace

amounts occurring in WAE and FeAm species in the upper part of the mottled clay

(Table 5.5). In contrast, Zr was more enriched in Org and FeCry than WAE and FeAm in

all regolith samples, though the Res still contained the highest concentration of Zr.

Unlike both Ti and Zr, the WAE and FeCry species contained more Th than Org and

FeAm and the highest concentration of Th was in the Res. In saprolite, a high amount of

Th was also determined in the WAE.

Table 5.5 Concentrations of Zr, Ti and Th in different sequential extraction species

Element concentrations (ppm)

Ti Zr Th

Detection limit 2.00 0.002 0.001

Method ICP-OES ICP-MS ICP-MS

Saprolite

JG1m_WAE b.d. 0.40 8.78

JG1m_Org b.d. 2.03 3.93

JG1m_FeAm b.d. 0.01 0.38

JG1m_FeCry b.d. 3.26 1.42

JG1m_Res 1592 140 9.30

Upper mottled clay

JG3m_WAE 2.00 0.26 2.67

JG3m_Org b.d. 1.32 0.56

JG3m_FeAm 3.00 0.02 0.19


JG3m_Res 3317 242 29.70

Duricrust

JG5m_WAE b.d. 0.29 1.14

JG5m_Org b.d. 1.72 0.65

JG5m_FeAm b.d. 0.04 0.06


JG5m_Res 4175 225 36.30

5.6 Discussion

5.6.1 Mode of occurrence of Zr and Th in the lateritic regolith

In the ferruginous duricrust, some Zr occurs in poorly crystalline phases associated with

Ce and Th, forming a rim or coating around Al/Fe matrix in the pore systems (Figure

5.8). This occurrence of Zr and Ce is not associated with Si as zircon (ZrSiO4) or P as

rhabdophane (LnPO4, where Ln denotes REE; Table 5.4), demonstrating that Zr was


98

precipitated as oxides or hydroxides (ZrO2∙nH2O or Zr(OH)4) in addition to being

present in zircon in intensely weathered regolith. The geochemical behaviour of Zr

including complexation, mobilization and precipitation, depends on pH and the

presence or absence of organic matter. At a pH of ca. 5, the Zr hydroxy-bicarbonate

(Zr(OH)4-HCO3-H2O) complex, which may be the most significant Zr complex in

natural water, is unstable and possibly decomposes to form Zr(OH)4 (Salminen, 2005;

Vos et al., 2006). When a high organic component is present Zr can also be adsorbed as

colloidal oxides or hydroxides and translocated in the profile (Duvallet et al., 1999).

The sequential extraction (Table 5.5) showed that as well as being hosted by zircon in

residue, Zr was also present in the species of FeCry (7.3 ppm) and Org (1.7 ppm) in the

matrix of ferruginous duricrust (JG5m). Thus, in this case, it is likely that released Zr

was included in neoformed crystalline (hydr)oxides and attached onto the walls of

Al/Fe-rich pores, a process that was enhanced by low pH and the presence of organic

matter.

It is accepted that the geochemical behaviour of Th is dominated by the Th4+

ion

(Langmuir and Herman, 1980), and thus it shows affinity with other tetravalent

elements such as Ce and Zr. This behaviour is seen in this study by the distribution of

Th as a trace component in secondary REE-bearing phosphates (e.g. rhabdophane and

florencite in Figure 5.9) in weathered lateritic regolith. The high concentrations of Th

(5 wt% in Table 5.4) determined in neoformed poorly crystalline phases in ferruginous

duricrust, suggest the formation of insoluble Th (hydr)oxides associating with Ce and

Zr. Trace amounts of Th, associated with the WAE (1.14 ppm) and FeCry species

(1.62 ppm), in the matrix of ferruginous duricrust (Table 5.5) suggest that Th was

affected by sorption or co-precipitation with Al- and Fe- oxides.

5.6.2 Sources of Zr in poorly crystalline phases in duricrust

In this study, the formation of poorly crystalline (hydr)oxide phases containing Zr

indicates some mobility of Zr in the supergene weathering environment. However, the

distance that Zr is mobilized from its original location is not clear, but it is reasonable to

propose that this was less than the sampling scale (centimetre scale) due to: (i) the very

low solubility of Zr(OH)4; (ii) an almost constant Zr/Hf ratio consistent with the parent

meta-granitoid; and (iii) the absence of eroded zircon grains. This then poses the

question: where did the mobilized Zr come from? Limited remobilization of Zr in

supergene environments may occur in strongly acidic and organic-rich media in podzols,


99

or F/Cl-rich coastal profiles (Colin et al., 1993). The susceptibility of zircon to

weathering can also be enhanced by mechanical fracturing during deformation and

damage to the crystal lattice (metamictization) from radioactive decay of incorporated U

and Th (Rubin et al., 1993). Therefore, old zircons (with high U and Th concentrations)

in paleosols are more likely to be susceptible to dissolution than undamaged zircon.

The appearance of metamict areas in zircon resembles the effects of chemical

weathering (Rubin et al., 1993) and is visible using a petrological microscope. The

majority of zircon grains in this regolith profile, however, still retained typical crystal

morphology and only two containing metamict areas were observed (Figure 5.7). Thus,

it is unlikely that the Zr contained in the neoformed (hydr)oxides was released from

zircon breakdown. In contrast, other igneous phases incorporating trace amounts of Zr,

such as thorite (up to 13 wt% Zr) and REE-bearing fluorocarbonates (ca. 0.07 wt% Zr

and 7.3 wt% F), would be more likely to break down, thus releasing Zr from the parent

meta-granitoids at the initial stages of weathering. The mobility of Zr would be further

enhanced by the F-rich solution released by the REE-rich fluorocarbonates during

breakdown. Therefore, the residence of Zr not only in zircon but also in other primary

igneous minerals, and the amount and distribution of these mineral phases in the parent

rock, are likely to be significant controls on the mobility of Zr in the lateritic regolith.

5.6.3 Partitioning of Th between gravel and matrix

Unlike Zr, Th has a more complicated mode of occurrence in the lateritic regolith.

Though most Th was contained in the resistant mineral phases (revealed by significant

concentrations of Th hosted by the Res in Table 5.5), trace to minor amounts of Th were

also detected in the WAE, Org, FeAm and FeCry species. In addition, strong partitioning

of Th into gravel in the ferruginous zone reflects the local translocation and

redistribution of Th in the profile. Similar enrichment of Th in iron nodules (gravel) was

also observed in the Nsimi lateritic profile, Cameroon (Braun et al., 2005). However,

the enrichment of Th in gravel is not consistent with the concentrations of Ti and Zr

determined in this study, as concentrations of Ti and Zr were not consistently higher in

gravel than matrix. This indicates that zircon, ilmenite, rutile and anatase were not the

only hosts for Th in gravel. In the duricrust Th also precipitated as Th (hydr)oxides

associating with Zr and Ce (Table 5.4); in the upper ferruginous zone and the A horizon

Th-hosting REE-rich phosphates distributing into iron nodules had been found (Figure

5.9). In addition, from the ferruginous mottled zone to the A horizon, pH varies from


100

4.4 to 5.3, concentrations of Al and Fe oxides in gravel are higher than in matrix; both

features may also contribute to the strong partitioning of Th into surface gravel. This is

in agreement with the strong sorption of Th by hematite and gibbsite within a similar

pH range (Cromieres et al., 1998; Zhang et al., 2006). In the mottled clay and saprolite

the silt size fraction had the highest concentrations of Th, which is likely to be the result

of the Th-hosting minerals being dominantly within the range of the silt size fraction

originally (e.g. secondary REE-phosphates). However, the reason for the abnormal

enrichment of Th in the lower mottled clay zone is not clear. To understand more fully

the abnormal accumulation of Th in the lower mottled clay and to evaluate whether this

is observed elsewhere, further research is needed.

5.6.4 Mobility of Ti in the JG profile

The weathering-resistant minerals ilmenite and anatase are the main hosts of Ti in the

profile (Figure 5.5); the absence of the igneous mineral, sphene (present in the parent

meta-granitoid), and the presence of fractured and eroded grains of ilmenite and rutile

are all evidence that Ti is mobile at the mineral assemblage scale. Trace amounts of Ti

were found in the WAE (2 ppm) and FeAm (3 ppm) species in the upper mottled clay

(Table 5.5) as well as with neoformed Zr-(hydr)oxide phases in ferruginous duricrust

(0.07-0.20 wt% in Table 5.4). These amounts, however, are negligible in comparison

with the concentration of Ti in the Res in the upper mottled clay (3317 ppm) and in the

duricrust (4175 ppm). A similar result, where Ti-phases were trapped within neoformed

clay minerals, was noted by Malengreau et al. (1995). In addition, enrichment of Ti

increased in the fine soil fractions (silt and clay) with increasing weathering intensity

(decreasing depth). Enrichment of Ti in the fine fraction was also found in ten

weathering and pedogenetic soil profiles developed on granitic rocks by Taboada et al.

(2006a). The relatively mostly constant Ti/Zr from saprolite to ferruginous mottled zone

suggests that Ti remains largely conservative at the sampling scale in the lower part of

the profile, although mobility of Ti at the mineral scale has been revealed. This implies

that the sphene and ilmenite in parent meta-granitoid break down and are replaced by

secondary ilmenite and rutile in-situ (or nearly so) in the lower part of lateritic regolith

during initial to moderate weathering. The secondary ilmenite and rutile are then altered

into anatase and thus constrain any further mobility of Ti. This is supported by the

intergrowth of ilmenite and Ti oxides in the ferruginous mottled zone (Figure 5.5c).

Similar mineral transformations in intensely weathered lateritic regolith, from


101

ilmenite/rutile to anatase, have been reported before by Anand et al. (1991; 2002).

Fluctuation of Ti/Zr (in comparison with almost constant Zr/Hf) in the ferruginous

duricrust and upper ferruginous zone indicates that Ti fractionates from Zr during

extreme weathering and advanced lateritization. High concentrations of Ti partitioned

into iron nodules in the upper ferruginous zone (Table 5.2) are consistent with

cementation of Fe oxides and formation of iron nodules in the upper ferruginous zone.

This appears to have resulted in further alteration of ilmenite and rutile into anatase,

which was cemented with Al/Fe oxides and incorporated into iron nodules. Although

zircon and rutile from aeolian input are relatively stable in the upper part of regolith

profile (Brimhall et al., 1992; Brimhall et al., 1988), redox change during lateritization,

and associated changes in pH and Eh have a profound effect on the mobilization and

translocation of Ti that is further enhanced by the involvement of organic matter. This is

supported by the experiment of Thompson et al., (2006) who reported that 10% of total

Ti in a basaltic soil was mobilized as colloids at peak dispersion (related to a change in

pH accompanying redox oscillation); furthermore, Ti and Zr were also observed to be

mobile in the uppermost meter of lateritic regolith in Cameroon, attributed to the

presence of organic colloids (Braun et al., 2005).

5.6.5 Geological parent mineralogy vs. weathering conditions

It is widely known that the factors influencing the mobility of trace elements include: (i)

initial concentration and mineralogical host in the parent rocks, (ii) the susceptibility of

these hosts to subsequent alteration, and (iii) the ability of the solution to transport the

elements released (Rubin et al., 1993). In this study, the susceptibility of primary

igneous hosts of Zr, Ti and Th to weathering (e.g. thorite for Th and Zr, sphene for Ti

and Th, REE-rich fluorocarbonates for REE, Zr and Th) fundamentally controls the

subsequent mobility of these elements, and changes their abundance at the early stages

of weathering. Breakdown of thorite, sphene and REE-rich fluorocarbonates releases Ti,

Zr and Th into solution. Once in solution, Ti, Zr and Th re-enter the solid phase by

formation of new secondary minerals (e.g. ilmenite, rutile, rhabdophane, (hydr)oxide).

Formation of these secondary minerals with very low solubility further limits

mobilization of Ti, Zr and Th. As weathering proceeds, the initial control by igneous

host minerals in the parent rocks diminishes in importance, rather the weathering

intensity and characteristics of the solutions present play an increasingly important role

in translocation of Ti, Zr and Th. Low pH (3.2-5.3 throughout the profile), extreme


102

weathering intensity (CIA 65%-99%) and the presence of organic matter enhance the

mobility of Ti, Zr and Th. However, the greater stability of zircon relative to ilmenite

and rutile results in alteration of ilmenite and rutile into anatase, leaving zircon as a

residual mineral. The presence of organic matter, clay minerals and Fe

oxides/oxyhydroxides enhances the mobility of Th by formation and sorption of

complexes at the profile scale.


A geochemical and mineralogical study of the mobility and mode of occurrence of Ti,

Zr and Th in the intensely weathered lateritic profile at Jarrahdale, Western Australia,

was conducted. The mobilization and redistribution of Zr and Th at the sampling scale

was revealed by neoformed poorly crystalline Zr, Ce and Th (hydr)oxide phases

attaching onto the walls of Al/Fe-rich pores in the ferruginous duricrust. The source for

Zr and Th in these neoformed phases is proposed to be the breakdown of thorite and

REE-rich fluorocarbonates during the early stages of weathering. Distribution into

secondary REE-bearing phosphates (e.g. rhabdophane and florencite) as a trace

component in the regolith showed translocation of Th at the mineral assemblage scale,

whereas strong partitioning of Th into gravel rather than matrix reflects redistribution of

Th at the profile scale. Absence of primary sphene in the regolith and dissolution of

ilmenite and rutile in the ferruginous mottled zone suggest mineral transformation from

sphene, ilmenite and rutile to anatase at the mineral assemblage scale during intense

weathering. The limited range of Ti/Zr from saprolite to ferruginous mottled zone

indicates that Ti is mostly conservative during moderate weathering despite varying in

concentration. The fluctuation of Ti/Zr in the duricrust and upper ferruginous zone

suggests that Ti and Zr fractionate from each other and partition between gravel and

matrix during extreme weathering and advanced lateritization. Therefore, these

commonly considered immobile elements are mobile at a variety of scales, and special

attention should be paid when using these elements to calculate the flux mass,

especially under intensely weathered conditions or where there are particle size sorting

transport processes.

103

6 Distribution and fractionation of REE in intensely weathered

lateritic profiles in Western Australia

6.1 Abstract

Three intensely weathered lateritic profiles (GE, MQ I and MQ II) developed on

granitoids with dolerite dykes in Western Australia were studied to investigate

geochemical behaviour and fractionation mechanisms of rare earth elements (REE)

during intense weathering and lateritization. In three profiles, regolith developed from

the granitoid rather than the dolerite was confirmed by chondrite normalized REE

distribution patterns. Substantial depletion of REE in the regolith was observed,

especially in the GE profile. Chondrite normalized REE distribution patterns of regolith

from three profiles showed light REE (LREE)-enrichment, coupled with higher

depletions of LREE than HREE relative to the parent granitoids.

Monazite, allanite, apatite, zircon, ilmenite and sphene were important REE-hosting

mineral phases in the parent granitoids, with REE-rich fluorocarbonates restricted to the

MQ parent granitoids. Residual monazite and secondary rhabdophane were important

phosphates for retention of LREE, whereas zircon and ilmenite were significant HREE

selective hosts, in weathered MQ regolith. REE released by breakdown of

easy-weathering LREE-rich allanite and fluorocarbonate in parent granitoids at the early

stages of weathering may be partially leached away, or alternatively, be retained in the

regolith by formation of secondary phosphates, e.g. rhabdophane, and hence limited

further mobility.

In addition to being hosted by mineral phases, REE were also retained in weathered

regolith by association with clay minerals, Fe oxides/oxyhydroxides and organic matter.

Among five species of sequential extraction, the water soluble (including adsorbed and

exchangeable) species hosted up to 7.9% of total REE in the C horizon regolith of MQ I

profile; the amorphous Fe oxyhydroxide species contained 3.7% of total REE in the

duricrust of GE profile. A positive Ce anomaly (Ce*=6.1) in the duricrust of GE profile

was likely related to the redox change during formation of the duricrust. All of these

observations suggest that REE can mobilize during weathering and lateritization, to the

extent of becoming highly depleted in intensely weathered lateritic regolith. The

abundance, stability and composition of secondary LREE-rich phosphate minerals and

residual HREE-selective weathering-resistant minerals may control the fractionation of


104

REE in intensely weathered regolith.

6.2 Key word

Rare earth elements; laterite; weathering; regolith; fractionation;

6.3 Introduction

Chemical weathering of rocks and the formation of soils are important geochemical

processes, contributing to the redistribution of rare earth elements (REE) in the Earth’s

surface environment. A series of well documented studies have shown that REE can

mobilize and fractionate during weathering (Aubert et al., 2001; Braun et al., 1993;

Duddy, 1980; Feng, 2011; Harlavan and Erel, 2002; Harlavan et al., 2009; Nesbitt,

1979). Although REE have been widely studied, the geochemical behaviour of REE

during weathering cannot be generally applied because of: (i) the wide range of

REE-bearing minerals and their variable concentrations of REE; (ii) different

susceptibility of these minerals to solutions and variable solution chemistry; and (iii)

location-specific physicochemical and biological factors during weathering (Bao and

Zhao, 2008; Price et al., 1991).

Two factors are believed to be the main controls on the mobilization and fractionation

of REE: the type of primary REE-bearing minerals in the protolith and the weathering

conditions (Braun et al., 1990; Braun et al., 1998; Condie et al., 1995; Nesbitt, 1979).

However, the behaviour and fractionation mechanisms of REE during weathering are

still not fully understood. For example, preferential enrichment of LREE over HREE in

lateritic regolith has been widely reported (e.g. Braun et al., 1993; Braun et al., 1998;

Ndjigui et al., 2009), whereas stronger enrichment of HREE than LREE has also been

found in lateritic profiles (e.g. Beyala et al., 2009; Braun et al., 1990; Viers and

Wasserburg, 2004). Lateritic regolith represents one of the most common superficial

formations in the tropics, and is commonly diachronous, extending over tens of millions

of years (Dequincey et al., 2006). Lateritization is therefore of particular significance in

the study of translocation and fractionation of REE (Ji et al., 2004). Therefore, three

intensely weathered lateritic profiles developed over granitoids with a cross-cutting

dolerite dyke in Western Australia were investigated in this study. The aims are to: (i)

determine the abundance and residence of REE in the parent granitoids and lateritic

regolith; (ii) improve the understanding of the geochemical behaviour and fractionation

Chapter Six: Distribution and fractionation of REE in intensely weathered lateritic profiles in WA

105

mechanisms of REE during intense weathering and lateritization.

6.4 Methods and materials


This study was performed on regolith samples from three lateritic profiles (GE, MQ I

and MQ II) developed over granitoids with dolerite dyke in Western Australia (Chapter

Three). Regolith samples were separated into two subgroups based on grain size: gravel

(>2 mm, represented by suffix ‘g’) and matrix (<2 mm, represented by suffix ‘m’). The

exception to this subdivision was the duricrust (GE6) in the GE profile, a very hard

cemented material without corresponding loose matrix. Subsamples of gravel and

matrix and crushed duricrust were oven dried at 105 °C overnight, then ground to ≤200

µm prior to fusion in order to determine element concentrations. Following this division,

regolith raw bulk matrix from the MQ II profile was further separated into the following

three size fractions: clay (<2 µm), silt (2-20 µm) and sand (>20 µm) using the

sedimentation pipette and wet sieving methods (Day, 1965). Particle size fraction limit

recommended by the International Society of Soil Science (ISSS) has been adopted in

Australia (Marshall, 1947; Marshall, 2003; Prescott et al., 1934). Different particle size

fractions were rinsed with MilliQ water three times, oven dried at 105 °C overnight,

then ground to ≤200 µm prior to fusion.

To investigate the chemical species and association behaviour of trace elements, a

sequential extraction procedure was performed. Saprolite matrix (GE1m) and ground

duricrust (GE6) from the GE profile and regolith matrix (MQ1m, 3.6 m depth; MQ5m,

1.1 m depth; MQ8m, 0.5 m depth) from the MQ I profile were selected. An in-house

laboratory reference material was analysed together with the selected samples. Regolith

trace elements were operationally divided into five species (modified from Hall et al.,

1996): (i) water soluble, adsorbed, exchangeable and carbonate bound (WAE); (ii)

organic matter bound (Org); (iii) amorphous Fe-Mn oxyhydroxide bound (FeAm); (iv)

crystalline Fe-Mn oxide bound (FeCry); and (v) residual species (Res). Since carbonates

are unlikely to be present in the regolith being studied here due to low pH, they are not

considered relevant in this work, and the acronym AEC

(adsorbed-exchangeable-carbonate) used by Hall et al. (1996) is not used. Sulfides are

also scarce in the lateritic regolith, therefore it is assumed that species Org is mainly

hosted by organic matter complexes. A brief summary of the sequential extraction


106

procedures is shown in Table 5.1 and the detailed extraction method and chemical

preparation are listed in the Appendix 11.5. The residual samples and reference

materials were rinsed with MilliQ water three times, oven dried at 105 °C overnight and

ground to ≤200 µm prior to fusion.



(lithium metaborate:lithium tetraborate) and heating in a muffle furnace at 1050 °C for

40 minutes. Duplicate fusion beads were made on 10% of samples to check preparation

errors. After cooling, the fusion beads were dissolved in 100 mL of 10% analytical

grade HCl. The trace elements, including REE, in fusion beads of the gravel and matrix,

were analysed after an additional 10-fold dilution with 10 ppb Rh/Ir solution in 10 ml

polypropylene tubes using a Perkin-Elmer Elan 6000 inductively coupled plasma-mass

spectrometry (ICP-MS) at the University of Western Australia. Trace elements,

including REE, in sequential extractions were analysed at the Genalysis Laboratory

Services of Intertek Commodities in Maddington, Western Australia. Certified

international standard materials, including stream sediment reference material STSD-2,

STSD-4 (Canada Centre for Mineral and Energy Technology, CANMET) and an

in-house standard reference were prepared in the same way as the samples and analysed

together with samples to check the accuracy and precision. The variations of REE

between tested values and certified values were within 10% from La to Er and Yb,

except Tm and Lu (variation was within 20%, Appendix 11.6). Therefore, when

calculating fractionation of REE, La/Sm and La/Yb were used rather than La/Lu. The

concentrations of REE in gravel and matrix of regolith samples from three profiles are

listed in Table 6.1.






15kV accelerating voltage with a 3 nA beam current. Semi-quantitative modal

abundances of REE-bearing accessory minerals in parent rocks were calculated based

on SEM-BSE images of polished thin sections and chemical maps produced by EDS.

The relative volume percentages of REE-bearing minerals were calculated and selected

mineral density data (Deer et al., 1992) were used to convert volume percentage to


107

weight percentage. Given that the region of interest had more REE-bearing minerals,

calculated weight percentages will be overestimated. Although the accessory mineral

abundances obtained this way are only semi-quantitative, they can be used as clues for

assessing the mineral control and fractionation of REE in the parent rock. The chemical

composition of representative REE-bearing minerals was analysed by electron probe

micro-analyzer (EPMA, JEOL 8530) at 20 kV accelerating voltage and 5 nA beam

current. Software Probe for EPMA from Probe Software Inc. was used for setting up

and analysing the data. Standard references for microprobe calibration were synthetic

glass 612 from the National Institute of Standards and Technology (NIST), in-house

standard synthetic REE phosphates, rutile, zircon and thorite; standard Brazil monazite

was analysed with samples for cross checking. All microscopy analyses were conducted

at the Centre for Microscopy, Characterisation and Analysis (CMCA), University of

Western Australia. The detection limit of the EPMA for mineral analysis is listed in the

Appendix 11.8.

6.4.2 Calculation methods

6.4.2.1 Fractionation of REE and anomalies of Ce and Eu

In order to study the fractionation of REE, three groups are identified (Henderson,

1984): the light REE (LREE; from La to Nd), the middle REE (MREE: from Sm to Ho)

and the heavy REE (HREE: from Gd to Lu). Regolith REE distribution patterns are

normalized to average chondrite values (Anders and Grevesse, 1989) in order to show

the fractionation of REE during lithogenical weathering and the difference of REE

distribution between granitoids and dolerites; in addition, REE in regolith are also

compared with the parent rock (PR) in order to reveal relative enrichment or depletion.

The normalized ratios (La/Sm)PR and (La/Yb)PR are used for identifying fractionations

between LREE-MREE and LREE-HREE using the composition of the parent rock as a

reference. Cerium and Eu anomalies are calculated using the following equations

(subscript PR refers to parent rock):

Ce*=(Ce/CePR)/[(La/LaPR)

0.5×(Pr/PrPR)

0.5] (1)

Eu*=(Eu/EuPR)/[(Sm/SmPR)

0.5×(Gd/GdPR)

0.5] (2)

6.4.2.2 Weathering intensity-Chemical Index of Alteration (CIA)

To evaluate the intensity of chemical weathering quantitatively, the Chemical Index of


108

Alteration (CIA) (Nesbitt and Young, 1982) was used. The CIA calculates loss of

mobile elements relative to Al in bulk samples, providing a single parameter estimate of

the intensity of chemical weathering. The formula (Nesbitt and Young, 1982) is:



carbonates).

6.4.2.3 Mass balance calculation

To quantify net elements fluxes from pedogenic weathering, a geochemical mass

balance calculation was used (Brimhall et al., 1991). The formula for normalized

concentration (τi,j) in Equation (4) assumes that an immobile element (e.g., Zr, Th)

behaves conservatively during weathering and pedogenesis.

1))((,

,

,

,

, pj

wj

wi

pi

C

C

C

C

ji

(4)






Equation (4) provides a tool for estimating elemental loss or gain for a profile; however,


regolith and the underlying rock and a fully conserved reference element. Although the

mobility of Ti, Zr and Th is subject to debate (e.g. Braun et al., 1993; Cornu et al.,

1999), Zr is still considered to be conservative in the profiles studied based on the

consistent ratio Zr/Hf in the GE profile and the observation of uncorroded zircon

minerals by SEM in the MQ profiles.

6.4.2.4 Mass loading of REE in grain size fraction

To determine an element’s partitioning into different grain size fractions, a mean

element mass loading was calculated based on its concentration in a selected grain size

of known mass percentage (Sutherland, 2003).


109

GSFloading 100 (X i GSi

X i GSii1

n

)

(5)

Where:

Xi is the concentration of REE (ppm) in an individual grain size fraction (e.g. <2 um);

GSi is the mass percentage of an individual fraction, which has limits of 0–100%.

GSFloading is the element mass loading in a selected grain size and the summation of

GSFloading for each sample equals 100%.

Four classes of grain sizes (clay, silt, sand and gravel) were defined in all regolith in the

MQ II profile. Thus, if the REE concentration for a given fraction is very high but it

forms only a small portion of the overall sample mass, the contribution of this fraction

to the total sample REE mass will be minimal.

6.5 Results

6.5.1 Geochemical data of REE

6.5.1.1 REE concentrations and normalized patterns

In all three profiles, uniform REE distribution patterns of regolith samples normalized

to chondrite confirmed that the regolith was developed from weathering of the granitoid

rather than the dolerite (Figure 6.1, Figure 6.2 & Figure 6.3), in agreement with the

major elemental results discussed in Chapter Four; thus, granitoid GEPR2 was selected

as the parent rock in the GE profile, while in both MQ profiles, the average

concentrations of granitoids MQPR1 and MQPR2 were used as the concentrations of

the parent rock. The parent granitoids in three profiles had higher sum of concentrations

of REE (ΣREE, 137 ppm of GE parent granitoid and average 201 ppm of MQ parent

granitoids) than the dolerite (GE dolerite GEPR3 ΣREE=38 ppm; MQ dolerite MQPR3

ΣREE=61 ppm).

In the GE profile, a high deficiency of REE was observed throughout the regolith

profile compared with the parent granitoid (Table 6.1): Up to 96% ΣREE was lost in the

mottled clay (GE3, 10 m depth), followed by 83% loss in the duricrust (GE6, 3.5m

depth); and the saprolite (GE1, 12m depth) showed the least depletion of ΣREE (64%).

Although the REE distribution patterns of regolith normalized by chondrite showed a

LREE-enrichment, higher depletions of LREE than HREE relative to the parent


110

granitoids were observed, apart from a positive Ce anomaly (Ce*=6.1) in the duricrust

(GE6, 3.5m depth). The matrix showed more intense depletion of REE than the gravel

in the A horizon, whereas the matrix was less REE depleted than the gravel in the

subsurface regolith (<0.5 m depth) of the GE profile.

In the MQ I profile, the matrix from the A (MQ8-9, <0.5 m depth) and C horizons

(MQ1, 3.6 m depth) was enriched in ΣREE compared to the parent granitoid (Table 6.1).

The matrix generally had higher ΣREE than the gravel, apart from two B horizon

reoglith samples (MQ5-6). Gravel and matrix from the B horizon (0.9-3.3 m depth) lost

more ΣREE than the A (0.2-0.7 m depth) and C horizons (3.6 m depth). The REE

distribution patterns of regolith normalized by chondrite showed LREE-enrichment with

the highest depletion of REE at 1.1 m depth (MQ5). The matrix from MQ5 regolith had

depleted 76% ΣLREE, 67% ΣMREE and 58% ΣHREE, higher than the gravel with

depletion of 69% ΣLREE, 64% ΣMREE and 51% ΣHREE relative to the parent

granitoid.

In the MQ II profile, matrix from the A horizon (MQ15, 0.08 m depth) was enriched in

ΣREE (22%), whereas the regolith below was depleted in ΣREE compared with the

parent granitoid (Table 6.1). The REE distribution patterns normalized by chondrite

showed LREE-enrichment and a general reduction of REE with depth, with a significant

loss of REE below 0.6 m depth. The abundance and distribution of REE in regolith

changed at 1.1 m depth of MQ I profile and at 0.6 m depth of MQ II profile, implying a

mass movement involved in the upper parts of both profiles, and this is in agreement

with the discussion in Chapter Four based on major elemental analyses.

6.5.1.2 Fractionation of REE during intense weathering

In the GE profile, the LREE/MREE ratio (La/Sm)PR in regolith fluctuated with depth

but was mostly below 1.0 throughout the profile, except GE2 (1.2 in gravel and 1.4 in

matrix) at 11 m depth (Figure 6.4a and Table 6.1). The fractionation between La and

Sm in gravel was more severe than or similar to the corresponding matrix in most

regolith, apart from the upper mottled clay (GE4, 8.4 m depth). All regolith had

(La/Yb)PR ≤ 0.7, coupled with high deficiency of REE, suggesting higher loss of LREE

than HREE discussed above. The fluctuation of (La/Yb)PR was correlated negatively to

the index of weathering intensity (e.g. CIA) of regolith samples.

In the MQ I profile, fractionation between La and Sm first increased with depth, and


111

then decreased below 1.1 m depth (Figure 6.4b and Table 6.1). Fractionation between

La and Yb was more complexed than La and Sm, especially in the matrix. The highest

depletion of La was at 1.1 m depth with (La/Yb)PR = 0.5, whereas the highest depletion

of Yb at 3.3 m depth with (La/Yb)PR = 1.7.

In the MQ II profile, (La/Sm)PR was below 1.0, except in the surface matrix (MQ15,

0.08 m depth, (La/Sm)PR = 1.1) (Figure 5.4c and Table 6.1). Similar to (La/Sm)PR,

(La/Yb)PR in the B and C horizons was below 1.0 and up to 2.1 in matrix and 1.1 in

gravel of the A horizon.


112

Figure 6.1 REE distribution patterns of (a) rocks and regolith samples normalized by the


lateritic GE profile2 (GEPR2-granitoid; GEPR3-dolerite; GEA3-A horizon, 0.23 m

depth; GE6-duricrust, 3.5 m depth; GE5-ferruginous zone below duricrust, 7.0 m depth;

GE3-mottled clay, 10 m depth; GE1-saprolite, 12.5 m depth; ‘g’ denotes gravel and ‘m’

denotes matrix; from GE6 to GE1, weathering intensity decreased; GE2 and GE4

showed similar patterns with GE3, and GEA1 and GEA2 showed similar patterns with

GEA3, so were not plotted).

2 Several REE e.g. Tm appear to have somewhat abnormal values in the REE patterns of GE and MQ profiles

(normalized to the parent granitoid), which probably result from the compounded errors from the measurements and

the plots without log transformation (due to some extremely low concentrations of REE).


113

Figure 6.2 REE distribution patterns of (a) rocks and regolith samples normalized by the


lateritic MQ I profile (MQPR2-granitoid; MQPR3-dolerite; MQ7-A horizon, 0.7 m

depth; MQ5 and MQ3-B horizon, 1.1 m and 2.8 m depth respectively; MQ1-C horizon,

3.6 m depth; ‘g’ denotes gravel and ‘m’ denotes matrix; from MQ6 to MQ1, weathering

intensity decreased; MQ2 and MQ4 showed similar patterns with MQ3, and MQ8 and

MQ9 showed similar patterns with MQ7, so were not plotted).


114

Figure 6.3 REE distribution patterns (a) rocks and regolith samples normalized by the


lateritic MQ II profile (MQPR2-granitoid; MQPR3-dolerite; MQ15-A horizon, 0.08 m

depth; MQ13 and MQ12-B horizon, 0.6 m and 1.1 m depth respectively; MQ10-C

horizon, 2.0 m depth; ‘g’ denotes gravel and ‘m’ denotes matrix; from MQ13 to MQ10,

weathering intensity decreased; MQ11 showed similar patterns with MQ12, and MQ14

showed similar patterns with MQ15, so were not plotted).


115

Figure 6.4 normalized ratios (La/Sm)PR (LREE/MREE) and (La/Yb)PR (MREE/HREE)

and CIA of regolith samples against depth in three lateritic profiles: (a) GE profile; (b)

MQ I profile; (c) MQ II profile (dashed vertical line refers to no fractionation relate to

the parent granitoid).


116

6.5.1.3 Variation of REE with lateritization degree

The variation of REE in the GE profile is compared with a second index of weathering

intensity by using the concentration ratio of SiO2/(SiO2+Al2O3+Fe2O3) (S/SAF, Hill et

al., 2000) and to the degree of lateritization by using SiO2-Al2O3-Fe2O3 ternary plots

(Schellmann, 1981), and plotted in Figure 6.5.

The very low concentrations of REE in comparison to the parent rock suggested a major

loss during weathering and lateritization. Cerium did not show an anomaly except at the

strong lateritization stage where Ce*=6.1 in the duricrust. This Ce enrichment co-occurs

with total iron enrichment at a redox gradient in the duricrust (Chapter Four). Ytterbium

(Yb, a HREE) was less depleted than La, Ce, Sm and Y compared with parent

granitoids. Different concentration ranges of each REE among gravel, matrix and parent

granitoids suggested that depletion and partitioning of REE into different size grains

occurred during weathering and lateritization. Compared with parent grantioids,

(La/Yb)PR was lower than (La/Sm)PR, suggesting that fractionation between LREE and

HREE was stronger than fractionation between LREE and MREE under intense

leaching conditions. The (Y/Ho)PR was relatively consistent with changes in S/SAF,

reflecting that these two elements which generally show similar geochemical behaviour,

didn’t fractionate significantly; however, (Y/Ho)PR of regolith average ca. 0.8 relative to

the parent granitoid indicated depletion of both elements during intense weathering.

Since the MQ I and MQ II profiles were still in a weak lateritization stage and greatly

influenced by transported materials, variations of REE against S/SAF do not provide

any more information than the variation of REE against depth already presented above,

so are not plotted.

6.5.1.4 Mass balance calculation of REE

Mass balance of REE is calculated based on the weighted average concentrations of

REE in the matrix and gravel samples with Zr as the reference element, and plotted in

Figure 6.6.

In the GE profile (Figure 6.6a), the A horizon regolith (above 0.5 m depth) had τ(Zr, REE)

values ranging from −0.96 to −0.80; and the ferruginous zone (3.5-5 m depth) had

τ(Zr, REE) ranging from −0.99 to −0.87, apart from τ(Zr, Ce) = 0.9 in the duricrust (3.5 m

depth). The τ(Zr, REE) slightly increased from La to Yb in the same regolith above lower


117

mottled clay (11.4 m depth); in the lower mottled clay and saprolite, τ(Zr, Sm) and τ(Zr, Gd)

had the lowest values.

