See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/242255710
Unravelling the regolith with geochemistry
Chapter · January 2006
CITATIONS READS
11 549
1 author:
Kenneth George Mcqueen
University of Canberra
176 PUBLICATIONS 1,163 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Longterm landscape evolution of Northwestern New South Wales View project
Ore deposits and ore deposit geochemistry View project
All content following this page was uploaded by Kenneth George Mcqueen on 30 January 2015.
The user has requested enhancement of the downloaded file.
Regolith 2006 - Consolidation and Dispersion of Ideas
UNRAVELLING THE REGOLITH WITH GEOCHEMISTRY
Kenneth G. McQueen
CRC LEME, Department of Earth and Marine Sciences, Australian National University, ACT, 0200
INTRODUCTION
Geochemical data from the regolith can be used in five practical ways:
• To detect anomalous abundances of target and pathfinder elements during mineral exploration (or for
other purposes);
• To distinguish different regolith materials (and help decipher their evolution);
• To determine the parent rock type;
• To estimate the degree of weathering and leaching of in situ profiles;
• To locate ferruginous zones and redox fronts, which are potential sites for accumulation/fixation of
target and pathfinder trace elements.
Much has been written on anomaly detection and there has been some investigation of geochemical methods
for identifying parent rocks from their weathering products (e.g. Hallberg 1984; Khider & McQueen 2005).
This article focuses on the other uses of geochemical data for unravelling the regolith.
DISTINGUISHING REGOLITH MATERIALS
The regolith is composed of minerals, mineraloids, biota, water, dissolved compounds and gases. Minerals
make up the largest part and determining the mineral composition is a key step in characterising different
regolith materials. However, it can be difficult to identify minerals in regolith because they are mostly very
fine grained or pulverised in drill cuttings (X-ray diffraction or infra red spectral analysis is generally
required). Minerals have a more or less definite chemical composition and if geochemical data are available
these can be used as a broad surrogate for the mineralogy.
The major primary source minerals for the regolith are feldspars, quartz, micas, mafic minerals and in some
cases carbonates and volcanic glass (not strictly a mineral). These are ultimately transformed to quartz,
kaolinite and iron and aluminium oxide/oxyhydroxides. The accompanying chemical changes involve
addition of water, progressive loss of Na+, K+, Ca2+, Mg2+ (some Si4+) and retention of Si4+, Al3+ and Fe3+.
The K/Al and Mg/Al ratios provide a simple index for characterising different regolith materials where they
are at different stages of mineralogical/chemical evolution on the journey to end-stage regolith. Figure 1
shows the compositional distribution in terms of this index for three different types of regolith sampled down
three drill holes in the Cobar region. Also shown are the typical fields for these types of regolith, based on the
current data and previous work in the Kalgoorlie region (Johnson & McQueen 2001).
Figure 1: K/Al vs. Mg/Al ratio
plot (wt. proportions) for three
different types of regolith from
the Cobar region of western New
South Wales. Samples are 4 kg
bulks from 1 m composites
collected in drill holes CBAC
215, CBAC 217 and BRAC 1.
Arrows indicate the major
mineralogical controls on ratio
trends. Also shown are the typical
ratio fields for these types of
regolith. Analyses by XRF.
The three different regolith materials in these examples are in situ saprolite/saprock, lacustrine clays (difficult
to distinguish visually from weathered saprolite) and younger ferruginous alluvium/colluvium. These
materials have different regolith histories that are reflected in their mineral and chemical composition. The
lacustrine clays were deposited in the Late Mesozoic to Early Cenozoic and appear to have been derived
230
Regolith 2006 - Consolidation and Dispersion of Ideas
from a deeply weathered landscape (Chan et al. 2004; Chan 2005). They were well sorted during erosion and
transport to produce a kaolinite-smectite-quartz dominant sediment. The saprolite/saprock contains
significant muscovite and illite, which have been variably altered to kaolinite, depending on degree of in situ
weathering and depth in the profile. The younger alluvium/colluvium was deposited in the later Cenozoic,
when climatic conditions were significantly drier and chemical weathering less intense. Erosion of less
altered profiles, more limited sorting and low levels of post-depositional weathering produced material that
retained significant amounts of weakly altered phyllosilicates. These sediments can also contain local
concentrations of dolomitic calcrete, which will be apparent in their Mg/Al ratio. There is some
compositional overlap between the three materials and the clearest distinction is between saprolite and
transported clays. This approach to chemically distinguishing regolith types will work within particular
regions where there are regolith components characterised by different parent rock compositions, different
degrees of weathering or different histories of sorting or remixing/homogenisation during transport. The
differences can be established with an orientation survey.
