Vibrational Spectroscopy 97 (2018) 140–145

Contents lists available at ScienceDirect

Vibrational Spectroscopy
journal homepage: www.elsevier.com/locate/vibspec

Archaeological pottery from Nasca culture studied by Raman and Mössbauer T

spectroscopy combined with X-ray diffraction
⁎
H.B. Lucas, H.J.A. Silva, C.M.S. Tasayco, P. Munayco, J.L.B. Faria
Universidade Federal do Mato Grosso, Cuiabá, MT, Brazil

A R T I C LE I N FO A B S T R A C T

Keywords: In the present paper, we report on a study of archaeological fragments from Nasca ceramics using Raman and
Nasca pottery Mössbauer spectroscopy combined with X-ray diffraction (XRD). By combining results obtained by these
Firing temperature methods it is possible to quantitatively determine the paints composition, temperature and environment during
Pigments the firing. The samples were collected from the Ceremonial Centre of Cahuachi in Southern coast of Peru. Raman
Raman spectroscopy
spectroscopy allows us to determine the composition of the different pigments used in the preparation of Nasca
Mössbauer spectroscopy
ceramic. The results show that the composition of the white pigments is formed by rutile and anatase while the
black and red pigments are formed by amorphous carbon and hematite, respectively. The Mössbauer spectra
were measured at room temperature (RT) and show the presence of components associated with Fe3+ indicating
an oxidizing environment during the manufacturing process of the ceramic. The analysis is complemented by
data obtained by X-ray diffraction suggesting firing temperatures around 950 °C, in agreement with Raman
measurements.

1. Introduction
Recently the Raman and Mössbauer spectroscopy has become a
standard method for the characterization of archaeological materials.
One of its main applications is in the study of ceramic sherds: the
Raman spectroscopy can be determined of the composition of the pig-
ments; the effect of firing on clay can be studied by following the
transformations of the components in the Mössbauer spectra attributed
to the different iron-containing phases. 57Fe Mössbauer and Raman
studies [1,2] combined with other techniques such as X-ray diffraction
(XRD), mineralogical and chemical analysis, can help to gather a rather
complete picture of the method of production of the pottery. This in-
formation, combined with the one classically provided by archae-
ologists, such as details of excavation, stratigraphic, stylistic and ty-
pological dating, among others, is very important to establish the
technological level of the culture that produced the pottery. Here we
report on a study of the Nasca ceramics. The Nasca Culture emerged
during the Early Intermediate Period (100 B.C. to 650 A.D.) and was
centered in the Ica and Nasca valleys of south coastal Peru. The Nasca
people practiced intensive agriculture in a precarious area character-
ized by frequent droughts, earthquakes and flash flooding. The Nasca
ceramic is distinguished by the use of polychrome slip paints applied to
both effigy vessels as well as a broad range of utilitarian shapes [3]. The
Nasca iconography is characterized by a wide variety of naturalistic as
well as fantastic motifs, some of which are clearly recognizable as birds,
fish, plants, and others portraying strange creatures exhibiting both
human and animal characteristics [3]. These motifs were painted on a
variety of different shapes: bowls, jars, cups, and bottles with double
spouts connected by a flat handle. There are also modeled vessels in the
form of humans, plants, animals, and the same mysterious creatures
seen in the paint [4,3]. The firing of the ceramics Nasca was done in
open air, and not in kilns. The aim of the Mössbauer study of ceramic
materials is to obtain information on the firing conditions (temperature
and environment). The Mössbauer spectra yield characteristic para-
meters, which make it possible to infer the oxidation state of the iron
and the symmetry of its environment. All information one can draw
from Mössbauer spectra of ceramics depends on the chemical varia-
bility of iron, which may occur as Fe2+ or Fe3+ in a variety of silicate
and oxide phases, depending on the redox conditions during the firing
[5,6]. Raman spectroscopy is used to study ceramic manufactures.
Raman spectroscopy is a spectroscopic technique that allows non-de-
structive analysis, to obtain qualitative information about the in-
vestigated samples. It is used to study the composition of the colored
pigments used in ceramics. Although the technique is widely known
and used, investigation of ceramics with Raman spectroscopy is a
challenge due to its weak effect and the heterogeneity of the ceramic
ware. With complementary methods, such as X-ray diffraction, non-iron
bearing minerals and oxides (e.g. hematite, magnetite and wüstite)

⁎
Corresponding author.
E-mail address: hulk@fisica.ufmt.br (J.L.B. Faria).