In the MQ I profile (Figure 6.6b), REE were less depleted than the GE profile, and even

slightly enriched in the regolith at 0.5 m depth with τ(Zr, Ce) = 0.09, τ(Zr, Pr) = 0.05 and

τ(Zr, Sm) = 0.05. In the B horizon (0.9-3.3 m depth), REE (not including Lu) were more

depleted (τ(Zr, REE) between −0.72 and −0.19) than the A (τ(Zr, REE) between −0.47 and

0.06) and C horizons (τ(Zr, REE) between −0.21 and 0.05). The C horizon regolith (3.6 m

depth) had τ(Zr, Er) = 0.05 and τ(Zr, Tm) = 0.01.

In the MQ II profile (Figure 6.6c), REE mass flux consistently decreased with depth

except for the C horizon (2.0 m depth). The surface regolith was enriched in REE with

τ(Zr, Ce) = 0.15 and τ(Zr, Pr) = 0.10; however, the regolith below was REE depleted.

11

8

Table 6.1 concentrations of REE in parent rock and lateritic regolith of the GE and MQ profiles

Sample D1 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Y Th ΣREE (La/Sm)PR (La/Yb)PR 2Ce* 2Eu*

Unit m ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm

GE profile

Rock

GEBPRA 27.5 45.7 4.13 14.1 2.62 0.96 2.70 0.37 2.06 0.40 0.95 0.17 1.22 b.d. 11.1 32.1 103

GEBPRB 31.4 50.1 4.72 16.5 3.06 0.90 3.23 0.41 2.10 0.45 1.21 0.18 1.38 b.d. 12.2 25.6 116

GEPR1A 9.69 14.2 1.19 4.72 1.04 0.09 1.60 0.29 1.94 0.43 1.49 0.23 1.40 b.d. 12.4 10.4 38.4

GEPR1B 29.9 50.4 4.96 17.3 3.66 1.24 4.17 0.60 3.62 0.71 2.01 0.25 1.67 b.d. 19.7 14.2 120

GEPR23 35.5 59.2 5.67 19.0 3.77 0.90 4.68 0.69 3.26 0.63 1.72 0.28 2.12 b.d. 19.1 21.8 137

GEPR3 4.00 10.8 1.37 7.55 2.45 0.82 2.56 0.48 3.03 0.56 1.98 0.30 1.72 b.d.4 17.2 1.38 103

Regolith

GEA1g 5.08 13.8 1.17 4.56 0.97 0.18 0.88 0.15 0.89 0.19 0.53 0.08 0.64 b.d. 4.70 69.7 29.1 0.6 0.5 1.4 0.9

GEA1m 0.09 3.86 8.43 0.78 2.68 0.69 0.15 0.64 0.12 0.82 0.18 0.50 0.11 0.64 b.d. 4.86 7.56 19.6 0.6 0.4 1.2 1.0

GEA2g 5.47 15.0 1.26 4.83 1.06 0.22 1.02 0.17 0.96 0.21 0.61 0.08 0.65 b.d. 4.93 60.0 31.5 0.5 0.5 1.4 1.0

GEA2m 0.12 2.69 5.88 0.55 1.88 0.54 0.13 0.57 0.13 0.59 0.15 0.53 0.09 0.51 b.d. 4.03 7.04 14.2 0.5 0.3 1.2 1.1

GEA3g 5.12 15.1 1.15 4.44 0.96 0.19 0.92 0.15 0.87 0.20 0.59 0.09 0.67 b.d. 4.70 69.4 30.4 0.6 0.5 1.5 0.9

GEA3m 0.23 3.53 7.16 0.63 2.07 0.48 0.13 0.51 0.11 0.89 0.20 0.57 0.09 0.79 b.d. 5.09 6.26 17.2 0.8 0.3 1.1 1.2

GE6 3.5 1.43 17.5 0.32 1.26 0.22 0.11 0.65 0.09 0.48 0.12 0.26 0.07 0.54 b.d. 2.96 94.2 23.1 0.7 0.2 6.1 1.4

GE5g 1.58 4.16 0.41 1.70 0.55 0.21 0.40 0.07 0.45 0.10 0.33 0.06 0.52 b.d. 2.61 110 10.5 0.3 0.2 1.2 2.1

GE5m 7.0 2.69 6.77 0.51 2.41 0.46 0.23 0.69 0.11 0.53 0.12 0.32 0.08 0.53 b.d. 3.18 32.5 15.5 0.6 0.3 1.4 1.9

GE4g 6.88 11.1 0.89 3.06 0.71 0.16 0.67 0.12 0.76 0.21 0.61 0.11 0.74 b.d. 4.64 32.4 26.0 1.0 0.6 1.1 1.1

GE4m 8.4 8.68 14.4 1.04 3.44 1.06 0.21 0.92 0.13 0.94 0.20 0.63 0.12 0.91 b.d. 5.76 34.7 32.7 0.9 0.6 1.1 1.0

GE3g 0.99 1.74 0.16 0.61 0.23 b.d. 0.25 0.06 0.26 0.09 0.24 0.06 0.38 b.d. 1.77 18.1 5.08 0.4 0.2 1.1 0.0

GE3m 10.0 1.30 1.93 0.21 0.64 0.18 0.11 0.19 0.04 0.31 0.09 0.28 0.08 0.48 b.d. 2.31 15.0 5.83 0.8 0.2 0.9 2.7

GE2g 4.57 6.50 0.59 1.99 0.42 0.07 0.41 0.08 0.52 0.12 0.39 0.07 0.50 b.d. 3.01 30.4 16.2 1.2 0.5 1.0 0.8

GE2m 11.4 6.47 9.32 0.96 2.92 0.51 0.15 0.64 0.12 0.64 0.19 0.49 0.10 0.66 b.d. 4.22 31.4 23.2 1.4 0.6 0.9 1.2

GE1g 10.0 16.7 1.56 5.52 1.16 0.17 1.14 0.18 1.10 0.25 0.77 0.13 0.89 b.d. 6.42 59.4 39.5 0.9 0.7 1.0 0.7

GE1m 12.5 12.6 21.1 1.98 6.82 1.41 0.27 1.45 0.21 1.27 0.28 0.94 0.16 1.24 b.d. 8.76 62.0 49.7 0.9 0.6 1.0 0.9

11

9



MQ profile

Rock

MQPR1 57.0 89.0 8.27 29.1 4.90 1.08 5.66 0.69 3.95 0.90 2.44 0.39 2.60 0.11 25.6 37.5 206

MQPR2 54.7 84.3 7.65 26.9 4.61 1.15 4.55 0.67 4.07 0.70 2.48 0.35 2.84 0.08 24.6 39.4 195

MQPR3 8.29 17.9 2.44 11.2 4.90 1.23 4.05 0.72 4.25 0.81 2.47 0.30 2.14 0.08 21.6 5.75 60.8

Regolith

MQ I

MQ9g 43.0 74.5 6.57 22.3 3.75 0.69 3.51 0.47 2.73 0.57 1.60 0.24 1.40 0.22 14.6 35.5 162 1.0 1.5 1.1 0.8

MQ9m 0.2 62.9 108 9.66 31.6 4.88 0.80 5.30 0.71 3.52 0.68 2.04 0.23 1.84 0.08 20.1 49.6 232 1.1 1.7 1.1 0.7

MQ8g 42.6 73.0 6.69 22.4 3.83 0.83 3.21 0.49 2.86 0.63 1.67 0.27 1.52 0.25 15.6 38.1 160 0.9 1.4 1.1 1.0

MQ8m 0.5 74.5 126 11.1 37.8 6.74 0.96 6.66 0.92 4.35 0.82 2.53 0.32 2.00 0.15 26.7 54.2 275 0.9 1.8 1.1 0.6

MQ7g 29.0 51.9 4.79 16.5 2.95 0.62 2.54 0.42 2.46 0.57 1.57 0.26 1.63 0.27 13.0 53.9 115 0.8 0.9 1.1 1.0

MQ7m 0.7 44.8 78.2 7.27 24.2 4.25 0.67 4.02 0.54 2.80 0.56 1.81 0.22 1.64 0.09 16.6 57.6 171 0.9 1.3 1.1 0.7

MQ6g 19.7 36.8 3.51 12.6 2.56 0.59 2.26 0.38 2.39 0.55 1.63 0.27 1.74 0.27 12.6 65.3 85.2 0.7 0.6 1.1 1.1

MQ6m 0.9 18.9 37.0 3.33 12.2 2.62 0.57 2.39 0.36 2.80 0.49 1.63 0.24 1.60 0.13 15.2 64.5 84.4 0.6 0.6 1.1 1.0

MQ5g 15.3 28.8 2.61 9.25 1.73 0.36 1.53 0.27 1.72 0.39 1.13 0.20 1.23 0.21 8.4 66.7 64.7 0.8 0.6 1.1 1.0

MQ5m 1.1 10.5 22.9 1.89 6.76 1.70 0.32 1.46 0.25 1.41 0.33 1.06 0.19 1.10 b.d. 8.24 63.5 49.9 0.5 0.5 1.3 0.9

MQ4g 12.0 20.1 1.78 5.99 1.12 0.22 0.94 0.18 1.14 0.30 0.95 0.15 1.01 0.16 8.06 12.9 46.1 0.9 0.6 1.1 1.0

MQ4m 2.2 27.1 48.9 4.05 14.2 2.81 0.31 2.41 0.26 1.36 0.27 1.01 0.16 1.14 0.04 9.47 32.7 104 0.8 1.2 1.1 0.5

MQ3g 15.7 25.1 2.32 7.68 1.55 0.33 1.11 0.18 1.12 0.28 0.82 0.14 0.98 0.19 7.34 19.0 57.5 0.9 0.8 1.0 1.1

MQ3m 2.8 34.3 56.9 5.13 16.2 3.03 0.39 2.72 0.35 1.74 0.36 0.84 0.14 1.09 0.08 10.1 31.7 123 1.0 1.5 1.0 0.6

MQ2g 13.5 22.0 2.02 6.71 1.29 0.25 1.06 0.18 1.07 0.25 0.78 0.14 0.82 0.16 6.16 12.9 50.3 0.9 0.8 1.0 0.9

MQ2m 3.3 32.8 52.2 4.77 16.0 2.93 0.33 2.24 0.33 1.35 0.28 0.86 0.12 0.92 0.05 8.98 24.2 115 1.0 1.7 1.0 0.6

MQ1g 34.9 58.0 5.28 17.7 3.11 0.87 2.83 0.46 2.76 0.63 1.93 0.30 2.00 0.31 16.5 28.4 131 1.0 0.8 1.0 1.3

MQ1m 3.6 54.7 90.3 8.40 27.5 4.45 0.90 4.50 0.71 4.36 0.82 2.78 0.38 2.48 0.28 26.0 40.5 203 1.0 1.1 1.0 0.9

MQ II

MQ15g 41.1 70.9 6.26 21.4 3.78 0.77 3.21 0.50 2.86 0.65 1.84 0.30 1.77 0.30 16.8 35.2 156 0.9 1.1 1.1 1.0

MQ15m 0.08 66.6 115 10.2 32.6 5.37 0.67 5.38 0.67 3.45 0.62 1.90 0.23 1.51 b.d. 19.0 43.9 244 1.1 2.1 1.1 0.6

12

0



MQ14g 34.2 57.2 5.42 18.7 3.33 0.80 2.85 0.45 2.64 0.66 1.79 0.28 1.87 0.31 15.6 37.6 131 0.9 0.9 1.0 1.1

MQ14m 0.25 49.1 81.3 7.79 26.1 4.66 0.81 4.31 0.67 3.34 0.71 2.26 0.32 2.25 0.06 18.7 50.9 184 0.9 1.1 1.0 0.8

MQ13g 23.3 49.4 4.05 14.3 2.62 0.52 2.31 0.38 2.31 0.53 1.64 0.28 1.83 0.30 11.9 48.8 104 0.8 0.6 1.2 0.9

MQ13m 0.6 27.8 54.2 4.20 14.0 3.15 0.52 3.10 0.41 2.20 0.49 1.66 0.23 2.00 0.07 13.3 47.7 114 0.8 0.7 1.2 0.7

MQ12g 8.55 15.5 1.31 4.65 1.01 0.24 0.84 0.16 1.09 0.26 0.82 0.16 1.00 0.18 6.78 20.9 35.8 0.7 0.4 1.1 1.2

MQ12m 1.1 10.7 19.1 1.57 5.33 0.99 0.35 1.28 0.16 0.98 0.23 0.78 0.16 0.85 b.d. 7.36 26.6 42.4 0.9 0.6 1.1 1.4

MQ11g 6.88 10.7 0.96 3.39 0.86 0.30 0.65 0.11 0.68 0.18 0.58 0.11 0.71 0.13 4.74 10.0 26.2 0.7 0.5 1.0 1.8

MQ11m 1.6 7.62 12.9 1.13 4.08 1.16 0.27 0.68 0.14 0.89 0.18 0.63 0.14 0.99 b.d. 6.28 18.8 30.8 0.6 0.4 1.1 1.4

MQ10g 6.05 10.1 0.90 2.97 0.72 0.25 0.47 0.08 0.49 0.12 0.40 0.08 0.52 0.10 3.25 7.12 23.3 0.7 0.6 1.1 1.9

MQ10m 2.0 10.3 16.3 1.43 4.63 1.46 0.20 0.77 0.11 0.83 0.14 0.54 0.11 0.60 b.d. 5.64 13.6 37.4 0.6 0.8 1.0 0.8

RSD5 Min 0.5 0.4 0.1 0.1 0.0 0.0 0.1 0.0 0.2 0.0 0.0 0.0 0.1 0.1 1.1 0.8

RSD Max 2.3 1.2 0.3 1.3 0.3 0.1 0.1 0.0 0.5 0.0 0.1 0.0 0.3 0.2 1.2 1.0

1D denotes depth (m);

2Ce


0.5×(Pr/PrPR)

0.5]; Eu


0.5×(Gd/GdPR)

0.5];

3Subscript PR refers to the parent granitoids: GEPR2, average MQPR1 and MQPR2 were used as the parent granitoids in corresponding profile;

4b.d. refers to below detection limit; ‘g’ represents gravel and ‘m’ represents matrix.

5RSD refers to the range of relative standard deviations of the duplicates/triplicates analysed by ICP-MS.

12

1

Figure 6.5 SiO2-Al2O3-Fe2O3 ternary plots and associated variation of REE concentrations and ratios against the S/SAF weathering index for the GE

profile.

12

2

Figure 6.6 Mass balance calculations of REE against depth for three lateritic profiles, based on weighted average concentrations of REE in matrix and

gravel, using Zr as the reference element: (a) GE Profile; (b) MQ I Profile; (c) MQ II Profile (vertical dashed line refers to mass balance τ(Zr,REE) = 0;

Only selected REE are plotted here, as the remaining REE have similar patterns).


123

6.5.2 Mineralogy of REE in the parent rock

The parent granitoids of the GE profile contained accessory minerals such as monazite

and allanite, which controlled the abundance and distribution of REE (Figure 6.7 and

Table 6.2). Monazite (<0.05 wt%) was usually incorporated into quartz or feldspar or

intergrown with apatite in GE parent granitoids. However, two different types of

monazite were identified by EPMA (Figure 6.7 and Table 6.2). Type 1 monazite (Figure

6.7a & b) contained an average 52 wt% ΣREE, 5.6 wt% Th and 0.3 wt% U with high

(La/Yb)PR (average 5.9, up to 19.5). Type 2 monazite (Figure 6.7c & d) was

characterized by a much higher concentration of Th (average 23 wt%) than the Type 1,

but lower concentrations of ΣREE (average 23 wt%) and (La/Yb)PR (average 1.0). Both

types of monazite had no apparent Ce anomalies (Ce* ranged from 0.9-1.1) and

moderate Eu anomalies (Eu* ranged from 0.3-0.7).

In addition to monazite, allanite (<0.03 wt%) was another important REE-rich

accessory mineral in GE parent granitoids (Figure 6.7e). The average concentration of

ΣREE in allanite was ca. 18 wt%, lower than Type 1 monazite, and with 0-0.02 wt% Th

and/or U. The ΣREE concentration was dominated by LREE with an average (La/Yb)PR

of 4.9 and without Ce anomaly, but had a moderate negative Eu anomaly (average 0.6).

Other accessory minerals such as zircon and ilmenite hosted trace to minor

concentrations of ΣREE. Zircon (<0.02 wt%) contained 0.15-3 wt% ΣREE, with HREE

predominating (average (La/Yb)PR = 0.01). Ilmenite (<0.03 wt%) contained trace to

minor amounts of REE with a preference for Dy and Yb. Thorite, intergrown with

zircon or as individual grain surrounded by feldspars (Figure 6.7f), also contained a

minor amount of ΣREE (0.3-5.2 wt%). Feldspar may contain negligible concentrations

of ΣREE (ca. 1.2 wt%) with LREE enrichment; in this study the Eu anomaly was not

determined as Sm was below the detection limit of EPMA.


124

(a) (b)

(c) (d)

(e) (f)


granitoids of the GE profile (scale bar all 10 µm). (a) and (b) Type 1 monazite

surrounded by feldspar; (c) Type 2 monazite surrounded by feldspar; (d) minute grains

of Type 2 monazite included in feldspars; (e) allanite surrounded by feldspars; (f)

hollow thorite surrounded by feldspars. (Aln: allanite; Ap: apatite; Fsp: feldspar; Kfs:

feldspar-K; Mnz: monazite; Qz; quartz; Thr: thorite).


125

In the parent granitoids of both MQ profiles, accessory minerals such as monazite,

fluorocarbonate, allanite, apatite, zircon, ilmenite and sphene contained abundant REE

(Figure 6.8 and Table 6.3). Monazite (<0.03 wt%) (Figure 6.8a & b) contained an

average of 51 wt% ΣREE, with a preference for LREE ((La/Yb)PR ranged from 2.7 to

7.5) and negative Eu anomalies (Eu* ranged from 0.2 to 0.8). The concentrations of

ΣREE in MQ monazite were similar to the Type 1 monazite in the GE parent granitoids.

In the MQ parent granitoids, REE-rich fluorocarbonates (<0.02 wt%) (Figure 6.8b)

were observed, occurring as an intergrown mineral with monazite and/or zircon,

incorporated into feldspars and having 50-58 wt% ΣREE with a preference for LREE

(average (La/Yb)PR = 5.4), slightly positive Eu anomalies (Eu* ranged from 1.0-1.4),

and trace to minor amount of Th (0.3-5.3 wt%) and U (ca. 0.06 wt%). These

fluorocarbonates were not observed in GE parent granitoids but observed in the JG

parent granitoids (Chapter Seven). In addition, allanite (<0.01 wt%) (Figure 6.8c) also

hosted average 10 wt% ΣREE with the range of (La/Yb)PR 1.1-2.8.

Apatite (<0.31 wt%) contained 0.1-0.7 wt% ΣREE without apparent LREE or HREE

selectivity. Ilmenite (<0.33 wt%) contained 0-0.6 wt% ΣREE and showed a preference

for Dy (up to 0.47 wt%). Sphene (also called titanite) was observed in MQ parent

granitoids (<0.26 wt%) as an individual crystal or intergrown with ilmenite or allanite

(Figure 6.8d). It contained an average 0.06 wt% Yb. Zircon (<0.03 wt%) contained

ca. 0.1 wt% ΣREE, with preference for Yb.

6.5.3 Mineralogy of REE in the regolith

In the weathered MQ regolith, phosphate phases were identified as the main hosts for

REE, as well as zircon and ilmenite (Figure 6.9 and Table 6.4). Allanite and apatite

were absent in the B horizon regolith. REE-rich phosphates were preferentially enriched

in LREE ((La/Yb)PR average 5.3) with an average 1,000-fold enrichment of REE

(ca. 53 wt% ΣREE) relative to the parent granitoids (average 200 ppm), reflecting

strong mineralogical control of the distribution and fractionation of REE.

Residual monazite and secondary rhabdophane were inferred as the REE-rich

phosphates in the regolith based on chemical composition. Residual monazite (Figure

6.9a, b & c) had higher concentrations of Th (3.8-8.6 wt%) and Pb (0.4-1.1 wt%) with

grain sizes between 10-50 µm, similar to the concentrations of Th (average 6.4 wt%)

and Pb (average 0.6 wt%) in monazite in MQ parent granitoids. Secondary rhabdophane


126

(Figure 6.9d, e & f) had lower concentrations of Th (0.4-2.8 wt%) and Pb (0.1-0.3 wt%)

with varied grain size (<60 µm) and the compositional sum of oxides below 100%

(probably due to hydration). Florencite and xenotime were not observed in the MQ

regolith; this does not mean they are not present, as they could have a sub-micron grain

size and be present in very low abundance.

In addition, zircon grains containing average 0.3 wt% ΣREE, with a preference for

HREE as well as Ce, and minor contents of Th (0.1-0.2 wt%) and U (0.1-0.4 wt%) were

observed in the MQ regolith. Thorite was rare and only one grain was found in the C

horizon of MQ II profile (2.0 m depth) (Figure 6.9g), where it contained ca. 9 wt%

ΣREE with a preference for HREE. Ilmenite with 0.02-0.08 wt% Yb was determined by

EPMA, but other REE were below detection limits. Ilmenite was usually intergrown

with rutile/anatase (Figure 6.9h); however, concentrations of REE in Ti-oxides were

below detection limits. In addition, trace concentrations of ΣREE (less than 100 ppm)

were determined in Fe oxides at 1.1 m depth of MQ I profile.


127

(a) (b)

(c) (d)


granitoids of the MQ profiles (scale bars vary and are present on each image): (a)

monazite surrounded by quartz; (b) monazite, REE-rich fluorocarbonate together with

zircon, incorporated into feldspars; (c) allanite surrounded by feldspar and sphene; (d)

sphene intergrown with ilmenite, surrounded by feldspars; (Aln: allanite; Fc: REE-rich

fluorocarbonate; Fsp: feldspar; Hbl: hornblende; Ilm: ilmenite; Kfs: feldspar-K; Mnz:

monazite; Qz; quartz; Spn: sphene; Zrn: zircon).

12

8

(a) (b) (c) (d)

(e) (f) (g) (h)

Figure 6.9 Backscatter electron images of REE-bearing minerals in regolith of the MQ profiles (scale bars vary and are present on each image): (a) and

(b) corroded residual Th-rich monazite; (c) Th-rich residual monazite surrounded by hornblende and clay matrix; (d) secondary rhabdophane

surrounded by Fe-rich REE-bearing phosphates incorporated in quartz; (e) and (f) varied grain sizes of Th-poor secondary rhabdophane; (a) REE-rich

thorite; (b) eroded ilmenite and intergrown Ti oxides.

12

9

Table 6.2 Concentrations of REE and associated elements from EPMA analyses of representative minerals in parent granitoids from the GE profile

No. 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Min. Aln Aln Aln MnzI MnzI MnzI MnzII MnzII MnzII Zrn Zrn Zrn Thr Ilm Fsp

Si 15.9 15.6 16.0 0.38 0.78 0.47 3.19 3.11 4.20 15.8 13.5 13.4 9.83 0.02 18.0 Zr b.d. b.d. b.d. b.d. b.d. b.d. 0.40 0.52 0.18 48.8 43.6 38.0 6.63 b.d. b.d.

Ti 0.21 0.12 0.29 b.d. b.d. b.d. 0.14 0.09 b.d. b.d. 0.26 2.33 b.d. 32.3 0.82

Pb b.d. b.d. b.d. 0.40 0.66 0.82 0.31 0.14 0.07 b.d. 0.03 0.05 3.12 b.d. b.d.

Th 0.02 0.02 b.d. 4.90 5.80 6.32 27.3 25.3 16.2 0.02 0.42 0.73 35.8 b.d. b.d.

U 0.02 0.01 0.02 0.20 0.19 0.35 0.55 0.36 0.36 0.04 0.21 0.48 19.6 b.d. b.d.

Al 10.0 9.73 9.07 b.d. b.d. b.d. b.d. 0.91 1.74 b.d. 0.67 1.27 b.d. b.d. 11.4

Y 0.36 0.29 0.91 0.62 1.50 1.18 2.41 1.80 1.97 b.d. 1.15 1.00 0.93 b.d. b.d.

La 3.99 4.55 4.57 14.3 12.5 13.3 6.15 5.36 6.79 b.d. 0.05 b.d. b.d. b.d. 0.33

Ce 8.02 8.79 7.43 24.5 24.4 23.6 8.22 8.86 12.5 b.d. 0.31 0.30 b.d. b.d. 0.62

Pr 0.82 0.87 0.77 2.42 2.56 2.61 0.83 0.90 1.21 b.d. 0.05 0.07 b.d. b.d. b.d.

Nd 2.68 2.77 2.30 8.18 8.82 7.61 2.67 2.63 3.75 b.d. 0.25 0.30 b.d. b.d. 0.21

Sm 0.40 0.37 0.38 1.25 1.49 1.48 0.62 0.49 0.74 b.d. 0.09 0.09 b.d. b.d. b.d.

Eu 0.07 0.06 0.06 0.19 0.21 0.16 0.10 0.08 0.12 b.d. 0.02 0.01 b.d. b.d. 0.01

Gd 0.33 0.29 0.29 0.81 1.02 1.02 0.57 0.44 0.62 b.d. 0.12 0.13 0.04 b.d. 0.03

Dy b.d. b.d. b.d. 0.20 0.42 0.35 0.43 0.27 0.43 b.d. 0.10 0.06 0.10 0.37 b.d.

Er b.d. b.d. b.d. b.d. 0.02 b.d. 0.08 0.02 0.03 0.03 b.d. b.d. 0.06 b.d. b.d.

Tm b.d. b.d. b.d. b.d. 0.02 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Yb 0.03 0.06 0.14 0.04 0.23 0.21 0.38 0.27 0.36 0.12 0.31 0.23 0.13 0.07 b.d.

Lu b.d. b.d. b.d. 0.02 0.03 0.02 b.d. b.d. 0.02 b.d. 0.04 0.04 b.d. b.d. b.d.

Fe 8.65 8.93 10.2 0.15 0.03 0.03 0.60 0.68 0.59 0.01 1.47 2.76 0.10 31.9 10.1

Ca 8.32 7.88 8.49 0.34 b.d. 0.33 4.10 4.17 2.92 b.d. 0.02 0.25 0.27 b.d. 14.2

Sr 0.05 0.04 0.02 b.d. b.d. b.d. 0.05 0.07 b.d. 0.48 0.38 0.33 b.d. b.d. 0.20

K b.d. b.d. b.d. b.d. b.d. b.d. 0.47 0.61 b.d. b.d. 0.01 0.08 b.d. b.d. b.d.

P 0.01 0.02 0.01 13.6 13.0 13.1 11.3 11.5 12.2 b.d. b.d. b.d. b.d. b.d. 0.01

F 0.18 0.16 0.21 1.40 1.08 1.28 1.09 1.13 0.88 0.14 0.35 0.27 0.38 b.d. b.d.

O 36.3 35.6 35.8 27.4 27.3 26.9 26.0 28.4 30.6 35.2 32.5 32.8 21.6 30.8 40.1

total 96.4 96.3 97.2 101.4 102.1 101.1 98.1 98.1 99.1 100.6 95.9 95.1 98.6 95.5 96.1

Tb and Ho were below detection limit (b.d.); Na was also below the detection limit except No. 18 (0.57 wt%); Fc: REE-rich fluorocarbonate; Mnz: monazite; Aln: allanite; Ilm:

ilmenite; Spn: sphene; Ap: apatite; Zrn: zircon.

13

0

Table 6.3 Concentrations of REE and associated elements from EPMA analyses of representative minerals in parent granitoids from the MQ profile

No. 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Min. Fc Fc Fc Mnz Mnz Mnz Aln Aln Aln Ilm Ilm Ilm Spn Spn Spn Ap Ap Ap Zrn

Si 1.62 1.88 1.66 1.04 1.24 0.67 16.7 15.6 16.8 b.d. 0.04 0.23 14.3 14.5 14.8 0.15 0.35 0.07 15.7 Zr 1.02 0.71 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 47.8

Ti 0.04 0.04 0.26 b.d. b.d. b.d. 0.05 0.06 0.14 33.6 35.1 38.4 21.8 22.2 20.8 0.04 b.d. b.d. 0.04

Pb 0.18 0.29 0.03 0.63 0.50 0.45 0.02 b.d. b.d. b.d. 0.06 0.15 b.d. b.d. b.d. b.d. b.d. b.d. 0.08

Th 5.26 4.99 0.27 7.91 8.76 5.55 0.02 0.02 0.01 b.d. b.d. 0.07 b.d. b.d. b.d. b.d. 1.96 b.d. 0.24

U 0.07 0.06 0.05 0.23 0.22 0.11 0.02 b.d. 0.02 0.02 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.35

Al 0.48 0.61 0.69 b.d. b.d. b.d. 11.7 10.3 10.7 b.d. b.d. b.d. 1.05 1.02 1.61 b.d. b.d. b.d. b.d.

Y 0.88 0.82 0.23 1.57 2.02 1.07 0.38 0.80 0.09 0.01 b.d. b.d. b.d. b.d. 0.03 0.39 0.51 0.20 0.16

La 22.1 20.5 12.3 14.5 13.6 15.1 1.89 3.87 2.36 b.d. b.d. 0.04 b.d. b.d. b.d. 0.05 0.05 b.d. b.d.

Ce 17.0 19.3 36.6 24.1 22.5 25.7 3.61 6.68 4.03 b.d. b.d. 0.05 b.d. b.d. b.d. 0.15 0.13 b.d. b.d.

Pr 2.31 2.41 2.05 2.15 2.19 2.39 0.38 0.82 0.41 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Nd 7.10 7.52 6.22 6.51 6.84 7.13 1.19 2.00 1.18 b.d. b.d. b.d. b.d. b.d. b.d. 0.12 0.15 0.06 b.d.

Sm 0.74 0.86 0.81 0.97 1.12 1.05 0.10 0.17 0.11 b.d. b.d. b.d. b.d. b.d. b.d. 0.05 0.05 0.03 b.d.

Eu 0.26 0.27 0.21 0.16 0.15 0.17 0.02 0.02 0.02 b.d. b.d. b.d. b.d. b.d. b.d. 0.01 b.d. b.d. b.d.

Gd 0.45 0.49 0.45 0.67 0.71 0.66 0.16 0.31 0.08 b.d. b.d. b.d. b.d. b.d. b.d. 0.07 0.08 0.02 b.d.

Tb b.d. b.d. b.d. b.d. 0.07 0.03 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Dy b.d. b.d. b.d. 0.38 0.44 0.27 b.d. b.d. b.d. b.d. 0.14 0.47 b.d. b.d. b.d. 0.12 0.09 0.04 b.d.

Er b.d. b.d. b.d. 0.07 0.10 0.03 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.02 b.d. b.d. b.d.

Yb 0.19 0.23 0.10 0.18 0.25 0.18 0.09 0.18 0.04 b.d. 0.05 b.d. 0.06 0.06 0.07 0.05 0.02 0.02 0.10

Fe 5.46 4.71 4.70 b.d. b.d. b.d. 8.78 9.51 10.0 27.9 23.5 18.0 0.47 0.38 0.62 0.12 0.48 0.16 0.55

Mg 0.11 0.13 0.20 b.d. b.d. b.d. 0.05 0.11 0.03 0.10 0.03 0.07 b.d. b.d. b.d. b.d. b.d. b.d. 0.01

Ca 2.80 2.71 1.76 b.d. 0.28 0.11 11.1 8.12 11.1 0.05 b.d. b.d. 17.6 17.9 18.4 35.4 35.5 36.2 0.07

K 0.02 0.03 0.08 b.d. b.d. b.d. 0.05 0.06 b.d. b.d. b.d. b.d. 0.04 b.d. b.d. 0.02 0.01 b.d. b.d.

P 0.10 0.13 0.03 12.4 12.1 13.2 b.d. 0.02 0.01 b.d. b.d. b.d. b.d. 0.01 b.d. 18.2 17.6 17.5 b.d.

S 0.09 0.11 0.05 b.d. 0.05 0.01 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.01 b.d. b.d.

F 4.69 5.31 7.19 1.11 b.d. 1.24 b.d. b.d. b.d. b.d. b.d. b.d. 0.38 0.22 1.30 3.29 2.94 4.61 b.d.

O 13.3 13.3 12.1 26.8 27.2 27.4 37.8 35.7 37.5 30.5 30.3 31.2 38.8 39.5 39.2 36.7 36.6 35.4 35.1

total 86.3 87.4 88.0 101.5 100.4 102.5 94.0 94.4 94.8 92.1 89.2 88.7 94.6 95.9 96.9 95.0 96.5 94.3 100.2

Ho, Tm and Na were below the detection limit; Lu was also below the detection limit, except No. 28 monazite (0.03 wt%) and No.32 allanite (0.03 wt%).

13

1

Table 6.4 Concentrations of REE and associated elements from EPMA analyses of representative minerals in lateritic regolith from the MQ profile

No. 44 45 46 47 48 49 50 51 52 53 54 55 56

Min. Mnz Mnz Mnz Rbp Rbp Rbp Ilm Ilm Ilm Zrn Zrn Zrn Thr

Si 0.86 1.09 0.75 0.45 0.26 0.15 b.d. b.d. 0.04 14.8 15.2 15.2 9.76 Zr b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 44.9 46.0 47.1 6.48

Ti b.d. b.d. b.d. b.d. b.d. b.d. 33.5 31.8 30.8 0.06 b.d. b.d. 0.05

Pb 1.05 0.95 0.91 0.05 0.06 0.10 b.d. 0.02 b.d. b.d. b.d. b.d. 0.05

Th 7.24 8.61 7.02 0.70 1.23 1.18 b.d. b.d. b.d. 0.18 0.08 b.d. 38.4

U 0.24 0.26 0.17 0.02 b.d. b.d. 0.02 0.02 b.d. 0.14 0.19 0.13 0.40

Al b.d. b.d. b.d. 0.09 0.04 b.d. 0.09 b.d. b.d. 0.21 0.05 b.d. 0.38

Y 2.14 1.56 1.59 0.13 0.31 0.27 b.d. b.d. b.d. 0.67 0.37 b.d. 3.11

La 13.9 14.4 13.7 15.7 16.0 16.6 b.d. b.d. b.d. b.d. b.d. 0.04 b.d.

Ce 23.5 23.6 24.4 26.8 27.5 28.0 b.d. b.d. b.d. 0.37 0.09 b.d. 1.09

Pr 2.70 2.67 2.77 2.54 2.44 2.75 b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Nd 6.87 6.39 7.09 7.82 7.88 8.41 b.d. b.d. b.d. 0.04 b.d. b.d. 0.46

Sm 1.05 0.90 1.09 1.23 1.13 1.24 b.d. b.d. b.d. 0.02 b.d. b.d. 0.20

Eu 0.06 0.05 0.10 0.20 0.20 0.19 b.d. b.d. b.d. b.d. b.d. b.d. 0.07

Gd 0.68 0.56 0.67 0.89 0.64 0.68 b.d. b.d. b.d. 0.08 0.03 b.d. 0.30

Dy 0.51 0.37 0.37 0.09 0.06 0.14 b.d. b.d. b.d. 0.08 0.03 b.d. 0.23

Er 0.09 0.04 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.03 b.d.

Yb 0.26 0.22 0.20 0.07 0.05 0.14 0.02 0.04 0.05 0.07 0.10 0.09 0.52

Lu 0.03 b.d. 0.02 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.06

Fe 0.08 0.04 0.25 0.03 1.56 b.d. 31.8 35.1 34.8 0.56 0.52 0.07 0.06

Mg b.d. b.d. b.d. b.d. b.d. b.d. 0.10 0.09 b.d. 0.01 b.d. b.d. 0.02

Ca b.d. b.d. b.d. b.d. 0.01 b.d. b.d. b.d. 0.03 b.d. 0.01 b.d. 0.09

Sr b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.41 0.43 0.47 b.d.