DEGREE OF WEATHERING/LEACHING
The Chemical Index of Alteration (CIA) has been widely used to quantify degree of rock weathering (Nesbitt
& Young 1982). This index, CIA = 100 x Al2O3/(Al2O3 + CaO + Na2O + K2O), reflects the breakdown of
feldspars and mica to kaolinite, but it has a major drawback in that it estimates the total history of chemical
weathering from the primary source rock, i.e. including that already present in sedimentary rocks prior to
further weathering. It is thus difficult to apply this index as a direct measure of the in situ weathering of a
particular regolith sample. However, it can be useful for comparing samples within profiles developed on a
variably weathered common rock type (Figure 2).
Figure 2: Plot of CIA vs. sample depth for saprolite/saprock
samples from four profiles developed on siltstone/sandstone
units from the Cobar region. Due to the presence of calcite
(i.e. non-silicate CaO) in some of these samples, the mole
proportions of CaO used in the CIA have been set equal to
those of Na2O where they otherwise exceed this value (i.e.
assuming CaO in plagioclase, McLennan 1993).
Weathering trends within a profile or group of profiles on
common rock types can also be distinguished using the
relative mole proportions of Al2O3, (CaO+Na2O+K2O) and
(total Fe2O3+MgO). Figure 3 shows an example of an A-
CNK-FM plot for samples from one of the Cobar regolith
profiles. This approach is complicated by inclusion of Fe,
which can be accumulated at redox boundaries or rapidly
mobilised from Fe-bearing carbonates in weathering
sedimentary rocks. Formation of secondary dolomite can
also complicate Mg contents of strongly weathered regolith
(not corrected for in the MgO mole proportion).
Figure 3: A-CNK-FM plot (Nesbitt & Young
1989) of regolith samples from profile CBAC
217. The connecting line shows the
compositional path from saprock (closed
lowermost square) to progressively weathered
saprolite (open squares). Also shown is the trend
of igneous rock compositions from the
Fe2O3+MgO apex to feldspar and the typical
weathering trend for sedimentary rocks.
231
Regolith 2006 - Consolidation and Dispersion of Ideas
The Rare Earth Elements (REE) are commonly used as petrogenetic indicators because of their similar
chemical properties, typically low solubilities and assumed resistance to fractionation in supracrustal
environments. However a number of studies have shown that under some weathering conditions REE are
significantly mobilised and fractionated (e.g. Nesbitt 1979; Duddy 1980; Banfield & Eggleton 1989; Sharma
& Rajamani 2000). Analysis of REE distributions through deeply weathered profiles in the Cobar region
indicates significant leaching in the upper parts and enrichment in the lower zone close to the weathering
front (Figure 4). The LREE, particularly Ce, appear to have been the most mobile under the weathering
conditions that pertained in this region. This pattern is less marked in the least weathered profiles and there is
also some indication that profiles in different settings have different patterns of REE mobility and
enrichment. The REE may thus provide a basis for evaluating the extent and intensity of chemical weathering
and leaching. Comparison of Ce (the most mobilised LREE) with Nd (another LREE with similar properties
and initial distribution) is one possibility for a REE mobility and weathering index. Examples of Ce/Nd plots
for different weathered profiles in the Cobar region are shown in Figure 5. Departure of the Ce/Nd ratio from
the initial ratio (typically around 1.5 to 2.5 for sedimentary rocks) indicates the relative depletion or
enrichment of Ce and provides an indication of the chemical leaching of profiles.
Figure 4: REE patterns for samples from different depths in weathered in situ saprolite from the Cobar
region (examples are from the CBAC 215 and Wilga Tank profiles, REE analysis by ICP MS on dissolved
fused glass discs). Data are normalised to North American Shale composite (Gromet et al. 1984).
Figure 5: Variation in Ce/Nd ratios with depth in two weathered profiles from the Cobar region. The
bedrock is interbedded siltstone-sandstone with Ce/Nd ratio of ca. 2.2. Departure from this ratio shows
the relative depletion and enrichment (mobility) of Ce relative to Nd.
