Journal of Geochemical Exploration 108 (2011) 21–26

Contents lists available at ScienceDirect

Journal of Geochemical Exploration

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j g e o ex p

Geochemical characterization of the mining district of Linares (Jaen, Spain) by means

of XRF and ICP-AES
L. Arenas, M. Ortega, M.J. García-Martínez ⁎, E. Querol, J.F. Llamas
Departamento de Ingeniería Química y Combustibles, E.T.S.I. de Minas de Madrid, Universidad Politécnica de Madrid, C/ Alenza, 4. 4ª planta. 28003, Madrid, Spain

a r t i c l e i n f o a b s t r a c t

Article history: Linares (Jaen, Spain) has been subjected to an intense mining and metallurgic activity during 2500 years.
Received 11 February 2010 Basically, lead and silver have been extracted from galena; as well as copper and zinc. Different studies have
Accepted 2 September 2010 been conducted to evaluate trace element concentrations to obtain relationships with anthropogenic and
Available online 8 September 2010
geologic factors. The analytical results of these studies have been obtained with ICP-AES. This analytical
technique is relatively expensive and inaccessible in zones with scarce economic means. In this study, XRF
Keywords:
XRF
analyses have been conducted and the obtained results are presented, as well as a comparison with the
ICP-AES previously obtained ICP results. In both cases, a cluster analysis has been made to try to identify the same
Trace elements relations in the target area. 5 groups have been identified, mainly related with lithology. Only 10 of the 122
Mining grid squares (each one of 1 km2) are classified in different groups if the cluster analysis is conducted with XRF
Sulphides results or ICP results. ICP classifies better where these grid squares are located, mostly, in contact zones of
different lithologies.
© 2010 Elsevier B.V. All rights reserved.

1. Introduction Linares (Martínez, 2002). In this work, ICP-AES analytical technology

was used for the characterization of the samples. This analytical
The place of study is located in the mining district of Linares. It technology is not always available, especially in developing countries.
covers a surface of 126 km2 that has been divided in 126 squared For that reason, the analysis of this study has been carried out by means
patterns, each one with an area of 1 km2. The mining district is located of another more accessible and economic analytical technology. This
on a substratum in which several geological units can be distin- technology is the X-ray fluorescence (XRF) analysis whose needs are
guished, mainly by granites and phyllites of the Paleozoic, sandstones much more attainable both for laboratory equipment and sample
and lutites of the Trias and marls of the Miocene (Martínez et al., preparation. It is difficult and time consuming to undertake a total
2007) (Fig. 1). elemental analysis following the ICP-AES method because the sample
The wide and intense mining, mineralurgical (gravimetry and must usually be dissolved. It is well known that sometimes it is difficult
flotation) and metallurgical activities developed in the mining zone of to dissolve sediments and soils, as some parts may be filtered out during
Linares (south of Spain) have generated collections of sterile of mine the sample preparation procedure. Clearly, in the sample treatment, loss
in dumps of blocks, rafts of washer thins, collections of washer thicks of some elements or sample contamination may occur. X-ray fluores-
and collections of slags. Both the mining activities and the geological cence analysis is a non-destructive method, possessing the advantage
natural conditions make predictable that the concentrations of some that the total elemental content of the samples can be obtained without
elements in the soil overcome the normal ranges in undisturbed areas. any complicated sample pre-treatment (Somogyi et al., 1997). ICP-AES
The area has been characterized historically due to the exploitation of has high initial and maintenance costs and it uses high intakes of Ar and
veins of galena (PbS), though the paragenesis also includes other electricity. Nowadays, this technique is being competitive with XRF due
minerals as sphalerite (ZnS), chalcopyrite (CuFeS2), and baryte to economy of scale in some laboratories. Another advantage of XRF is
(BaSO4) (Azcárate and Argüelles, 1971; Azcárate, 1977; Fontboté, that it can be used on site to identify major and minor elements.
1982; Lillo, 1992, 2003). The comparison of the results obtained using both analytical
It exists, therefore, a real need to evaluate the levels of concentration technologies allows to obtain the results with its “background”
of the trace elements in the soils of the region of Linares. In the year associated values, the definition of anomalous concentrations inside
2002, a doctoral thesis on environmental geochemistry was defended in the characteristics of the emplacement that is being studied, as well
as the identification of the contributions of trace elements from
⁎ Corresponding author. Tel.: + 34 91 336 32 43; fax: + 34 336 69 48. different sources (natives, anthropics, agriculturalists …). The
E-mail address: [email protected] (M.J. García-Martínez). spatial study of the values distribution has been carried out with