P 12.5 12.1 12.7 13.1 12.5 12.8 b.d. b.d. 0.01 b.d. b.d. b.d. 1.29

S b.d. b.d. b.d. b.d. 0.02 b.d. b.d. b.d. b.d. 0.02 b.d. b.d. 0.03

F 1.29 1.12 1.08 1.53 1.02 1.38 b.d. b.d. b.d. 0.15 b.d. b.d. 0.04

O 26.7 26.5 26.9 26.6 26.3 26.3 31.6 31.4 30.6 33.3 33.9 34.0 23.8

total 101.7 101.3 101.7 98.1 99.2 100.1 97.1 98.4 96.3 96.1 97.0 97.2 100.5

Tb, Ho, Tm, Na were below detection limit; K was below the detection limited except No.47 (0.09 wt%). Mnz: monazite; Rbp: rhabdophane; Ilm: ilmenite; Zrn: zircon; Thr: thorite.


132

6.5.4 REE in grain size fractions and chemical extractions of regolith

In the MQ II profile, the silt fraction (2-20 µm) had the highest concentrations ΣREE,

followed by clay (<2 µm), except the surface regolith which had a higher concentration

of REE in clay than in silt (Figure 6.10, data presented in Appendix 11.9).

Concentrations of ΣREE in each grain size fraction generally decreased with depth until

1.1 m depth, and then increased to saprolite except in gravel (> 2 mm). Concentrations

of ΣREE in gravel decreased with depth until saprolite.

Given the mass percentage of each grain size fraction, sand (0.02-2 mm) contained the

greatest mass of REE, especially HREE at 0.1-0.3 m depth (Figure 6.11). Although silt

generally had the highest concentrations of ΣREE, the relatively low mass percentage

compared with the other fractions minimises its contribution. The clay fraction had the

highest mass of REE, followed by gravel at 0.6-1.1 m depth; these two fractions account

for 71-80% LREE, 70-99% MREE and 65-99% HREE. At 1.6-2.0 m depth, the sand

fraction had the highest mass of REE, composing 44-60% LREE, 37-53% MREE and

38-59% MREE, except Eu. The gravel fraction contained significant mass of Eu

(58-70%) at 1.6-2.0 m depth.

The sequential extraction experiment revealed the REE speciation in representative

lateritic regolith of the GE and MQ I profile (Table 6.5). The Res contained the highest

percentages of ΣREE. In addition to the Res, the WAE hosted minor percentages of

ΣREE (4.6% of ΣREE in the WAE of the GE saprolite and 7.9% in the MQ I C horizon).

The ΣREE percentage associated with the WAE in the saprolite (4.6%) of the GE

profile was higher than the ΣREE percentage associated with the WAE in the duricrust

(2.0%), where the ΣREE percentages in the FeAm (3.7%) and FeCry (2.7%) were higher.

Similarly, the ΣREE percentage associated with the WAE in the C horizon regolith

(7.9%) of the MQ I profile was higher than the ΣREE percentage associated with the

WAE in the A horizon regolith (5.6%), where the ΣREE percentage in the Org was

higher (10.7%), although the Res hosted the highest ΣREE percentage (80%). In the

duricrust of the GE profile, an average of 3.7% ΣLREE, 4.4% ΣMREE and 1.8%

ΣHREE were associated with the FeAm, higher than 3.0% ΣLREE, 1.4% ΣMREE and

0.5% ΣHREE associated with the FeCry.

13

3

Figure 6.10 Concentrations of selected REE (La, Ce, Sm, Dy, and Yb) in grain size fractions of the MQ II profile.

Figure 6.11 Mass loading of selected REE (La, Ce, Sm, Dy, and Yb) in grain size fractions of the MQ II profile (MQ15, 0.1 m depth; MQ14, 0.3 m

depth; MQ13, 0.6 m depth; MQ12, 1.1 m depth; MQ11, 1.6 m depth; MQ10, 2.0 m depth. Only selected REE are plotted here, as the remaining REE

have similar patterns).

0

20

40

60

80

MQ15 MQ14 MQ13 MQ12 MQ11 MQ10

La

(p

pm

)

0

40

80

120

160


Ce

(p

pm

)

0

2

4

6


Sm

(p

pm

)

sand silt clay matrix gravel

0

2

4

6


Dy

(p

pm

)

0

2

4

6


Yb

(p

pm

)

0%

20%

40%

60%

80%

100%


La

(w

t%)

0%

20%

40%

60%

80%

100%


Ce

(w

t%)

0%

20%

40%

60%

80%

100%


Sm

(w

t%)

gravel sand silt clay

0%

20%

40%

60%

80%

100%


Dy

(w

t%)

0%

20%

40%

60%

80%

100%


Yb

(w

t%)

13

4

Table 6.5 Concentrations of REE in sequential extractions of representative regolith in the GE and MQ I profiles

Sample Element concentrations (ppm)

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Y Mn Fe ΣREE%1

d.l. 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 1.00 5.00

GE1m WAE 0.518 0.876 0.109 0.377 0.074 0.017 0.063 0.010 0.056 0.013 0.035 0.005 0.034 0.006 0.136 b.d. 30.0 4.56

GE1m Org 0.016 0.024 0.003 0.011 0.004 b.d. 0.002 b.d. 0.003 b.d. 0.002 b.d. 0.004 0.001 0.013 b.d. 125 0.15

GE1m FeAm 0.175 0.514 0.038 0.139 0.028 0.006 0.025 0.004 0.027 0.006 0.018 0.003 0.018 0.003 0.155 2.00 615 2.09

GE1m FeCry 0.055 0.192 0.007 0.025 0.005 0.001 0.005 b.d. 0.006 0.001 0.004 b.d. 0.004 0.001 0.038 3.00 696 0.64

GE1m Res 11.5 18.7 1.90 6.10 1.30 b.d. 1.00 0.20 1.20 0.30 0.90 0.20 1.00 0.20 7.80 26.0 5304 92.6

GE6 WAE 0.086 0.263 0.021 0.079 0.022 0.005 0.017 0.003 0.015 0.003 0.008 0.001 0.006 0.001 0.024 5.00 8.00 1.96

GE6 Org 0.007 0.040 0.003 0.011 0.004 b.d. 0.002 b.d. 0.002 b.d. 0.002 b.d. 0.004 0.001 0.008 b.d. 56.0 0.28

GE6 FeAm 0.059 0.699 0.023 0.094 0.029 0.007 0.023 0.004 0.023 0.004 0.012 0.002 0.011 0.002 0.067 15.0 1272 3.67

GE6 FeCry 0.037 0.626 0.007 0.032 0.010 0.002 0.008 0.001 0.007 0.001 0.004 b.d. 0.004 b.d. 0.022 2.00 5329 2.73

GE6 Res 2.10 17.3 0.45 1.55 0.45 b.d. 0.45 0.10 0.65 0.20 0.55 0.10 0.70 0.10 4.55 22.5 22.5% 91.4

MQ1m WAE 3.473 5.617 0.640 2.081 0.342 0.081 0.281 0.048 0.274 0.058 0.163 0.024 0.131 0.019 1.056 2.00 59.0 7.89

MQ1m Org 0.221 0.361 0.041 0.126 0.022 0.005 0.018 0.003 0.023 0.005 0.016 0.003 0.017 0.003 0.140 b.d. 69.0 0.52

MQ1m FeAm 1.618 2.829 0.301 0.963 0.168 0.037 0.147 0.027 0.167 0.036 0.105 0.017 0.097 0.014 0.977 14.0 1884 3.89

MQ1m FeCry 0.493 0.888 0.093 0.294 0.056 0.015 0.051 0.011 0.073 0.017 0.048 0.008 0.049 0.007 0.435 8.00 1738 1.25

MQ1m Res 38.5 66.0 6.50 19.6 3.10 0.50 2.40 0.50 2.80 0.60 1.80 0.30 2.00 0.30 18.0 67.0 8759 86.4

MQ5m WAE 2.556 5.709 0.583 2.164 0.451 0.126 0.369 0.065 0.387 0.087 0.251 0.039 0.234 0.035 1.316 2.00 b.d. 14.6

MQ5m Org 0.293 0.776 0.066 0.195 0.036 0.009 0.025 0.005 0.035 0.009 0.036 0.010 0.082 0.013 0.244 b.d. 156 1.77

MQ5m FeAm 0.451 2.294 0.115 0.399 0.085 0.024 0.067 0.014 0.094 0.021 0.062 0.010 0.062 0.009 0.365 2.00 1456 4.13

MQ5m FeCry 0.087 0.232 0.021 0.075 0.018 0.005 0.014 0.003 0.020 0.004 0.014 0.003 0.019 0.003 0.072 4.00 5267 0.58

MQ5m Res 19.0 32.0 3.20 9.60 1.60 0.10 1.20 0.20 1.30 0.30 0.90 0.20 1.00 0.20 8.50 29.0 3.52% 79.0

MQ8m WAE 4.316 5.358 0.915 3.378 0.686 0.209 0.661 0.112 0.618 0.134 0.346 0.045 0.230 0.032 2.295 45.0 13.0 5.61

MQ8m Org 7.306 16.82 1.312 3.733 0.670 0.169 0.489 0.099 0.644 0.143 0.429 0.080 0.478 0.066 4.298 21.0 824 10.7

MQ8m FeAm 2.711 5.825 0.473 1.443 0.257 0.063 0.203 0.039 0.234 0.048 0.121 0.016 0.077 0.010 1.055 49.0 2172 3.79

MQ8m FeCry 0.170 0.360 0.030 0.096 0.018 0.005 0.014 0.003 0.018 0.004 0.011 0.002 0.012 0.002 0.095 9.00 4249 0.25

MQ8m Res 67.1 113 11.2 34.3 5.20 0.40 3.50 0.50 3.00 0.60 1.60 0.20 1.40 0.20 17.1 69.0 1.24% 79.7

1ΣREE% refers to percentage of ΣREE in each extraction species; d.l.: detection limit.


135

6.6 Discussion

6.6.1 Evolution of REE-bearing minerals during intense weathering

The high deficiencies of REE relative to the parent rock in weathered subsurface

regolith from the three profiles studied, especially in the GE profile, are likely to be the

result of the characteristics of REE inherited from the parent granitoids and the

weathering environment.

Easily-weathered LREE-rich minerals in the parent granitoids, such as allanite and

fluorocarbonate (only observed in MQ granitoids), can quickly break down by

hydrolysis and dissolution during the early stages of weathering. The absence of

fluoroapatite in the MQ regolith suggests its dissolution during weathering. The REE

liberated from dissolving minerals can either be partially leached away in solutions

(Braun et al., 1990), or precipitated in the form of secondary mineral phases (Braun et

al., 1998; Lottermoser, 1990; Nedachi et al., 2005). In the present study, strong

depletion of REE in the GE regolith (Figure 6.1b) suggests leaching of REE during

intense weathering, whereas secondary rhabdophane observed in the B horizon of MQ

regolith (Figure 6.9) reflects the retention of REE by formation of secondary phosphate

minerals. The PO43-

released by dissolution of fluoroapatite plays an important role in

the formation of secondary LREE-rich phosphates according to the following reaction:

Ca5(PO4)3F + 3LREE3+

+ 3H2O → 3(LREE)PO4∙H2O + 5Ca2+

+ F-

(fluoroapatite → rhabdophane)

The breakdown of primary igneous minerals and precipitation of secondary minerals not

only changes the abundance, distribution and fractionation of REE; the growth of

secondary REE-bearing phosphates also constrains further mobility of REE (Braun et

al., 1993). At the same time, weathering-resistant minerals, e.g. monazite, zircon and

ilmenite, become residually enriched in the regolith (Table 6.4). As weathering

intensifies and lateritization proceeds, some REE-bearing weathering-resistant minerals,

e.g. monazite and ilmenite, may be further altered or eroded (Taunton et al., 2000a).

High concentrations of Th and U determined in monazite crystals (Table 6.2 & Table


136

6.3) can lead to a disordering of the structure due to self-irradiation and persistent

intensive weathering may enhance the dissolution rate. Weathering and dissolution of

phosphate minerals such as apatite and monazite have important effects on REE budget

in weathered regolith (Aubert et al., 2001). Residual monazite, secondary rhabdophane,

zircon and ilmenite have been shown to be the most important REE-bearing mineral

phases controlling REE concentrations in weathered MQ regolith (Table 6.4).

In addition to weathering-resistant minerals and secondary phosphate minerals, REE

can also be retained in the regolith by sorption to Al- and/or Fe- oxides or organic

matter (Coppin et al., 2002; Cullers et al., 1987; Land et al., 1999; Laveuf and Cornu,

2009; Piasecki and Sverjensky, 2008; Quinn et al., 2006; Sonke and Salters, 2006).

Sorption of REE onto clay minerals was identified by sequential extraction experiments

(Table 6.5): minor concentrations of REE (2.2 ppm ΣREE in the GE saprolite and

13.2 ppm ΣREE in the MQ I C horizon) were found in the WAE species, and the ΣREE

percentage in WAE increased from the C to B horizon in the MQ I profile but decreased

from saprolite to duricrust in the GE profile, both in agreement with the change of the

clay weight proportion in each profile. In addition, 1.0 ppm ΣREE were determined in

the FeAm extraction in the duricrust of GE profile and 3.7 ppm ΣREE determined in B

horizon regolith (MQ5, 1.1 m depth) of MQ I profile (Table 6.5). This suggests that

trace to minor amounts of REE may be conserved in the ferruginous regolith by

sorption onto or coprecipitation with Fe oxyhydroxides.

In addition, the surface regolith in both MQ profiles (0.2-0.5 m depth in MQ I profile

and 0.08 m depth in MQ II profile) accumulated REE at concentrations higher than in

the parent granitoid (Table 6.1); however, the reason for this is unclear. This surface

regolith is thought to include transported materials, so enrichment of REE by lateral

transportation under the influence of soil creep or colluviation has been considered.

However, REE concentrations in most of the weathered subsurface regolith are lower

than parent granitoids, thus transported materials from upper slope is expected to show

similar REE depletion; and given the likely dilation of surface regolith, concentrations

higher than in the parent granitoids cannot be explained solely by lateral transportation.


137

Therefore, another mechanism is therefore required, possibly biogeochemical recycling.

Enrichment of La, Ce, Sm, and Th in outer bark and, to a lesser extent, in needles and

twigs of pine trees was reported in southeast of Australia (Arne et al., 1999). In addition,

some ferns, mosses and lichens are also known to accumulate REE (Chiarenzelli et al.,

2001; Tyler, 2004). Further research is required to determine the process, or processes,

causing the accumulation of REE in surface regolith of MQ profiles.

6.6.2 Reason for stronger depletion of LREE over HREE

Subsurface matrix and gravel from the GE and MQ II profile had (La/Yb)PR < 1.0,

coupled with a high deficiency of REE relative to the parent granitoids, showing

stronger depletion of LREE than HREE; subsurface matrix and gravel from the MQ I

profile, however, had varied (La/Yb)PR, reflecting that the fractionation of REE

controlled by grain size was more important in MQ regolith during weathering.

In the GE profiles, stronger depletion of LREE than HREE is believed to result from

extreme weathering conditions. Breakdown of LREE-rich allanite induced LREE

releasing into solution; although REE released may partially precipitate as secondary

mineral rhabdophane and/or florencite, persistent extreme weathering conditions may

further enhance the REE depletion, especially LREE, due to the susceptibility of

LREE-rich phosphates. This is supported by the previous studies: A dissolution

experiment of REE in granitoids by Harlavan and Erel (2002) showed that the

dissolution of allanite dominated the release of REE and weakened the LREE

enrichment in the first stage of weathering. Secondary REE-bearing phosphates

(including rhabdophane and florencite) could be dissolved by organic complexation

and/or uptake by biological cells (Taunton et al., 2000b), especially in the upper profile.

Although the MQ II regolith was less weathered than the GE profile, the relatively

shallow regolith depth (2.0 m depth) may facilitate weathering of the phosphates and

this is supported by the dissolution of phosphates observed by the SEM-BSE images.

The stability of weathering-resistant minerals, e.g. zircon, may enhance the retention of

HREE, especially in highly weathered environments, such as the GE profile.


138

The C horizon regolith (MQ10) from the MQ II profile is taken as an example to

estimate the residual accumulation of HREE. The concentration of Zr in the bulk sample

was 136.3 ppm (calculated by the weight percentage of gravel and matrix) and thus the

weight percentage of zircon is 0.028% (assuming all Zr is contained by zircon). Given

the average composition of La (0.05%) and Yb (0.22%) in zircon grains in MQ regolith

determined by EPMA, the zircon hosts ca. 0.14 ppm La and ca. 0.63 ppm Yb.

Compared with 9.2 ppm La and 0.58 ppm Yb in this bulk regolith (calculated by the

weight percentage of gravel and matrix), it suggests that zircon dominates the

abundance of Yb in this regolith sample.

In addition to the different stabilities of LREE- and HREE-bearing minerals, pH may

influence REE mobility during weathering as well (Henderson, 1984). REE are well

known to be more easily leached from regolith in acidic conditions than under neutral or

alkaline conditions (Brown et al., 1955; Henderson, 1984; Nesbitt, 1979) and

preferential enrichment of HREE in the lateritic regolith is widely reported to be due to

higher stability of HREE carbonate complexes compared with LREE (Braun et al., 1993;

Braun et al., 1990; Koppi et al., 1996; Laveuf et al., 2008; Nesbitt, 1979).

However, when the pH ranges from 4.6-6.1, (as this study 4.8-6.1 in GE regolith and

4.5-5.6 in MQ regolith), REE can occur as Ln3+

, LnH2PO42+

, LnSO4+, LnF

2+ and/or

LnCO3+, depending on the presence and concentrations of the complex ligands (Wood,

1990). The complex LnCO3+ preferentially enriched in LREE over HREE, whereas

Ln(CO3)2-

is HREE selective (Cantrell and Byrne, 1987), and highly pH-dependent

(Wood, 1990); at pH<6 the predominant REE species will be the Ln3+

ion (Wood, 1990).

Fluoride complexes tend to favour HREE rather than LREE (Wood, 1990); however,

due to their low concentrations and the relatively acidic weathering conditions, fluoride

complexes would not be expected to have been important carriers in the fractionation of

REE during weathering. Although the pH of regolith cannot represent the pH conditions

when weathering took place, it may act as a clue for the imprint of pH on the signature

and fractionation of REE.

Therefore, stronger depletion of LREE over HREE is believed to be a combined


139

function of weathering of LREE-rich minerals and residual accumulation of HREE-rich

minerals in the regolith under low pH leaching conditions and persistent moderate to

extreme weathering conditions.

6.6.3 Fractionation of REE in weathered regolith

The differences in weathering rates of REE-bearing minerals, and in the stability of

complexes between LREE, MREE and HREE during weathering, result in REE

fractionation in weathered regolith. Preferential breakdown of LREE-rich accessory

minerals reveals that fractionation of REE starts at the early stage of weathering.

Residual accumulation of weathering-resistant HREE-rich minerals e.g. zircon and

ilmenite may lead to (La/Yb)PR < 1.0 in regolith, especially in intensely weathered

profiles, e.g. the GE regolith and the upper B horizon regolith in both MQ profiles with

high CIA. In the MQ I profile, gravel had (La/Yb)PR < 1.0 whereas matrix had

(La/Yb)PR > 1.0 in the lower part; this may result from the dilution effect of quartz in

gravel (gravel had higher concentration of Si than matrix) and the preferential

enrichment of secondary phosphates in the matrix during the early stages of weathering.

Mineral phases controlling the abundance and fractionation of REE in intensely

weathered regolith is revealed by the sequential extraction experiment (Table 6.5): Res

species hosted >90% REE in saprolite and duricrust of the GE profile, and ca. 86% and

80% REE in the C and A horizon of the MQ I profile. The abundance and stability of

LREE-rich minerals and HREE-rich minerals, however, are greatly affected by the

weathering conditions (e.g. weathering time and intensity). This might be the reason

why the depth profile of (La/Yb)PR has a negative correlation with weathering intensity

in the GE profile which had undergone persistent extreme weathering (Figure 6.4).

6.6.4 Ce and Eu anomaly

In the GE profile, Ce had a positive anomaly in the duricrust (Ce*=6.1) compared with

the parent granitoid, reflecting that Ce fractionated from the other REE and was

enriched in the duricrust. Sequential extraction showed that 0.7 ppm Ce was associated


140

with FeAm and 0.63 ppm Ce with FeCry species (Table 6.5), suggesting that Ce likely

precipitated with Fe oxides and/or adsorbed onto Fe oxyhydroxides during formation of

either cerianite (CeO2) under oxidised conditions (Angelica and Dacosta, 1993; Braun et

al., 1990; Braun et al., 1998), or insoluble Ce-rich phosphate when pH changed (Bau,

1999). The formation of cerianite and valence change of Ce can be the result of redox

change which may be related to seasonal fluctuation of the water table under alternating

arid and wet periods (Braun et al., 1993; Ji et al., 2004). This process is consistent with

total iron enrichment and high concentrations of Fe in FeAm (1272 ppm) and FeCry

(5329 ppm) extractions in the duricrust (Table 6.5). In contrast to previous studies, this

occurrence of the positive Ce anomaly is not relevant to reduction and oxidation of Mn

(Duzgoren-Aydin and Aydin, 2009) because the concentration of Mn was very low

(Table 6.5) in both the FeAm (15 ppm) and FeCry (2 ppm), especially when compared to

Fe in the duricrust.

In both MQ profiles, no apparent Ce anomalies were observed in secondary phosphates

(Ce* ranged from 1.0-1.1) or regolith samples (1.0-1.3) compared with the parent

granitoid. Thus Ce is believed to occur mainly in a trivalent form and hence is not

fractionated from the other REE during weathering of the MQ regolith. The MQ

regolith is still in the weak lateritization stage and a lack of persistent intensive leaching

and/or changes in redox environment result in no apparent Ce anomaly.

In addition, weathering of feldspar and sphene may result in negative Eu anomalies in

the saprolite (Bea, 1996; Condie et al., 1995; Panahi et al., 2000); breakdown of

REE-rich fluorocarbonates, which had weak positive Eu anomalies (Eu* ranged 1.0-1.4),

possibly induces negative Eu anomalies as well in the MQ regolith. Due to higher

solubility, Eu (II) will be more easily leached than the other trivalent REE from the

minerals rich in Eu and depleted in the regolith (Van der Weijden and Van der Weijden,

1995).

6.6.5 Grain size fractionation and chemical speciation of REE

In the three profiles studied, REE were more enriched in matrix than gravel, except in


141

the A horizon of the GE profile (Table 6.1). Matrix enrichment of REE rather than

gravel suggests that cementation of clay and Fe oxides did not play a significant role in

scavenging REE from matrix during gravel formation. The higher concentrations of

REE in silt and clay fractions than the sand fraction of the MQ II profile (Figure 6.10)

may suggest: (i) the importance of secondary REE-bearing minerals in the silt fraction;

(ii) adsorption of REE by clay minerals; and (iii) a dilute effect of quartz in the sand

fraction. In addition, according to mass loading calculations (Figure 6.11), a significant

mass of REE was found in sand fraction at 1.6-2 m and 0.1-0.3 m depth, in contrast to

the large mass of REE contained in clay fraction at 0.6-1.1 m depth in MQ II profile.

This relative enrichment may be for the reason that regolith is less weathered at the

bottom than the upper part of profile, and thus more REE is contained by large

particle-size minerals such as monazite and ilmenite. As weathering proceeds and

advances, the formation of secondary minerals and sorption by clay or Fe

oxyhydroxides becomes important for retention of REE in regolith. Apparently, REE

mobilizes and partitions into different size minerals during weathering.

Significant percentages of REE associated with the WAE extraction may suggest that

some amount of REE is bio-available and occurring as free Ln3+

ion or being adsorbed

in regolith. High concentrations of REE in the WAE extraction may be also related to

relatively low pH in regolith, which favours the conversion of metals from precipitated

forms into dissolved forms (Harter, 1983). In addition, an average of 3.7% ΣLREE, 4.4%

ΣMREE and 1.8% ΣHREE were associated with FeAm in the duricrust of the GE profile,

higher than 3.0% ΣLREE, 1.4% ΣMREE and 0.5% ΣHREE in FeCry (Table 6.5),

suggesting that amorphous Fe oxyhydroxide plays a more important role for retention of

REE than crystalline Fe oxide during ferruginization. Up to 11.9% Ce, 17.1% Er and

23.5% Tm in the Org extraction, higher than the WAE, FeAm and FeCry species, at 0.5 m

depth in MQ I profile (Table 6.5) indicates that organic ligands are particularly

important for hosting REE in the upper part of profile; similar results have been

reported previously (Aubert et al., 2004; Land et al., 1999).


142


In this chapter, the distribution and fractionation of REE in three intensely weathered

lateritic profiles (GE, MQ I and MQ II) developed over granitoids with dolerite dykes in

Western Australia were investigated and the conclusions are as follows:

(i) The regolith of all three profiles developed from granitoid rather than dolerite were

confirmed by chondrite normalized REE distribution patterns.

(ii) High deficiencies of REE, especially LREE, were observed in the regolith of three

profiles, especially in the GE and MQ II profiles.

(iii) Preferential weathering of LREE-rich minerals including monazite, and residual

accumulation of weathering-resistant minerals such as zircon, result in a stronger

depletion of LREE than HREE under persistent intense weathering.

(iv) Mineral phases control the fractionation of REE, which greatly depends on the

abundance and stability of LREE-rich minerals (monazite and rhabdophane) and

HREE-rich minerals (zircon and ilmenite) in intensely weathered lateritic regolith.

(v) In addition to mineral phases, trace to minor amounts of REE can be retained in the

regolith by association with clay minerals, Fe oxides/oxyhydroxides and organic matter

as revealed by sequential extractions.

(vi) A positive Ce anomaly (Ce*=6.1) in the duricrust in the GE profile may result from

the redox change during formation of the duricrust.

All this information suggests that REE can be mobilized during weathering and

lateritization, even becoming highly depleted from intensely weathered lateritic regolith.

This is important when using REE as tracers for geochemical processes, especially in

extremely weathered settings.

143

7 Mode of occurrence of REE in an intensely weathered lateritic

profile in Western Australia

7.1 Abstract

The mineralogy and geochemistry of rare earth elements (REE) were studied in an

intensely weathered lateritic profile (JG) developed on meta-granitoids in Jarrahdale,

Western Australia, in order to understand the geochemical pathways of REE during

intense weathering and the effects of Fe oxides and oxyhydroxides on the mode of

occurrence of REE during advanced lateritization. Investigations were made by

combined use of bulk chemical analyses, the electron microprobe and the synchrotron

x-ray fluorescence microprobe.

Great depletion of ΣREE (ca. 94%) in the saprolite was observed, due to near-complete

dissolution of fluorocarbonates, thorite, apatite, etc., suggesting high mobility of REE

during weathering. Evidence for the translocation of REE includes the formation of

secondary phosphate minerals rhabdophane and florencite in regolith. These secondary

phosphate minerals are absent from the parent meta-granitoids and play a significant

role in trapping REE, especially LREE, released from the parent meta-granitoids during

weathering processes. Residual accumulation of the weathering-resistant minerals

zircon, ilmenite, rutile and anatase is also important for retention of REE in regolith,

especially HREE. The abundance and stability of LREE-rich secondary phosphates and

HREE-rich weathering-resistant minerals control the fractionation of REE in intense

weathered lateritic regolith.

In the ferruginous zone, Ce has fractionated from the other REE, showing a high

enrichment (Ce* up to 25 in the duricrust gravel). The occurrence of poorly crystalline

Ce (hydr)oxide phases as a rim along Fe-rich pores in the duricrust or in the boundary

of Al/Fe clay layer in iron nodules in the ferruginous zone is further evidence for this

fractionation, suggesting formation of hydrous cerianite (CeO2∙nH2O) under oxidised

conditions.


144

In addition, trace concentrations of Yb (0.02-0.12 wt%) were determined in the iron

core and clay layer of iron nodules. Fine-grained (<10 µm) REE-bearing phosphate

crystals were precipitated with crystalline Fe phases in the duricrust, or occurred in

kaolinitic layers of iron nodules. These observations suggest that Fe phases are

important regolith components influencing the redistribution, fractionation and mode of

occurrence of REE in extremely weathered lateritic regolith.

7.2 Key words

Rare earth elements; mode of occurrence; laterite; weathering; iron nodules;

7.3 Introduction

It is well documented that REE can mobilize, redistribute and fractionate during rock

weathering (e.g. Aide and Pavick, 2002; Aubert et al., 2001; Banfield and Eggleton,

1989; Braun et al., 1993; Koppi et al., 1996; Laveuf and Cornu, 2009; Nesbitt, 1979;

Tyler, 2004). However, the precise sequence of events relating to the behaviour of REE

in rock weathering processes and pedogenesis are still not unambiguously understood,

especially in iron nodules/duricrust formed in intensely weathered lateritic profiles,

defined by high concentrations of Fe oxides and oxyhydroxides.

Differences in the weathering rates of REE-bearing minerals and the complex stability

of REE during weathering processes result in REE fractionation. It is proposed that the

fractionation of REE during weathering is predominantly constrained by weathering

conditions and primary REE-bearing minerals in the system (Aubert et al., 2001; Braun

et al., 1998; Ji et al., 2004; Nesbitt, 1979). Diverging views exist on the impact of

weathering intensity on the fractionation of REE, and at which stage of weathering REE

starts to fractionate from each other. For example, Banfield and Eggleton (1989), Price

et al. (1991) and Sharma and Rajamani (2000) demonstrated that the relative

concentrations of different REE change dramatically during the initial stage of

weathering, while Middelburg et al. (1988) and Compton et al. (2003) recognized that

the migration and differentiation of REE occurs at advanced stages of weathering. In

Chapter Seven: Mode of occurrence of REE in an intensely weathered lateritic profile in WA

145

addition, although iron oxides are known to have high surface areas thus rendering them

very efficient sinks for many cationic trace elements (Singh and Gilkes, 1992), their effects

on the translocation and fractionation of REE in lateritic regolith during weathering and

lateritization are still not well understood.

Therefore, this paper investigates the mode of occurrence and fractionation of REE in

an intensely weathered lateritic profile (JG) developed on meta-granitoid in Jarrahdale

in Western Australia. Geochemical analyses of bulk regolith, electron probe

micro-analyzer (EPMA) and synchrotron X-ray Fluorescence Microscopy (SXFM)

techniques were used for spatial characterization and quantitative analysis of the mode

of occurrence of REE in parent meta-granitoids and regolith samples.



This study was performed on a regolith profile (JG) developed over meta-granitoids in

Jarrahdale, Western Australia (for profile descriptions and sampling details, see Chapter

Three). Regolith samples were separated into two subsample groups based on grain size:

gravel (>2 mm, represented by suffix ‘g’) and matrix (<2 mm, represented by suffix

‘m’). The exception to this subdivision was the saprolite (JG1) and mottled clay (JG2-3),

which had only matrix without gravel. All gravel and matrix subsamples were oven

dried at 105 °C overnight and ground to ≤200 µm prior to fusion in order to determine

element concentrations. In addition, particle size analysis and sequential extraction

experiments were carried out, as described in detail in Chapter Eight.



(lithium metaborate:lithium tetraborate) and heating in a muffle furnace at 1050 °C for

40 minutes. Duplicate fusion beads were made on 10% of samples to check

reproducibility and preparation errors. After cooling, the fusion beads were dissolved in

100 mL of 10% analytical grade HCl. The trace elements, including REE, in fusion


146

beads of the gravel and matrix were determined after an additional 10-fold dilution with

10 ppb Rh/Ir solution in 10 mL polypropylene tubes by inductively coupled

plasma-mass spectrometry (ICP-MS, Perkin-Elmer Elan 6000, Toronto, Canada) at the

University of Western Australia. The trace elements, including REE, in particle size

fractions and sequential extractions were analysed at Genalysis Laboratory Services,

Maddington, Western Australia. Certified international standard materials, including

stream sediment reference material STSD-2, STSD-4 (Canada Centre for Mineral and

Energy Technology, CANMET) and an in-house standard reference were prepared in

the same way as the samples and analysed together with samples to check accuracy and

precision. The variation of REE between tested values and certified values was within

10%, except Lu (variation was 19.8%, Appendix 11.6). The concentrations of REE in

gravel and matrix of regolith samples from three profiles are listed in Table 7.1.






15kV accelerating voltage with a 3 nA beam current. Semi-quantitative modal

abundances of REE-bearing accessory minerals in parent rocks were calculated based

on SEM-BSE images of polished thin sections and chemical maps produced by EDS.

The relative volume percentages of REE-bearing minerals were calculated and selected

mineral density data (Deer et al., 1992) were used to convert volume percentage to

weight percentage. Given that the regions examined by SEM had more REE-bearing

minerals, calculated weight percentages will be overestimated. Although the accessory

mineral abundances obtained in this way are semi-quantitative, they can be used as a

basis for assessing the mineral control and fractionation of REE in the parent rock. The

chemical composition of representative REE-bearing minerals was analysed by JEOL

8530 electron probe micro-analyzer (EPMA) at 20 kV accelerating voltage and 5 nA

beam current. Software Probe for EPMA from Probe Software Inc. was used for setting

up and analysing the data. Standard reference materials for microprobe calibration were


147

synthetic glass 612 from the National Institute of Standards and Technology (NIST),

in-house standard synthetic REE phosphates, rutile, zircon and thorite; in addition,

standard Brazil monazite was analysed with samples for cross checking. All microscopy

analyses were conducted at the Centre for Microscopy, Characterisation and Analysis

(CMCA), University of Western Australia. The detection limit of the EPMA for

individual elements in mineral grains is presented in the Appendix 11.10. All

microscopy analyses were conducted at the Centre for Microscopy, Characterisation and

Analysis (CMCA), University of Western Australia.

Resin-impregnated polished thin-sections of iron nodules on pure quartz slides

(50.8×25.4 mm, ProSciTech) were also examined using SXFM. The elemental mapping

by SXFM equipped with a Maia detector was conducted on the XFM beamline at the

Australian Synchrotron, Melbourne, Australia. The chemical associations of Fe, Mn Ti,

Ce etc. in iron nodules were imaged by scanning the sample stage using 14.5 keV beam

energy (X-ray wavelength 0.85508 Å) with in-house Ni foils as standards. The pixel

step size was set to 2×2 μm, with a dwell time of 2 ms and beam size of 2 μm.

Post-imaging analyses and full spectra of selected regions were generated to separate

the overlapping contributions from interfering elements using Geopixe software (Ryan

et al., 2010).


7.4.2.1 Fractionation of REE and anomalies of Ce and Eu

In order to study the fractionation of REE, three groups are identified: the light REE

(LREE; from La to Nd), the middle REE (MREE: from Sm to Ho) and the heavy REE

(HREE: from Gd to Lu) (Henderson, 1984). Regolith REE distribution patterns were

normalized to average chondrite values (Anders and Grevesse, 1989) in order to show

the fractionation of REE during lithogenical weathering; in addition, regolith REE

distribution patterns were compared with the parent rock (PR) in order to reveal relative

enrichment and depletion. The normalized ratios (La/Sm)PR and (La/Yb)PR were used

for identifying fractionations between LREE-MREE and LREE-HREE using the


148

composition of the parent rocks as a reference. Cerium and Eu anomalies were

calculated using the following equations (subscript PR refers to parent rock):

Ce*=(Ce/CePR)/[(La/LaPR)

0.5×(Pr/PrPR)

0.5] (1)

Eu*=(Eu/EuPR)/[(Sm/SmPR)

0.5×(Gd/GdPR)

0.5] (2)

7.4.2.2 Weathering intensity-Chemical Index of Alteration (CIA)

The Chemical Index of Alteration (CIA) (Nesbitt and Young, 1982) was used as a

quantitative estimate of the intensity of chemical weathering. The CIA calculates loss of

mobile elements relative to Al in bulk samples, providing a single parameter estimate of

the intensity of chemical weathering. The formula (Nesbitt and Young, 1982) is:



carbonates). The CIA values of regolith samples were listed in Table 4.1 in Chapter

Four.

7.4.2.3 Mass balance calculation

A geochemical mass balance calculation (Brimhall et al., 1991) was calculated to assess

the absolute loss or gain of REE during weathering. The formula for normalized

concentration (τi,j) in Equation (4) assumes that an immobile element (e.g. Zr) behaves

conservatively during weathering and pedogenesis.

