Radiation Physics and Chemistry 95 (2014) 65–68

Contents lists available at ScienceDirect

Radiation Physics and Chemistry

0969-80

http://d

n Corr

E-m

journal homepage: www.elsevier.com/locate/radphyschem

Mineralogical analysis of clays in hardsetting soil horizons, by X-rayfluorescence and X-ray diffraction using Rietveld method

L.V. Prandel a,n, S.C. Saab a, A.M. Brinatti a, N.F.B. Giarola b, W.C. Leite a, F.A.M. Cassaro a

a Laboratory of Soil Physics and Environmental Sciences, Department of Physics, State University of Ponta Grossa (UEPG), Av. Carlos Cavalcanti 4748, 84030-

900 Ponta Grossa, PR, Brazilb Department of Soil Science and Agricultural Engineering, State University of Ponta Grossa (UEPG); Av. Gal. Carlos Cavalcanti, 4748, 84030-900 Ponta

Grossa, PR, Brazil

H I G H L I G H T S

c Elemental composition of soil samples through X-Ray fluorescence.c Mineralogical quantification through X-ray diffraction and Rietveld method.c Oxisol and Ultisol, Brazil ‘Barreiras’ formation.c High amounts of Si and Al oxides and low amounts of Fe and Ti oxides.c Predominance of kaolinite in the clay fraction.

a r t i c l e i n f o

Article history:

Received 29 September 2012

Accepted 8 December 2012Available online 25 December 2012

Keywords:

Oxisols

Ultisols

Cohesion

Kaolinite

6X/$ - see front matter & 2012 Elsevier Ltd. A

x.doi.org/10.1016/j.radphyschem.2012.12.017

esponding author. Tel.: þ55 42 3220 3044; fa

ail address: luisprandel@hotmail.com (L.V. Pr

a b s t r a c t

Diffraction and spectroscopic techniques have been shown to be suitable for obtaining physical and

mineralogical properties in polycrystalline soil samples, and also in their precursor compounds. For

instance, the X-ray fluorescence (XRF) spectroscopy allows obtaining the elemental composition of an

investigated sample, while the X-ray diffraction (XRD) technique permits obtaining qualitative and

quantitative composition of the soil minerals through the Rietveld method (RM). In this study Yellow

Latosol (Oxisol), Yellow Argisol (Ultisol) and Gray Argisol (Ultisol) soil samples, classified as ‘‘hardsetting

soils’’, extracted from areas located at Northeast and Southeast of Brazilian coast were investigated. The

soils and their fractions were analyzed in an EDX-700 and an XRD-6000 (Cu Ka radiation). XRF results

indicate high percentages of Si and Al, and small percentage of Fe and Ti in the investigated samples. The

DRX data and RM indicate that there was a predominance of kaolinite and halloysite minerals (kaolin

group minerals) in the clay fractions, which are presumably responsible for the formation of kaolinitic

plasma in these soils. Also, the obtained results showed that the XRF, XRD techniques and RM were very

helpful for investigating the mineralogical composition of a hardsetting soil.

& 2012 Elsevier Ltd. All rights reserved.

1. Introduction

According to Du et al. (2008), soil is a complex mixture ofseveral compounds, and due to this complexity in its composition,for the study of physical, chemical and mineralogical properties ofthe soil, which are of great importance for agriculture andenvironment, the use of spectroscopic and diffraction techniquesis very relevant.

Currently, X-ray diffraction (XRD) and the Rietveld method(RM) (Rietveld, 1967, 1969) data is used in the mineral character-ization and quantification of Brazilian soils (Brinatti et al., 2010;

ll rights reserved.

x: þ55 42 3220 3042.

andel).

Alves et al., 2007). These results, together with X-ray fluorescencedata (XRF), provide information on the mineral and chemicalcomposition of complex mixtures found in soils (Pantenburget al., 1992).