FERRUGINISATION
Ferruginisation in the regolith is controlled by reduction/oxidation of iron in solution. The resulting iron
oxides/oxyhydroxides overprint other regolith materials and for this reason Fe content is not a good
parameter to use in some other aspects of regolith characterisation. However, ferruginisation can indicate the
position of redox boundaries, which are important chemical transition zones in the regolith. Ferruginisation
also promotes trace element fixation due to the abilities of goethite and hematite to adsorb and/or incorporate
232
Regolith 2006 - Consolidation and Dispersion of Ideas
many of these elements (particularly the target and pathfinder elements used in geochemical exploration such
as As, Ba, Bi, Cu, Pb, Sb, Zn). Ferruginisation and the location of potential redox boundaries can be
estimated by visual observation, but this is prone to error due to the ability of small amounts of finely divided
hematite to strongly colour clay-rich regolith. Short wave infra-red spectral analysis (e.g. using the ASD) can
give a more reliable estimate of iron oxides/oxyhydroxides.
The Fe2O3 content of regolith is an obvious potential indicator of ferruginisation, however this is affected by
the relative abundance of quartz and other Fe-free minerals in the sample (i.e. a very small increase in Fe2O3
could be very significant in a weathered quartz-rich sandstone). Comparing total Fe against an ‘immobile’
element can provide a more precise index, the difficulty in strongly weathered regolith is finding an
‘immobile’ or ‘near immobile’ element. Probably the best candidate is Zr, as it is largely concentrated in
zircon, a resistate mineral that is remarkably stable under most weathering conditions. A problem is that
zircons are not uniformly distributed in many rocks, particularly sedimentary rocks where the sediments have
undergone sorting. Also Zr is a trace element, meaning small variations in its abundance can make a large
change to comparative ratios. Titanium is also considered a relatively immobile element and is more
abundant and widely hosted than Zr, although in sedimentary rocks a significant proportion is hosted in rutile
and ilmenite, which can also be affected by sedimentary sorting. Titanium oxyhydroxide is the least soluble
of the hydroxides and carbonates of the major and minor elements in near neutral dilute solutions (Garrels &
Christ 1965). Variation in the Fe/Ti ratios for weathered profiles from the Cobar region highlights the zones
of ferruginisation and potential redox boundaries (Figure 6).
Figure 6: Variation in the Fe/Ti ratio through four saprolite profiles from the Cobar region. The top two
profiles show ferruginsation at depth reflecting a redox zone. The lower examples show near surface
ferruginisation
Of the major elements Al is typically the least mobile during weathering, so comparison of total Fe and Al
content may also be useful in detecting the level of ferruginisation, particularly in weathered sedimentary
rocks with highly variable quartz content. Aluminium is more likely to be included in the element suite for
geochemical exploration programs when using multi-acid digest and ICP OES analysis. Even if Ti is
included in this suite, the analyses will be less than true values due to partial or non digestion of rutile and
ilmenite. Figure 7 shows a plot of Fe/Al vs. Fe/Ti for saprolite samples from the Cobar region (analysed by
XRF, which gives total values for both Al and Ti). There is generally good positive correlation of these
233
Regolith 2006 - Consolidation and Dispersion of Ideas
ratios, suggesting that for these weathered profiles Al is relatively immobile. The least weathered saprock
samples plot in the centre of the range of values with a trend to lower ratios with leaching of Fe from
saprolite and a trend to higher ratios with ferruginisation. In a single profile, the change from a negative
(leaching) trend to a positive (ferruginisation) trend will typically mark the redox boundary as reflected by
iron oxide/oxyhydroxide precipitation.
Figure 7: Plot of Fe/Al vs. Fe/Ti
(wt. proportions) for saprolite
samples from the four weathered
profiles in the Cobar region. These
show evidence of both iron
leaching and ferruginisation.
DISCUSSION
Different weathering and erosional
regimes can produce regolith with
different mineralogical and
chemical characteristics. In this
way landscape evolution leaves a
geochemical signature. This
signature can in turn be used to
differentiate regolith materials
with different regolith/landscape
histories. Element mobility and
dispersion in the regolith varies with the extent of chemical weathering and leaching. Changes in major
element and key trace element contents down regolith profiles can provide clues to the extent and history of
chemical weathering and help predict zones of trace element depletion and accumulation. Ferruginsation,
resulting from groundwater mobilisation of Fe2+ under reducing conditions, followed by oxidation to Fe3+
and precipitation of goethite and hematite, is a particularly important control on the accumulation of target
and pathfinder elements used in geochemical exploration. Some basic interpretation of the major and trace
element contents of drill cuttings can locate ferruginous zones and redox boundaries in the regolith profile.
Clearly it is important to collect and analyse regolith samples through complete weathering profiles. In a
major exploration program it may not be possible to sample every drill hole in this way but sufficient holes
should be sampled to provide a good understanding of the three dimensional chemical structure of the
regolith. This knowledge can then be used to select the most appropriate regolith horizon/s for sampling.