1))((,

,

,

,

, pj

wj

wi

pi

C

C

C

C

ji

(4)






Equation (4) provides a tool for estimating elemental loss or gain within a profile;

however, mass balance equations have two critical assumptions: a genetic relationship


149

between regolith and the underlying rock and a fully conserved reference element.

Although the mobility of Ti, Zr and Th is subject to debate (Braun et al., 1993; Cornu et

al., 1999), Zr is considered conservative at the sampling scale in this study (see Chapter

Five) and is used as the reference element to calculate the mass flux.

7.5 Results

7.5.1 Bulk geochemical data of REE

7.5.1.1 REE concentration and normalized pattern

The concentrations of REE and derived relevant fractionation parameters are presented

in Table 7.1. The parent meta-granitoids contained an average of 102 ppm ΣREE (sum

of concentrations of REE), lower than the average upper continental crust (UCC, 146

ppm, McLennan, 1995).

The A horizon regolith had an average of 65.2 ppm ΣREE in gravel and 58.6 ppm in

matrix. The duricrust contained the highest ΣREE, 236 ppm in gravel and 50.2 ppm in

matrix, and the saprolite contained the lowest (6.45 ppm ΣREE). Yttrium is not a REE

but showed similar behaviour to La, and was more enriched in matrix than gravel,

whereas other REE from Ce to Er were more enriched in gravel than matrix. The ΣREE

decreased with depth in regolith matrix from duricrust (JG5m, 3 m depth,

ΣREE 50.2 ppm) to saprolite (JG1, 10 m depth, ΣREE 6.45 ppm), but were significantly

enriched in gravel from the duricrust (JG5g, 3 m depth, ΣREE 236 ppm, Ce*=25.3) and

ferruginous mottled zone (JG4g, 5 m depth, ΣREE 117 ppm, Ce*=13.9). The saprolite

(JG1) had the most depleted REE, especially LREE and MREE. The chondrite

normalized REE distribution patterns of regolith samples (Figure 7.1) showed a

similarity to the underlying meta-granitoids, suggesting inheritance from the protolith,

and thus the average concentrations of meta-granitoids JGPR1 and JGPR2 were used as

the parent rock concentrations. Although the chondrite-normalized REE distribution

patterns of regolith samples still showed LREE-enrichment, the regolith REE

distribution patterns normalized by the parent meta-granitoid (Figure 7.1) showed


150

HREE enrichment, especially the saprolite, suggesting stronger depletion of LREE and

MREE than HREE in this zone.

7.5.1.2 Fractionation of REE during intense weathering

In the JG profile, the LREE/MREE ratio (La/Sm)PR and the LREE/HREE ratio

(La/Yb)PR are plotted versus depth and the La concentration to illustrate the

fractionation of REE with spatial distribution and degrees of depletion (Figure 7.2 &

Figure 7.3).

The saprolite, matrix of ferruginous zone and A horizon regolith showed higher

depletion of LREE than MREE with (La/Sm)PR below 0.8, whereas in the mottled clay

and gravel of ferruginous mottled zone and duricrust, (La/Sm)PR ranged from 1.2-1.4,

suggesting stronger depletion of MREE than LREE and a local redistribution of La and

Sm between gravel and matrix in ferruginous mottled zone and duricrust.

Apart from the upper mottled clay, matrix and gravel from other regolith had (La/Yb)PR

below 0.8, showing stronger depletion of LREE than HREE. In the upper mottled clay

(JG3, 6.5 m depth), the (La/Yb)PR was 1.2, indicating a more severe depletion of HREE

than LREE in this horizon.

The saprolite (10 m depth) showed the strongest depletion of La than other regolith and

thus had significant fractionation of LREE from MREE ((La/Sm)PR = 0.4) and HREE

((La/Yb)PR = 0.2), although it had undergone moderate weathering (CIA = 0.65).

In addition, most regolith samples had negative Eu anomalies (Eu*<1) except the

saprolite (JG1, Eu*=1.0, Table 7.1). The anomaly increased from the saprolite to the

lowest value (0.6) in ferruginous regolith.


151

Figure 7.1 REE distribution patterns of (a) meta-granitoids and regolith samples

normalized by the average chondrite composition; and (b) regolith samples normalized

by the parent meta-granitoid in the JG profile; (JGPR2-meta-granitoid; JG10-A horizon,

0.4 m depth; JG6-upper ferruginous zone, 1.5 m depth; JG5-duricrust, 3.0 m depth;

JG3-upper mottled clay, 6.5 m depth; JG1-saprolite, 10 m depth; ‘g’ denotes gravel and

‘m’ denotes matrix).


152

Figure 7.2 Normalized ratios (La/Sm)PR (LREE/MREE) and (La/Yb)PR (MREE/HREE)

of regolith samples against depth in the JG profile (dashed vertical line at 1.0 shows no

fractionation relative to parent rock).

Figure 7.3 Plots of (La/Sm)PR and (La/Yb)PR vs. La for the JG profile, illustrating the

degrees of depletion and fractionation of REE.

0

2

4

6

8

10

12

0.0 0.5 1.0 1.5(La/Sm)PR

de

pth

(m

)0

2

4

6

8

10

12

0.0 0.5 1.0 1.5(La/Yb)PR

de

pth

(m

)

matrix gravel

(a) (b)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

0 5 10 15 20 25 30 35

La (ppm)

(La

/Sm

) PR

JG8g

JG2

JG9m

JG10gJG1

JG3

JG5gJG4g

JG5m

JG4m

JG6m JG8m

JG7gLa more depleted than Sm

JG7m

JG6g

JG10mJG9g

REE fractionation

increases

La depletion increases

JGPR1

JGPR2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0 5 10 15 20 25 30 35

La (ppm)

(La

/Yb

) PR

JG9g

JG7gJG6m

JG9mJG6g

JG1

JG3

JG5g

JG2

JG5mJG4m JG7m

JG8m

JG4g

JG8g

JG10g

JG10m

La less depleted than Yb

REE fractionation

increases

La more depleted than Yb

La depletion increases

JGPR2

JGPR1


153

7.5.1.3 Variation of REE with lateritization degree

The variation of REE and derived fractionation parameters in the JG profile are

compared with a second index of weathering intensity by using the concentration ratio

SiO2/(SiO2+Al2O3+Fe2O3) (S/SAF) (Hill et al., 2000) and with the degree of

lateritization by using SiO2-Al2O3-Fe2O3 ternary plots (Schellmann, 1981) (Figure 7.4).

Gravel was more enriched in Sm (MREE) and Yb (HREE) than matrix; however, the

concentration of La (LREE) in gravel was similar to matrix; initial scatter of (La/Sm)PR

and (La/Yb)PR suggested that fractionation of REE occurred at the early (weak) stage of

lateritization. (Y/Ho)PR was relatively consistent; however, (Y/Ho)PR in gravel with

strong lateritization was lower than other regolith with lower degrees of lateritization,

which may suggest fractionation of Y and Ho at advanced stages of weathering and

lateritization.

15

4

Table 7.1 Concentrations of REE and derived fractionation parameters in parent meta-granitoids and lateritic regolith from the JG profile

Sample1 unit JG7g JG7m JG8g JG8m JG9g JG9m JG10g JG10m JG6g JG6m JG5g JG5m JG4g JG4m JG3 JG2 JG1 JGPR1 JGPR2 RSD

2

CIA % 99.2 99.1 98.7 92.9 99.3 92.7 99.3 92.2 99.4 98.7 98.6 98.1 95.4 95.3 94.1 86.2 64.7 47.0 47.0

Depth m 0.02 0.15 0.3 0.4 1.5 3.0 5.0 6.5 8.6 10.0

Proportion3 % 0.24 0.76 0.60 0.40 0.51 0.49 0.69 0.31 0.82 0.18 0.47 0.53 0.40 0.60 W

4 W W W W

La ppm 10.1 14.3 12.0 14.9 11.8 12.1 8.52 9.52 7.53 7.78 6.02 6.08 4.98 7.76 10.5 5.22 1.31 25.2 30.6 0.0-0.4

Ce ppm 32.0 28.7 32.3 29.6 34.4 26.4 30.9 21.8 27.0 28.8 224 37.2 107 19.0 14.8 7.48 2.38 44.4 50.4 0.0-0.4

Pr ppm 2.43 2.54 3.07 2.65 2.95 2.24 2.16 1.87 1.72 1.56 0.70 0.89 0.63 1.13 1.26 0.58 0.24 4.14 4.48 0.0-0.1

Nd ppm 9.52 9.16 11.9 10.5 11.5 8.14 8.57 6.72 6.52 5.66 2.35 3.02 1.94 4.27 3.61 1.67 1.09 13.8 14.2 0.0-0.3

Sm ppm 1.96 1.76 2.46 2.00 2.34 1.59 1.80 1.61 1.42 1.06 0.37 0.57 0.31 0.90 0.56 0.31 0.26 2.12 2.05 0.0-0.1

Eu ppm 0.37 0.30 0.46 0.34 0.42 0.31 0.34 0.23 0.28 0.20 0.09 0.10 0.08 0.15 0.12 0.06 0.07 0.42 0.59 0.0

Gd ppm 1.79 1.88 2.06 1.68 2.08 1.50 1.78 1.47 1.41 1.27 1.59 0.67 0.69 0.73 0.54 0.22 0.29 1.75 1.93 0.0-0.1

Tb ppm 0.29 0.27 0.36 0.26 0.34 0.24 0.25 0.25 0.22 0.21 0.08 0.07 0.04 0.12 0.07 0.05 0.03 0.24 0.21 0.0

Dy ppm 1.54 1.58 1.89 1.75 1.75 1.48 1.42 1.26 1.40 1.20 0.38 0.52 0.39 0.70 0.41 0.27 0.19 0.97 0.95 0.0-0.1

Ho ppm 0.33 0.42 0.38 0.35 0.39 0.34 0.31 0.30 0.29 0.27 0.07 0.12 0.08 0.15 0.11 0.07 0.05 0.27 0.23 0.0

Er ppm 0.99 1.03 1.04 0.98 1.06 1.00 0.84 0.86 0.89 0.90 0.23 0.32 0.24 0.56 0.30 0.32 0.19 0.93 0.65 0.0-0.1

Tm ppm 0.15 0.18 0.16 0.18 0.16 0.18 0.14 0.17 0.15 0.18 0.05 0.07 0.04 0.10 0.06 0.06 0.04 0.17 0.14 0.0

Yb ppm 1.06 1.21 1.16 1.23 1.11 1.21 0.97 1.20 1.02 1.10 0.33 0.54 0.35 0.74 0.36 0.51 0.26 1.28 0.93 0.0-0.1

Lu ppm 0.18 0.21 0.20 0.20 0.19 0.19 0.17 0.22 0.18 0.20 0.06 0.08 0.06 0.16 0.09 0.09 0.07 0.26 0.21 0.0

ΣREE ppm 62.6 63.5 69.4 66.6 70.5 56.8 58.2 47.4 50.0 50.4 236 50.2 117 36.5 32.8 16.9 6.45 95.9 108

(La/Sm)PR5 0.38 0.61 0.37 0.56 0.38 0.57 0.35 0.44 0.40 0.55 1.21 0.80 1.21 0.64 1.40 1.25 0.37

(La/Yb)PR 0.38 0.47 0.41 0.48 0.42 0.39 0.35 0.32 0.29 0.28 0.74 0.45 0.56 0.42 1.16 0.41 0.20

15

5

Sample1 unit JG7g JG7m JG8g JG8m JG9g JG9m JG10g JG10m JG6g JG6m JG5g JG5m JG4g JG4m JG3 JG2 JG1 JGPR1 JGPR2 RSD

2

6Ce

* 1.50 1.10 1.23 1.09 1.35 1.17 1.67 1.19 1.73 1.91 25.3 3.69 13.9 1.49 0.94 0.99 0.98

Eu* 0.76 0.64 0.80 0.73 0.74 0.78 0.73 0.58 0.76 0.68 0.43 0.63 0.63 0.71 0.81 0.87 0.95

Y ppm 6.98 11.56 7.76 10.3 7.55 9.77 5.81 9.27 6.42 7.57 2.03 3.20 2.07 4.50 2.70 2.13 1.34 9.72 7.27 0.0-0.2

Th ppm 122 22.0 121 21.9 138 22.3 138 23.9 94.0 37.6 196 45.0 119 35.6 48.0 167 31.0 16.4 18.5 0.2-3.3

Zr ppm 348.3 471.7 353.9 425.0 353.6 506.8 346.1 490.1 349.3 444.7 348.9 291.4 292.0 481.7 341.4 164.0 105.4 159.7 158.8

Hf ppm 8.67 13.2 11.2 11.4 9.95 13.6 10.6 12.3 9.84 12.8 11.0 8.13 7.58 12.6 10.6 4.87 2.81 4.12 5.45

Ti7 ppm 7440 7441 8857 6471 7668 6875 6658 6653 9927 3772 3336 4434 3695 5248 3765 1678 1050 1300 1267

1suffix ‘g’ represents gravel and suffix ‘m’ represents matrix;

2RSD is the range of relative standard deviations of the duplicates/triplicates analysed by ICP-MS;

3proportion refers to weight proportion of matrix and gravel;

4W refers to whole rock analysis;

5Suffix PR refers to the parent meta-granitoids: arithmetic means of concentrations from JGPR1 and JGPR2 were used;

6Ce


0.5× (Pr/PrPR)

0.5]; Eu


0.5× (Gd/GdPR)

0.5];

7Concentrations of Ti, Zr and Hf were determined by ICP-OES (details see Chapter Five).

15

6

Figure 7.4 SiO2-Al2O3-Fe2O3 ternary plots and associated variation of REE concentrations and ratios against the S/SAF weathering index for the JG

profile.

15

7

Table 7.2 Element concentrations from EPMA analyses of representative minerals in parent meta-granitoids (Figure 7.6) of the JG profile

No. 3 4 57 58 2 59 5 6 60 61 62 1 63

Min. Fc Fc Fc Fc Thr Ap Ilm Spn Fsp Mag Mag Zrn Zrn

Si 0.77 5.61 2.20 3.09 11.3 0.01 b.d. 13.9 18.4 0.02 0.25 15.2 15.2

Zr b.d. 0.27 b.d. 0.16 9.79 b.d. b.d. b.d. b.d. b.d. b.d. 45.4 47.1

Ti 0.02 0.09 0.03 0.08 0.76 b.d. 32.2 22.2 0.01 0.05 b.d. 0.04 b.d.

Pb b.d. 0.08 0.23 0.19 0.46 b.d. b.d. b.d. b.d. 0.04 b.d. 0.07 0.11

Th 2.06 2.31 0.64 6.44 32.9 b.d. b.d. b.d. b.d. b.d. 0.02 0.10 0.08

U 0.09 0.23 0.18 0.20 7.00 b.d. 0.02 b.d. b.d. 0.03 0.04 0.22 0.33

Al 0.48 1.26 0.44 0.94 0.35 b.d. b.d. 1.13 12.1 0.02 0.02 0.03 b.d.

Y 0.04 0.08 0.06 0.10 1.17 0.02 b.d. b.d. b.d. b.d. b.d. 0.13 0.13

La 20.0 16.2 20.8 16.3 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Ce 30.6 23.4 27.7 24.2 1.83 0.06 b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Pr 2.51 1.90 2.85 2.52 0.26 0.04 b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Nd 5.42 4.33 5.97 5.30 1.13 b.d. b.d. b.d. b.d. b.d. b.d. 0.11 b.d.

Sm 0.54 0.46 0.52 0.55 0.45 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Eu 0.19 0.14 0.06 0.08 0.05 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Gd 0.22 0.15 0.27 0.28 0.39 b.d. b.d. b.d. b.d. b.d. b.d. 0.02 b.d.

Dy b.d. b.d. b.d. b.d. 0.17 b.d. b.d. b.d. b.d. b.d. b.d. 0.03 b.d.

Yb 0.06 0.04 b.d. 0.07 0.25 0.05 0.06 b.d. 0.05 0.04 0.18 0.13 b.d.

Lu b.d. b.d. b.d. b.d. 0.04 b.d. b.d. b.d. b.d. b.d. 0.04 0.03 b.d.

Fe 1.40 2.90 2.96 3.57 0.93 0.02 34.1 0.42 8.59 71.9 72.1 0.64 0.31

Ca 4.90 5.26 2.89 2.68 0.32 38.4 0.20 19.1 15.5 0.02 0.06 0.08 0.01

Sr b.d. b.d. b.d. b.d. 0.01 0.02 b.d. 0.08 0.18 b.d. 2.24 0.39 0.44

K b.d. 0.04 0.04 0.01 b.d. b.d. b.d. b.d. b.d. b.d. 0.03 0.01 b.d.

P 0.02 0.02 0.02 0.05 0.50 17.6 b.d. 0.01 b.d. b.d. 0.02 b.d. b.d.

F 7.09 10.6 6.31 6.77 1.19 4.54 b.d. 0.50 b.d. b.d. 0.45 b.d. b.d.

O 11.6 14.9 12.9 13.5 24.4 36.3 31.5 39.2 40.4 20.7 21.3 33.7 34.1

total 88.5 90.7 87.5 87.4 95.9 97.1 98.2 96.6 95.3 92.9 96.8 96.3 97.8

Tb, Ho and Na were below detection limits (b.d.); Fc: REE-rich fluorocarbonate; Thr: thorite; Ap: apatite; Ilm: ilmenite; Spn: sphene; Fsp: feldspar; Mag: magnetite; Zrn: zircon.

15

8

Table 7.3 Element concentrations from EPMA analyses of REE-bearing phosphates in lateritic regolith (Figure 7.8) of the JG profile

No. 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79

Min Rbp Rbp Rbp Rbp Rbp Rbp Rbp Rbp Rbp Rbp Rbp Rbp Rbp Rbp Rbp Flo d.l.1

Si 0.96 0.72 0.08 1.04 0.74 0.66 1.20 3.34 0.05 0.48 0.32 0.74 0.15 0.03 b.d. 0.46 0.01

Pb 0.46 0.81 0.54 1.21 0.22 0.20 0.26 0.01 0.09 0.91 0.27 1.40 0.15 0.75 0.17 0.13 0.02

Th 9.26 6.04 3.93 7.94 8.94 7.82 10.0 3.24 2.22 5.29 3.50 9.84 4.19 1.91 2.18 3.05 0.02

U 0.16 0.07 0.90 0.19 0.14 0.14 0.15 0.06 0.15 0.31 0.47 0.13 0.58 0.98 0.30 0.02 0.02

Al 0.03 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.10 b.d. 0.24 0.82 b.d. b.d. 18.5 0.01

Y 0.17 0.27 1.61 2.39 0.13 0.81 0.64 0.10 0.79 1.44 0.06 0.43 1.45 1.44 3.02 0.23 0.02

La 12.1 16.4 11.8 10.5 10.1 12.5 12.6 13.0 13.8 14.1 14.1 12.7 11.7 13.6 11.8 8.19 0.04

Ce 23.8 25.5 23.2 21.6 23.0 24.0 23.7 25.0 26.9 24.2 26.6 22.4 22.9 24.4 24.3 16.8 0.04

Pr 2.47 2.25 2.55 2.61 2.75 2.49 2.38 2.72 2.91 2.34 2.73 2.13 2.45 2.42 2.56 1.90 0.04

Nd 8.50 6.57 9.07 8.96 10.7 8.49 8.12 9.08 9.78 7.40 9.23 6.51 8.46 8.48 8.81 1.74 0.04

Sm 1.17 0.76 1.58 1.70 1.80 1.36 1.21 2.07 1.42 1.09 1.29 0.90 1.96 1.47 1.55 0.59 0.02

Eu 0.17 0.15 0.21 0.15 0.18 0.15 0.15 0.17 0.33 0.16 0.15 0.12 0.32 0.35 0.18 0.16 0.01

Gd 0.57 0.34 1.05 0.65 0.57 0.60 0.58 1.07 0.75 0.93 0.89 0.71 1.53 1.36 1.54 0.48 0.02

Dy b.d. b.d. 0.45 0.54 b.d. b.d. b.d. b.d. 0.24 0.27 b.d. b.d. 0.52 0.39 0.74 0.04 0.02

Ho b.d. b.d. b.d. 0.39 b.d. b.d. b.d. 0.42 b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.02

Yb 0.14 0.13 0.35 0.35 0.13 0.14 0.14 0.13 0.17 0.25 0.07 0.09 0.33 0.35 0.34 0.14 0.02

Lu b.d. b.d. 0.03 0.04 b.d. b.d. b.d. b.d. b.d. 0.04 b.d. b.d. b.d. 0.04 0.10 b.d. 0.02

Fe 0.59 0.46 0.45 0.60 0.45 b.d. b.d. b.d. 0.09 0.77 0.92 1.08 0.45 0.33 0.49 0.47 0.01

Ca 0.65 0.30 0.24 0.50 0.25 0.63 0.50 0.04 0.12 0.09 0.34 0.92 0.13 0.01 b.d. 0.05 0.01

P 12.2 12.1 12.9 12.3 12.8 12.7 12.3 8.02 13.6 12.5 12.9 12.3 12.0 13.0 13.3 7.71 0.01

S 0.01 b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.08 b.d. b.d. b.d. 0.15 b.d. 0.01 0.06 0.01

F b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 1.15 1.20 1.12 1.37 1.34 1.19 b.d. 0.09

O 26.9 26.5 26.7 27.4 27.5 27.8 27.9 24.3 27.9 26.4 26.9 26.9 25.7 26.1 26.8 31.3

total 100.3 99.4 97.7 100.9 100.3 100.5 101.8 92.8 101.4 100.2 101.9 100.7 97.2 98.8 99.4 92.1

1d.l. refers to detection limit; Tb and Tm, K, Mg and Na were below detection limit (b.d.); Rbp: rhabdophane; Flo: florencite.

15

9

Table 7.4 Element concentrations from EPMA analyses of weathering-resistant minerals in lateritic regolith of the JG profile

No. 8 80 81 7 82 9 83 84 85 86 87 88

Min Ilm Ilm Ilm Ilm Ilm TiO TiO Zrn Zrn Zrn Zrn Thr

Si b.d. 0.01 b.d. b.d. b.d. 0.25 0.16 14.8 14.6 14.3 14.4 9.47

Zr b.d. b.d. b.d. b.d. b.d. 0.38 0.17 44.4 48.0 48.9 47.3 b.d.

Ti 32.1 37.2 30.6 33.1 31.9 51.0 51.9 0.17 b.d. 0.29 0.02 0.03

Pb b.d. 0.02 b.d. b.d. b.d. 0.02 0.05 b.d. b.d. b.d. b.d. 4.07

Th b.d. 0.05 b.d. 0.01 b.d. 0.14 0.05 0.56 b.d. 0.17 b.d. 45.7

U 0.17 b.d. 0.03 b.d. 0.02 0.14 0.31 0.18 0.04 0.10 0.15 19.0

Al 0.01 b.d. b.d. b.d. b.d. 2.68 2.05 0.59 0.01 0.03 0.14 b.d.

Y b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.40 b.d. 0.47 b.d. 0.50

Ce b.d. b.d. b.d. b.d. b.d. 0.05 b.d. 0.27 0.05 0.31 b.d. b.d.

Pr b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.04 b.d. b.d. b.d. b.d.

Nd b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.21 b.d. 0.12 b.d. b.d.

Sm b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.09 b.d. 0.05 b.d. b.d.

Eu b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.03 b.d. 0.02

Gd b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.09 b.d. 0.06 b.d. 0.08

Dy b.d. 0.79 b.d. 1.04 2.18 b.d. b.d. 0.05 b.d. b.d. b.d. 0.08

Yb 0.07 b.d. 0.08 b.d. b.d. 0.02 0.03 0.14 b.d. 0.20 0.14 0.23

Lu b.d. b.d. b.d. b.d. b.d. b.d. b.d. 2.47 b.d. b.d. 0.02 b.d.

Fe 33.4 25.2 34.5 27.9 27.7 3.91 3.59 0.75 1.52 10.6 1.07 0.06

Mg 0.07 0.03 0.03 0.03 0.06 0.01 b.d. b.d. b.d. 0.02 b.d. b.d.

Ca b.d. b.d. b.d. b.d. b.d. 0.04 0.04 b.d. b.d. 0.06 b.d. b.d.

Sr b.d. b.d. b.d. b.d. b.d. b.d. b.d. 0.44 0.50 0.26 0.49 b.d.

P b.d. b.d. b.d. b.d. b.d. 0.03 0.07 b.d. b.d. b.d. b.d. 0.19

S b.d. b.d. b.d. b.d. b.d. 0.08 0.04 b.d. b.d. 0.15 b.d. b.d.

O 31.3 32.3 30.3 30.3 29.7 38.2 38.0 34.0 34.0 34.8 33.6 20.4

total 97.1 95.6 95.5 92.4 91.5 96.9 96.4 99.8 98.7 99.6 97.4 99.8

La, Tb, Ho, Er and Tm, Na and K were below detection limit (b.d.); Ilm: ilmenite; TiO: titanium oxides (rutile/anatase); Zrn: zircon; Thr: thorite.


160

7.5.1.4 Mass balance of REE

Mass balances of REE were calculated at each sampling depth, based on weighted

average concentrations of matrix and gravel samples, using Zr as the reference element.

Depth profiles of mass fluxes of REE are plotted in Figure 7.5.

The A horizon regolith (above 0.5 m depth) had τ(Zr, La) between −0.9 and −0.8, τ(Zr, Sm)

between −0.7 and −0.5 and τ(Zr, Yb) ca. −0.6. In the ferruginous zone (1.5-5.0 m depth),

τ(Zr, La) was ca. −0.9, τ(Zr, Sm) was between −0.9 and −0.7 and τ(Zr, Yb) was between −0.8

and −0.6; τ(Zr, Ce) had a wide range: from −0.8 in the upper ferruginous zone (1.5 m

depth) increasing to 0.2 in the duricrust (3 m depth), and then decreasing to −0.4 in the

ferruginous mottled zone (5 m depth). In the lower mottled clay zone (8.6 m depth),

τ(Zr, La) was −0.8, close to τ(Zr, Sm), but lower than τ(Zr, Yb) (−0.6); In the saprolite (10 m

depth), τ(Zr, La) was −0.9, suggesting more depletion of La than Sm (τ(Zr, Sm) = −0.8) and

Yb (τ(Zr, Yb) = −0.7).

Figure 7.5 Mass balance calculations of REE against depth in the JG profile, based on

weighted average concentrations of REE in matrix and gravel, using Zr as the reference

element (vertical dashed line refers to mass balance τ(Zr,REE) = 0; Only selected REE are

plotted here, as the remaining REE have similar patterns).

0

2

4

6

8

10

-1.0 -0.5 0.0 0.5

de

pth

(m

)

τ(Zr,La)

τ(Zr,Ce)

τ(Zr,Sm)

τ(Zr,Dy)

τ(Zr,Yb)

τ(Zr,Y)

(enriched)

(depleted)

A horizon

duricrust

saprolite

mottled clay

ferruginous

mottled zone

upper

ferruginous

zone


161

7.5.2 Mineralogy of REE in the parent rock

In the parent meta-granitoids of the JG profile, accessory minerals contained abundant

REE (Figure 7.6 and Table 7.2). The REE concentrations of selected REE-bearing

minerals are presented in Table 7.2.

In the JG parent meta-granitoids, fluorocarbonates were the most important accessory

minerals (<0.06 wt%, Figure 7.6a, b & c), containing average 55 wt% ΣREE (103-fold

enrichment of REE above the average value of parent meta-granitoids), with an average

of 2.8 wt% Th and 0.2 wt% U (Table 7.2). Fluorocarbonates were strongly LREE

selective (Figure 7.7), with (La/Yb)PR ranging from 9.4-15.8, and occurred either as ca.

hundred-micron grain size individual crystals, or as (sub)micron grains intergrown with

other minerals (Figure 7.6a & g). Both moderate negative Eu anomalies (Eu*

average

0.7) and positive Eu anomalies (Eu* ca. 2.1) were observed in fluorocarbonates without

a significant Ce anomaly (Ce* ranged from 0.8 to 1.1). These fluorocarbonates may be

produced by reaction with late- or post-magmatic fluids, or occur as secondary

carbonates, the product of the decomposition of allanite and epidote (Bea, 1996).

Thorite (<0.04 wt%, Figure 7.6d) with partial Zr substitution was another important

accessory mineral, containing average 5.0 wt% ΣREE with a preference for HREE

((La/Yb)PR ca. 0.05), slightly positive Ce anomalies (Ce* ca. 1.4), and moderate

negative Eu anomalies (Eu* ca. 0.5). Sphene (also called titanite, <0.65 wt%, Figure

7.6e) was abundant in the JG meta-granitoids, occurring chiefly as individual crystals or

as intergrowths with ilmenite. It contained variable concentrations of ΣREE up to

0.7 wt%, and no systematic preference for LREE or HREE was observed. Zircon

(<0.03 wt%) also contained varied concentrations of ΣREE, from negligible to 0.3 wt%

with HREE predominating (La, however, was below the EPMA detection limit).

Magnetite (<1.0 wt%, Figure 7.6f) contained up to 0.2 wt% ΣREE with a preference for

HREE (La below EPMA detection limit). Apatite (<0.15 wt%, Figure 7.6g & h) had

ca. 0.15 wt% ΣREE without LREE or HREE selectivity and a moderate negative Ce

anomaly was observed (Ce*=0.4). Ilmenite (<0.23 wt%, Figure 7.6e & h) commonly

had a preference for Yb (up to 0.06 wt%). Feldspars contained up to 0.05 wt% ΣREE.


162

7.5.3 Mode of occurrence of REE in lateritic regolith

7.5.3.1 Mineral phases containing REE in the lateritic regolith

Selected REE-bearing minerals in the duricrust and iron nodules were analysed and

mapped by electron microprobe, and the resulting images are presented in Figure 7.8.

The REE mainly existed in two types of mineral phases in the intensely weathered

regolith: secondary REE-bearing phosphates and primary weathering-resistant minerals;

both are discussed in more detail below.

Secondary REE-bearing phosphates (Figure 7.8 and Table 7.3) are important hosts for

REE in intensely weathered lateritic regolith, especially LREE, as these elements are

essential structural components in these minerals. These REE-bearing phosphates in

regolith are believed to be the secondary weathered products of fluorocarbonates,

allanite, and especially apatite, because no REE-bearing phosphates e.g. monazite, were

observed in the JG parent meta-granitoids. These secondary phosphates were identified

as rhabdophane and florencite based on EPMA analyses, and were the main form of

REE-bearing mineral phases observed in the regolith (Table 7.3), playing an important

role in trapping REE during weathering. Secondary rhabdophane and florencite are

predominantly LREE hosts, as ((La/Yb)PR ranged from 1.2-7.5. Conversely, xenotime is

significant for retaining HREE (Bea, 1996) and was observed as micron-size crystals in

the duricrust; however, this size range is below the spatial resolution of the electron

microprobe, and thus the compositional results cannot be separated from the

interference of Al, Si and Fe in the nearby clay, quartz and Fe oxides. Most of the

REE-bearing secondary phosphates are in the size range 1-10 µm and are distributed in:

(i) in the clay layers rather than the iron cores of iron nodules in the ferruginous zones

(Figure 7.8, Figure 7.9 & Figure 7.10); and (ii) in the clay matrix of iron nodules from

the A horizon (Figure 7.8). Note that iron nodules from the A horizon are different from

the iron nodules in the ferruginous zone, as they are non-concentric, having a kaolinitic

matrix cemented with Fe oxides without layering (Appendix 11.4).

These secondary rhabdophane minerals usually contained a 103-fold enrichment of

ΣREE over the average JG parent meta-granitoids. Rhabdophane lacked an apparent Ce


163

anomaly, but had variable Eu anomalies (Eu* ranged from 0.4-1.3), reflecting the strong

mineralogical control on the redistribution of REE in the intensely weathered lateritic

regolith, similar to Braun’s study (1993). High Th concentrations in the secondary

rhabdophane and florencite in JG regolith were also determined, consistent with high Th

concentrations in LREE-rich fluorocarbonates in meta-granitoids.

Weathering-resistant minerals (Table 7.4) such as zircon, ilmenite, rutile and anatase,

present in trace to minor concentrations in lateritic regolith, are also important hosts for

REE. The REE, especially HREE, included in these weathering-resistant minerals are

not commonly expected to be extensively mobilized during pedogenesis, although

erosion and dissolution may occur under very intense weathering (Taunton et al.,

2000a). In weathered regolith, and especially in extremely weathered lateritic regolith,

zircon is the most important of these minerals to host significant concentrations of REE.

Zircon in regolith contained up to 3.4 wt% ΣREE, with a preference for HREE or Ce

(0.05 wt%). Zircon also contained varied concentrations of Th (0-0.56 wt%) and U

(0.04-0.18 wt%). In addition, ilmenite was an important HREE-selective mineral in the

ferruginous regolith, especially for Dy and Yb, containing up to 2 wt% Dy and/or

0.08 wt% Yb whereas other REE were below detection limits. Ilmenite and rutile

concentrated Yb, containing up to 0.03 wt%. Thorite (ThSiO4), though very rare in the

lateritic regolith (only one grain was observed), contained 0.64 wt% ΣREE with a

preference for HREE.

16

4

(a) (b) (c) (d)

(e) (f) (g) (h)

Figure 7.6 Backscatter electron images of REE-bearing accessory minerals in parent meta-granitoids of the JG profile: (a) micron-size fluorocarbonate

intergrown with 100 µm-size fluorocarbonate; (b) and (c) REE-rich fluorocarbonates; (d) thorite rich in REE and Zr; (e) sphene intergrown with

ilmenite surrounded by feldspars; (f) magnetite surrounded by quartz; (g) apatite intergrown with a tiny crystal REE-bearing fluorocarbonate; (h)

apatite intergrown with ilmenite; (Qz-quartz; Ap-apatite; Ilm-ilmenite; Spn-sphene; Fsp-feldspar; Mag-magnetite).


165

Figure 7.7 REE distribution patterns of fluorocarbonate and thorite normalized by the

parent meta-granitoids in the JG profile (Fc: fluorocarbonate; Thr: thorite).

0.0

0.2

0.4

0.6

0.8

La Ce Pr Nd Sm Eu Gd Yb

RE

E/p

are

nt

me

ta-g

ran

ito

id

Fc

Fc

Fc

Fc

Thr


166

(a) (b)

(c) (d)

(e) (f)

Figure 7.8 Backscatter electron images of REE-bearing secondary phosphate minerals

in regolith samples of the JG profile: (a) and (b) secondary rhabdophane in the clay

layer of iron nodules in the ferruginous zone at 1.5m depth; (c) and (d) secondary

rhabdophane in clay matrix of iron nodules in A horizon at 0.4m depth; (e) and (f) are

secondary florencite locating in the clay layer of iron nodules in ferruginous zone at

1.5m depth; (Zrn-zircon; dark circles are secondary REE-bearing phosphates).


167

(a)

(i

(ii


168

(b)

(i

(ii

Figure 7.9 Images of REE-bearing secondary phosphates located in the clay layer of

iron nodules at 1.5m depth in the JG profile; element compositions of the phosphate are

listed in the Table 7.3: (a) labelled as the No. 63 rhabdophane; (b) labelled as the No. 64

rhabdophane; (i is backscatter image: the rectangular box indicates the area mapped by

EPMA, and the bright spot in the rectangular box is the fine-grained (<10 µm)

secondary rhabdophane; and (ii is the corresponding microprobe mapping: CP refers to

backscatter scan image, and SL refers to secondary scan image.

16

9

(a) (b) (c)

Figure 7.10 Mapping of secondary rhabdophane in iron nodule at 1.5 m depth of the JG profile. (a) backscatter image; (b) elemental mapping of the

rectangular black box in (a) by EPMA; (c) RGB post-imaging of the rectangular white box in (a), collected by SXFM with Maia 384/96 detector, using

Geopixe software with Ni foils as the in-house reference standards; mapping area 3.6×0.8 mm, scan duration 30 min, Maia run number 18745; (the

blue circles in (a) are the bright spots of Ti minerals in (b); the green circles in (a) are the green spots of Ti minerals in (c); red circle is the micron-size

secondary rhabdophane; element compositions are listed in the Table 7.3 and labelled as the No. 65 rhabdophane).