The soils to be analyzed are of a hardsetting character, a termwhich is used in Brazil to describe sub-superficial mineralhorizons (B) of soils which present significant increase in cohe-sion amongst its particles, becoming hard when dry and friablewhen humid (Lima Neto et al., 2009). Such soils were incorpo-rated to the current Brazilian System of Soil Classification asYellow Latosol (Oxisol), Yellow Argisol (Ultisol) and Gray Argisol(Ultisol) (Embrapa, 1999), and are part of the advanced process ofweathering of tertiary sediments in the Barreiras formation (LimaNeto et al., 2009), which comprises the whole Southeastern andNortheastern coast of Brazil.

L.V. Prandel et al. / Radiation Physics and Chemistry 95 (2014) 65–6866

These soil compact horizons, when dry, hamper the plant rootdevelopment, regarding water and air content available; they arephysical inhibitors, associated to high aluminum saturation (LimaNeto et al., 2009), they also present low iron content, and quartzgrains surrounded mainly by a dense and continuous kaoliniticplasma (Giarola et al., 2009; Melo et al., 2002).

Studies carried out on the hardsetting character of soils arestill rudimentary due to lack of defined parameters which mightindicate the different cohesion degrees (Giarola et al., 2009). It isbelieved that some cohesion parameters will be explainedthrough physic-chemical and mineralogical analyses of these soilsand their fractions.

2. Experimental methods

Five samples of soils considered to be of a hardsetting characterwere collected and classified (Table 1). Around 20 g of each samplewith average density (2.6070.03) g/cm3 were mashed, submitted toa dispersion and sedimentation physical fractioning process, and thefractions coarse sand (cs: 53–1000 mm), fine sand (fs: 20–53 mm), silt(sl: 2–20 mm), and clay (cl:r2 mm) were extracted.

All samples were reduced to diameters lower than 53 mmand analyzed in a dispersive energy XRF spectrometer (EDX-700)Shimadzu, with an Rh source. The spectra were obtained at 300 stime intervals [in Na–Sc (15 kV) and Ti–U (50 kV) energy bands],in vacuum, semi-quantitative mode. The samples were alsoexamined by XRD, in an XRD-6000 diffractometer Rigaku (modelUltima IV), with Cu Ka radiation, as a function of the 2y angle (8 to801, with 0.021 step, and 5 s time per step).

The RM refinement was carried out in the general structureanalysis system (GSAS) software, developed by Larson and VonDreele (2004), accessed through the EXPGUI interface, developedby Toby (2001). The GSAS uses the RM as a refining tool based onthe calculation of square minimums until it reaches the bestadjustment between the intensity of the diffraction pattern

Table 1Samples of Brazilian hardsetting soils and their fractions [coarse sand (cs), fine sand (f

Sample Collect place Frac. Percentage (%)

(1) Yellow argisol Coruripe/Alagoas

(101070 3300S 361 1003300W)

cs 61.5371.33

fs 2.3970.08

sl 2.2470.55

cl 33.7070.25

Whole soil

(2) Yellow latosol Cruz das Almas/Bahia

(1214001200S 3910600700W)

cs 64.1770.33

fs 2.8870.42

sl 0.8970.07

cl 32.0970.31

Whole soil

(3) Yellow argisol Porto Seguro/Bahia

(1612700000S 3910305400W)

cs 52.2870.49

fs 2.3570.15

sl 1.6470.23

cl 43.8870.32

Soil whole

(4) Yellow argisol Aracruz/Espırito Santo

(1914901200S 4011602200W)

cs 52.8870.30

fs 2.2170.06

sl 1.5570.09

cl 43.8670.08

Soil whole

(5) Gray argisol Pacajus/Ceara

(0411002200S 3812703900W)

cs 67.1970.80

fs 3.8470.09

sl 0.6070.18

cl 28.3770.31

Soil whole

n Ca oxide was found only in the clay fraction: (1) 12.972.1; (2) 15.772.0; (3) 14

observed, yobs, and the calculated pattern, ycalc:

Sy ¼X

i

wi yobs�ycalc

� �2, ð1Þ

where wi(¼ 1=ffiffiffiffiffiffiffiffiyobsp

) is the function weight in the ith step (Young,2002; Rietveld, 1967, 1969).