0375-6742/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.gexplo.2010.09.002
22 L. Arenas et al. / Journal of Geochemical Exploration 108 (2011) 21–26

Fig. 1. Main lithologies in the area, principal mining district zones (mines, foundries, and industrial parks) and grid square distribution (J. Martínez et al., 2007).

the help of different statistical multivariant techniques and of the analysis equipment (Philips, PW 1404, XRF Spectrometer). The
geostatistical tools. anode of the tube of X-rays is made of Sc–Mo, so the results for these
It is of interest the development of methodologies for geochemical two elements have not been taken into account.
characterization, using different analytical technologies, with a The analysis of the samples by means of ICP-AES was conducted in
general application, not only restricted to the district of Linares, ACME laboratories (ACME Analytical Laboratories Ltd.) in Vancouver
giving response to a new approach of increasing development in the (Canada). The sample preparation was carried out with a total acid
study and resolution of environmental problems. attack. One gram of the fraction smaller than 63 μm is attacked
chemically with HNO3, HClO4, HF and HCl in an open Teflon reactor.
2. Methods and materials Every sample in dissolution was analyzed with ICP-AES.

2.1. Sample collection 3. Analysis of results

The samples were collected for the accomplishment of a doctoral Table 1 summarises the results for the elements analyzed: mean,
thesis (Martínez, 2002) carried out during a previous pilot study that median, standard deviation, minimum, and maximum values for the
indicated the need to gather five increases of 1500 g each from every original data are given.
squared pattern soil by means of an Auger sampler. These already Before carrying out the cluster analysis, a study of the "enrichment
collected and analyzed samples were used for this study. factor" (E.F.) of the different elements analyzed in the zone of study
was conducted. The enrichment factor is the factor by which it is
2.2. Sample preparation and analysis necessary to multiply the normal abundance of an element (the
Clarke) in the terrestrial crust to obtain an economically exploitable
The sample preparation for the analysis by means of XRF was concentration. For example, the normal abundance of lead in the
carried out in the Centralized laboratory of the Madrid School of terrestrial crust is 0.0001% and its enrichment factor is 2000. So a lead
Mining Engineers using pressed tablets. In this technique, 6–8 g of deposit should contain 0.2% of lead. However, the feasibility of
sample is compressed under a pressure of 2–3 t/cm2, to give a tablet of exploiting a deposit depends on many other factors such as geology of
31.5 mm of diameter and 4–5 mm of thickness. A binder is almost the sector, product pricing, cost of purification, geography, climate,
always used to avoid crumbling. In the present study, to carry out the infrastructure, national politics, political trust, socio-economic stabil-
above mentioned preparation of the pressed tablet, 8 g of the sample ity, interest rate, industrial growth and so on.
is mixed with 3.5 cm3 of elvacite (binder) solution. This mixture was The study of the enrichment factor started with the mean of every
inserted in a press (Research & Industrial Instruments Company, element in the zone of study and the calculation of the enrichment
30 ton press C-30) to obtain a tablet that, later on, was introduced in factor afterwards, the latter determined as the relation between a) the
 L. Arenas et al. / Journal of Geochemical Exploration 108 (2011) 21–26 23

Table 1
Statistical data: mean, minimum, maximum, median, and standard deviation. Values of concentration are expressed in mg kg− 1.