170

7.5.3.2 Mode of occurrence of Ce in the duricrust and iron nodules

Cerium was fractionated from the other REE and significantly enriched in iron nodules

(up to 200 ppm). Element mapping of the duricrust and iron nodules shows that, in

addition to being hosted by secondary phosphate and weathering-resistant minerals, Ce

also precipitated as poorly crystalline phases: (i) as a rim along the Fe-rich pores in the

duricrust (Figure 7.11); (ii) as a rim along the boundary between clay layers (Al-rich)

and iron layers (Fe-rich) in iron nodules (Figure 7.12); and (iii) as joint matrix between

two iron cores within one large nodule (Figure 7.13). Quantitative microprobe analysis

revealed that the concentration of Ce in the rims was up to 1.5 wt% and varied with

location, whereas most of other REE except Gd, Sm and Nd were below detection

limit (Table 7.5). However, the size range of the rims (width<1 µm) is below the spatial

resolution of the microprobe, and therefore the compositional results cannot be

separated from the interference of Al and Fe in the matrix. Despite this problem,

quantitative analysis by microprobe of selected several areas revealed that Ce did not

always exist with P or Si, and the sum of oxides was <100, likely reflecting a hydrous

Ce (hydr)oxide. In addition, a correlation between Th, Zr and Ce was also identified in

both element mapping and chemical analyses (Chapter Five).

In addition to amorphous Fe, crystalline Fe was also observed to be important for

retention of REE, including Ce. A crystalline Fe oxide in the duricrust contained

(sub)micron-size Ce-rich phosphates (Figure 7.14). This may suggest that crystalline Fe

can partially control Ce redistribution by secondary mineral intergrowth, whereas

amorphous Fe controls Ce occurrence by sorption and coprecipitation of Ce

(hydr)oxide.


171

(a)

(b)


Al/Fe-rich pores in the duricrust; (a) is backscatter image with rectangular box

indicating the area mapped by EPMA; and (b) presents the corresponding EPMA

mapping; element concentrations of spot analysis of Ce-rich rim are listed in Table 7.5.


172

(a)

(b)


boundary between clay and iron layers in iron nodules; (a) is backscatter image with

rectangular box indicating the area mapped by EPMA; and (b) presents the

corresponding EPMA mapping; element concentrations of spot analysis of Ce-rich rim

are listed in Table 7.5; the slightly paler rectangular zone in (a) resulted from previous

beam scans by the electron microprobe.


173

(a)

(b)

Figure 7.13 Cerium fractionated from other REE and occurring as joint matrix between

two iron cores within one large nodule; (a) is backscatter image with rectangular box

indicating the area mapped by EPMA; and (b) presents the corresponding EPMA

mapping; element concentrations of spot analysis of Ce-rich rim are listed in Table 7.5;

the slightly paler rectangular zone in (a) resulted from previous beam scans by the

electron microprobe.

17

4

(a) (b) (c)

Figure 7.14 Backscatter electron images of crystalline Fe oxides intergrown with micron-size Ce-rich secondary phosphates in the duricrust: (a)

crystalline Fe oxides; (b) and (c) Ce-rich secondary phosphates.

Table 7.5 Element concentrations in Figure 7.11, Figure 7.12 & Figure 7.13 of the duricrust and iron nodules in the JG profile

El.*

Element concentrations (wt%)

Ce P Si Al Fe Gd Yb Ca U Th Pb S O total

Figure 7.11 1.16 b.d. 0.64 10.7 49.5 b.d. 0.09 0.01 0.03 0.55 0.05 0.14 25.3 89.3

Figure 7.12 0.49 0.02 0.58 26.8 27.4 0.03 b.d. 0.03 0.03 0.23 0.03 0.06 32.6 88.3

Figure 7.13 1.49 0.06 0.49 28.5 19.4 0.12 b.d. 0.10 0.03 0.68 0.02 0.09 32.1 83.1

*Apart from Ce, Gd and Yb, other REE and Y were below the microprobe detection limit, so are not listed above; the spots for analysis were located at the brightest areas of the

Ce-mapping images.


175

7.5.3.3 Determination of Yb onto iron oxide core

In order to understand more fully the effect of Fe oxides on the distribution and

fractionation of REE, 16 random spots in the iron core and clay layer of iron nodules

from the A horizon (0.4 m depth) and the upper ferruginous zone (1.5 m depth) were

analysed by EPMA. Voids in the iron nodules were avoided. Trace concentrations of Yb

(0.02-0.12 wt%) were determined in seven spots, including six spots in the iron core

and one spot in the clay layers (Table 7.6). Only one spot, in the clay layer, contained

trace concentrations of Pr.

Table 7.6 REE concentrations of random spots in iron core and clay layer in iron

nodules from the A horizon and upper ferruginous zone of the JG profile

El.

Concentration (wt%)

iron

core

iron

core

iron

core

iron

core

iron

core

iron

core

clay

layer

Si 0.27 0.26 0.60 0.24 0.26 0.54 0.83

Zr 0.03 b.d. 0.03 0.03 b.d. b.d. b.d.

Ti 0.27 0.08 0.27 0.43 0.83 0.04 0.44

Pb 0.03 0.06 0.04 b.d. 0.03 0.04 b.d.

Th 0.01 b.d. 0.02 0.02 0.08 0.03 b.d.

U 0.03 0.02 0.03 0.03 0.03 0.04 b.d.

Al 3.22 2.99 6.87 15.3 8.48 4.32 35.7

Pr b.d. b.d. b.d. b.d. b.d. b.d. 0.02

Nd b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Yb 0.11 0.04 0.07 0.04 0.02 0.12 0.05

Lu b.d. b.d. b.d. b.d. b.d. b.d. b.d.

Fe 63.0 63.9 58.1 49.8 54.4 57.8 2.42

Ca 0.01 b.d. 0.02 0.01 b.d. 0.02 0.03

Sr b.d. b.d. b.d. b.d. b.d. b.d. 0.01

Na b.d. b.d. b.d. b.d. 0.02 b.d. b.d.

K b.d. b.d. 0.02 0.01 b.d. 0.01 b.d.

P 0.01 0.02 0.03 0.01 0.02 0.01 b.d.

S 0.05 0.02 0.07 0.05 0.14 0.10 0.03

As 0.07 0.06 0.06 0.02 0.04 0.06 b.d.

F b.d. b.d. 0.13 b.d. b.d. 0.15 0.11

O 21.6 21.4 23.8 28.6 24.3 21.2 33.7

total 88.7 88.8 90.2 94.6 88.6 84.57 73.36


176

7.6 Discussion

7.6.1 Geochemical pathways and fractionation of REE

Based on bulk geochemistry and the mode of occurrence of REE, abundant easily

weathered accessory minerals in the parent meta-granitoids e.g. fluorocarbonates,

thorite, apatite, are thought to have broken down during the early stages of weathering,

greatly changing the abundance and distribution pattern of REE. In the saprolite (10 m

depth) ca. 94% ΣREE released by dissolution of the accessory minerals has been

leached away or transported via solutions, and only ca. 6% ΣREE is retained; of that 6%,

ca. 5.1% REE is retained in the saprolite by mineral phases (e.g. secondary phosphates

or weathering-resistant minerals), and the remaining ca. 0.9% ΣREE has been retained

by association with other phases e.g. clay minerals, organic matter and Fe

oxides/oxyhydroxides (revealed by sequential extraction in Chapter Eight). Strong

depletion of ΣREE was also shown in the mottled clay zones (6.5-8.6 m depth,

68%-83%) and the A horizon (32%-53%), except for the duricrust (3 m) and the

ferruginous mottled zone (6.5 m depth) because of their anomalous enrichment of Ce

(Table 7.1). Therefore, significant amounts of REE, excluding Ce, have been leached

out of the profile rather than being translocated at the profile scale, suggesting high

mobility of REE under advanced weathering and strong lateritization. This is in contrast

to commonly reported accumulation of REE at the base of lateritic profiles (Beyala et

al., 2009; Braun et al., 1993; Dequincey et al., 2006; Nesbitt, 1979). The acidic

condition found in the weathered matrix of this study (a range of pH 3.2-5.3 from the

saprolite to the ferruginous zone) is lower than or close to the pH of natural rainfall (ca.

4.5-5.6, Charlson and Rodhe, 1982), which may have enhanced strong leaching of REE

during weathering.

Formation of secondary phosphate minerals e.g. rhabdophane and florencite, constrains

further mobility of REE (Braun et al., 1993), especially LREE (because of their LREE

selectivity, with average (La/Yb)PR = 3.2), and play an important role in redistribution

and fractionation of REE. Though Tripathi and Rajamani (2007) proposed that

secondary minerals are not particularly known to produce strong REE fractionation, a

strong preference for LREE in secondary rhabdophane and florencite was observed in

the lateritic JG regolith (Table 7.3). Unlike the LREE-hosting minerals, most

HREE-selective minerals are weathering-resistant and have undergone residual

accumulation in the weathered regolith. Thus, the low values of (La/Yb)PR in the saprolite


177

(0.2) and the lower mottled clay (0.4) suggests stronger depletion of LREE than HREE

relative to the parent meta-granitoids. As weathering intensifies, however, some

HREE-rich minerals, e.g. ilmenite, may be partially dissolved. The dissolution and even

removal of ilmenite (Chapter Five) under intense-extreme weathering and strong

lateritization may change the fractionation of REE and increase (La/Yb)PR. This might

be the reason for (La/Yb)PR in the upper mottled clay being higher than the lower

mottled clay (1.2 and 0.4 respectively), as the weathering of upper mottled clay

(CIA=94%) is more intense than lower mottled clay (CIA=86%). Further support for

the partial breakdown of ilmenite changing the fractionation of REE was provided by

the sequential extraction experiments (Chapter Eight): in the saprolite, where ca. 92%

ΣHREE was hosted by mineral phases (Res species), and this value has decreased to

ca. 88% in the upper mottled clay and ca. 82% in the duricrust. This reflects the mineral

control on the translocation of REE and the important effects of weathering on

redistribution and fractionation of REE. Therefore, the abundance and fractionation of

REE in regolith are essentially weighted mean of the abundances and compositions of

LREE-rich secondary phosphates and HREE-rich weathering-resistant minerals, which

are predominantly controlled by the weathering conditions (including weathering

intensity, weathering time, accessibility to solution and pH).

7.6.2 Enrichment mechanism of Ce in ferruginous zone

In the JG profile, significant Ce anomalies were observed in the ferruginous zone (Ce*

ranged from 1.5-25.3), especially in the duricrust (Ce*=25.3 in gravel) relative to the

parent meta-granitoids (Table 7.1); except Ce, the other REE are commonly depleted

(Figure 7.5). This enrichment of Ce is consistent with the total Fe enrichment and the

occurrence of neoformed Ce-(hydr)oxide phases rimming along Fe-rich pores in the

duricrust. The co-accumulation of both Fe (III) and Ce (IV) probably reflects a redox

boundary and existence of oxidising conditions. Similar situations have been reported

previously (Angelica and Dacosta, 1993; Braun et al., 1990; Braun et al., 1998).

During lateritic weathering, goethite, hematite, and maghemite form in the ferruginous

zone. Repetitive dissolution-precipitation of Fe oxyhydroxides produces the duricrust

and concentric iron nodules by seasonal growth during alternative wet-dry periods

(Chapter Four). Under oxidizing conditions, Ce is likely to be fractionated from the

other REE and precipitated as Ce (IV) and attaching onto Fe oxyhydroxides. The net

result of these processes will be positive Ce anomalies in the ferruginous zone


178

(including the duricrust).

No apparent Ce anomalies were observed in secondary phosphates in this study (Ce*

ranged from 0.95-1.02), suggesting that the replacement of accessory REE-bearing

minerals by secondary phosphates is not related to a redox gradient. Dissolution of

accessory minerals released REE at the early stages of weathering; some REE

(including Ce) precipitate as secondary phosphates, or complexed by different phases

e.g. clay minerals and retained in regolith, whereas some REE may dissolve in solutions.

Once in solution, pH, Eh and ligand concentrations are important controls on the

solubility of Ce. Fluctuation of water tables can induce redox change; consequently, Ce

may be oxidised to form (hydr)oxides or precipitated with ferric minerals in the

duricrust. Similar situations in lateritic regolith have been reported by Braun et al.,

(1990). Alternatively, repetitive dissolution and precipitation of Fe oxyhydroxides may

shift pH and Eh conditions; in consequence, Ce may fractionate from other REE by

surface precipitation with Fe oxyhydroxides during seasonal growth of iron nodules.

This is supported by the occurrence of Ce as a rim coating along the boundary between

Al-rich and Fe-rich layers in iron nodules.

7.6.3 Effects of Fe oxides/oxyhydroxides on mode of occurrence of REE

Although Fe oxides are known to be efficient sinks for heavy metals due to their large

surface areas (Nedel et al., 2010; Singh and Gilkes, 1992), 58%-82% ΣREE excluding

Ce are depleted in the ferruginous zone of the JG profile, similar to the mottled clay

(67%-83% ΣREE depletion excluding Ce) and the saprolite (93% ΣREE depletion

excluding Ce). This may be the result of persistent intense acidic leaching of

REE-bearing minerals under alternative wet-dry periods.

The mass proportion of amorphous Fe oxyhydroxides was 2.0%, lower than the mass

proportion of crystalline Fe oxides (4.7%), however, amorphous Fe oxyhydroxides

contained a higher mass proportion of ΣREE (4.5%, excluding Ce) than the crystalline

Fe oxides (mass proportion of ΣREE 1.0%, excluding Ce) in the duricrust matrix

(Chapter Eight). This suggests that amorphous Fe oxyhydroxide is more efficient at

scavenging REE than crystalline Fe oxide, a finding that is supported by previous

studies (Compton et al., 2003; Land et al., 1999; Laveuf and Cornu, 2009). In the JG

duricrust (Figure 8.4), amorphous Fe oxyhydroxide has higher (La/Sm)PR (0.16) and

(La/Yb)PR (0.25) than the crystalline Fe oxide (La/Sm)PR (0.37) and (La/Yb)PR (0.49),


179

reflecting the tendency for amorphous Fe oxyhydroxides to be more selective for Sm

(MREE) and Yb (HREE). The (La/Yb)PR < 1.0 in the amorphous and crystalline Fe

extractions is in consistent with trace concentrations of Yb (0.02-0.12 wt%) determined

in the iron cores and the clay layers (Table 7.6). This finding is supported by Marmier et

al.’s experiments (1997; 1999) showing that surface complexation of Yb occurred on

hematite and magnetite between pH5 and 7.

Although REE showed different degrees of association with Fe oxides and

oxyhydroxides in this study, the mechanism of these associations is proposed to be

different:

(i) Poorly crystalline Ce (hydr)oxide phases as a rim along Fe-rich pores in the duricrust

reflect the oxidation and surface precipitation of Ce with Fe oxyhydroxides when redox

changes.

(ii) The association of REE with extracted Fe species and the determination of trace

concentrations of Yb (0.02-0.12 wt%) in the iron core and the clay layers are likely to

be the result of surface complexation, substitution and/or co-precipitation.

(iii) Minor amounts of REE in the extracted crystalline Fe oxides are likely the result of

surface precipitation of secondary REE-bearing phosphates with Fe oxides during

duricrust formation. This is supported by the observation of micron-size REE-bearing

phosphate crystals occurring within the crystalline Fe oxides in the duricrust (Figure

7.14).

(iv) Alternatively, the occurrence of REE-bearing phosphates in the clay layer of the

iron nodules might result from iron nodules sequestering secondary REE-bearing

phosphates during their seasonal growth and cementation at advanced stages of

lateritization.

Therefore, Fe oxides and oxyhydroxides can play important roles in redistribution and

fractionation of REE during intense weathering and lateritization.


A lateritic profile (JG) locating in Jarrahdale, Western Australia was investigated for the

mode of occurrence and geochemical behaviour of REE under intense weathering and

advanced lateritization.

In the saprolite, ca. 94% ΣREE was released by dissolution of accessory minerals in the


180

parent meta-granitoids e.g. fluorocarbonates, thorite and apatite; ca. 5.1% ΣREE was

retained by formation of secondary phosphates and residual accumulation of

weathering-resistant minerals, and ca. 0.9% ΣREE was retained by association with clay

minerals, organic matter and Fe oxides/oxyhydroxides. The formation of secondary

phosphate minerals e.g. rhabdophane and florencite, which are absent in the parent

meta-granitoids, constrains further mobility of REE, especially LREE. The residual

accumulation of weathering-resistant minerals e.g. zircon and rutile/anatase are

important hosts for retention of HREE, especially in extremely weathered ferruginous

zones. Thus, the abundance and stability of LREE-rich secondary phosphates and

HREE-rich weathering-resistant minerals control the fractionation of REE in intensely

weathered lateritic regolith.

In the ferruginous zone, Ce fractionated from the other REE and was abnormally

enriched (Ce*=25.3 in the duricrust gravel). This Ce enrichment is in agreement with

the occurrence of poorly crystalline Ce (hydr)oxide phases as rims along Al/Fe-rich

pores in the duricrust, or along the boundary of Al/Fe-rich layers in iron nodules of the

ferruginous zone. The fractionation of Ce is the result of surface precipitation of Ce (IV)

phases with Fe oxyhydroxides under oxidization and/or changes of pH and Eh

conditions during advanced stages of lateritization.

Trace concentrations of Yb (0.02-0.12 wt%) determined in the iron cores and clay

layers are consistent with the association between REE and amorphous and crystalline

Fe extracted species. They suggest that Fe phases are effective for retention of REE in

lateritic regolith. Fine-grained (<10 µm) REE-bearing phosphates occurred in the clay

layer of iron nodules, most likely to be the result of sequestering by the iron nodules

during their formation.

Therefore, the significant mineralogical control and high mobility of REE during

intense lateritic weathering are important considerations when using REE as tracers of

geochemical processes in intensely weathered environments. The sensitivity of REE to

weathering conditions, especially Ce to redox change, suggests a potential for REE to

be used as complementary geochemical clues along with Fe to investigate the

lateritization processes and to understand the role of Fe oxides and oxyhydroxides on

scavenging trace metals.

181

8 Particle size fractionation and chemical speciation of REE in a

lateritic profile in Western Australia

8.1 Abstract

The rare earth elements (REE) are commonly used as indicators of geochemical and

pedological processes. To better understand the distribution and partitioning of REE in

different particle size fractions and chemical species, an intensely weathered lateritic

profile developed on meta-granitoids in Jarrahdale, Western Australia was investigated.

High concentrations of REE were found in silt and clay fractions. Given the variation in

mass percentages of different particle size fractions, however, gravel and sand contained

56%-98% of the mass of REE in the ferruginous zone. In the saprolite and mottled clay,

clay had the highest mass loading of light REE (LREE) in contrast to the highest mass

loadings of heavy REE (HREE) found in sand. In the ferruginous zone, gravel was the

predominant host for Ce, whereas most of other REE were contained in the gravel and

sand fractions, suggesting that Ce fractionated from other REE and precipitated with, or

was adsorbed by, Fe oxides/oxyhydroxides during formation of duricrust and iron

nodules. The residual species contained the highest percentages of total REE revealed

by sequential extraction, indicating that the abundance and distribution of REE are

controlled by weathering-resistant minerals in intensely weathered regolith. Water

soluble (including adsorbed and exchangeable) species was the fraction hosting the

second highest percentages of total REE, suggesting the important effect of adsorption

by clay and potential bio-availability. The low pH of the profile is believed to account

for the high proportion of REE in this species. The amorphous Fe oxyhydroxide and

crystalline oxide extractions preferentially hosted LREE and MREE over HREE,

whereas the organic matter species was important in complexing HREE. The

distribution and fractionation of REE in different particle size fractions and chemically

extractable species can be used to better understand geochemical behaviour of REE in

intensely weathered lateritic profiles.

8.2 Key words

Particle size; speciation; rare earth element; laterite; weathering; regolith;


182

8.3 Introduction

Physical and chemical weathering of rocks and minerals leads to soil formation. During

this processes, mineral transformations result in a mass flux change of elements within

the mineral assemblage and among particle size fractions, which yields information on

element partitioning and transportation within a profile. The concentration of metals in

soils increases with decreasing particle size (Acosta et al., 2009; Al-Rajhi et al., 1996;

Ljung et al., 2006) because fine particles usually have a larger specific surface area

capable of retaining higher amounts of metals (Wang et al., 2006), or alternatively,

metals are co-precipitated with fine-grained secondary minerals. In the course of

weathering, weathered regolith shifting into smaller particle sizes can result in the

relative accumulation of the REE as refractory elements (Caspari et al., 2006).

However, the substantial influence that the particle size exerts on the abundance and

redistribution of REE in lateritic regolith is not well known. Most of the studies on the

geochemical behaviour of REE during supergene weathering concentrate on bulk

regolith. Therefore, a systematic understanding of the occurrence of REE in different

grain size fractions of lateritic regolith is needed. Understanding grain size effects

would assist pedological interpretation of the fate of REE, and assessment of plant

availability of REE under natural environmental conditions.

Currently, two different approaches are widely used for determining trace element

location and speciation in uncontaminated soils: physical fractionation (e.g. Acosta et

al., 2011; Fichter et al., 1998; Taboada et al., 2006a) and chemical methods (especially

sequential selective extraction, e.g. Aubert et al., 2004; Land et al., 1999). Although the

sequential extraction method suffers from relying on operationally defined fractions and

lack of a standard method for specific trace elements, it is still considered useful for

investigation of element associated phases in soils (Aubert et al., 2004; Cao et al., 2000).

The reactivity or mobility of REE largely depends on their chemical speciation in

weathered profiles, however, few studies have dealt with the speciation of REE in

non-contaminated soils (Aubert et al., 2004), especially in natural weathered profiles.

The objective of this study is to determine the distribution and fractionation of REE in

various particle size fractions and chemical species. The concentrations of REE in

different particle size fractions and chemical species are quantified and fractionation of

REE with respect to particle size distribution and chemical speciation are discussed.

Chapter Eight: Particle size fractionation and chemical speciation of REE in the JG profile in WA

183



This study was performed on a lateritic profile (JG) developed over meta-granitoid

rocks near Jarrahdale, Western Australia. Pre-treatment and analytical methods of the

parent rocks and regolith samples were explained in detail in Chapter Seven. Regolith

samples were separated into two subsample groups: gravel (>2 mm, represented by

suffix ‘g’) and matrix (<2 mm, represented by suffix ‘m’), with the exception of mottled

clay and saprolite, which have only matrix fractions. The fractions of matrix and gravel

were oven dried at 105 °C overnight and ground to ≤ 200 µm prior to fusion in order to

determine trace element concentrations. The regolith matrix was further separated into

the following three size fractions recommended by the International Society of Soil

Science (ISSS) (Marshall, 1947; Marshall, 2003; Prescott et al., 1934): clay (<2 µm),

silt (2-20 µm) and sand (>20 µm) using the sedimentation and wet sieving methods

(Day, 1965). Different particle size fractions were rinsed with MilliQ water three times,

oven dried at 105 °C overnight and ground to ≤ 200 µm prior to fusion.

To investigate chemical species and association behaviour of trace elements, a

sequential extraction procedure was performed. The matrix fraction (< 2 mm) from the

saprolite (JG1m), upper mottled clay (JG3m) and duricrust (JG5m) were selected. An

in-house laboratory reference material was prepared together with selected samples.

Regolith trace elements were extracted as five species (modified from Hall et al., 1996):

(i) water soluble, adsorbed, exchangeable and carbonates bound (WAE); (ii) organic

matter and sulphide bound (Org); (iii) amorphous Fe-Mn oxyhydroxide bound (FeAm);

(iv) crystalline Fe-Mn oxide bound (FeCry); and (v) residual species (Res). Since

carbonates were unlikely to be present in the regolith being studied here due to low pH,

species WAE is considered to include mainly water soluble, adsorbed or exchangeable

elements. Sulfides are also scarce in the lateritic regolith, therefore it is assumed that

species Org is mainly hosted by organic matter complexes. A brief summary of the

method is shown in Table 5.1 in Chapter Five. The residual samples and reference

materials were rinsed with MilliQ water three times, oven dried at 105 °C overnight and

ground to ≤200 µm prior to fusion in order to determine trace element concentrations.

The fusion beads were made by mixing 0.1 g (to an accuracy of 0.1 mg) of finely

ground sample or reference material with 0.7 g 12:22 Norrish flux (lithium

metaborate:lithium tetraborate) and heating in a muffle furnace at 1050 °C for 40


184

minutes. Duplicate fusion beads were also made on 10% of samples to check

preparation errors. After cooling, the fusion beads were dissolved in 100 mL of 10%

analytical grade HCl. The trace elements including REE were determined by

inductively coupled plasma-mass spectrometry (ICP-MS) in Genalysis Laboratory

Services of Intertek Commodities in Maddington, Western Australia. Certified

international standard materials, including stream sediment reference material STSD-2,

STSD-4 (Canada Centre for Mineral and Energy Technology, CANMET) and an

in-house standard material were prepared in the same way as the samples and analysed

together with samples to check the accuracy and precision. The variation between tested

values and expected values was within 10% of the certified values. The concentrations

of REE in different particle size fractions and chemical species are given in Table 8.1

and Table 8.2 respectively.


8.4.2.1 Fractionation of REE

In order to study the fractionation of REE, three groups are identified (Henderson,

1984): the light REE (LREE; from La to Nd), the middle REE (MREE: from Sm to Ho)

and the heavy REE (HREE: from Gd to Lu). The normalized ratios (La/Sm)PR and

(La/Yb)PR were used for identifying fractionations between LREE-MREE and

LREE-HREE using the average composition of parent meta-granitoids as a reference.

8.4.2.2 Calculation of REE mass loading in particle size fraction

To index an element’s partitioning into different particle size fractions, a mean element

mass loading was calculated based on the element’s concentration in a selected grain

size of known mass percentage (Sutherland, 2003).

GSFloading 100 (X i GSi

X i GSii1

n

)

Where:

Xi is the concentration of REE (ppm) in an individual grain size fraction (e.g. <2 µm);

GSi is the mass percentage of an individual fraction, which has limits of 0-100%.

GSFloading is the element mass loading in a selected grain size and the summation of

GSFloading indices for each soil sample equals 100%.

In the ferruginous zone, four classes of particle sizes (clay, silt, sand and gravel) were


185

defined and three in the mottled clay zone and the saprolite (clay, silt and sand). Thus, if

the REE concentration for a given fraction is very high but it forms only a small portion

of the overall sample mass, the contribution of this fraction to the total sample REE

loading will be minimal.

8.5 Results

8.5.1 Concentrations of REE in different particle size fractions

In the lateritic JG profile, silt and clay fractions generally contained the highest

concentrations of REE, except in the saprolite (Figure 8.1). In the ferruginous zone

(from JG6 to JG4, 1.5-5 m depth), clay contained the highest concentrations of LREE

(from La to Nd), followed by the silt fraction. In the duricrust (3 m depth) and

ferruginous mottled zone (5 m depth), however, gravel was abnormally enriched in Ce.

Concentrations of LREE in matrix were slightly higher than in sand in the ferruginous

zone. In the mottled clay (6.5-8.6 m depth) and the saprolite (10 m depth), the relative

concentrations of LREE from high to low were: silt > clay > sand.

MREE (from Sm to Ho) had different distribution patterns between the particle size

fractions. From Sm to Gd, closer to LREE, the highest concentrations were in the clay

fraction in the ferruginous zone but in the silt fraction in the mottled clay zone. From Tb

to Ho, closer to HREE, silt fraction had the highest concentrations except duricrust and

upper ferruginous zone.

HREE (from Er to Lu) and Y showed mostly consistent distribution patterns. The silt

fraction contained the highest concentration of HREE throughout the profile followed

by the clay fraction in the ferruginous zone. In the saprolite and mottled clay, both clay

and sand fractions had similar HREE concentrations.

8.5.2 Mass loading of REE in different particle size fractions

Given the mass percentage of each particle size, the mass loading of selected REE in

each particle size fraction was plotted in Figure 8.2. Although silt and clay fractions had

the highest concentrations of REE, their relatively low mass percentage compared with

other fractions minimized the enrichment.

In the ferruginous zone, gravel dominated the distribution and abundance of Ce, with up


186

to 84% Ce in the duricrust. In the upper ferruginous zone (1.5 m depth), gravel and sand

accounted for more than 95% mass of REE, decreasing to ca. 80% in duricrust (3 m

depth) and ca. 60% in ferruginous mottled zone (5 m depth). In the duricrust, the mass

loading of each REE was higher in the clay fraction than the silt fraction. In the

ferruginous mottled zone, however, REE were fractionated: the mass loadings of LREE

and MREE were higher in the clay fraction whereas the mass loadings of HREE were

higher in the silt fraction.

From the upper mottled clay to the saprolite (JG3-JG1, 6.5-10 m depth), the regolith

does not contain gravel. The clay fraction was the most important host for LREE in

these zones (6.5-10 m depth), especially in the upper mottled clay zone (6.5 m depth)

with ca. 48%-50% LREE was in the clay fraction. Higher mass loadings of HREE

(46%-61%), however, were found to be in the sand fraction in the saprolite and mottled

clay. The mass loading of REE in the silt fraction increased with depth from upper

mottled clay to saprolite.

8.5.3 Speciation of REE from sequential extraction

The sequential extraction experiment revealed the percentages of ΣREE (the total REE

concentration) in each chemical species of representative lateritic regolith in the JG

profile (Figure 8.3). Generally, the ΣREE distribution percentage followed the order:

Res > WAE > FeAm > FeCry and Org. The Res and WAE species dominated the

distribution and abundance of REE, accounting for 89%-98% ΣLREE, 87%-97%

ΣMREE and 91%-98% ΣHREE. The saprolite Res had higher percentages of ΣMREE

(85%) and ΣHREE (92%) than upper mottled clay (75% ΣMREE and 88% ΣHREE) and

duricrust (66% ΣMREE and 82% ΣHREE) and the percentages decreased from saprolite

to duricrust. In addition, the saprolite WAE had higher percentage of ΣLREE (13%)

than upper mottled clay (9%) and duricrust (9%). The percentages of ΣMREE (12%)

and ΣHREE (7%) in the saprolite WAE were lower than in the WAE of upper mottled

clay (20% ΣMREE and 9% ΣHREE) and duricrust (21% ΣMREE and 9% ΣHREE).

The duricrust Org had higher percentage of ΣREE (1.5%) than the Org in the saprolite

(0.6%) and upper mottled clay (0.2%), especially HREE. The percentages of total REE

hosted in the FeAm phase of the duricrust (6.6%) were also higher than in the FeAm

phases of the saprolite (1.4%) and upper mottled clay (1.7%). Similarly, the FeCry phase

in the duricrust also had higher percentages of total REE (3.3%) than the total REE


187

percentage in the FeCry of saprolite (0.8%) and upper mottled clay (0.6%). In addition,

in the duricrust the percentage of REE in the FeAm phase (6.6%) was higher than the

percentages of REE in the FeCry (3.3%) and Org (1.5%) phases.

18

8

Figure 8.1 Concentrations of REE in grain size fractions in the JG profile

Figure 8.2 Mass loading of REE in grain size fractions in the JG profile (JG6-upper ferruginous zone, 1.5 m depth; JG5- duricrust, 3 m depth;

JG4-ferruginous mottled zone, 5 m depth; JG3-upper mottled clay, 6.5 m depth; JG2-lower mottled clay zone, 8.6 m depth; JG1-saprolite, 10 m depth.

Only selected REE are plotted here; other REE showed similar patterns).

18

9

Figure 8.3 Distribution of REE percentages in sequential extractions of the representative regolith of the JG profile. (Res: residual; FeCry: crystalline Fe

oxides; FeAm: amorphous Fe oxyhydroxides; Org: organic matter; WAE: water soluble, adsorbed and exchangeable. JG5- duricrust, 3 m depth;

JG3-upper mottled clay, 6.5 m depth; JG1-saprolite, 10 m depth. (Some REE concentrations were below the detection limit of ICP-MS and are not

presented here).

19

0

Table 8.1 Concentrations of REE in grain size fractions of the JG profile

sample Element concentrations (ppm)

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Y

d.l. 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1

Upper ferruginous zone JG6sand 7.2 21.9 1.5 5.0 1.1 0.2 1.0 0.2 1.1 0.3 0.8 0.1 0.9 0.2 6.8

1.5 m depth JG6silt 14.6 60.1 3.1 10.3 2.0 0.3 1.8 0.3 2.0 0.5 1.5 0.3 1.8 0.4 13.1

JG6clay 17.2 121 4.3 14.8 2.9 0.6 2.5 0.4 2.2 0.5 1.3 0.2 1.2 0.2 13.1

JG6matrix 7.8 28.8 1.6 5.7 1.1 0.2 1.3 0.2 1.2 0.3 0.9 0.2 1.1 0.2 7.6

JG6gravel 7.5 27.0 1.7 6.5 1.4 0.3 1.4 0.2 1.4 0.3 0.9 0.2 1.0 0.2 5.3

Duricrust JG5sand 5.7 35.8 0.8 2.4 0.5 b.d. 0.4 b.d. 0.5 b.d. 0.3 b.d. 0.4 b.d. 2.5

3 m depth JG5silt 13.8 61.1 2.4 7.7 1.6 0.3 1.1 0.2 1.2 0.3 0.9 0.2 1.2 0.2 7.3

JG5clay 14.9 63.2 2.9 10.1 2.1 0.4 1.5 0.2 1.3 0.3 0.8 0.1 1.0 0.2 5.8

JG5matrix 6.1 37.2 0.9 3.0 0.6 0.1 0.7 0.1 0.5 0.1 0.3 0.1 0.5 0.1 3.2

JG5gravel 6.0 224 0.7 2.4 0.4 0.1 1.6 0.1 0.4 0.1 0.2 0.0 0.3 0.1 2.0

Ferruginous mottled zone JG4sand 4.5 16.0 0.7 2.0 0.4 b.d. 0.3 b.d. 0.3 b.d. 0.2 b.d. 0.3 b.d. 1.8

5 m depth JG4silt 20.2 40.7 3.4 10.8 2.2 0.4 1.7 0.3 1.9 0.5 1.6 0.3 2.2 0.5 12.6

JG4clay 25.1 40.5 4.6 14.8 2.8 0.5 2.0 0.3 1.7 0.4 1.0 0.2 1.1 0.2 7.8

JG4matrix 7.8 19.0 1.1 4.3 0.9 0.1 0.7 0.1 0.7 0.2 0.6 0.1 0.7 0.2 4.5

JG4gravel 5.0 107 0.6 1.9 0.3 0.1 0.7 0.0 0.4 0.1 0.2 0.0 0.4 0.1 2.1

Upper mottled clay JG3sand 7.0 9.9 1.0 2.5 0.4 b.d. 0.3 b.d. 0.4 0.1 0.4 b.d. 0.7 0.2 2.9

6.5 m depth JG3silt 27.0 38.2 3.5 9.7 1.7 0.3 1.2 0.2 1.4 0.4 1.1 0.2 1.7 0.3 10.2

JG3clay 22.6 30.2 2.8 7.7 1.1 0.2 0.8 0.1 0.8 0.2 0.4 b.d. 0.5 0.1 4.0

JG3matrix 10.5 14.8 1.3 3.6 0.6 0.1 0.5 0.1 0.4 0.1 0.3 0.1 0.4 0.1 2.7

Lower mottled clay JG2sand 3.4 4.6 0.4 1.0 0.3 b.d. 0.2 b.d. 0.2 b.d. 0.2 b.d. 0.3 b.d. 1.3

8.6 m depth JG2silt 12.3 17.3 1.5 4.1 0.9 0.1 0.8 0.1 0.8 0.2 0.7 0.1 1.0 0.2 5.6

JG2clay 9.1 12.1 1.0 2.4 0.4 b.d. 0.3 b.d. 0.3 b.d. 0.2 b.d. 0.3 b.d. 1.6

JG2matrix 5.2 7.5 0.6 1.7 0.3 0.1 0.2 0.1 0.3 0.1 0.3 0.1 0.5 0.1 2.1

19

1

sample Element concentrations (ppm)


d.l. 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1

Saprolite JG1sand 0.3 0.6 b.d. 0.3 0.2 b.d. 0.1 b.d. 0.1 b.d. 0.1 b.d. 0.2 b.d. 1.0

10 m depth JG1silt 1.3 2.3 0.2 0.9 0.3 b.d. 0.3 b.d. 0.3 b.d. 0.3 b.d. 0.5 0.1 2.2

JG1clay 1.2 2.0 0.2 0.8 0.2 b.d. 0.1 b.d. 0.1 b.d. 0.1 b.d. 0.2 b.d. 0.8

JG1matrix 1.3 2.4 0.2 1.1 0.3 0.1 0.3 0.0 0.2 0.0 0.2 0.0 0.3 0.1 1.3

Average meta-granitoids 27.9 47.4 4.3 14.0 2.1 0.5 1.8 0.2 1.0 0.2 0.8 0.2 1.1 0.2 8.5

d.l. refers to detection limit; b.d. refers to below detection limit.