Modern analytical methods allow for relatively cheap and efficient collection of multi-element data. In many
cases much of it is not fully utilised. It is possible to get more information from these data to improve
interpretation of regolith materials and their controls on trace element distributions. The results presented
here are based on high quality XRF analyses. Investigation of data acquired by the cheaper and more widely
used (by the mineral exploration industry) multi acid ‘total digest’ method, followed by analysis using ICP
OES, reveal that these data can be applied but with some caution due to the less than total analysis of K and
to a lesser extent Al. Lower K results are due to the formation of insoluble potassium perchlorate during
digestion. It should be possible to address this issue and improve the useability of these data.
CONCLUSIONS
Three simple chemical indices, which are easily applied, can be used to get a better understanding of the
regolith. These are
• A compositional index based on the K/Al and Mg/Al ratios for differentiating regolith derived from
distinctly different parent materials or with a contrasting history of formation;
• A weathering/leaching index using variation in CIA values within profiles; and
• A ferruginisation index based on Fe/Ti ratios (or alternatively Fe/Al) of regolith samples through the
weathered profile.
Far from being immobile, REE can be significantly redistributed during weathering, particularly under acidic
conditions, to the point where their depletions in one part of the profile and subsequent enrichments in other
parts can provide an index for the intensity and style of chemical weathering.
234
Regolith 2006 - Consolidation and Dispersion of Ideas
REFERENCES
BANFIELD J.F. & EGGLETON R.A. 1989. Apatite replacement and rare earth mobilisation, fractionation and
fixation during weathering. Clays and Clay Minerals 37, 113-127.
CHAN R.A. 2005. Girilambone regolith and landscape evolution. In: ROACH I.C., ed. Regolith 2005 – Ten
Years of CRC LEME. CRC LEME, pp. 43-48
CHAN R.A. GREENE R.S.B., HICKS M., LE GLEUHER M., MCQUEEN K.G., SCOTT K.M. & TATE S.E. 2004.
Regolith architecture and geochemistry of the Byrock area, Girilambone region, north-western NSW.
CRC LEME Open File Report 159, 71 pp.
DUDDY I.R. 1980. Redistribution and fractionation of rare-earth and other elements in a weathering profile.
Chemical Geology 30, 363-381.
GARRELS R. M. & CHRIST C. I. 1965. Solution, Minerals and Equilibria. Harper & Rowe, New York, pp.
403-429.
GROMET I.P., DYMEK R.F., HASKIN L.A. & KOROTEV R.L. 1984. The “North American Shale Composite”:
its compilation, major and trace element characteristics. Geochimica et Cosmochimica Acta 48, 2469-
2482.
JOHNSON C.B. & MCQUEEN K.G. 2001. The nature of gold-bearing palaeochannel sediments in the Gidji
area north of Kalgoorlie, Western Australia. Quaternary International 82, 51-62.
HALLBERG J.A. 1984. A geochemical aid to igneous rock identification in deeply weathered terrain. Journal
of Geochemical Exploration 20, 1-8.
KHIDER, K & MCQUEEN K.G. 2005. Geochemical discrimination of weathered bedrock in the Hermidale-
Byrock region of western NSW. In: ROACH I.C., ed. Regolith 2005 – Ten Years of CRC LEME. CRC
LEME, pp. 170-175.
MCLENNAN S.M. 1993. Weathering and global denudation. Journal of Geology 101, 295-303.
NESBITT H.W. 1979. Mobility and fractionation of rare earth elements during weathering of a granodiorite.
Nature 279, 206-210.
NESBITT H.W. & YOUNG G.M. 1982. Early Proterozoic climates and plate motions inferred from major
element chemistry of lutites. Nature 299, 715-717.
NESBITT H.W. & YOUNG G.M. 1989. Formation and diagenesis of weathering profiles. Journal of Geology
97, 129-147.
SHARMA A. & RAJAMANI V. 2000. Major element, REE, and other trace element behaviour in amphibolite
weathering under semiarid conditions in southern India. The Journal of Geology 108, 487-496.
Acknowledgments: The work reported here was supported by the Australian Government’s Cooperative
Research Centre Program, within CRC LEME. I thank my co-workers and students from CRC LEME who
helped collect some of the data used in this synthesis and interpretation. Regolith drilling was carried out by
the NSW Department of Primary Industries, as part of a collaborative project with CRC LEME, and by
Cobar Management Pty Ltd. Geochemical analyses were performed by Geoscience Australia, Canberra and
ALS Chemex, Brisbane. I thank Tony Eggleton, Ian Roach, Ross Taylor and an anonymous reviewer for
their comments and suggestions on the article.
235
View publication stats