Mean Min Max Median Standard deviation Mean Min Max Median Standard deviation

Mo.icp 1.88 1.50 15.00 1.50 1.52 Ca.icp 34636.07 1700.00 207600.00 13300.00 43599.36
Mo.xrf 56.14 35.00 94.00 53.00 13.76 Ca.xrf 37745.49 2970.00 211200.00 15250.00 45502.04
Cu.icp 141.17 14.00 1654.00 53.50 233.60 P.icp 867.05 330.00 2920.00 780.00 444.95
Cu.xrf 812.73 400.00 2100.00 660.00 492.97 P.xrf 798.00 541.00 1340.00 771.00 235.01
Pb.icp 4148.98 25.00 38419.00 1346.50 7134.92 La.icp 68.94 17.00 145.00 69.50 22.89
Pb.xrf 4352.47 36.00 62600.00 1315.00 8424.30 Cr.icp 50.02 28.00 150.00 49.00 15.43
Zn.icp 110.43 27.00 953.00 74.00 121.75 Cr.xrf 300.00 300.00 300.00 300.00 NA
Zn.xrf 139.79 26.00 880.00 104.00 137.62 Mg.icp 6179.51 2500.00 12300.00 5900.00 2336.66
Ag.icp 2.09 0.50 29.00 0.50 3.88 Mg.xrf 9305.41 4800.00 17500.00 9055.00 2795.37
Ag.xrf 24.00 20.00 30.00 23.00 4.32 Ba.icp 862.92 120.00 3464.00 703.50 655.50
Ni.icp 20.43 10.00 40.00 20.00 5.83 Ba.xrf 1249.30 241.00 11400.00 770.00 1525.80
Co.icp 12.27 5.00 39.00 10.00 6.15 Ti.icp 3320.49 1500.00 6400.00 3000.00 1083.07
Co.xrf 41.92 20.00 69.00 41.00 11.48 Ti.xrf 5979.51 2660.00 9510.00 5990.00 1419.27
Mn.icp 1236.05 194.00 3981.00 1094.50 813.41 Al.icp 63549.18 30100.00 114200.00 57350.00 19989.32
Mn.xrf 1420.09 390.00 3930.00 1350.00 758.37 Al.xrf 83959.02 41300.00 119100.00 82200.00 16978.73
Fe.icp 32254.92 15900.00 60500.00 31050.00 8873.06 Na.icp 4902.46 1500.00 16400.00 3300.00 3624.90
Fe.xrf 36656.56 18300.00 65500.00 35100.00 10404.70 Na.xrf 3710.16 660.00 13400.00 2470.00 2797.99
As.icp 25.33 3.00 162.00 17.00 26.31 K.icp 25508.20 10100.00 38300.00 24800.00 6828.17
As.xrf 77.50 56.00 100.00 75.00 17.56 K.xrf 29540.98 10800.00 45900.00 28850.00 8018.01
U.xrf 20.00 20.00 20.00 20.00 NA W.icp 5.99 2.50 70.00 5.00 7.94
Th.icp 26.16 4.00 80.00 21.00 15.51 Zr.icp 104.20 36.00 238.00 100.00 33.45
Th.xrf 134.50 128.00 141.00 134.50 9.19 Zr.xrf 786.05 141.00 1760.00 793.00 293.44
Sr.icp 149.25 60.00 523.00 121.50 88.72 Sn.icp 5.65 1.50 51.00 4.00 5.96
Sr.xrf 163.43 79.00 490.00 140.00 84.17 Sn.xrf 60.50 34.00 87.00 60.50 37.48
Cd.icp 0.77 0.50 5.00 0.50 0.63 Y.icp 22.36 9.00 48.00 19.50 8.96
Sb.icp 11.14 3.00 186.00 3.00 21.50 Y.xrf 65.65 30.00 119.00 62.50 21.09
Sb.xrf 70.25 35.00 179.00 52.50 46.30 Nb.icp 9.82 5.00 24.00 8.00 4.54
Bi.icp 3.18 3.00 9.00 3.00 0.92 Nb.xrf 50.00 50.00 50.00 50.00 0.00
V.icp 72.48 43.00 124.00 72.50 13.36 Be.icp 3.76 1.00 6.00 4.00 1.43
V.xrf 93.05 52.00 175.00 91.50 19.34 Sc.icp 11.80 6.00 25.00 10.00 4.16