19

2

Table 8.2 Concentrations of REE in different chemical extractions of representative regolith in the JG profile

Sample Element concentrations (ppm)

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Mn Fe ΣREE%1

d.l. 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 1.00 5.00

JG1m WAE 0.140 0.290 0.040 0.149 0.032 0.007 0.021 0.003 0.017 0.004 0.013 0.002 0.018 0.003 2.00 55.00 12.5

JG1m Org 0.008 0.009 0.002 0.006 0.003 b.d. 0.001 b.d. 0.002 b.d. 0.001 b.d. 0.002 b.d. b.d. 130.0 0.6

JG1m FeAm 0.016 0.032 0.004 0.013 0.004 b.d. 0.003 b.d. 0.003 b.d. 0.002 b.d. 0.003 b.d. 3.00 862.0 1.4

JG1m FeCry 0.023 0.014 0.001 0.003 0.001 b.d. 0.001 b.d. 0.001 b.d. b.d. b.d. 0.001 b.d. 4.00 173.0 0.8

JG1m Res2 1.1 1.9 0.2 0.7 0.2 b.d. 0.2 b.d. 0.2 b.d. 0.2 b.d. 0.3 b.d. 14 6773 84.8

JG3m WAE 0.938 0.905 0.168 0.531 0.083 0.018 0.086 0.014 0.070 0.016 0.041 0.006 0.031 0.005 2.00 86.00 9.2

JG3m Org 0.022 0.019 0.004 0.013 0.002 b.d. 0.002 b.d. 0.002 b.d. 0.001 b.d. 0.003 b.d. b.d. 66.00 0.2

JG3m FeAm 0.130 0.200 0.028 0.098 0.019 0.004 0.018 0.003 0.016 0.003 0.010 0.002 0.009 0.002 2.00 519.0 1.7

JG3m FeCry 0.053 0.079 0.008 0.028 0.005 b.d. 0.004 b.d. 0.004 b.d. 0.002 b.d. 0.002 b.d. 3.00 757.0 0.6

JG3m Res 9.2 13 1.1 2.8 0.5 b.d. 0.3 b.d. 0.3 b.d. 0.2 b.d. 0.4 0.2 29 5554 88.3

JG5m WAE 0.422 2.744 0.149 0.586 0.146 0.031 0.102 0.015 0.073 0.014 0.037 0.005 0.030 0.005 3.00 158.0 9.1

JG5m Org 0.031 0.563 0.015 0.058 0.014 0.003 0.009 0.001 0.006 0.001 0.005 0.001 0.012 0.003 b.d. 83.00 1.5

JG5m FeAm 0.127 2.553 0.054 0.221 0.060 0.012 0.042 0.007 0.038 0.008 0.020 0.003 0.020 0.003 b.d. 438.0 6.6

JG5m FeCry 0.049 1.450 0.011 0.039 0.010 0.002 0.006 0.001 0.007 0.001 0.004 b.d. 0.004 b.d. 6.00 1014 3.3

JG5m Res 6..0 27 0.8 2.5 0.5 b.d. 0.3 b.d. 0.4 b.d. 0.3 b.d. 0.4 b.d. 31 19810 79.6

1ΣREE% refers to percentage of ΣREE in each extraction species;

2The detection limit (d.l.) of the Res species, determined by the fusion method, is 0.1 ppm.


193

8.6 Discussion

Although sequential extraction schemes do not extract chemically discrete forms of

elements, the data have revealed variation of the REE distribution in different particle

size fractions and chemical species. Most of REE were hosted by the Res, indicating

that both the abundance and distribution of REE are controlled by weathering-resistant

minerals in intensely weathered regolith. SEM imaging and EPMA analyses show that

LREE are mostly hosted by secondary phosphates ca. 2-20 µm-size, e.g. rhabdophane

and florencite, and HREE are mainly contained in weathering-resistant minerals of

varied grain size (1-100 µm), e.g. zircon and anatase in the lateritic regolith (Chapter

Seven). The high concentration of REE in the silt fraction (2-20 µm) is in good

agreement with the REE-bearing mineral size in the regolith, especially LREE-rich

secondary minerals, indicating morphological and mineralogical change from

REE-bearing accessory minerals e.g. apatite, fluorocarbonates and thorite in the parent

meta-granitoids to secondary rhabdophane and florencite during intense weathering

and lateritization (Chapter Seven). In addition, in the duricrust an abnormal enrichment

of Ce was observed, especially in gravel. It suggests that Ce fractionated from other

REE and less likely to be mobile than the other REE during formation of duricrust and

iron nodules. This agrees with (i) high concentrations of Ce in the FeAm and FeCry in

the duricrust (Table 8.2); and (ii) precipitation and neoformation of Ce-(hydr)oxides as

a rim between Al/Fe layer boundary or along the Al/Fe-rich pore walls (Figure 7.13 in

Chapter Seven).

A significant proportion of REE bound to the WAE species in natural uncontaminated

soils is not common in previous studies; although some of the WAE-extractable REE

may also be colloidal, and therefore the WAE fraction might be overestimated, it

suggests that some amount of REE is bio-available in the regolith studied here. High

deficiency of REE in the profile, especially in the saprolite, may partially attribute to

low soil pH which favours the conversion of metals from precipitated forms into

dissolved forms (Cao et al., 2001; Harter, 1983). In an acidic environment, such as this

(pH ranges from 3.2 to 4.0 from saprolite to mottled clay in Table 4.1), the


194

predominant REE species in solution is the free Ln3+

ion (Ln denotes REE). In the

duricrust, REE may also partially occur as LnHCO32+

complexes due to the slightly

higher pH (4.7) and organic complexes due to relatively higher dissolved organic

matter (total carbon 0.30%) than in the mottled clay (total carbon 0.08%). Extraction of

adsorbed or exchangeable REE in a spodosol profile has been reported to be closely

related to pH, in the range 4.2 to 6.5 (Land et al., 1999).

In addition, the high proportion of REE bound to the WAE is probably relevant to high

concentrations of REE in the clay fraction. Kaolinite and halloysite were identified in

the saprolite and mottled clay. The transformation from kaolinite to halloysite during

weathering is accompanied by an increase in hydration, a decrease in Si/Al ratio and an

increasing cation exchange capacity (CEC) (Tari et al., 1999). The clay fraction of the

mottled clay zone had (La/Sm)PR ranging from 1.5-1.7 and (La/Yb)PR from 1.2-1.8

relative to the parent meta-granitoids (Figure 8.4), suggesting that clay acts an

important role in trapping REE, especially LREE. This is also supported by Cullers et

al. (1987) who showed that heavy minerals (biotite, hornblende and sphene) in a soil

developed from a granitic parent material appeared to be altering and making LREE

available to the clay minerals forming in the soil. However, opposite fractionation

(HREE more sorbed than LREE) onto kaolinite has also been reported (Coppin et al.,

2002). Adsorption of REE by clay is controlled by the nature of the clay minerals, pH,

ionic strength, the presence of additional ligands such as carbonate or organic

complexes, surface coverage, and effects specific to the characteristics of the different

REE (Coppin et al., 2002; Fendorf and Fendorf, 1996; Koeppenkastrop and Decarlo,

1992; Koeppenkastrop and Decarlo, 1993; Laveuf and Cornu, 2009; Piasecki and

Sverjensky, 2008; Takahashi et al., 1999). As well as these controls, differences in clay

mineralogy can affect fractionation of REE (Laveuf and Cornu, 2009), potentially

explaining the contradictory signatures of REE adsorbed by clay minerals. Usually,

REE adsorption increases with increasing pH (Coppin et al., 2002), which may explain

the increasing concentrations of REE in the WAE and clay fraction from saprolite to

duricrust. The (La/Sm)PR (ranging from 0.1-0.4) and (La/Yb)PR (ranging from 0.1-0.2)

in all particle size fractions of saprolite suggest that La was substantially fractionated


195

from Sm and Yb and greatly depleted from the saprolite. It is consistent with the

breakdown of LREE-rich accessory minerals (e.g. fluorocarbonates and thorite) at the

early stages of weathering. In addition, a high mass of HREE is observed in sand from

saprolite to upper mottled clay, in contrast to most of LREE being present in clay in

these zones. This may suggest that relatively large-grained (ca. 100 µm) and

weathering-resistant minerals, e.g. zircon, anatase or ilmenite contained significant

amounts of HREE, or alternatively, HREE may be adsorbed onto larger-grained

metal-oxide surfaces, e.g. rutile, hematite (Piasecki and Sverjensky, 2008).

The FeAm and FeCry had higher percentages of LREE and MREE than HREE

throughout the regolith studied. In the duricrust the FeAm phase had a preference for

MREE whereas the FeCry showed a preference for LREE. Since there are negligible

variations in complexation constants for the acetate ligand with various REE (Wood,

1993), the fractionation in extraction of the FeCry is not caused by the extractant

solution (Land et al., 1999) in which the only other solute is NH2OH∙HCl. The reasons

why FeAm and FeCry species show different preference for LREE and MREE is not

clear, but are believed to be related to pH and the presence of other ligands such as

organic complexes (Piasecki and Sverjensky, 2008; Quinn et al., 2006). The

fractionation between LREE, MREE and HREE in Fe oxides is subject to debate

(Laveuf and Cornu, 2009) and varied fractionation with enrichments of LREE

(Koeppenkastrop and Decarlo, 1993), MREE (Bau, 1999; Land et al., 1999) or HREE

(Elderfield and Greaves, 1981; Marker and Deoliveira, 1994) have been observed. For

example, Land et al. (1999), studying a spodosol profile, observed an enrichment of

MREE in the FeAm and a clear HREE enrichment relative to the LREE in FeCry and Org

species. The differences in REE fractionation between species probably also arise from

the various proportions of the different types of Fe- and Mn- oxides present (Laveuf

and Cornu, 2009). The affinity of Ce with Fe oxides indicates surface sorption and

oxidation/coprecipitation of Ce onto Fe oxides, which have been examined by many

authors (Bau and Koschinsky, 2009; Davranche et al., 2004; Nedel et al., 2010).

The Org species plays an important role in complexing HREE in this study, especially


196

in the duricrust, in contrast to the FeAm which have a preference for MREE. Affinity of

HREE for organic materials has been observed before (Aubert et al., 2004; Land et al.,

1999). Organic ligands form complexes with HREE which are more stable than those

with LREE (Henderson, 1984; Sonke and Salters, 2006). The proportion of total REE

hosted by the FeAm is higher than both Org and FeCry species, indicating amorphous Fe

oxyhydroxide plays a more important role than other solid components in controlling

the mobility and bioavailability of REE in lateritic regolith.

Figure 8.4 Normalized ratios of (La/Sm)PR and (La/Yb)PR in particle size fractions and

sequential extractions in the JG profile (solid vertical lines indicate ratio=1.0, no

fractionation of REE relative to the parent meta-granitoids).

0

2

4

6

8

10

0.0 0.5 1.0 1.5 2.0

(La/Sm)PR

de

pth

(m)

0

2

4

6

8

10

0.0 0.5 1.0 1.5 2.0

(La/Yb)PRd

ep

th(m

)

sand

silt

clay

matrix

gravel

0

2

4

6

8

10

0.0 0.5 1.0 1.5 2.0

(La/Sm)PR

de

pth

(m)

0

2

4

6

8

10

0.0 0.5 1.0 1.5

(La/Yb)PR

de

pth

(m)

AEC

Org

Am

Cry

Res


197


A systematic study of particle size fractionation and chemical fractionation of REE in a

lateritic weathered profile developed on meta-granitoids in Jarrahdale, Western

Australia showed that most of the REE (by mass) had partitioned into coarse-grained

material (gravel and sand), despite the high concentrations in fine-grained (silt and clay)

fractions. This partitioning by grain size was not consistent, however, across the REE

series, with significant fractionation occurring; for example, in the lower profile most

LREE mass was in the clay (<2 µm) fraction, but most HREE were associated with

sand (>20 µm). The most significant fractionation of REE was shown by a strong Ce

anomaly in ferruginous duricrust, consistent with formation of both ferruginous

materials and the Ce enrichment by oxidative processes such as precipitation of ferric

minerals. Particle size, sequential extraction, and electron microprobe data were

consistent with REE occurrence being dominated, in intensely weathered regolith, by

mineral phases resistant to weathering. The dominance of residual forms in sequential

extracts supported this conclusion, but the existence of significant REE in water

soluble, exchangeable or adsorbed forms was surprising and was likely to be related to

the low pH of regolith materials. This study demonstrates that the distribution and

fractionation of REE within different particle size fractions and chemically extractable

species can be used as clues for better understanding geochemical behaviour of REE in

intensely weathered lateritic profiles. Both have potential implication for pedological

interpretation of the fractionation of REE during weathering and lateritization,

especially when a particle size sorting process is involved.

199

9 Conclusion and future work

9.1 Conclusion

With this thesis, I set out to improve on the current understanding of geochemical

behaviour and fractionation mechanisms of rare earth elements (REE) during

weathering and lateritization. Four intensely weathered lateritic profiles (GE, MQ I,

MQ II and JG) developed on granitoids with dolerite dykes were investigated.

Substanital depletion of base cations, great loss of Si and enrichment of Al and Fe in

the GE and JG profiles were revealed by mass balance calculations using Zr as the

conservative reference element. Significant geochemical processes e.g. intense

leaching of cations, kaolinization, desilication and ferruginization had occured during

the weathering and lateritization history as suggested by the bulk chemical and

mineralogical data and principal component analysis.

In intensely weathered lateritic regolith, REE (except Ce) were significantly depleted,

compared to the parent granitoids, especially in the GE and JG profiles where the

depletion was up to 94%, reflecting high mobility of REE under extreme weathering.

This is an important consideration when using REE as tracers of geochemical

processes, especially in intensely weathered environments. Stronger depletion, relative

to parent rock, of light REE (LREE) over both middle REE (MREE) and heavy REE

(HREE) was also observed in the regolith, although chondrite-normalized REE

patterns still showed LREE-enrichment.

Breakdown of abundant weathering-susceptible LREE-rich minerals in parent

granitoids e.g. allanite and REE-rich fluorocarbonate contributes to initial depletion of

REE, especially LREE, at early stages of weathering. The REE released may be

partially leached away by solutions, or alternatively, precipitated as secondary

LREE-rich phosphate minerals e.g. rhabdophane and florencite. Formation of

secondary phosphates indicates translocation of REE at the mineral scale, which is

important for retention of REE, especially LREE, as this limits their further mobility


200

during weathering. Residual accumulation of weathering-resistant minerals, e.g. zircon,

becomes more important as HREE hosts during advanced weathering and

lateritization.

The importance of stable mineral phases in controlling the occurrence of REE in

intensely weathered regolith is revealed by the dominance of residual species in the

sequential extraction. Therefore, the abundance and fractionation of REE in regolith

essentially correspond to the weighted mean of the abundance and composition of

LREE-rich secondary phosphates and HREE-rich weathering-resistant minerals, which

is closely associated with the weathering conditions (including weathering intensity,

weathering time, accessibility to solution and pH).

In addition to mineral phases, REE were retained in regolith by association with clay

minerals, Fe oxides/oxyhydroxides and organic ligands. Trace to minor amounts of

REE were hosted in water soluble (including adsorbed and exchangeable) species,

amorphous Fe oxyhydroxide and crystalline Fe oxide species, and organic matter

species, as revealed by the sequential extraction. Trace concentrations of Yb

(0.02-0.12 wt%) were substituted with and/or co-precipitated onto the iron nodules.

Fine-grained secondary REE-bearing phosphates were precipitated with crystalline Fe

oxides in the duricrust or incorporated into clay layers of iron nodules. The association

between Fe oxides/oxyhydroxides and REE suggests that Fe oxides and oxyhydroxides

are important for redistribution of REE at advanced stages of weathering.

Positive Ce anomalies in the duricrust of the GE (Ce*=6.1) and JG (Ce

*=25.3) profiles

were observed. This enrichment and fractionation of Ce was evidenced by neoformed

poorly crystalline (hydr)oxides associated with Zr and Th forming a rim on the walls of

Al/Fe-rich pores in the duricrust or along the boundary between Al-rich and Fe-rich

rims in iron nodules. Fractionation of Ce is evidently governed by oxidative processes,

which is consistent with precipitation of the ferric minerals observed in the duricrust.

The absence of any apparent Ce anomalies in secondary phosphates (Ce* ranged from

0.95-1.02) suggests that replacement of accessory REE-bearing minerals by secondary

phosphates will not result in a Ce anomaly.

Chapter Nine: Conclusion and future work

201

In particle size fractionations, silt and clay size fractions generally had higher

concentrations of REE than the sand size fraction, which usually contained lower

concentrations of REE but a higher mass of REE; this indicates the importance of

secondary phosphate formation, adsorption of REE by clay minerals, the dilution effect

of quartz and the presence of weathering-resistant minerals. In sequential extractions of

duricrust, crystalline Fe oxide species showed a preference for LREE, whereas

amorphous Fe oxyhydroxide species favoured MREE and organic matter species

favoured HREE. This suggests the sorption/complexation by different ligands affect

the fractionation of REE during weathering.

In addition, the affinity of Ce with Zr and Th in neoformed phases in the duricrust of

the JG profile suggests that Zr and Th are mobile at the sampling scale. Breakdown of

thorite and REE-bearing fluorocarbonates is believed to be the source for mobile Zr

and Th during early stages of weathering. Redistribution of Th into secondary

phosphates as a trace component and strong partitioning into gravel rather than matrix

showed translocation of Th at mineral assemblage and profile scales. The absence of

primary sphene crystals and the presence of partially weathered ilmenite and rutile in

the ferruginous mottled zone of the JG profile suggests Ti-hosting mineral sphene,

ilmenite and rutile transforms to anatase during intense weathering. The fluctuation of

Ti/Zr in the ferruginous zone, in contrast to the consistency of Zr/Hf throughout the

profile (within the range of the parent granitoids), suggests that Ti and Zr fractionate

from each other and partition between gravel and matrix during extreme weathering

and advanced lateritization. This proves that Ti, Zr and Th are mobile at a variety of

scales, despite their accepted use as reference elements for studying element mass flux

change.

9.2 Future work

First, this research lends support to the hypothesis that Ce fractionates from other REE

and is enriched in the duricrust and iron nodules through neoformation of poorly

crystalline phases associated with Zr and Th. Quantitative analysis revealed that Ce is

not associated with silicate or phosphate in at least two locations, and thus, the most


202

likely phase is proposed to be a (hydr)oxide. However, direct determination of the Ce

valence has not been achieved, and the cerianite phase has not been absolutely

identified by SXRD, most likely due to the poorly crystalline morphology and very

low concentration. Furthermore, the size of the rim (sub-micron) is also below the

resolution of the microprobe, and thus it is difficult to avoid interference from Si, Al

and Fe when using this technique. Given these constraints, it was not possible in this

study to conclusively determine the mode of occurrence of Ce, Zr and Th as

(hydr)oxides only. Therefore, it is suggested that better resolution imaging should be

obtained of micro-morphological characteristics of the neoformed phases using nano

secondary ion mass spectrometry (NanoSIMS) and/or transmission electron

microscopy (TEM). Furthermore, electron diffraction patterns may be obtained by

selected area electron diffraction with TEM (SAED-TEM). The oxidation states and

quantification of elements at sub-micron resolution can be achieved by electron

energy-loss spectroscopy (EELS) and synchrotron near edge x-ray absorption fine

structure spectroscopy (NEXAFS). All these information would greatly enhance our

understanding of the nature of Ce-Zr-Th affinity and the influence of Fe oxides and

oxyhydroxides on sequestering of REE. More detailed mineralogical information will

provide insight into geochemical signatures of Ce, Zr and Th during lateritic

weathering and yield useful information for the geochemical history of ferruginization

and for the scavenging mechanism of trace elements by iron nodules.

Second, the interpretation of preferential depletion of LREE and MREE over HREE is

hindered by the inadequate knowledge of aqueous geochemistry of REE. Lack of

systematic experimental complexation data of the whole series of REE at a wide range

of pH, and in the presence of multi-complex phases, constrains further understanding

of the mobilization and fractionation of REE in lateritic regolith. Significantly, REE in

aqueous phases have not been investigated in this study due to time and financial

limitations. The investigation of the abundance and fractionation of REE in soil

solutions, groundwater and surface water will greatly improve our understanding of the

release and migration of REE during water/rock interactions. In addition, by coupling

isotopic data (e.g. Sm and Nd) in soil and water samples, a better understanding of the


203

genesis and the evolution history of the lateritic regolith and the geochemical pathways

as a function of water/rock interactions can be achieved.

Third, accumulation of REE in surface soils in both MQ profiles is also worthy of

further research. These topsoils are believed to include transported materials. Though

enrichment of REE by lateral transportation under the influence of soil creep or

colluviation has been considered, biogeochemical recycling is also likely. Thus, a

further study of REE concentrations in above- and below-ground tissues of the surface

plants would test an alternative hypothesis for the enrichment of REE in surface soils.

Fourth, the reason for the abnormal enrichment of Th in the lower mottled clay is not

clear with the current geochemical data. Particle size analysis suggests that this

accumulation is specifically in the clay and silt size fractions, but not quite correlated

to the concentrations of Zr, Ti and REE. Therefore, re-sampling of the regolith is

required to investigate whether this abnormal enrichment is at a profile scale or just a

sampling scale; furthermore, quantitative investigation of (i) Th-hosting mineralogy

based on polished thin section, and (ii) chemical speciation based on the sequential

extraction, are proposed to study the mode of occurrence of Th and the reason for the

repeatable extreme accumulation. The isotopic ratios of Th would also provide an

insight to the source of this enrichment.

Fifth, this study concentrated on lateritic profiles developed from similar granitoids. It

is accepted that the parent rock is a fundamental control on the subsequent

geochemical behaviour of REE during weathering; therefore, comparable research of

the geochemical behaviour of REE developed over contrasting rock types (e.g. dolerite)

would be interesting as a complementary study, and would improve our understanding

of the role of parent rock and weathering conditions in mobilizing REE.

Finally, despite lateral transportation at the sampling scale being considered minor in

this study, lateral redistribution at the landscape scale as a component of pedogenetic

processes have not been investigated. It would be worthwhile to investigate the REE

subject to both vertical and horizontal redistribution at a landscape or catchment scale.


204

This is particularly relevant and important given the significant depletion of REE in the

intensely in-situ weathered regolith studied. This will further explain whether this

depletion is complete dispersion or translocation of REE and investigate the location of

accumulation of REE at both landscape and catchment scales if possible.

9.3 Summary

In summary, the substantial loss of REE from the four lateritic profiles studied is an

important consideration in the use of REE as tracers in studying pedogenesis and

weathering, especially in intensely weathered regolith. Primary accessory minerals in

the parent rock play a fundamental role in the abundance and mobilization of REE;

however, the subsequent mobility and fate of REE in these minerals is strongly

influenced by the pedological processes and weathering conditions. Primary

weathering-resistant minerals (e.g. zircon and anatase for HREE) and secondary

phosphates (e.g. rhabdophane and florencite for LREE) are predominant hosts for REE

in lateritic regolith and the weighted average of mineral abundance and REE

concentrations essentially dominates the redistribution and fractionation of REE in

regolith horizons and/or profiles. The difficulties in quantifying the abundance and

differentiating the REE signatures of these minerals constituting the soil horizons

restrict the understanding of the mobilization and fractionation of REE at multi scales

in the course of weathering. This thesis therefore applied a variety of suitable bulk and

in situ analytical techniques (e.g., electron microprobe, synchrotron X-ray fluorescence

microprobe, synchrotron X-ray powder diffraction) to arrive at conclusions on the

mobilization, fractionation and mode of occurrence of REE and associated elements at

both mineral and profile scales.

The results suggest that the redistribution of REE in the profiles studied is derived

from successive mobilization steps. First, REE are initially released from differential

weathering of primary accessory minerals, transported in solution and then partially

deposited as secondary phosphates in the reoglith. Second, some weathering-resistant

primary accessory minerals and secondary phosphates are dissolved and altered,

further releasing REE into solution under extreme weathering. Third, REE released in


205

solution are absorbed and/or complexed by Fe oxides/oxyhydroxides, organic matter

and clay minerals in different horizons and retained in the regolith.

Variations of REE affinity for Fe oxides/oxyhydroxides demonstrate that the signature

of REE can be affected by Fe oxides and oxyhydroxides:

(i) Cerium was abnormally enriched as poorly crystalline (hydr)oxides

attaching onto the Al/Fe-rich pore walls in the ferruginous duricrust during

ferruginization.

(ii) Minor concentrations of REE (up to 10%) were associated with Fe

oxides/oxyhydroxides species in duricrust matrix extraction and trace

concentrations of Yb (up to 0.12 wt%) were determined by EPMA in cores

of iron nodules.

(iii) Micron-size REE-bearing phosphates co-precipitated with crystalline Fe

oxides in duricrust and distributed in the clay layer of iron nodules.

The association between REE, especially Ce, and Fe oxides/oxyhydroxides contributes

to the understanding of the occurrence and behaviour of REE in lateritic regolith and

sheds lights on the redox change and the weathering and lateritization history.

The importance of secondary phosphate and clay minerals on retention of REE is also

revealed by higher concentrations of REE in silt and clay size fractions than sand size

fraction in particle size analyses, which is consistent with higher concentrations of

REE determined in the Res and WAE than other species. The particle size fractionation

reflects the partitioning of REE being controlled by the size of REE-hosting phases

during weathering. This is an important consideration when a particle size sorting

process (e.g. transportation/mass movement/sedimentation) is involved.

In addition, the mobilization of Ti and Zr at the mineral scale, and Th at the profile

scale, was illustrated. Strong partitioning of Th into gravel, and apparent fractionation

of Ti and Zr in the ferruginous zone, suggest that these relatively conservative

elements can be mobile at different scales under extreme weathering and advanced

lateritization. This is often overlooked when using these elements as internal reference


206

elements to calculate element mass flux changes in supergene environment. This study

adds to and improves our understanding of the geochemical behaviour and mode of

occurrence of Ti, Zr and Th during supergene weathering.

207

10 References

Acosta, J.A., Cano, A.F., Arocena, J.M., Debela, F., Martínez-Martínez, S., 2009.

Distribution of metals in soil particle size fractions and its implication to risk

assessment of playgrounds in Murcia City (Spain). Geoderma 149(1-2), 101-109.

Acosta, J.A., Martínez-Martínez, S., Faz, A., Arocena, J., 2011. Accumulations of

major and trace elements in particle size fractions of soils on eight different

parent materials. Geoderma 161(1-2), 30-42.

Aide, M., Smith-Aide, C., 2003. Assessing soil genesis by rare-earth elemental analysis.

Soil Science Society of America Journal 67(5), 1470-1476.

Aide, M.T., Aide, C., 2012. Rare earth elements: Their importance in understanding

soil genesis. ISRN Soil Science 2012, 1-11.

Aide, M.T., Heberlie, L., Statler, P., 1999. Soil genesis on felsic rocks in the St.

Francois Mountains. II. The distribution of elements and their use in

understanding weathering and elemental loss during genesis. Soil Science

164(12), 946-959.

Aide, M.T., Pavick, Z., 2002. Rare earth element mobilization and migration in a

Wisconsin spodosol. Soil Science 167(10), 680-691.

Al-Rajhi, M.A., Al-Shayeb, S.M., Seaward, M.R.D., Edwards, H.G.M., 1996. Particle

size effect for metal pollution analysis of atmospherically deposited dust.

Atmospheric Environment 30(1), 145-153.

Anand, R.R., Butt, C.R.M., 2000. Approaches to geochemical exploration in lateritic

and related terrains: A comparison of Australian and African terrains. Southern

Africa - Australia Mineral Sector Synergies Symposium, Proceedings, 2000.

Anand, R.R., Butt, C.R.M., 2003. Distribution and evolution of 'laterites' and lateritic

weathering profiles, Darling Range, WA. Australian Geomechanics 38(4), 41-58.

Anand, R.R., Butt, C.R.M., 2010. A guide for mineral exploration through the regolith

in the Yilgarn Craton, Western Australia. Australian Journal of Earth Sciences

57(8), 1015-1114.

Anand, R.R., Gilkes, R.J., 1987. The association of maghemite and corundum in

Darling Range laterites, Western Australia. Australian Journal of Soil Research


208

25(3), 303-311.

Anand, R.R., Gilkes, R.J., Roach, G.I.D., 1991. Geochemical and mineralogical

characteristics of bauxites, Darling Range, Western Australia. Applied

Geochemistry 6(3), 233-248.

Anand, R.R., Hough, R.M., Phang, C., Norman, M.D., 2006. The formation of

ferruginous pisoliths and the mobility of gold and pathfinder elements in the

Yilgarn Craton, Western Australia. Geochimica et Cosmochimica Acta 70(18),

A15-A15.

Anand, R.R., Paine, M., 2002. Regolith geology of the Yilgarn Craton, Western

Australia: Implications for exploration. Australian Journal of Earth Sciences

49(1), 3-162.

Anand, R.R., Robertson, I.D.M., Wilford, J., Phang, C., Wildman, J.E., Shu, L.,

Munday, T.J., Jones, M.R., 2010. Regolith and exploration geochemical strategy

for the Mount Isa Inlier and surroundings. Smart Science for Exploration and

Mining.

Anders, E., Grevesse, N., 1989. Abundances of the elements-meteoritic and solar.

Geochimica et Cosmochimica Acta 53(1), 197-214.

Angelica, R.S., Dacosta, M.L., 1993. Geochemistry of rare-earth elements in surface

lateritic rocks and soils from the Maicuru complex, Para, Brazil. Journal of

Geochemical Exploration 47(1-3), 165-182.

Arne, D.C., Stott, J.E., Waldron, H.M., 1999. Biogeochemistry of the Ballarat East

Goldfield, Victoria, Australia. Journal of Geochemical Exploration 67(1-3), 1-14.

Åström, M., Corin, N., 2003. Distribution of rare earth elements in anionic, cationic

and particulate fractions in boreal humus-rich streams affected by acid sulphate

soils. Water Research 37(2), 273-280.

Aubert, D., Probst, A., Stille, P., 2004. Distribution and origin of major and trace

elements (particularly REE, U and Th) into labile and residual phases in an acid

soil profile (Vosges Mountains, France). Applied Geochemistry 19(6), 899-916.

Aubert, D., Stille, P., Probst, A., 2001. REE fractionation during granite weathering

and removal by waters and suspended loads: Sr and Nd isotopic evidence.


Chapter Ten: Reference

209

Banfield, J.F., Eggleton, R.A., 1989. Apatite replacement and rare earth mobilization,

fractionation, and fixation during weathering. Clays and Clay Minerals 37(2),

113-127.

Bao, Z.W., Zhao, Z.H., 2008. Geochemistry of mineralization with exchangeable REY

in the weathering crusts of granitic rocks in south China. Ore Geology Reviews

33(3), 519-535.

Barrea, R.A., Bonzi, E.V., 2001. Rare earth experimental L X-ray fluorescence

cross-sections at 13 and 14 keV with synchrotron radiation. X-Ray Spectrometry

30(1), 3-7.

Bau, M., 1999. Scavenging of dissolved yttrium and rare earths by precipitating iron

oxyhydroxide: Experimental evidence for Ce oxidation, Y-Ho fractionation, and

lanthanide tetrad effect. Geochimica et Cosmochimica Acta 63(1), 67-77.

Bau, M., Koschinsky, A., 2009. Oxidative scavenging of cerium on hydrous Fe oxide:

Evidence from the distribution of rare earth elements and yttrium between Fe

oxides and Mn oxides in hydrogenetic ferromanganese crusts. Geochemical

Journal 43(1), 37-47.

Bea, F., 1996. Residence of REE, Y, Th and U in granites and crustal protoliths;

Implications for the chemistry of crustal melts. Journal of Petrology 37(3),

521-552.

Bea, F., Montero, P., 1999. Behavior of accessory phases and redistribution of Zr, REE,

Y, Th, and U during metamorphism and partial melting of metapelites in the

lower crust: An example from the Kinzigite Formation of Ivrea-Verbano, NW

Italy. Geochimica et Cosmochimica Acta 63(7-8), 1133-1153.

Beauvais, A., 1999. Geochemical balance of lateritization processes and climatic

signatures in weathering profiles overlain by ferricretes in Central Africa.

Geochimica et Cosmochimica Acta 63(23-24), 3939-3957.

Bednar, A.J., Gent, D.B., Gilmore, J.R., Sturgis, T.C., Larson, S.L., 2004. Mechanisms

of thorium migration in a semiarid soil. Journal of Environmental Quality 33(6),

2070-2077.

Beyala, V.K.K., Onana, V.L., Priso, E.N.E., Parisot, J.C., Ekodeck, G.E., 2009.

Behaviour of REE and mass balance calculations in a lateritic profile over


210

chlorite schists in South Cameroon. Chemie Der Erde-Geochemistry 69(1),

61-73.

Bidoglio, G., Gibson, P.N., Haltier, E., Omenetto, N., 1992. XANES and laser

fluorescence spectroscopy for rare-earth speciation at mineral-water interfaces.

Radiochimica Acta 58-59, 191-197.

Bierlein, F.P., Waldron, H.M., Arne, D.C., 1999. Behaviour of rare earth and high field

strength elements during hydrothermal alteration of meta-turbidites associated

with mesothermal gold mineralization in central Victoria, Australia. Journal of

Geochemical Exploration 67(1-3), 109-125.

Bourman, R.P., 1993. Perennial problems in the study of laterite-A review. Australian

Journal of Earth Sciences 40(4), 387-401.

Braun, J.J., Ngoupayou, J.R.N., Viers, J., Dupre, B., Bedimo, J.P.B., Boeglin, J.L.,

Robain, H., Nyeck, B., Freydier, R., Nkamdjou, L.S., Rouiller, J., Muller, J.P.,

2005. Present weathering rates in a humid tropical watershed: Nsimi, South

Cameroon. Geochimica et Cosmochimica Acta 69(2), 357-387.

Braun, J.J., Pagel, M., 1994. Geochemical and mineralogical behavior of REE, Th and

U in the Akongo lateritic profile (SW Cameron). Catena 21, 173-177.

Braun, J.J., Pagel, M., Herbillon, A., Rosin, C., 1993. Mobilization and redistribution

of REEs and thorium in a syenitic lateritic profile: A mass balance study.


Braun, J.J., Pagel, M., Muller, J.P., Bilong, P., Michard, A., Guillet, B., 1990. Cerium

anomalies in lateritic profiles. Geochimica et Cosmochimica Acta 54(3),

781-795.

Braun, J.J., Viers, J., Dupre, B., Polve, M., Ndam, J., Muller, J.P., 1998. Solid/liquid

REE fractionation in the lateritic system of Goyoum, east Cameroon: The

implication for the present dynamics of the soil covers of the humid tropical

regions. Geochimica et Cosmochimica Acta 62(2), 273-299.

Breemen, N., Buurman, P., 1998. Soil formation. Dordrecht; Boston: Kluwer Academic

Publishers.

Brimhall, G.H., Chadwick, O.A., Lewis, C.J., Compston, W., Williams, I.S., Danti,

K.J., Dietrich, W.E., Power, M.E., Hendricks, D., Bratt, J., 1992. Deformational


211

mass-transport and invasive processes in soil evolution. Science 255(5045),

695-702.

Brimhall, G.H., Compston, W., Williams, I.S., Reinfrank, R.F., Lewis, C.J., 1994.