Intensities calculated for each phase (p), with starting crystal-line structures obtained from Downs and Hall-Wallace (1993)data basis, are represented through Eq. (2).

ycalc ¼ AbSR

X

p

Sp

X

K

½LK9FK92f 2yi�2yKð ÞASPK �Pþybi, ð2Þ

where Ab is the absorption factor; SR a function which models thesurface roughness effect; Sp the scale factor; K is Miller index toeach Bragg reflection; LK contains the Lorentz, polarization andmultiplicity factors; f is the diffraction profile function given bythe Pseudo-Voigt function; yi is the scattering angle; yK is thediffraction angle in the Bragg reflecting lattice plane; AS is theasymmetry function; PK is the preferential orientation function;9FK9

2 is the square of the structure factor and ybi is the backgroundradiation intensity (Young, 2002).

The refinement quality is indicated through the minimizationand convergence of terms R-pattern (Rp), R-weighted pattern (Rwp ),R-expected (Re), goodness-of-fit (S¼Rwp/Re) and R-structure factor(RF

2) (Young, 2002; Larson and Von Dreele, 2004).The procedure adopted for the soil samples RM refinement

was: (i) simultaneous and individual adjustment of Sp; (ii)individual refinement of each Sp, 9FK9

2 and function ybi; (iii)individual refinement of some f parameters of each phase (U, V

and W of FWHM), (iv) adjustment of PK by March–Dollasefunction for kaolinite, in (0 0 1) and (0 0 2) planes, and quartz,in (0 1 0) and (0 0 1) planes; (v) simultaneous adjustments of Sp,Ab, Sr and peak anisotropy terms (Sijk); (vi) PK refinement throughspherical harmonics; and (vii) finalization with adjustment of Sp,

s), silt (sl), and clay (cl)]: percentage, texture and XRF semiquantitative results.

Texture Oxide (g/kg)*

SiO2 Al2O3 Fe2O3 TiO2

Sandy clay loam 973.671.2 – 3.470.1 7.570.1

926.671.9 – 15.171.2 28.170.3

582.678.3 317.275.7 35.472.7 48.170.4

388.871.8 533.373.5 32.570.8 22.170.4

444.174.4 492.377.0 28.970.6 22.670.4

Sandy clay loam 969.971.2 – 5.770.1 7.170.1

924.172.6 – 20.570.5 30.870.9

884.275.1 – 45.972.5 49.373.8

373.473.4 507.970.2 76.771.3 16.270.1

427.873.9 466.176.5 74.371.5 18.770.5

Sandy clay 981.371.3 – 2.270.1 4.470.1

915.774.0 – 8.870.5 33.171.3

615.871.9 277.972.9 29.872.0 51.476.7

361.175.2 543.776.2 36.571.2 24.070.7

403.075.3 524.976.2 34.870.7 26.170.5

Sandy clay 95.671.2 – 2.270.1 –

914.375.1 – 17.872.8 33.972.0

752.572.8 121.872.9 42.771.5 59.072.7

378.372.0 514.274.7 54.371.7 24.970.6

414.472.6 489.473.0 54.870.3 27.470.5

Sandy clay loam 98.171.2 – – –

951.872.8 – 7.670.8 11.970.6

937.679.3 – 15.675.8 18.073.2

403.873.0 506.876.6 25.170.4 11.870.2

471.873.2 477.874.0 22.17 0.6 11.370.3

.670.5; (4) 17.772.5; (5) 27.976.2.