concentration of the element with regard to the Ytrium concentration always higher for XRF than for ICP. The fact that ICP gives, sometimes,
in the zone of study and b) the relation of the Clarke of the element a higher enrichment factor than XRF is not related with greater
with regard to the Ytrium in acid rocks: concentrations of the element but with the fact that concentration of
Ytrium by means of ICP is 22 ppm whereas by means of XRF is 66 ppm
½Element concentration (3 times more). The only exception is Pb (in which almost all the
½Ytrium concentration
E:F: = ½Clarke of the element in acid rocks
: results obtained by means of XRF are over the limit of detection of the
½Clarke of Ytrium in acid rocks technique, probably due to the high occurrence in the zone) and it is
the only case in which XRF's average (4352 ppm) and ICP's average
The values established by Mason and Moore (1982) were used as (4149 ppm) are very similar.
the Clarke for every element in the acid rocks. Lawson and Winchester The results obtained suggest that a more detailed comparison
(1979) established a scale to determine the origin of the element that study of both techniques for the elements enriched in the zone of
is studied. The above mentioned authors indicated that an enrichment study is advisable:
factor above 500 indicates serious pollution of anthropic origin,
■ Pb, Cu, and Sb. For these elements, a good correlation exists
whereas if it is minor to 10, it indicates that the above mentioned
between the results obtained with both techniques. In such graphs
element comes from the rock (natural origin).
the values are very close to the line in which, theoretically, all the
The elements shown to be more enriched in the emplacement
represented points should be: a line starting at 0 with a slope of 1.
were, in this order:
Ideally, the results obtained for each cell by means of XRF should
a) According to ICP results: Pb (1856 times more), Sb (55 times overlap the results obtained by means of ICP-AES for each sample.
more), Ag (23 times more), As (15 times more), and Cd (10 For Cu and Sb, there are some “y” values with x = 0; this behaviour
times more) and also a slight enrichment is observed in Cu (4.2 owes to the fact that small concentrations are being detected with
times more) and Mo (2.06 times more). ICP-AES (more sensitive technique) which cannot be detected
b) According to XRF results: Pb (663 times more), Sb (54 times with XRF because they are below the detection limit of the
more), Ag (49 times more), and As (16 times more) and also a equipment.
slight enrichment is observed in Cu (8.3 times more). By means of ■ Ag, As, and Y. The dispersion graphs show a widespread trend of
XRF, Cd has not been detected and Mo result is invalidated because points concentrated parallel to the “y” axis and another
the anode of the XRF tube was made from Mo. numerous concentration in the right side of the graphs, close to
the “x” axis (high XRF's concentrations with low ICP-AES's
The first thing remarkable is the disparity between XRF and ICP-
concentrations). These graphs indicate either a weakly positive
AES values. Whereas for Sb and As the enrichment values of both
correlation or no correlation at all. These graphs show two
technologies are similar, for Pb there is three times more enrichment
differentiated trends:
by means of ICP, and for Ag and Cu there is twice more enrichment by
means of XRF. - Some of them show that, certain elements, in most of the
As it can be observed in Fig. 2, all the analyzed elements with samples, only are detected by means of ICP-AES but not by
anomalies have a greater upper limit of detection by means of XRF means of XRF; they are those elements (Ag and As) in which
than by means of ICP-AES (except for Pb). Therefore the mean is almost all the points are parallel to the "y" axis with "x = 0".
24 L. Arenas et al. / Journal of Geochemical Exploration 108 (2011) 21–26

Fig. 2. Graphs showing the correlation between XRF and ICP-AES for the elements enriched in the zone: Pb, Sb, Ag, As, Cu, and Mo. Y has also been included.