Darwinian zircons as provenance tracers in dust-size exotic components in

laterites: mass balance and SHRIMP ion microprobe results. In: A.J.

Ringrosevoase, G.S. Humphreys (Eds.), Soil Micromorphology: Studies in

Management and Genesis. Developments in Soil Science, pp. 65-81.

Brimhall, G.H., Lewis, C.J., Ague, J.J., Dietrich, W.E., Hampel, J., Teague, T., Rix, P.,

1988. Metal enrichment in bauxites by deposition of chemically mature aeolian

dust. Nature 333(6176), 819-824.

Brimhall, G.H., Lewis, C.J., Ford, C., Bratt, J., Taylor, G., Warin, O., 1991.

Quantitative geochemical approach to pedogenesis-Importance of parent material

reduction, volumetric expansion, and eolian influx in lateritization. Geoderma

51(1-4), 51-91.

Brookins, D., 1988. Eh-pH diagrams for geochemistry. Springer - Verlag, Berlin.

Brown, D.J., Helmke, P.A., Clayton, M.K., 2003. Robust geochemical indices for

redox and weathering on a granitic laterite landscape in central Uganda.


Brown, R.E., Parker, H.M., Smith, J.M., 1955. Disposal of liquid wastes to the ground,

Internal Conference on the Peaceful Uses of Atomic Energy, United Nations,

New York, pp. 669-675.

Brugger, J., Etschmann, B., Pownceby, M., Liu, W.H., Grundler, P., Brewe, D., 2008.

Oxidation state of europium in scheelite: Tracking fluid-rock interaction in gold

deposits. Chemical Geology 257(1-2), 26-33.

Buettner, K.M., Valentine, A.M., 2012. Bioinorganic chemistry of titanium. Chemical

Reviews 112(3), 1863-1881.

Bureau of Meterology, 2012. Climate statistics for Australian locations.

Cantrell, K.J., Byrne, R.H., 1987. Rare-earth element complexation by carbonate and

oxalate ions. Geochimica et Cosmochimica Acta 51(3), 597-605.

Cao, X.D., Chen, Y., Gu, Z.M., Wang, X.R., 2000. Determination of trace rare earth

elements in plant and soil samples by inductively coupled plasma-mass


212

spectrometry. International Journal of Environmental Analystical Chemistry

76(4), 295-309.

Cao, X.D., Chen, Y., Wang, X.R., Deng, X.H., 2001. Effects of redox potential and pH

value on the release of rare earth elements from soil. Chemosphere 44(4),

655-661.

Caspari, T., Baumler, R., Norbu, C., Tshering, K., Baillie, I., 2006. Geochemical

investigation of soils developed in different lithologies in Bhutan, Eastern

Himalayas. Geoderma 136(1-2), 436-458.

Castor, S.B., Hedrick, J.B., 2006. Rare earth elements. In: J.E. Kogel, N.C. Trivedi,

J.M. Barker (Eds.), Industrial Minerals and rocks. Society for Mining,

Metallurgy and Exploration, Littleton, pp. 769-792.

Cerny, P., Francis, S., Peter, M., 1989. Lanthanides, tantalum, and niobium: mineralogy,

geochemistry, characteristics of primary ore deposits, prospecting, processing,

and applications: proceedings of a workshop in Berlin in November 1986.

Special publication of the Society for Geology Applied to Mineral Deposits.

Springer, Berlin.

Charlson, R.J., Rodhe, H., 1982. Factors controlling the acidity of natural rainwater.

Nature 295(5851), 683-685.

Chen, J.R., Chao, E.C.T., Back, J.M., Minkin, J.A., Rivers, M.L., Sutton, S.R., Cygan,

G.L., Grossman, J.N., Reed, M.J., 1993. Rare-earth element concentrations in

geological and synthetic samples using synchrotron X-ray fluorescence analysis.

Nuclear Instruments & Methods in Physics Research Section B-Beam

Interactions with Materials and Atoms 75(1-4), 576-581.

Chiarenzelli, J., Aspler, L., Dunn, C., Cousens, B., Ozarko, D., Powis, K., 2001.

Multi-element and rare earth element composition of lichens, mosses, and

vascular plants from the Central Barrenlands, Nunavut, Canada. Applied


Colin, F., Alarcon, C., Vieillard, P., 1993. Zircon: An immobile index in soils?

Chemical Geology 107, 273-276.

Compton, J.S., White, R.A., Smith, M., 2003. Rare earth element behavior in soils and

salt pan sediments of a semi-arid granitic terrain in the Western Cape, South


213

Africa. Chemical Geology 201(3-4), 239-255.

Condie, K.C., Dengate, J., Cullers, R.L., 1995. Behavior of rare-earth elements in a

paleoweathering profile on granodiorite in the front range, Colorado, USA.


Coppin, F., Berger, G., Bauer, A., Castet, S., Loubet, M., 2002. Sorption of lanthanides

on smectite and kaolinite. Chemical Geology 182(1), 57-68.

Cornu, S., Lucas, Y., Lebon, E., Ambrosi, J.P., Luizao, F., Rouiller, J., Bonnay, M.,

Neal, C., 1999. Evidence of titanium mobility in soil profiles, Manaus, central

Amazonia. Geoderma 91(3-4), 281-295.

Costa, M.L., 1997. Lateritization as a major process of ore deposit formation in the

Amazon region. Exploration and Mining Geology 6(1), 79-104.

Cromieres, L., Moulin, V., Fourest, B., Guillaumont, R., Giffaut, E., 1998. Sorption of

thorium onto hematite colloids. Radiochimica Acta 82, 249-255.

Cullers, R.L., Barrett, T., Carlson, R., Robinson, B., 1987. Rare-earth element and

mineralogic changes in holocene soil and stream sediment-a case study in the wet

mountains, Colorado, USA. Chemical Geology 63(3-4), 275-297.

Dardenne, K., Denecke, M.A., Schäfer, T., 2003. Spectroscopic and structural

investigation of the Lu incorporation limit in the 2-line ferrihydrite (2LFh)

transformation products, Argonne national laboratory, Argonne.

Davranche, M., Pourret, O., Gruau, G., Dia, A., 2004. Impact of humate complexation

on the adsorption of REE onto Fe oxyhydroxide. Journal of Colloid and Interface

Science 277(2), 271-279.

Day, P.R., 1965. Particle fraction standard particle size analysis. Methods of Soil

Analysis. Part 1. Physical and Mineralogical Properties. American Society of

Agronomy, Madison.

Deer, W.A., Howie, R.A., Zussman, J., 1992. An introduction to the rock forming

minerals. Longman Scientific & Technical New York.

Delattre, S., Utsunomiya, S., Ewing, R.C., Boeglin, J.-L., Braun, J.-J., Balan, E., Calas,

G., 2007. Dissolution of radiation-damaged zircon in lateritic soils. American

Mineralogist 92(11-12), 1978-1989.

Dequincey, O., Chabaux, F., Clauer, N., Sigmarsson, O., Liewig, N., Leprun, J.C.,


214

2002. Chemical mobilizations in laterites: Evidence from trace elements and

U238-U234-Th230 disequilibria. Geochimica et Cosmochimica Acta 66(7),

1197-1210.

Dequincey, O., Chabaux, F., Leprun, J.C., Paquet, H., Clauer, N., Larque, P., 2006.

Lanthanide and trace element mobilization in a lateritic toposequence: Inferences

from the Kaya laterite in Burkina Faso. European Journal of Soil Science 57(6),

816-830.

Ding, Y.C., Liu, G.Q., Wang, X.R., 2002. Environmental chemical behaviors of

rare-earth elements in soil and their biological effects. Agricultural Environment

Protection (in Chinese) 21(6), 567-571.

Duddy, I.R., 1980. Redistribution and fractionation of rare-earth and other elements in

a weathering profile. Chemical Geology 30(4), 363-381.

Dupre, B., Viers, J., Dandurand, J.L., Polve, M., Benezeth, P., Vervier, P., Braun, J.J.,

1999. Major and trace elements associated with colloids in organic-rich river

waters: Ultrafiltration of natural and spiked solutions. Chemical Geology

160(1-2), 63-80.

Duvallet, L., Martin, F., Soubies, F., Salvi, S., Melfi, A.J., Fortune, J.P., 1999. The

mobility of zirconium and identification of secondary Zr-bearing phases in

bauxite from Pocos de Caldas, Minas Gerais, Brazil: A mass-balance and x-ray

absorption spectroscopic study. Canadian Mineralogist 37(3), 635-651.

Duzgoren-Aydin, N.S., Aydin, A., 2009. Distribution of rare earth elements and

oxyhydroxide phases within a weathered felsic igneous profile in Hong Kong.

Journal of Asian Earth Sciences 34(1), 1-9.

Elderfield, H., Greaves, M.J., 1981. Negative cerium anomalies in the rare-earth

element patterns of oceanic ferromagnanese nodules. Earth and Planetary

Science Letters 55(1), 163-170.

Fendorf, S., Fendorf, M., 1996. Sorption mechanisms of lanthanum on oxide minerals.

Clays and Clay Minerals 44(2), 220-227.

Feng, J.L., 2011. Trace elements in ferromanganese concretions, gibbsite spots, and the

surrounding terra rossa overlying dolomite: Their mobilization, redistribution and

fractionation. Journal of Geochemical Exploration 108(1), 99-111.


215

Fernández-Caliani, J.C., Cantano, M., 2010. Intensive kaolinization during a lateritic

weathering event in South-West Spain Mineralogical and geochemical inferences

from a relict paleosol. Catena 80(1), 23-33.

Fichter, J., Turpault, M.P., Dambrine, E., Ranger, J., 1998. Localization of base cations

in particle size fractions of acid forest soils (Vosges Mountains, N-E France).

Geoderma 82(4), 295-314.

Foo, M.P.F., 1999. Host of gold and ore related elements in ferruginous materials in

mineralized environments - Implications for exploration Curtin University of

Technology, Perth (unpubl.).

Forster, H.J., 1998. The chemical composition of REE-Y-Th-U-rich accessory minerals

in peraluminous granites of the Erzgebirge-Fichtelgebirge region, Germany, Part

I: The monazite-(Ce)-brabantite solid solution series. Am. Miner. 83(3-4),

259-272.

Goldschmidt, V.M., 1937. The principles of distributionof chemical elements in

minerals and rocks. Journal of the Chemical Society 49, 655-673.

Gouveia, M.A., Prudencio, M.I., Figueiredo, M.O., Pereira, L.C.J., Waerenborgh, J.C.,

Morgado, I., Pena, T., Lopes, A., 1993. Behavior of REE and other trace and

major elements during weathering of granitic-rocks, Evora, Portugal. Chemical

Geology 107(3-4), 293-296.

Gozzard, J.R., 2007. Geology and landforms of the Perth Region. Western Australia

Geological Survey, Perth.

Hall, G.E.M., Vaive, J.E., Beer, R., Hoashi, M., 1996. Selective leaches revisited, with

emphasis on the amorphous Fe oxyhydroxide phase extraction. Journal of

Geochemical Exploration 56(1), 59-78.

Hardy, M., Cornu, S., 2006. Loccation of natural tracce elements in silty soils using

particle-size fractionation. Geoderma 133(3-4), 295-308.

Harkins, W.D., 1917. The evolution of the elements and the stability of complex atoms.

I. A new periodic system which shows a relation between the abundance of the

elements and the structure of the nuclei of atoms. Journal of the American

Chemical Society 39(5), 856-879.

Harlavan, Y., Erel, Y., 2002. The release of Pb and REE from granitoids by the


216

dissolution of accessory phases. Geochimica et Cosmochimica Acta 66(5),

837-848.

Harlavan, Y., Erel, Y., Blum, J.D., 2009. The coupled release of REE and Pb to the soil

labile pool with time by weathering of accessory phases, Wind River Mountains,

WY. Geochimica et Cosmochimica Acta 73(2), 320-336.

Harter, R.D., 1983. Effect of soil-pH on adsorption of lead, copper, zinc, and nickel.

Soil Science Society of America Journal 47(1), 47-51.

Haskin, L.A., Paster, T.P., 1979. Geochemistry and mineralogy of the rare earths,

Handbook on the physics and chemistry of rare earths, North-Holland.

Henderson, P. (Ed.), 1984. Rare earth element geochemistry. Developments in

geochemistry. Elsevier Science Publishers, Amsterdam.

Hill, I.G., Worden, R.H., Meighan, I.G., 2000. Yttrium: The immobility-mobility

transition during basaltic weathering. Geology 28(10), 923-926.

Hodson, M.E., 2002. Experimental evidence for mobility of Zr and other trace

elements in soils. Geochimica et Cosmochimica Acta 66(5), 819-828.

Hu, Z.Y., Haneklaus, S., Sparovek, G., Schnug, E., 2006. Rare earth elements in soils.

Communications in Soil Science and Plant Analysis 37(9-10), 1381-1420.

Huang, C.M., Gong, Z.T., 2001. Geochemical implication of rare earth elements in

process of soil development. Journal of Rare Earths 19(1), 57-62.

Ito, M., Messenger, S., 2009. Rare earth element measurements of melilite and fassaite

in allende cai by nanosims. Meteoritics & Planetary Science 44, A97-A97.

Ji, H.B., Wang, S.J., Ouyang, Z.Y., Zhang, S., Sun, C.X., Liu, X.M., Zhou, D.Q., 2004.

Geochemistry of red residua underlying dolomites in karst terrains of

Yunnan-Guizhou Plateau II. The mobility of rare earth elements during

weathering. Chemical Geology 203(1-2), 29-50.

Jiang, S.Y., 2000. Controls on the mobility of high field strength elements (HFSE), U,

and Th in an ancient submarine hydrothermal system of the Proterozoic Sullivan

Pb-Zn-Ag deposit, British Columbia, Canada. Geochemical Journal 34(5),

341-348.

Johannesson, K., Lyons, W., Stetzenbach, K., Byrne, R., Lyons, H., Stetzenbach, W.,

Byrne, B., Byrne, K., Byrne, J., Byrne, H., 1995. The solubility control of rare


217

earth elements in natural terrestrial waters and the significance of PO43-

and

CO32-

in limiting dissolved rare earth concentrations: A review of recent

information. Aquatic Geochemistry 1(2), 157-173.

Johannesson, K.H., Stetzenbach, K.J., Hodge, V.F., Lyons, W.B., 1996. Rare earth

element complexation behavior in circumneutral pH groundwaters: Assessing the

role of carbonate and phosphate ions. Earth and Planetary Science Letters

139(1-2), 305-319.

Johnson, D.L., 1990. Biomantle evolution and the redistribution of earth materials and

artifacts. Soil Science 149, 84-102.

Johnson, D.L., Balek, C.L., 1991. The genesis of Quaternary landscapes with stone

lines. Physical Geography 12, 385-395.

Kabata-Pendias, A., Pendias, H., 2001. Trace elements in soils and plants. 3rd Edition

ed. CRC Press London.

Kahmann, J.A., Seaman III, J., Driese, S.G., 2008. Evaluating trace elements as

paleoclimate indicators: Multivariate statistical analysis of late Mississippian

Pennington Formation Paleosols, Kentucky, USA. Journal of Geology 116,

254-258.

Kanazawa, Y., Kamitani, M., 2006. Rare earth minerals and resources in the world.

Journal of Alloys and Compounds 408, 1339-1343.

Kew, G.A., Gilkes, R.J., 2007. Properties of regolith beneath lateritic bauxite in the

Darling Range of south Western Australia. Australian Journal of Soil Research

45(3), 164-181.

Koeppenkastrop, D., Decarlo, E.H., 1992. Sorption of rare-earth elements from

seawater onto synthetic mineral particles-an experimental approach. Chemical

Geology 95(3-4), 251-263.

Koeppenkastrop, D., Decarlo, E.H., 1993. Uptake of rare-earth elements from solution

by metal-oxides. Environmental Science & Technology 27(9), 1796-1802.

Koppi, A.J., Edis, R., Field, D.J., Geering, H.R., Klessa, D.A., Cockayne, D.J.H., 1996.

Rare earth element trends and cerium-uranium-manganese associations in

weathered rock from Koongarra, northern territory, Australia. Geochimica et

Cosmochimica Acta 60(10), 1695-1707.


218

Kurtz, A.C., Derry, L.A., Chadwick, O.A., Alfano, M.J., 2000. Refractory element

mobility in volcanic soils. Geology 28(8), 683-686.

Land, M., Ohlander, B., Ingri, J., Thunberg, J., 1999. Solid speciation and fractionation

of rare earth elements in a spodosol profile from northern Sweden as revealed by

sequential extraction. Chemical Geology 160(1-2), 121-138.

Langmuir, D., Herman, J.S., 1980. The mobility of thorium in natural-waters at

low-temperatures. Geochimica et Cosmochimica Acta 44(11), 1753-1766.

Laveuf, C., Cornu, S., 2009. A review on the potentiality of rare earth elements to trace

pedogenetic processes. Geoderma 154(1-2), 1-12.

Laveuf, C., Cornu, S., Juillot, F., 2008. Rare earth elements as tracers of pedogenetic

processes. Comptes Rendus Geoscience 340(8), 523-532.

Lee, B.D., Graham, R.C., Laurent, T.E., Amrhein, C., 2004. Pedogenesis in a wetland

meadow and surrounding serpentinitic landslide terrain, northern California,

USA. Geoderma 118(3-4), 303-320.

Lichte, M., 2000. Bioturbation responsible for the concentration of stonelines?

Zeitschrift Fur Geomorphologie 44(4), 533-534.

Linnen, R.L., Samson, L., Breaks, F.W. (Eds.), 2005. Rare-element geochemistry and

mineral deposits Geological Association of Canada, Ontario.

Liu, C.Q., Zhang, H., 2005. The lanthanide tetrad effect in apatite from the Altay No. 3

pegmatite, Xingjiang, China: An intrinsic feature of the pegmatite magma.

Chemical Geology 214(1-2), 61-77.

Ljung, K., Selinus, O., Otabbong, E., Berglund, M., 2006. Metal and arsenic

distribution in soil particle sizes relevant to soil ingestion by children. Applied


Lottermoser, B.G., 1990. Rare-earth element mineralization within the Mt-Weld

carbonatite laterite, Western-Australia. Lithos 24(2), 151-167.

Ma, J.L., Wei, G.H., Xu, Y.G., Long, W.G., Sun, W.D., 2007. Mobilization and

re-distribution of major and trace elements during extreme weathering of basalt

in Hainan Island, South China. Geochimica et Cosmochimica Acta 71(13),

3223-3237.

Ma, Y.J., Huo, R.K., Xu, Z.F., 2004. REE Behavior and Influence factors during


219

geochemcial weathering. Advancements in Earth sciences (in Chinese) 19(1),

87-94.

Macfarlane, A.W., Danielson, A., Holland, H.D., Jacobsen, S.B., 1994. REE chemistry

and Sm-Nd systematics of late Archean weathering profiles in the Fortescue

Group, Western Australia. Geochimica et Cosmochimica Acta 58(7), 1777-1794.

Malengreau, N., Muller, J.P., Calas, G., 1995. Spectroscopic approach for investigating

the status and mobility of Ti in kaolinitic materials. Clays and Clay Minerals

43(5), 615-621.

Marker, A., Deoliveira, J.J., 1994. Climatic and morphological control of rare-earth

element distribution in weathering mantles on alkaline rocks. Catena 21(2-3),

179-193.

Marmier, N., Delisee, A., Fromage, F., 1999. Surface complexation modeling of

Yb(III), Ni(II), and Cs(I) sorption on magnetite. Journal of Colloid and Interface

Science 211(1), 54-60.

Marmier, N., Dumonceau, J., Fromage, F., 1997. Surface complexation modeling of

Yb(III) sorption and desorption on hematite and alumina. Journal of Contaminant

Hydrology 26(1-4), 159-167.

Marmier, N., Fromage, F., 1999. Comparing electrostatic and nonelectrostatic surface

complexation modeling of the sorption of lanthanum on hematite. Journal of

Colloid and Interface Science 212(2), 252-263.

Marshall, T.J., 1947. Mechanical composition of soil in relation to field descriptions of

texture, Council for Scienfic and Industrial Research, Australia.

Marshall, T.J., 2003. Particle-size distribution of soil and the perception of texture.

Australian Journal of Soil Research 41(2), 245-249.

Masuda, A., Kawakami, O., Dohmoto, Y., Takenaka, T., 1987. Lanthanide tetrad effects

in nature: two mutually opposite types, W and M. Geochemical Journal 21(3),

119-124.

McDonough, W.F., Sun, S.S., 1995. The composition of the Earth. Chemical Geology

120(3-4), 223-253.

McLennan, S.M., 1989. Rare earth elements in sedimentary rocks: Influence of

provenance and sedimentary processes. In: B.R. Lipin, G.A. McKay (Eds.),


220

Geochemistry and mineralogy of rare earth elements (Series: Reviews in

mineralogy). Mineralogical Society of America., Washington, D.C. .

McLennan, S.M., 1994. Rare-earth element geochemistry and the "tetra" effect.


McLennan, S.M., 1995. sediments and soils: Chemistry and abundances. Rock physics

& phase relations: A handbook of physical constants. Third ed. American

Geophysical Union.

McLennan, S.M., Nance, W.B., Taylor, S.R., 1980. Rare-earth element-thorium

correlations in sedimentary rocks, and the composition of the continental crust.


Melfi, A.J., Subies, F., Nahon, D., Formoso, M.L.L., 1996. Zirconium mobility in

bauxites of southern Brazil. Journal of South American Earth Science 9(3-4),

161-170.

Middelburg, J.J., Vanderweijden, C.H., Woittiez, J.R.W., 1988. Chemical processes

affecting the mobility of major, minor and trace-elements during weathering of

granitic-rocks. Chemical Geology 68(3-4), 253-273.

Minarik, L., Zigova, A., Bendl, J., Skrivan, P., St'astny, M., 1998. The behaviour of

rare-earth elements and Y during the rock weathering and soil formation in the

Ricany granite massif, Central Bohemia. Science of the Total Environment

215(1-2), 101-111.

Monecke, T., Kempf, U., Monecke, J., Sala, M., Wolf, D., 2002. Tetrad effect in rare

earth element distribution patterns: A method of quantification with application

to rock and mineral samples from granite-related rare metal deposits.


Mourier, B., Poulenard, J., Chauvel, C., Faivre, P., Carcaille, C., 2008. Distinguishing

subalpine soil types using extractable Al and Fe fractions and REE geochemistry.

Geoderma 145(1-2), 107-120.

Nakai, I., Terada, Y., Itou, M., Sakurai, Y., 2001. Use of highly energetic (116 keV)

synchrotron radiation for X-ray fluorescence analysis of trace rare-earth and

heavy elements. Journal of Synchrotron Radiation 8, 1078-1081.

Nance, W.B., Taylor, S.R., 1976. Rare-earth element patterns and crustal evolution I.


221

Australian post-Archean sedimentary rocks. Geochimica et Cosmochimica Acta

40(12), 1539-1551.

Ndjigui, P.D., Bilong, P., Bitom, D., 2009. Negative cerium anomalies in the saprolite

zone of serpentinite lateritic profiles in the Lomie ultramafic complex,

South-East Cameroon. Journal of African Earth Sciences 53(1-2), 59-69.

Ndjigui, P.D., Bilong, P., Bitom, D., Dia, A., 2008. Mobilization and redistribution of

major and trace elements in two weathering profiles developed on serpentinites

in the Lomie ultramafic complex, South-East Cameroon. Journal of African Earth

Sciences 50(5), 305-328.

Nedachi, Y., Nedachi, M., Bennett, G., Ohmoto, H., 2005. Geochemistry and

mineralogy of the 2.45 Ga Pronto paleosols, Ontario, Canada. Chemical Geology

214(1-2), 21-44.

Nedel, S., Dideriksen, K., Christiansen, B.C., Bovet, N., Stipp, S.L.S., 2010. Uptake

and release of cerium during Fe-oxide formation and transformation in Fe(II)

solutions. Environmental Science & Technology 44(12), 4493-4498.

Nesbitt, H.W., 1979. Mobility and fractionation of rare-earth elements during

weathering of a granodiorite. Nature 279(5710), 206-210.

Nesbitt, H.W., Markovics, G., 1997. Weathering of granodioritic crust, long-term

storage of elements in weathering profiles, and petrogenesis of siliciclastic

sediments. Geochimica et Cosmochimica Acta 61(8), 1653-1670.

Nesbitt, H.W., Young, G.M., 1982. Early Proterozoic climates and plate motion

inferred from major element chemistry of lutites. Nature 299(5885), 715-717.

Ohlander, B., Land, M., Ingri, J., Widerlund, A., 1996. Mobility of rare earth elements

during weathering of till in northern Sweden. Applied Geochemistry 11, 93-99.

Ohta, A., Kagi, H., Nomura, M., Tsuno, H., Kawabe, I., 2009. Coordination study of

rare earth elements on Fe oxyhydroxide and Mn dioxides: Part I. Influence of a

multi-electron excitation on EXAFS analyses of La, Pr, Nd, and Sm. American

Mineralogist 94(4), 467-475.

Panahi, A., Young, G.M., Rainbird, R.H., 2000. Behavior of major and trace elements

(including REE) during Paleoproterozoic pedogenesis and diagenetic alteration

of an Archean granite near Ville Marie, Quebec, Canada. Geochimica et


222


Parsons, R.B., Herriman, C.A., 1966. Morphology and genesis of six 'Red Hill' soils in

the Oregon coast range. Soil Science Society of America Journal Proceeding 30,

90-93.

Patino, L.C., Velbel, M.A., Price, J.R., Wade, J.A., 2003. Trace element mobility

during spheroidal weathering of basalts and andesites in Hawaii and Guatemala.


Perrier, N., Gilkes, R.J., Colin, F., 2006. Heating Fe oxide-rich soils increases the

dissolution rate of metals. Clays and Clay Minerals 54(2), 165-175.

Piasecki, W., Sverjensky, D.A., 2008. Speciation of adsorbed yttrium and rare earth

elements on oxide surfaces. Geochimica et Cosmochimica Acta 72(16),

3964-3979.

Poitrasson, F., Chenery, S., Bland, D.J., 1996. Contrasted monazite hydrothermal

alteration mechanisms and their geochemical implications. Earth and Planetary

Science Letters 145(1-4), 79-96.

Prescott, J.A., Taylor, J.K., Marshall, T.J., 1934. The relationship between the

mechanical composition of the soil and the estimate of texture in the field,

Transaction 1st Commission International Society of Soil Science, pp. 143-153.

Price, R.C., Gray, C.M., Wilson, R.E., Frey, F.A., Taylor, S.R., 1991. The effects of

weathering on rare-earth element, Y and Ba abundances in Tertiary basalts from

southeastern Australia. Chemical Geology 93(3-4), 245-265.

Quinn, K.A., Byrne, R.H., Schijf, J., 2006. Sorption of yttrium and rare earth elements

by amorphous ferric hydroxide: Influence of solution complexation with

carbonate. Geochimica et Cosmochimica Acta 70(16), 4151-4165.

R Development Core Team, 2011. R: A language and environment for statistical

computing. 3-900051-07-0, Vienna, Austria.

Rao, W.B., Gao, Z.M., Yang, Z.S., Luo, T.Y., 2004. Geology and geochemistry of the

Shangmanggang red clay-type gold deposit in West Yunnan. Journal of


Rayment, G.E., Higginson, F.R., 1992. Australian laboratory handbook of soil and

water chemical methods Australia Soil and Land Survey Handbook, 3. Inkata


223

Press, Melbourne.

Reiller, P., Moulin, V., Casanova, F., Dautel, C., 2002. Retention behaviour of humic

substances onto mineral surfaces and consequences upon thorium (IV) mobility:

Case of iron oxides. Applied Geochemistry 17(12), 1551-1562.

Reimann, C., Filzmoser, P., Garrett, R.G., Dutter, R., 2008. Statistical Data Analysis

Explained: Applied Environmental Statistics with R. John Wiley & Sons,

Chichester, England.

Rubin, J.N., Henry, C.D., Price, J.G., 1993. The mobility of zirconium and other

"immobile" elements during hydrothermal alteration. Chemical Geology

110(1-3), 29-47.

Ruhe, R.V., 1958. Stone lines in soils. Soil Science 84, 223-231.

Ryan, C.G., Siddons, D.P., Kirkham, R., Dunn, P.A., Kuczewski, A., Moorhead, G., De

Geronimo, G., Paterson, D.J., de Jonge, M.D., Hough, R.M., Lintern, M.J.,

Howard, D.L., Kappen, P., Cleverley, J., 2010. The New Maia Detector System:

Methods For High Definition Trace Element Imaging Of Natural Material. In:

M.A.W.C.T. Denecke (Ed.), X-Ray Optics and Microanalysis, Proceedings. AIP

Conference Proceedings, pp. 9-17.

Sadleir, S.B., Gilkes, R.J., 1976. Development of bauxite in relation to parent material

near Jarrahdale, Western Australia. Journal of the Geological Society of Australia

23(4), 333-344.

Salminen, R., 2005. Geochemical atlas of Europe: Background information,

methodology and maps. Geological Survey of Finland.

Schaetzl, R.J., Anderson, S., 2005. Soils: Genesis and geomorphology. Cambridge

University Press, Cambridge; New York.

Schellmann, W., 1981. Considerations on the definition and classification of laterities,

Proceedings of the International Seminar on lateritization processes, Trivandrum,

India, pp. 1-10.

Schellmann, W., 1994. Geochemical differentiation in laterite and bauxite formation.

Catena 21(2-3), 131-143.

Schmidt, C., Rickers, K., Bilderback, D.H., Huang, R., 2007. In situ

synchrotron-radiation XRF study of REE phosphate dissolution in aqueous fluids


224

to 800 degrees °C. Lithos 95(1-2), 87-102.

Schroeder, P.A., Le Golvan, J.J., Roden, M.F., 2002. Weathering of ilmenite from

granite and chlorite schist in the Georgia Piedmont. American Mineralogist

87(11-12), 1616-1625.

Seco, F., Hennig, C., de Pablo, J., Rovira, M., Rojo, I., Martí, V., Giménez, J., Duro, L.,

Grivé, M., Bruno, J., 2009. Sorption of Th(IV) onto iron corrosion products:

EXAFS study. Environmental Science & Technology 43(8), 2825-2830.

Sharma, A., Rajamani, V., 2000. Major element, REE, and other trace element behavior

in amphibolite weathering under semiarid conditions in southern India. Journal

of Geology 108(4), 487-496.

Sheldon, N.D., Tabor, N.J., 2009. Quantitative paleoenvironmental and paleoclimatic

reconstruction using paleosols. Earth-Science Reviews 95(1-2), 1-52.

Singh, B., Cornelius, M., 2006. Geochemistry and mineralogy of the regolith profile

over the Aries kimberlite pipe, Western Australia. Geochemistry-Exploration

Environment Analysis 6, 311-323.

Singh, B., Gilkes, R.J., 1992. Properties and distribution of iron-oxides and their

association with minor elements in the soils of south-western Australia. Journal

of Soil Science 43(1), 77-98.

Smith, R.E., Anand, R.R., Alley, N.F., 2000. Use and implications of paleoweathering

surfaces in mineral exploration in Australia. Ore Geology Reviews 16(3-4),

185-204.

Sommer, M., Halm, D., Geisinger, C., Andruschkewitsch, I., Zarei, M., Stahr, K., 2001.

Lateral podzolization in a sandstone catchment. Geoderma 103(3-4), 231-247.

Sommer, M., Halm, D., Weller, U., Zarei, M., Stahr, K., 2000. Lateral podzolization in

a granite landscape. Soil Science Society of America Journal 64(4), 1434-1442.

Sonke, J.E., Salters, V.J.M., 2006. Lanthanide-humic substances complexation. I.

Experimental evidence for a lanthanide contraction effect. Geochimica et


Sutherland, R.A., 2003. Lead in grain size fractions of road-deposited sediment.

Environmental Pollution 121(2), 229-237.

Taboada, T., Cortizas, A.M., Garcia, C., Garcia-Rodeja, E., 2006a. Particle-size


225

fractionation of titanium and zirconium during weathering and pedogenesis of

granitic rocks in NW Spain. Geoderma 131(1-2), 218-236.

Taboada, T., Cortizas, A.M., Garcia, C., Garcia-Rodeja, E., 2006b. Uranium and

thorium in weathering and pedogenetic profiles developed on granitic rocks from

NW Spain. Science of the Total Environment 356(1-3), 192-206.

Takahashi, Y., Minai, Y., Ambe, S., Makide, Y., Ambe, F., 1999. Comparison of

adsorption behavior of multiple inorganic ions on kaolinite and silica in the

presence of humic acid using the multitracer technique. Geochimica et


Takahashi, Y., Shimizu, H., Usui, A., Kagi, H., Nomura, M., 2000. Direct observation

of tetravalent cerium in ferromanganese nodules and crusts by X-ray-absorption

near-edge structure (XANES). Geochimica et Cosmochimica Acta 64(17),

2929-2935.

Takahashi, Y., Yoshida, H., Sato, N., Hama, K., Yusa, Y., Shimizu, H., 2002. W- and

M-type tetrad effects in REE patterns for water-rock systems in the Tono

uranium deposit, central Japan. Chemical Geology 184(3-4), 311-335.

Tari, G., Bobos, I., Gomes, C.S.F., Ferreira, J.M.F., 1999. Modification of surface

charge properties during kaolinite to halloysite-7Å transformation. Journal of

Colloid and Interface Science 210(2), 360-366.

Taunton, A.E., Welch, S.A., Banfield, J.F., 2000a. Geomicrobiological controls on light

rare earth element, Y and Ba distributions during granite weathering and soil

formation. Journal of Alloys and Compounds 303, 30-36.

Taunton, A.E., Welch, S.A., Banfield, J.F., 2000b. Microbial controls on phosphate and

lanthanide distributions during granite weathering and soil formation. Chemical

Geology 169(3-4), 371-382.

Taylor, G., Shirtliff, G., 2003. Weathering: Cyclical or continuous? An Australian

perspective. Australian Journal of Earth Sciences 50(1), 9-17.

Taylor, S.R., McLennan, S.M., 1985. The continental crust, its composition and

evolution: An examination of the geochemical record preserved in sedimentary

rocks Blackwell Scientific Publications, Oxford, Melbourne.

Thanachit, S., Suddhiprakarn, A., Kheoruenromne, I., Gilkes, R.J., 2006. The


226

geochemistry of soils on a catena on basalt at Khon Buri, northeast Thailand.

Geoderma 135, 81-96.

Thompson, A., Chadwick, O.A., Boman, S., Chorover, J., 2006. Colloid mobilization

during soil iron redox oscillations. Environmental Science & Technology 40(18),

5743-5749.

Tilley, D.B., Eggleton, R.A., 2005. Titanite low-temperature alteration and Ti mobility.

Clays and Clay Minerals 53(1), 100-107.

Timofeeva, Y.O., 2008. Accumulation and fractionation of trace elements in soil

ferromanganese nodules of different size. Geochemistry International 46(3),

260-267.

Tripathi, J.K., Rajamani, V., 2007. Geochemistry and origin of ferruginous nodules in

weathered granodioritic gneisses, Mysore Plateau, Southern India. Geochimica et


Tsai, C.C., Sang, C.Z., 2000. Lithologic discontinuities in ultisols along a

toposequence in Taiwan. Soil Science 165(7), 587-596.

Tyler, G., 2004. Rare earth elements in soil and plant systems-A review. Plant and Soil

267(1-2), 191-206.

Van der Weijden, C.H., Van der Weijden, R.D., 1995. Mobility of major, minor and

some redox-sensitive trace-elements and rare-earth elements during weathering

of four granitoids in central Portugal. Chemical Geology 125(3-4), 149-167.

Vanbaalen, M.R., 1993. Titanium mobility in metamorphic systems - A review.


Viers, J., Dupre, B., Braun, J.J., Deberdt, S., Angeletti, B., Ngoupayou, J.N., Michard,

A., 2000. Major and trace element abundances, and strontium isotopes in the

Nyong basin rivers (Cameroon): Constraints on chemical weathering processes

and elements transport mechanisms in humid tropical environments. Chemical

Geology 169(1-2), 211-241.