Fig. 1. Refinements obtained using the RM (a)–(e). (f) Quantification of minerals. In the figure: Intensity is presented in counts per second (c.p.s.). yobs states for the

observed profile, ycalc states for the calculated profile, the yobs–ycalc represents the residual curve, and the Bragg peak positions for each mineral are indicated in vertical

lines. The refinement quality is indicated by the terms R-pattern (Rp), R-weighted pattern (Rwp ), R-expected (Re), goodness-of-fit (S) and R-structure factor (RF2). Mineral

abbreviations: Kaolinite (Ka); Goethite (Gt); Anatase (An); Gibbsite (Gb); Halloysite (Ha); Quartz (Qz); Rutile (Rt); Calcite (Ct) and Muscovite (Mu). (For interpretation of

the references to color in this figure legend, the reader is referred to the web version of this article.)

L.V. Prandel et al. / Radiation Physics and Chemistry 95 (2014) 65–68 67

9FK92 and atomic positions of majority phases and ybi function.

Some parameters, obtained through the refinement of a CeO2

standard sample were not refined: As, sample displacement, someterms of f, and 2y�Zero.

3. Results and discussion

The results of XRF analyses and soil samples and their fractions(Table 1) reveal Si oxide, present in all fractions, predominantly in

L.V. Prandel et al. / Radiation Physics and Chemistry 95 (2014) 65–6868

sand samples. Al oxide is only seen in clay and silt fractions.Finally, the percentage of Fe oxide, both in whole samples andtheir fractions, is higher in clays (in lower amounts whencompared to the Si and Al oxides). Ti, Ca, S and K oxide amountsare within the limits of detection by the equipment.

Preliminary analysis of diffractograms obtained from the XRDin continuous mode revealed high intensity peaks in (1 0 1) and(1 0 0) planes, referring to Quartz (SiO2), in sand and silt fractions(in 2y¼20.811 and 2y¼26.631). Likewise, there was increase inintensity in (0 0 1) and (0 0 2) planes, regarding the Kaolinite[Al2Si2O5(OH)4], in clay and silt (in 2y¼12.231 and 2y¼24.781).Other peaks appear in the silt and clay fractions, with lowerintensity though and correspond to the following minerals:Halloysite [Al2Si2O5(OH)4], Gibbsite [g-Al(OH)3], Anatase andRutile (TiO2), Goethite [a-FeO(OH)], Calcite (CaCO3) and Musco-vite [KAl2(Si3Al)O10(OH,F)2)].

In Fig. 1 are shown the refinements using the RM in the clayfraction samples: Yellow Argisol (Fig. 1a), Yellow Latosol (Fig. 1b),Yellow Argisol (Fig. 1c), Yellow Argisol (Fig. 1d) and Gray Argisol(Fig. 1d). The obtained adjustment indices varied from: 1.73 to2.21, to S; from 1.72 to 4.45%, to RF

2; from 5.71 to 7.43%, to Rwp;and from 4.42 to 5.74%, to Rp. The quantification of minerals,presented in a percentage of weight basis, in the clay fractionsamples are shown in the Fig. 1f. In average, it has been observedhigh amounts of Kaolinite (58.0070.11%) and Halloysite(28.2570.21%) in the investigated clay fraction.

Thus, most Si and Al oxides, found in the XRF, are in theKaolinite and Halloysite minerals, the remaining ones are in theQuartz and Gibbsite. Basically, all Fe oxide composes the Goethitephase. Significant amounts of Ti oxide are present in Anatase andRutile. All these characteristics are common to soils in advancedweathering processes (Giarola et al., 2009). The Ca oxide found inthe clay fraction were attributed to the presence of Calcitemineral. However, it should be better investigated as the presenceof calcite minerals in very weathered environments is quiteimprobable.