- For Y, the behaviour is more chaotic because, for the same ICP- The cluster analysis of the results obtained of the samples analyzed
AES value, according to the different samples, sometimes the by means of XRF and by means of ICP-AES can be observed in Figs. 3
element is detected by means of XRF but by others it is not; and, and 4. The variables that have been in use are the results obtained in
in general, the concentrations obtained by means of XRF are the analyses (in ppm). In case of XRF, the Ni, La, and Be results are
quite higher than the concentrations obtained by means of ICP- rejected because there are no valid cases and the statistics cannot be
AES. calculated. Sc and Mo have also been eliminated because the anode of
the XRF tube was made from these elements. In case of ICP-AES, U and
Au are eliminated because there are no valid cases. The multivariate
4. Multivariate methods analysis carried out with ICP-AES's results has already been
conducted, though using the logarithms of the results instead
In spite of the discrepancies found between both techniques, in (Martínez López et al., 2008).
which is observed that the enrichment factors are usually higher for The white squared patterns were excluded from the analysis
XRF than for ICP, an analysis of conglomerates (using fuzzy K-means provided that they correspond to pilot squared patterns (38, 58, and
conglomeration method and the Square Euclidean distance) was 110) that were sampled according to a different methodology (Martínez,
performed to show if the separation in natural groups matches both 2002). It was decided to exclude them from the analysis to avoid
techniques. Fuzzy K-means was chosen because it has become the interference in the results produced by a different sampling technique.
most well-known and powerful method in cluster analysis (Kuo-Lung Squared pattern 112 was also excluded from the analysis because its
and Yang, 2002). results were outliers due to some problems in the analysis by XRF.
 L. Arenas et al. / Journal of Geochemical Exploration 108 (2011) 21–26 25

The first thing that is remarkable in Figs. 3 and 4 is that the

clustering analysis has given very similar results independently of the
analytical technique; which indicates that, though XRF should be a
technology with worse detection limits, it is robust enough and these
results are comparable to the results obtained by means of ICP-AES. Of
the 122 squared patterns studied, only 10 of them (8%) are grouped
differently depending on the analytic technique. Subsequently a more
detailed analysis of these 10 squared patterns was done, showing that
these cells are located in zones of contact or of strong anthropic
activity. It is important to remember at this point that in XRF the
matrix effect influences greatly the results, therefore it is expected
that the XRF results are greatly influenced by the lithology whereas
ICP-AES's results, being independent from the matrix of the sample,
are more independent from the above mentioned matrix effect.