Viers, J., Dupre, B., Polve, M., Schott, J., Dandurand, J.L., Braun, J.J., 1997. Chemical

weathering in the drainage basin of a tropical watershed (Nsimi-Zoetele site,

Cameroon): Comparison between organic-poor and organic-rich waters.



227

Viers, K., Wasserburg, G.J., 2004. Behavior of Sm and Nd in a lateritic soil profile.


Vos, W.D., Tarvainen, T., Salminen, R., 2006. Geochemical atlas of Europe:

Interpretation of geochemical maps, additional tables, figures, maps, and related

publications. Geological Survey of Finland.

Walter, A.V., Nahon, D., Flicoteaux, R., Girard, J.P., Melfi, A., 1995. Behaviour of

major and trace elements and fractionation of REE under tropical weathering of a

typical apatite-rich carbonatite from Brazil. Earth and Planetary Science Letters

136, 591-602.

Wang, X.-S., Qin, Y., Chen, Y.-K., 2006. Heavy meals in urban roadside soils: Part 1.

Effect of particle size fractions on heavy metals partitioning. Environmental

Geology 50(7), 1061-1066.

Whitney, D.L., Evans, B.W., 2010. Abbreviations for names of rock-forming minerals.

American Mineralogist 95, 185-187.

Williams, C.T. (Ed.), 1996. Analysis of rare earth minerals. Chapter thirteen of Rare

earth minerals: chemistry, origin and ore deposits. Mineralogical Society series.

Springer.

Wood, S.A., 1990. The aqueous geochemistry of the rare-earth elements and yttrium: 1.

Review of available low-temperature data for inorganic complexes and the

inorganic REE sepciation of natrual-waters. Chemical Geology 82(1-2), 159-186.

Wood, S.A., 1993. The aqueous geochemistry of the rare-earth elements: critical

stability constants for complexes with simple carboxylic acids at 25°C and 1 bar

and their application to nuclear waste management. Engineering Geology 34(3-4),

229-259.

Yang, Y.G., Liu, S.R., Jin, Z.S., 2009. Laterization and its control to gold occurrence in

Laowanchang gold deposit, Guizhou Province, Southwest of China. Journal of


Zhang, G.L., Pan, J.H., Huang, C.M., Gong, Z.T., 2007. Geochemical features of a soil

chronosequence developed on basalt in Hainan Island, China. Revista Mexicana

De Ciencias Geologicas 24(2), 261-269.

Zhang, H.X., Dong, Z., Tao, Z.Y., 2006. Sorption of thorium(IV) ions on gibbsite:


228

Effects of contact time, pH, ionic Strength, concentration, phosphate and fulvic

acid. Colloids and Surfaces A: Physicochemical and Engineering Aspects

278(1-3), 46-52.

229

11 Appendices

Appendix 11.1 Abbreviation

AAG: the Association of Applied Geochemists;

A-CN-K: a ternary plot based on the concentrations of oxides of Al, Ca, Na and K, and

usually used together with the Chemical Index of Alteration to illustrate the

weathering intensity and the weathering trend;

A-CNK-FM: a ternary plot based on the concentrations of oxides of Al, Ca, Na, K, Fe

and Mg;

BSE: backscattering electron;

CANMET: the Canada Centre for Mineral and Energy Technology;

CEC: cation exchange capacity;

CIA: Chemical Index of Alteration, a weathering intensity index based on major

element molar proportion ratio;

CMCA: the centre for microscopy, characterisation and analysis in the University of

Western Australia;

CSC: the China Scholarship Council;

EDS: energy dispersive spectrometer;

EELS: electron energy-loss spectroscopy;

EPMA: electron probe micro-analyser;

FeAm: amorphous Fe oxyhydroxide species determined in sequential extraction

experiment;

FeCry: crystalline Fe oxide species determined in sequential extraction experiment;

GE: the first profile sampled beside the Great Eastern Highway in Western Australia;

GSFloading: the element mass loading in a selected grain size and the summation of the

indices for each soil sample equals 100%.

HREE: heavy rare earth elements according to atomic mass, refers from Eu to Lu in

this thesis;

ICDD: the International Centre for Diffraction Data;

ICP-MS: inductively coupled plasma mass spectroscopy;


230

ICP-OES: inductively coupled plasma-optical emission spectroscopy;

ISSS: the International Society of Soil Science;

JG: the fourth profile sampled in the Jarrahdale in Western Australia;

LREE: light rare earth elements according to atomic mass, refers from La to Sm in this

thesis;

MQ: the second and third profile sampled at the Mountain Quarry in Western

Australia;

MREE: middle rare earth elements according to atomic mass, refers from Pm to Ho

when REE was divided into three subgroups;

NanoSIMS: nano secondary ion mass spectrometry;

NASC: North American shales composite;

NIST: the National Institute of Standards and Technology;

Org: organic matter species determined in sequential extraction experiment;

PAAS: Post-Archean Australian Shale;

REE: rare earth elements, also known as lanthanides, sometime referring as ‘Ln’ in

chemical formula;

Res: residual species determined in sequential extraction experiment;

SAED-TEM: selected area electron diffraction with transmission electron microscopy;

S/SAF: the concentration ratio of SiO2/(SiO2+Al2O3+Fe2O3);

SE: secondary electron;

SEM: scanning electron microscope;

SIMS: secondary ion mass spectrometry ion probe;

SiO2-Al2O3-Fe2O3: a ternary plot to quantitatively illustrate the lateritization degree;

STSD: certified international standard stream sediment reference materials from

Canada Centre for Mineral and Energy Technology;

SXFM: synchrotron x-ray fluorescence microscopy;

SXRD: synchrotron x-ray powder diffraction;

TC: the total carbon;

TEM: transmission electron microscopy;

UCC: the upper crust composition;

UWA: the University of Western Australia;

Chapter Eleven: Appendices

231

WAE: water soluble, adsorbed, exchangeable and carbonate species determined in

sequential extraction experiment;

XRD: X-ray diffraction;

μ-XRF: micro X-ray fluorescence spectroscopy;

μ-XANES: micro X-ray absorption near edge fine structure spectroscopy;

23

2

Appendix 11.2 ICP-OES analyses of the reference standards determined repeatedly with samples for each analysis

Reference Major element concentrations

Al

308.215

Ca Fe

238.204

K

766.490

Mg

285.213

Na

589.592

Si

251.611

S Ti P Mn Zr

Unit % % % % % % % % % ppm ppm ppm

d.l. 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.1 0.1 0.1

Batch 11

STSD-2 8.33 2.92 5.07 1.74 1.88 1.29 25.3 0.04 0.47 0.14 0.11 178

STSD-2 8.38 2.94 5.14 1.76 1.88 1.33 25.3 0.05 0.47 0.14 0.11 181

STSD-2 8.05 2.66 4.96 1.61 1.71 1.26 22.1 0.09 0.46 0.12 0.10 171

av2 8.25 2.84 5.06 1.70 1.82 1.29 24.2 0.06 0.47 0.13 0.11 177

error(%)3 -3.2 -0.8 -3.7 -2.2 -2.6 2.5 -3.4 0.0 -3.1 1.0 -0.7 -4.6

RSD4 0.2 0.2 0.1 0.1 0.1 0.0 1.9 0.0 0.0 0.0 0.0 4.8

Batch 2

STSD-2 8.36 2.92 5.11 1.79 1.95 1.31 25.3 0.06 0.45 0.13 0.10 196

STSD-2 8.23 2.99 5.26 1.78 1.93 1.32 25.0 0.07 0.46 0.14 0.10 190

av 8.29 2.96 5.19 1.79 1.94 1.31 25.1 0.06 0.46 0.14 0.10 193

error(%) -2.7 3.4 -1.2 2.7 3.8 4.0 0.1 2.5 -4.5 4.4 -2.7 4.4

RSD 0.1 0.1 0.1 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 3.8

Batch 3

STSD-2 8.93 3.00 5.33 1.78 1.98 1.35 25.3 0.06 0.48 0.13 0.11 184

STSD-2 8.73 3.01 5.36 1.79 1.90 1.27 25.3 0.07 0.48 0.14 0.11 187

STSD-2 8.54 2.93 5.32 1.73 1.94 1.25 24.2 0.06 0.47 0.14 0.11 176

STSD-2 8.68 3.04 5.46 1.76 1.98 1.39 25.2 0.06 0.48 0.14 0.11 174

av 8.72 2.99 5.37 1.77 1.95 1.32 25.0 0.06 0.48 0.14 0.11 180

error(%) 2.4 4.7 2.3 1.4 4.3 4.4 -0.3 2.0 -1.0 4.3 2.1 -2.5

RSD 0.2 0.1 0.1 0.0 0.0 0.1 0.5 0.0 0.0 0.0 0.0 6.5

exp

5 STSD-2 8.52 2.86 5.25 1.74 1.87 1.26 25.1 0.06 0.48 0.13 0.11 185

23

3

Reference Major element concentrations

Al

308.215

Ca Fe

238.204

K

766.490

Mg

285.213

Na

589.592

Si

251.611

S Ti P Mn Zr

Unit % % % % % % % % % ppm ppm ppm

d.l. 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.1 0.1 0.1

OREAS 43p 5.02 0.29 16.78 1.88 0.59 0.19 26.7 0.02 0.29 0.05 0.07 245

OREAS 43p 4.94 0.31 17.25 1.91 0.59 0.10 26.5 0.02 0.29 0.05 0.07 240

av 4.98 0.30 17.01 1.89 0.59 0.14 26.6 0.02 0.29 0.05 0.07 242

exp 5.08 0.31 17.49 1.78 0.57 0.13 26.8 0.02 0.30 0.04 0.06 232

error(%) -1.2 -0.4 -9.1 6.4 1.2 1.1 -1.0 4.6 -1.7 7.3 3.1 5.7

RSD 0.1 0.0 0.3 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1

1Batch refers to each time determination of major element concentrations by ICP-OES based on the profile;

2av: average value;

3error(%)=[(sample determined value)-(expected value)]/(expected value)×100%;

4RSD: relative standard deviation;

5exp: expected/recommended value.

These footnotes are also used in Appendix 11.6.


234

Appendix 11.3 R script for principal component analysis of major elements

(‘#’ is the explanation of the command with italic font)

#view data table

major<-read.table("D:/XIN/experimental results(desktop)/analysis/sta/major

oxides.csv",sep=",", header = TRUE)

library(Rcmdr)

str(major)

print(major)

#PCA of centered log ratio transformed major element data

(R cannot recognize subscript, so the the subscript was not used in the oxides)

pc1<-prcomp(~Al2O3_clr +CaO_clr +Fe2O3_clr +K2O_clr +MgO_clr +Na2O_clr

+SiO2_clr +TiO2_clr +ZrO2_clr +P2O5_clr +MnO_clr, scale=TRUE, data=major)

summary(pc1)

print(pc1)

pc1$sd^2 # eigenvalues (variances)

plot(pc1) # scree plot

TypeSymb<-cbind(major$Type)

# biplot PC1 vs. PC2

biplot(pc1, col = c("#000000", "#999999"), xlabs=TypeSymb)

predict(pc1)[,1-8]

write.table(predict(pc1)[,1-8],"D:/XIN/experimental results(desktop)/analysis/sta/x.csv

", sep=",")


235

Appendix 11.4 Photographs of polished thin sections of iron nodules mounted on

quartz slides from the JG profile

From left to right are: JG9, the A horizon, 0.3 m depth; JG6, the upper ferruginous

zone, 1.5 m depth; JG10, the A horizon, 0.4 m depth. Nodules from the upper

ferruginous zone were red and concentrically zoned, with a core of hematite

surrounded by goethite and Al-rich rims, whereas nodules from the A horizon were

dark brown-black, non-concentrically zoned, with cemented clay matrix and Fe oxides

without layers, containing less gibbsite but more quartz.


236

Appendix 11.5 Detailed operation procedure of the sequential extraction method

Extraction WAE:

1. To 1 g of sample in a 50 mL screw-cap centrifuge tube, add 20 mL of 1.0 M

CH3COONa (at pH 5 with CH3COOH) and cap.

2. Vortex contents for 5-10 s and place in an end-to-end tumbler at 25 °C constant

temperature for 6 h.

3. Centrifuge for 15 min. at 3000 rpm and decant supernatant liquid into a labelled

test-tube. Rinse residue with 5 ml of MilliQ water, vortex and centrifuge again; repeat

and add supernatant rinses to the test-tube. Make up to the 30 mL mark with MilliQ

water and analyse.

4. Carry out a second 20 mL 1M CH3COONa leach of the residue, repeating steps 2

and 3.

Extraction Org:

5. To the residue, add 40 mL of 0.1 M Na4P2O7.

6. Vortex contents for 5-10 s and place the centrifuge tubes in an end-to-end tumbler

at 25 °C constant temperature for 1 h and then centrifuge.

7. Decant the supernatant into a new labeled test-tube.

8. Rinse the residue with 5 mL H2O, centrifuge; do this twice and add the supernatant

to the test-tube; mark up to 50 mL.

9. Repeat steps 5 to step 8.

Extraction FeAm:

10. To the residue from step 9, add 20 mL of 0.25 M NH2OH∙HCl in 0.25 M HCl, cap

and vortex for 5-10 s.

11. Place in a water bath at 60 °C for 2 h with cap loosened. Every 30 min., cap tightly

and vortex the contents.

12. Centrifuge for 15 min. and decant supernatant liquid into a labeled test-tube. Rinse

residue with 5 mL of water, vortex and centrifuge again; repeat and add supernatant


237

rinses to the test-tube. Make up to the 30 mL mark with MilliQ water and analyse.

13. Carry out a second 0.25 M NH2OH∙HCl leach of the residue but heat for only 30

min.

Extraction Fecry:

14. To the residue from step 13, add 30 mL of 1.0 M NH2OH∙HCl in 25% CH3COOH,

cap and vortex for 5-10 s.

15. Place in a water bath at 90 °C for 3 h with cap on tightly. Vortex contents every

30 min.

16. Centrifuge for 15 min. and decant supernatant liquid into a labeled test-tube. Rinse

residue with 10 mL of 25% CH3COOH, vortex and centrifuge again; repeat and add

supernatant rinses to the test-tube. Make up to the 50.0 mL mark with MilliQ water and

analyse.

17. Carry out a second 1.0 M NH2OH∙HCl leach of the residue but heat for only 1.5 h.

Extraction Res:

18. Wash the residues with MilliQ water and dry in the oven at 60 °C. Add 0.1000 g

dried residue sample and 0.7000 g 12:22 Norrish flux (Lithium metaborate/ Lithium

tetraborate) into a pure platinum crucible.

19. Fuse at 1050 °C for 40 min.

20. Remove crucible from furnace and allow cooling. Place crucible into a labelled

120 mL polypropylene screw cap vial.

21. Add 100 ml of 10% HCl using a calibrated dispenser.

22. Place on tumbler for ½ hour or until bead is fully dissolved. Occasionally the

ultrasonic bath may be required to speed up dissolution.

23. Sore the solutions in cool room at 5 ºC prior to analysis by ICP-MS the next day.

Total REE and Residue REE

24. Total and residual REE content was determined by sampling 0.1000 g of dried

residual soil with 0.7000 g flux fused at 1050 ºC in furnace.


238

Apparatus

All laboratory-ware should be of borosilicate glass and the centrifuge tubes of

polypropylene. Vessels in contact with samples or reagents should be cleaned by

soaking in 10% HCl (overnight) and rinsed repeatedly with distilled water and MilliQ

water before use.

For each batch of extractions, dry a separate 1 g sample in an oven (105±2 °C) until a

constant mass is achieved. From this, a correction ‘to dry mass’ is obtained which

should be applied to all analytical values reported (i.e., results should be quoted as

amount of metal per gram of dry sediment).

The tools include:

20 mL dispenser; 5 mL pipette; 50 mL dispenser; 30 mL dispenser; 10 mL pipette;

Reagents

All reagents should be of analytical-reagent grade or better. MilliQ water should be

used throughout.

Solution A (Sodium acetic, 1.0 M NaAc, buffer at pH5 with acetic acid HAc):

Dissolve 136.08 g NaAc∙3H2O (136.08 g/mol) in MilliQ water and dilute to 0.9 L in a

fume cupboard. Adjust to pH 5 by adding glacial acetic acid (HAc). Wash the

electrodes of pH meter thoroughly before placing it in the extracting solution. Make

the volume to 1 L with MilliQ water and store in sealed plastic containers.

Solution B (Sodium Pyrophosphate, 0.1 M Na4P2O7):

Dissolve 44.606 g Na4P2O7∙10H2O (446.06 g/mol) in MilliQ water and make the

volume to 1 L. Store in sealed plastic containers.

Solution C (hydroxylamine hydrochloride, 0.25 M NH2OH∙HCl, in 0.25 M HCl):

Dissolve 17.373 g NH2OH∙HCl (69.49 g/mol) in 200 mL MilliQ water and add

24.6 mL 32% HCl and make up to 1 L volume with distilled water. Prepare this

solution on the same day the extraction is carried out.


239

Solution D (hydroxylamine hydrochloride, 1.0 M NH2OH∙HCl, in 25% HAc):

Dissolve 69.49 g NH2OH∙HCl (69.49 g/mol) in 200 mL MilliQ water and add 250 mL

glacial acetic acid (HAc) and Make up to 1 L volume with distilled water. Prepare this

solution on the same day the extraction is carried out.

Blanks

Vessel blank. To one vessel from each batch, taken through the cleaning procedure, add

40 mL of solution A. Analyse this blank solution along with the sample solutions from

step 1.

Reagent blank. Analyse a sample of each batch of solutions A, B, C and D.

Procedural blank. With each batch of extractions, a blank sample (i.e., a vessel with no

sediment) should be carried through the complete procedure and analysed at the end of

each extraction step.

24

0

Appendix 11.6 ICP-MS analyses of the reference standards determined repeatedly with samples for each analysis

reference Trace element concentrations

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Y Th

unit ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm

Batch 1

STSD-2 56.1 90.0 11.3 43.5 8.0 2.19 8.49 1.27 6.32 1.34 3.44 0.54 3.41 0.56 33.8 18.7

STSD-2 53.4 84.5 10.5 40.2 7.6 2.07 7.80 1.23 6.07 1.14 3.54 0.44 3.35 0.62 35.9 17.6

av 54.7 87.3 10.9 41.9 7.8 2.13 8.15 1.25 6.20 1.24 3.49 0.49 3.38 0.59 34.9 18.2

error% -7.2 -6.1 -2.5 -2.7 -2.8 6.4 -3.0 -7.6 -4.6 -8.3 -8.2 -18.4 -8.7 -15.7 -5.8 5.6

RSD 1.9 3.9 0.6 2.3 0.3 0.1 0.5 0.0 0.2 0.1 0.1 0.1 0.0 0.0 1.5 0.8

1Batch 2

STSD-2 55.9 94.9 11.6 44.9 8.72 2.20 9.05 1.35 7.10 1.45 3.91 0.57 3.61 0.62 38.3 19.4

STSD-2 54.9 92.4 11.7 45.1 8.66 2.06 7.72 1.17 5.87 1.31 3.54 0.51 3.34 0.53 34.4 16.5

av 55.4 93.6 11.7 45.0 8.69 2.13 8.38 1.26 6.49 1.38 3.72 0.54 3.48 0.58 36.4 17.9

error% -6.1 0.7 4.1 4.7 8.6 6.5 -0.2 -6.9 -0.2 2.5 -2.0 -10.3 -6.0 -17.4 -1.7 4.3

RSD 0.7 1.8 0.1 0.2 0.0 0.1 0.9 0.1 0.9 0.1 0.3 0.0 0.2 0.1 2.7 2.1

Batch 3

STSD-2 53.5 87.7 11.1 42.6 8.27 2.03 7.87 1.26 6.27 1.35 3.72 0.53 3.51 0.56 33.8 19.1

STSD-2 56.1 92.6 11.4 43.8 8.34 2.06 7.84 1.27 6.39 1.35 3.63 0.54 3.40 0.55 34.2 18.2

STSD-2 55.5 91.5 11.3 43.9 8.15 1.96 7.71 1.26 6.33 1.35 3.56 0.52 3.49 0.55 33.8 18.2

STSD-2 55.2 89.9 11.3 43.3 8.29 1.99 7.96 1.29 6.80 1.46 3.88 0.58 3.58 0.58 36.2 17.7

av 55.1 90.5 11.3 43.4 8.26 2.01 7.85 1.27 6.45 1.38 3.70 0.54 3.50 0.56 34.5 18.3

error% -6.7 -2.7 0.7 1.0 3.3 0.5 -6.6 -5.9 -0.8 2.0 -2.7 -9.7 -5.5 -19.9 -6.8 6.4

RSD 1.1 2.1 0.1 0.6 0.1 0.0 0.1 0.0 0.2 0.1 0.1 0.0 0.1 0.0 1.1 0.6

24

1

reference Trace element concentrations

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Y Th

unit ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm 2Sequential

STSD-2 53.6 86.5 11.8 41.7 7.80 2.00 7.90 1.20 6.30 1.30 3.50 0.60 3.40 0.60 36.4 17.9

STSD-2 54.2 87.7 11.8 42.0 7.80 1.90 7.60 1.20 6.10 1.30 3.50 0.60 3.30 0.60 35.3 16.8

av 53.9 87.1 11.8 41.9 7.80 1.95 7.75 1.20 6.20 1.30 3.50 0.60 3.35 0.60 35.9 17.4

exp 59.0 93.0 11.2 43.0 8.00 2.00 8.45 1.30 6.50 1.35 3.78 0.62 3.70 0.70 37.0 17.2

error% -8.6 -6.3 5.4 -2.7 -2.5 -2.5 -8.3 -7.7 -4.6 -3.7 -7.4 -3.2 -9.5 -14.3 -3.1 0.9

RSD 0.4 0.8 0.0 0.2 0.0 0.1 0.2 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.8 0.8

exp STSD-2 59.0 93.0 11.2 43.0 8.00 2.00 8.40 1.35 6.50 1.35 3.80 0.60 3.70 0.70 37.0 17.2

1Batch 1 (GE profile) and Batch 2 (MQ profiles) were determined together by ICP-MS;

2Sequential refers to the determination of trace element concentrations in sequential extractions and particle size fractions by ICP-MS.

24

2

Appendix 11.7 EPMA detection limits of element concentrations in Ti-, Zr- and Th- bearing minerals in the JG profile

No. Min Element concentrations (wt%)

Si Zr Ti Pb Th U Al Y La Ce Pr Nd Sm Eu Gd Dy Yb Lu Fe Ca Sr K P F

1 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.03 0.03 0.03 0.02 0.01 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.04 0.12

2 Thr 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.02 0.12

3 Fc 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.07

4 Fc 0.01 0.03 0.01 0.02 0.01 0.02 0.01 0.01 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.08

5 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.02 0.03 0.01 0.02 0.04 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.11

6 Spn 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.02 0.02 0.01 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.14

7 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.03 0.04 0.01 0.02 0.04 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.12

8 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.03 0.03 0.01 0.02 0.04 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.11

9 Rt 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.03 0.02 0.01 0.02 0.02 0.01 0.02 0.01 0.00 0.01 0.01 0.01 0.15

Zrn-zircon; Thr-thorite; Fc-fluorocarbonate; Ilm-ilmenite; Spn-sphene; Rt-rutile.

Appendix 11.8 EPMA detection limits of element concentrations in minerals of parent granitoids and regolith samples from the GE and MQ profiles


Si Zr Ti Pb Th U Al Y La Ce Pr Nd Sm Eu Gd Tb Dy Er Tm Yb Lu Fe Mg Ca Sr Na K P S F

10 Aln 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.03 0.02 0.01 0.02 0.09 0.03 0.02 0.02 0.01 1.73 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.09

11 Aln 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.03 0.02 0.01 0.02 0.09 0.03 0.02 0.02 0.01 1.75 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.09

12 Aln 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.03 0.02 0.01 0.02 0.09 0.03 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.10

13 MnzI 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.03 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.09

14 MnzI 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.09

15 MnzI 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.10

16 MnzII 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.01 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.03 0.02 0.02 0.02 0.01 2.11 0.01 0.01 0.01 0.01 0.03 0.01 0.01 0.01 0.12

17 MnzII 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.01 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.03 0.02 0.02 0.02 0.02 2.11 0.01 0.01 0.01 0.01 0.03 0.01 0.01 0.01 0.12

18 MnzII 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.03 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.10

19 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.03 0.03 0.03 0.02 0.01 0.02 0.02 0.02 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.05 0.01 0.11

20 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.03 0.03 0.03 0.02 0.01 0.02 0.04 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.01 0.04 0.01 0.12

21 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.03 0.03 0.03 0.02 0.01 0.02 0.05 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.01 0.04 0.01 0.12

22 Thr 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.05 0.05 0.04 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.02 0.01 0.02 0.01 0.13

23 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.02 0.04 0.01 0.02 0.16 0.04 0.04 0.03 0.01 1.73 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.11

24 Fsp 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.04 0.03 0.03 0.02 0.02 0.01 0.02 0.09 0.03 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.13

25 Fc 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.05 0.02 0.01 0.02 0.07 0.03 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.07

26 Fc 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.05 0.02 0.01 0.02 0.07 0.03 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.08

27 Fc 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.06 0.03 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.07

28 Mnz 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.09

29 Mnz 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.09

30 Mnz 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.09

31 Aln 0.01 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.04 0.03 0.03 0.03 0.02 0.01 0.02 0.09 0.03 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.12

32 Aln 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.03 0.03 0.02 0.03 0.02 0.01 0.02 0.09 0.03 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.09

33 Aln 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.03 0.02 0.01 0.02 0.09 0.03 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.10

34 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.02 0.03 0.01 0.02 0.15 0.04 0.04 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.10

35 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.02 0.04 0.01 0.02 0.14 0.04 0.03 0.03 0.01 0.01 0.01 0.01 0.00 0.01 0.02 0.01 0.01 0.01 0.12

36 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.03 0.04 0.01 0.02 0.12 0.03 0.03 0.02 0.01 0.01 0.01 0.01 0.00 0.01 0.02 0.01 0.01 0.01 0.12


Si Zr Ti Pb Th U Al Y La Ce Pr Nd Sm Eu Gd Tb Dy Er Tm Yb Lu Fe Mg Ca Sr Na K P S F

37 Spn 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.02 0.02 0.01 0.02 0.03 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.13

38 Spn 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.02 0.02 0.01 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.13

39 Spn 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.02 0.02 0.01 0.02 0.03 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.07

40 Ap 0.01 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.02 0.02 0.01 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.14

41 Ap 0.01 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.03 0.02 0.01 0.02 0.03 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.14

42 Ap 0.01 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.03 0.02 0.02 0.01 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.13

43 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.03 0.03 0.02 0.01 0.02 0.03 0.02 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.05 0.01 0.12

44 Mnz 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.09

45 Mnz 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.09

46 Mnz 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.04 0.04 0.02 0.01 0.02 0.03 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.09

47 Rbp 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 2.14 0.01 0.02 0.01 0.01 0.04 0.01 0.01 0.01 0.10

48 Rbp 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.04 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.08

49 Rbp 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.03 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.02 0.02 2.14 0.01 0.02 0.01 0.01 0.04 0.01 0.01 0.01 0.09

50 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.02 0.03 0.01 0.02 0.16 0.04 0.04 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.10

51 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.03 0.03 0.01 0.02 0.17 0.04 0.04 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.10

52 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.04 0.03 0.03 0.01 0.02 0.18 0.04 0.04 0.03 0.01 1.75 0.01 0.01 0.01 0.01 0.03 0.01 0.01 0.01 0.11

53 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.03 0.03 0.03 0.02 0.01 0.02 0.03 0.02 0.01 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.04 0.01 0.11

54 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.03 0.03 0.03 0.02 0.01 0.02 0.03 0.02 0.01 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.04 0.01 0.11

55 Zrn 0.01 0.03 0.01 0.02 0.01 0.02 0.01 0.02 0.04 0.03 0.03 0.03 0.02 0.01 0.02 0.02 0.02 0.01 0.01 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.04 0.01 0.12

56 Thr 0.01 0.04 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.04 0.03 0.03 0.02 0.01 0.02 0.07 0.03 0.02 0.02 0.02 2.12 0.01 0.01 0.01 0.01 0.03 0.01 0.02 0.01 0.10

Aln: allanite; Ap: apatite; Fc: REE-rich fluorocarbonate; Fsp: feldspar；Ilm: ilmenite; Mnz: monazite, I and II refers to Type 1 or Type 2; Rbp: rhabdophane; Spn: sphene; Thr:

thorite; Zrn: zircon.

24

5

Appendix 11.9 Concentrations of REE in grain size fractions of the MQ II profile

Sample Element concentrations(ppm)


d.l. 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10

MQ15sand 49.9 85.6 8.6 26.1 4.3 0.4 3.6 0.8 5.5 1.4 4.6 0.8 5.1 0.8 44.3

MQ15silt 54.1 107.8 9.7 30.4 5.5 1.1 4.5 0.8 5.0 1.1 3.2 0.5 3.1 0.5 31.0

MQ15clay 55.1 133.4 9.6 30.6 5.6 1.5 5.0 0.9 5.5 1.2 3.3 0.5 2.7 0.4 33.4

MQ15matrix 66.6 114.9 10.2 32.6 5.4 0.7 5.4 0.7 3.5 0.6 1.9 0.2 1.5 b.d. 19.0

MQ15gravel 35.9 66.3 6.3 18.2 3.8 0.8 3.2 0.5 2.9 0.7 1.8 0.3 1.8 0.3 13.9

MQ14sand 37.3 61.5 6.1 18.6 2.7 0.2 2.0 0.3 2.3 0.5 1.7 0.3 1.7 0.3 16.5

MQ14silt 53.4 103.7 10.3 32.4 5.7 1.1 4.5 0.8 5.2 1.2 3.4 0.6 3.6 0.6 29.2

MQ14clay 31.7 51.2 6.8 22.1 4.5 1.2 3.6 0.7 4.6 1.0 3.0 0.5 2.9 0.5 25.3

MQ14matrix 49.1 81.3 7.8 26.0 4.7 0.8 4.3 0.7 3.3 0.7 2.3 0.3 2.2 0.1 18.7

MQ14gravel 29.9 53.5 5.4 15.9 3.3 0.8 2.9 0.5 2.6 0.7 1.8 0.3 1.9 0.3 12.9

MQ13sand 10.3 17.8 1.7 5.4 0.9 b.d. 0.6 0.1 0.6 0.1 0.4 b.d. 0.4 b.d. 3.8

MQ13silt 26.1 63.3 5.1 16.6 3.0 0.6 2.3 0.4 2.9 0.6 2.0 0.4 2.3 0.4 15.9

MQ13clay 12.6 40.6 3.0 10.3 2.2 0.6 1.9 0.4 2.3 0.5 1.7 0.3 1.9 0.3 13.2

MQ13matrix 27.8 54.2 4.2 14.0 3.1 0.5 3.1 0.4 2.2 0.5 1.7 0.2 2.0 0.1 13.3

MQ13gravel 20.4 46.2 4.1 12.1 2.6 0.5 2.3 0.4 2.3 0.5 1.6 0.3 1.8 0.3 9.9

MQ12sand 1.6 3.0 0.2 0.7 0.2 b.d. 0.2 b.d. 0.3 b.d. 0.2 b.d. 0.3 b.d. 2.1

MQ12silt 20.6 40.0 3.6 11.2 1.9 0.3 1.5 0.3 1.9 0.5 1.6 0.3 1.9 0.3 13.8

MQ12clay 9.7 21.2 1.8 6.0 1.2 0.3 1.0 0.2 1.3 0.3 1.0 0.2 1.1 0.2 9.1

MQ12matrix 10.7 19.1 1.6 5.3 1.0 0.4 1.3 0.2 1.0 0.2 0.8 0.2 0.9 b.d. 7.4

MQ12gravel 7.5 14.5 1.3 4.0 1.0 0.2 0.8 0.2 1.1 0.3 0.8 0.2 1.0 0.2 5.6

MQ11sand 5.6 10.0 0.9 2.9 0.5 b.d. 0.4 b.d. 0.5 0.1 0.4 b.d. 0.6 0.1 3.9

MQ11silt 25.0 41.7 4.2 13.1 2.1 0.3 1.6 0.3 2.0 0.5 1.6 0.3 2.0 0.3 14.6

MQ11clay 11.2 15.2 1.8 5.8 1.1 0.3 1.0 0.2 1.3 0.3 1.0 0.2 1.0 0.2 8.9

MQ11matrix 7.6 12.9 1.1 4.1 1.2 0.3 0.7 0.1 0.9 0.2 0.6 0.1 1.0 b.d. 6.3

MQ11gravel 6.0 10.0 1.0 2.9 0.9 0.3 0.6 0.1 0.7 0.2 0.6 0.1 0.7 0.1 3.9

24

6

Sample Element concentrations(ppm)


d.l. 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10

MQ10sand 6.2 10.5 1.0 3.2 0.6 b.d. 0.4 b.d. 0.4 0.1 0.3 b.d. 0.5 0.1 3.1

MQ10silt 41.1 69.6 6.7 20.7 3.2 0.3 2.1 0.4 2.4 0.6 1.9 0.3 2.3 0.4 17.1

MQ10clay 7.9 11.7 1.2 3.9 0.7 0.2 0.6 0.1 0.9 0.2 0.7 0.1 0.7 0.1 6.2

MQ10matrix 10.3 16.3 1.4 4.6 1.5 0.2 0.8 0.1 0.8 0.1 0.5 0.1 0.6 b.d. 5.6

MQ10gravel 5.3 9.4 0.9 2.5 0.7 0.3 0.5 0.1 0.5 0.1 0.4 0.1 0.5 0.1 2.7

24

7

Appendix 11.10 EPMA detection limits of element concentrations in REE-bearing minerals from the JG profile


Si Zr Ti Pb Th U Al Y Ce Pr Nd Sm Eu Gd Dy Yb Lu Fe Mg Ca Sr P S

57 Fc 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.03 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01

58 Fc 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.04 0.03 0.04 0.02 0.01 0.02 0.02 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01

59 Ap 0.01 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.02 0.02 0.02 0.01 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01

60 Fsp 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.03 0.02 0.02 0.01 0.01 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01

61 Mag 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.02 0.02 0.02 0.02 0.01 0.02 0.06 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01

62 Mag 0.01 0.03 0.01 0.02 0.01 0.02 0.01 0.01 0.03 0.02 0.04 0.03 0.03 0.05 0.07 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01

63 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.03 0.03 0.03 0.02 0.01 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.04 0.01

80 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.04 0.03 0.03 0.01 0.02 0.01 0.01 1.69 0.01 0.01 0 0.01 0.01 0.01

81 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.04 0.02 0.02 0.01 0.02 0.04 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01

82 Ilm 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.04 0.03 0.05 0.01 0.02 0.04 0.01 1.72 0.01 0.01 0.01 0.01 0.01 0.01

83 TiO 0.01 0.02 0.01 0.01 0.01 0.01 0 0.01 0.03 0.04 0.02 0.02 0.01 0.02 0.02 0.01 0.01 0.01 0.01 0 0.01 0.01 0.01

84 Zrn 0.01 0.03 0.01 0.02 0.01 0.02 0.01 0.02 0.03 0.03 0.03 0.02 0.01 0.02 0.02 0.01 1.89 0.01 0.01 0.01 0.01 0.04 0.01

85 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.03 0.03 0.03 0.02 0.01 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.05 0.01

86 Zrn 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.02 0.03 0.03 0.03 0.02 0.01 0.02 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.04 0.01

87 Zrn 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.03 0.03 0.03 0.02 0.01 0.02 0.02 0.01 0.02 0.01 0.01 0.01 0.01 0.05 0.01

88 Thr 0.01 0.03 0.01 0.02 0.02 0.02 0.01 0.02 0.05 0.04 0.04 0.03 0.02 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.02

Ap: apatite; Fc: REE-rich fluorocarbonate; Fsp: feldspar; Ilm: ilmenite; Mag: magnetite; TiO: titanium oxides (rutile/anatase); Thr: thorite; Zrn: zircon.

The occurrence and behaviour of rare earth and associated … · The occurrence and behaviour of rare earth and associated elements in lateritic regolith profiles in Western Australia

Documents