4. Conclusion

Obtained results shown that the XRF, XRD and RM techniqueswere efficient in investigating mineralogical properties of soilswith hardsetting characteristics. Considering that the samples didnot receive any pre treatment, i.e., soil samples were investigatedin their natural condition, the Rietveld method (RM) presentedgood indexes of quality of refinement. More specifically: theparameter S varied from 1.73 to 2.21; RF

2 varied from 1.72 to4.45%; Rwp varied from 5.71 to 7.43%, and Rp varied from 4.42 to

5.74%. The obtained quantification showed that the goethiteappears in larger amounts in Yellow Latosol as compared to theYellow Argisols. Nonetheless, this last mineral was not detected inthe Gray Argisol. This is an expected result as soils containinggoethite present a yellow color. Finally, kaolinite and halloysite(kaolin group minerals) were presumably responsible for theformation of kaolinitic plasma in the investigated soils, as theseminerals were predominant in the investigated soil samples, inaverage, approximately 86% of kaolinite and halloysite, in com-parison with 14% of others minerals.

Acknowledgements

The authors acknowledge ‘‘Coordenac- ~ao de Aperfeic-oamentode Pessoal de Nıvel Superior’’ (CAPES) for the PhD Scholarship andthe Department of Physics (UEPG) for the support and financialassistance.

References

Alves, M.E., Mascarenhas, Y.P., French, D.H., Vaz, C.P.M., 2007. Rietveld-basedmineralogical quantitation of deferrified oxisol clays. Soil Res. 45, 224–232.

Brinatti, A.M., Mascarenhas, Y.P., Pereira, V.P., Partiti, C.S.M., Macedo, A., 2010.Mineralogical characterization of a highly-weathered soil by the Rietveldmethod. Sci. Agric. 67, 454–464.

Downs, R.T., Hall-Wallace, M., 1993. American mineralogist crystal structuredatabase. Am. Mineral. 88, 247–250.

Du, C., Linker, R., Shaviv, D., 2008. Identification of agricultural mediterranean soilsusing mid infrared photoacoustic spectroscopy. Geoderma 143, 85–90.

Empresa Brasileira de Pesquisa Agropecuaria, (EMBRAPA), 1999. Brazilian SoilClassification System. Embrapa Informac- ~ao Tecnologica, Brasılia.

Giarola, N.F.B., Lima, H.V., Romero, R.E., Brinatti, A.M., Silva, A.P., 2009. Crystal-lography and mineralogy of the clay fraction of hardsetting horizons in soils ofcoastal tablelands in Brazil. R. Bras. Ci. Solo. 33, 33–40.

Larson, A.C., Von Dreele, R.B., 2004. GSAS—general structure analysis system. LosAlamos Nat. Lab., 224.

Lima Neto, J.A., Ribeiro, M.R.R., Correa, M.M., Souza Junior, V.S., Lima, J.F.W.F.,Ferreira, R.F.A.L., 2009. Characterization and genesis of cohesive horizons inoxisols and ultisols of the coastal tablelands of Alagoas state, Brazil. R. Bras. Ci.Solo. 33, 1001–1011.

Melo, V.F., Schaefer, C.E.G.R., Singh, B., Novais, R.F., Fontes, M.P.F., 2002. Chemicaland crystal properties of kaolinite and iron oxides in the Barreiras groupsediments in Aracruz, Espırito Santo state Brazil. R. Bras. Ci. Solo 26, 53–64.

Pantenburg, F.J., Beier, T., Hennrich, F., Mommsen, H., 1992. The fundamentalparameter method applied to X-ray fluorescence analysis with synchrotronradiation. Nucl. Instrum. Methods Phys. Res., Sect. B 68, 125–132.

Rietveld, H.M., 1969. A profile refinement method for nuclear and magneticstructures. J. Appl. Crystallogr. 2, 65–71.

Rietveld, H.M., 1967. Line profiles of neutron powder-diffraction peaks forstructure refinement. Acta Crystallogr. 22, 151–152.

Toby, B.H., 2001. EXPGUI, a graphical user interface for GSAS. J. Appl. Cryst. 34,210–213.

Young, R.A., 2002. The Rietveld Method. Int. Union Crystallogr. newsl. OxfordUniversity Press, New York, p. 298.