➢ Squared pattern 3. According to the lithology map, this cell is in a

Fig. 3. Squared patterns coloured according to the five different groups obtained as a
zone of Trias–Miocene contact. In Trias, there are no shales,
result of the clustering analysis (method Ward, squared Euclidean distance) of the common in the rest of Trias, but an alluvium appears. ICP groups
concentrations of the elements analyzed by XRF. The number on the top of each cell this cell with Trias whereas XRF groups it with granite. Though the
stands for the number of the squared pattern and the number below stands for the cell is close to the granitic zone, the alluvium of the Trias and the
cluster's number.
Miocene clearly predominates, for what it seems to be better
grouped according to ICP's results. As for the anthropic activity,
though there are oak groves, it is necessary to emphasize that it is
Five different groups can be observed in Figs. 3 and 4: in the proximities of “Cerro Pelado” which would indicate a strong
mining activity. It is possible that the cell has residues proceeding
1. Cluster 1 cells correspond to granite outcrops. The activity that from this near hill, which would indicate a strong influence of the
prevails is mining and, in a minor way, mineralurgy, oak groves, mineralization (especially high concentrations in aluminium) that
and crops. might be the cause why XRF groups it with granite (zone with
2. Cluster 5 cells are placed mainly on the Miocene. The principal mineralization).
activity is the culture of olive groves, grain farming and the urban ➢ Squared pattern 4. According to the lithology map, this cell is
one (given its proximity to Linares). located in a zone of contact of quaternary granite with outcrop of
3. Cluster 3 cells are placed principally on Triassic areas, zones that the typical mineralization of the zone. Alluvium-Q of Trias appears
have been affected by mining industry, metallurgy, mineralurgy, in minor proportion. According to XRF this cell would be on granite
crops and in a minor way by olive groves. (high concentrations in Fe, Co, and Ti in comparison with ICP; it
4. Cluster 4 cells are placed mainly in zones of contact of Granite with also presents high concentrations of Nb in comparison with the
Trias; in squared patterns where mining activity and crops have been rest of the cells, where it has not been detected by means of XRF)
carried out and, in minor way, mineralurgical and urban activities. whereas according to ICP-AES it would be located in zones of
5. Cluster 2 cells are located principally on zones of Trias–Miocene's contact between granite and Trias. It is also in the proximities of a
contact or in zones highly impacted, where mining or metallurgical hill where a strong mining activity is expected. It is, therefore,
activities have been located. In these squared patterns crops and better classified by ICP.
mining were the main activities; emphasizing some cells where ➢ Squared pattern 17. It is located in a zone of contact between
mineralurgy and metallurgy and also some urban activity have alluvium-Q and shales of Trias and granite with mineralization of
been carried out (squared patterns 25, 51, 53, 63, and 91). Quaternary. According to XRF the cell would be grouped by those
of Trias. According to ICP it would remain classified as a zone of
contact Trias–granite. ICP classifies better the cell, although the
fact that XRF classifies it as Trias (low concentration in zirconium
in comparison with ICP) seems to indicate that the anthropic
activity related to the mineralization typical of the zone does not
seem to have a very strong influence in this cell, though it is
located very close to the hill with the above mentioned
mineralization outcrop.
➢ Squared patterns 45 and 46. Zone of contact of Trias with Granite.
According to XRF they would remain classified as granite (the cell 45
with major concentrations in Y, Zr, and Ti in comparison with ICP and
the cell 46 with major content in Co, V, Ti, Zr, Mn, Al and Y).
According to ICP, it would remain classified under a zone of contact
Trias–granite. ICP seems to classify better the cell but, given the
obtained results by means of XRF, the cells seem to have a strong
anthropic activity. In these squared patterns there are mining traces.
➢ Squared pattern 55. In this squared pattern Trias, granite and
mineralization of the zone could be identified. According to XRF it
would be in Trias (minor concentration of Fe, K, Na, Mn and Ba
Fig. 4. Squared patterns coloured according to the five different groups obtained as a than ICP). According to ICP it would be in a zone of contact Trias–
result of the clustering analysis (method Ward, squared Euclidean distance) of the
concentrations of the elements analyzed by ICP. The number on the top of each cell
granite. Therefore, it might say that the cell would be better
stands for the number of the squared pattern and the number below stands for the classified by means of ICP. XRF seems to indicate the absence of a
cluster's number. strong anthropic activity related to the mining activity of the zone.
26 L. Arenas et al. / Journal of Geochemical Exploration 108 (2011) 21–26

➢ Squared pattern 72. Zone of contact of Trias with granite and the two lithologies depending on which matrix dominates in the collected
mineralization of the zone. According to XRF, it would be in a zone sample. It is also necessary to keep in mind that ICP is capable of
of contact of Trias and Miocene (major concentrations of Al, Ti, V, detecting more elements (lower detection limits), which helps at the
Ca, Mg, As, Fe, and K than ICP). According to ICP, it would be in moment of classifying better the cells. Though ICP is more precise at
Trias. Though none of the two classifications is completely correct, the moment of identifying zones of mixing lithologies, this technique
it seems that it remains better classified according to ICP. XRF's is more expensive both in the preparation and in the analysis of the
results seem to indicate the absence of a strong anthropic activity. samples. Therefore, it is possible to conclude that both technologies
➢ Squared patterns 73 and 87. They are located in zones of contact of allow a good identification of the different lithologies present in
Trias with granite. According to XRF, they would be in granite (the mining districts.
cell 73 with major concentrations in Co, Al and Ti in comparison
with ICP and the cell 87 with major contents in Y, Sr, Co and Ti). References
According to ICP it would be located in zones of contact Trias–
Azcárate, J.E., 1977. Geological map and explanatory memorandum of the 905 sheet
granite. ICP better classifies the cells, though in the cell 73
(Linares), 1:50.000. Geological and Mining Institute of Spain.
predominates the granite. Azcárate, J.E., Argüelles, A., 1971. Tectonic evolution and vein structures in the mining
➢ Squared pattern 103. It is located in Trias. According to XRF it district of Linares. Congress Spanish–Portuguese–American of Economic Geology.
would remain classified under a zone of contact Trias–granite Madrid, Volume I, Section 4, pp. .17–32.
Fontboté, J.M., 1982. Geological map and explanatory memorandum of the 70 sheet
(major concentrations of Al, Zr and Ti than ICP). According to ICP- (Linares), scale 1:200, 000. Geological and Mining Institute of Spain.
AES it would be classified under a Trias–Miocene contact zone. Kuo-Lung, Wu., Yang, Miin-Shen, 2002. Alternative c-means clustering algorithms.
None of these two classifications are correct, though XRF Pattern Recognit. 35, 2267–2278.
Lawson, D.R., Winchester, J.W., 1979. A standard crustal aerosol as a reference for
approaches more to the proximities of the cell. It would possibly elemental enrichment factors. Atmos. Environ. 10, 925–930.
be an anthropic activity that impedes the correct classification of Lillo, J., 1992. Geology and geochemistry of Linares-La Carolina Pb-Ore field
this zone in Miocene. This cell is close to the former mine of “Los (Southeastern border of the Hesperial Massif). Ph D. Thesis, Univ. Leeds.
Lillo, J., 2003. Hydrothermal alteration in the Linares-La Carolina Ba–Pb–Zn–Cu–(Ag)
Salidos” and mining traces have been identified. vein district, Spain: mineralogical data from El Cobre vein. Trans. Inst. Min. Metal.
Sect. B 111, B114–B118.
5. Conclusions J. Martínez, 2002. Geochemical and environmental characterization of soils in the
mining sector of Linares. Ph D. Thesis, Universidad Politécnica de Madrid, 2002.
Martínez López, J., Llamas Borrajo, J., de Miguel García, E., Rey Arrans, J., Estévez, Mª.C.,
In a study where the cells in which the classifications of the results Sáez Castillo, A.J., 2008. Multivariate analysis of contamination in the mining
obtained by means of ICP and XRF are different, it is observed that the district of Linares (Jaén, Spain). Appl. Geochem. 23, 2324–2336.
Martínez, J., Llamas, J., de Miguel, E., Rey, J., Hidalgo, M.C., 2007. Determination of the
principal differences are in zones of contact between two lithologies
geochemical background in a metal mining site: example of the mining district of
(9 of 10 differently grouped cells). Therefore, these differences are Linares (South Spain). J. Geochem. Explor. 94, 19–29.
independent to a great extent from the pollution of the zone. ICP is Mason, B., Moore, C.B., 1982. Principles of Geochemistry, 4th Edition. John Wiley & Sons,
capable of differentiating better these zones of contact. XRF (probably New York. 344 pp.
Somogyi, A., Braun, M., Posta, J., 1997. Comparison between X-ray fluorescence and
due to the matrix effect) usually has problems at the moment of inductively coupled plasma atomic emission spectrometry in the analysis of
identifying these zones of contact and, sometimes, it selects one of the sediment samples. Spectrochim. Acta B 52, 2011–2017.