https://doi.org/10.1016/j.vibspec.2018.06.010
Received 26 April 2018; Received in revised form 29 May 2018; Accepted 17 June 2018
Available online 19 June 2018
0924-2031/ © 2018 Elsevier B.V. All rights reserved.
H.B. Lucas et al.
Fig. 1. Representative samples (C1 and C2) collected from the Ceremonial
Centre of Cahuachi.
present in ceramics can often be identified, which allows conclusions as
to the raw materials used and may give further clues as to the firing
conditions.
2. Materials and methods
The fragments of archaeological Nasca Ceramic were collected from
the Ceremonial Centre of Cahuachi in Southern coast of Peru. Fig. 1
Fig. 2. Raman spectrum obtained on the ceramic painting with predominant white in section 1 of the selected area.
141
Vibrational Spectroscopy 97 (2018) 140–145
shows two representative fragments samples. The samples are analyzed
by Raman and Mössbauer spectroscopy combined with X-ray diffrac-
tion. By combining results obtained by these methods it is possible to
quantitatively determine the paints composition, firing temperature
and environment during the firing.
The Raman spectroscopy is performed using a model LabRam
HR800 spectrometer (Horiba Inc.) to obtain the Raman spectra, with
objective lens of magnification 50× and 633 nm excitation line (He-Ne
laser), and 15 mW power output. We performed 7 scans with 3 s of
exposure for each measure in the range 80–1750 cm−1. To prevent
heating of the spot area has been selected optical density 0.6 filter for
darker pigments. The spectra were collected by a CCD cooled to −70 °C
(Peltier system). The Mössbauer spectra of the samples were recorded
in transmission geometry at room temperature (RT) in a high velocity
range (12 mm/s), using a 57 Co(Rh) source and a spectrometer with
512 channels. The drive velocity was calibrated with the same source
and a metallic iron foil at RT. The average recording time was 12 h per
sample. Mössbauer absorbers containing approximately 125 mg/cm2 of
the powdered ceramic material were used. Isomer shifts are given re-
lative to α-Fe. The NORMOS code [7] was used for the spectrum ana-
lysis and fitting with appropriate superposition of Lorentzian lines
grouped into quadrupole doublets and magnetic sextet.
The X-ray diffraction measurements were carried out on powder
samples using a Bruker D8 Discover Advance diffractometer with Cu X-
ray tube. The diffraction patterns were measured in steps of 0.05° with
accounting time of 3 s for each step. The analysis of diffraction patterns
is performed using the MAUD software [8] with a pseudo-Voigt func-
tion to describe the peaks and a Caglioti function to describe the full
width at half maximum (FWHM), as a function of the angle.
3. Results and discussion
3.1. Raman spectroscopy
White pigment – Fig. 2 shows the spectrum collected from the
white pigment of the sample C1 in one of the predominantly white
areas. The most intense vibrational modes, 443 and 611 cm−1, are as-
sociated with modes found in rutile ore (TiO2, tetragonal). A sharp peak
at 146 cm−1 is due to anatase (rutile polymorph) mode and 236 cm−1
band is associated with a non-ordinary Raman effect (second order
scattering). Rutile is a mineral from which up to 70% of titanium oxide
H.B. Lucas et al.
Fig. 3. Raman spectrum obtained on the ceramic painting with predominant white in section 2 of the selected area.
Fig. 4. Raman spectrum on black pigment (straight line), in comparison with typical coal spectrum (dotted line).
is extracted. Due to its high brightness and high refractive index, tita-
nium oxide are widely used in white pigments [9]. However, its oc-
currence is less than the white calcite, commonly found in other cave
paintings. The anatase occurs naturally in kaolin clays in several places
around the world, e.g. China [10,6], where this substance is found in
ancient arts of Chinese porcelain dating from 1300 BC. Anatase (in the
stable form of high-temperature rutile) converts into titanium oxide
(IV) at about 850 °C. Therefore, this polymorphs can be used to de-
termine the firing temperature of ceramics [11].
In other sections of the same area with white pigmentation (Fig. 3),
a different spectrum was obtained, which was associated with gypsum
or hydrated calcium sulfate (CaSO4 · 2H2O). SO4−2 ions play an im-
portant role in gypsum spectra, contributing with factor group. Here are
predicted 9 modes for SO4 (ΓSO vib
4
= A1 + E + 2F2 , Td point group) and
four typical vibrations (ν1 to ν4). The characteristic vibrational mode at
1009 cm−1 is associated with the tetrahedral formed by the symmetric
142
Vibrational Spectroscopy 97 (2018) 140–145
stretching (ν1) from SO4 group, while vibrational bands in 403 and
496 cm−1 are associated with internal modes of sulfate (ν2). The band
in 643 cm−1 is labeled as Bg mode in (ν4). Finally, ν3 mode is located in
1137 cm−1 [12]. In this case, the gypsum can be used as the basis for
the manufacture of pigment used in rock painting, as is reported in the
literature [13].
Black pigment – In particular, the darker pigment on the painting
had special attention due to the effects of laser thermal dispersion.
Fig. 4 shows the spectrum obtained over the region of ceramic with
black paint compared to the typical spectrum of coal. We can observe
the concordance of the main modes of bands of amorphous carbon. This
is consistent since the provision of charred plant material is abundant,
especially for people who have control over the use of fire. Vibrational
modes in 1311 cm−1 and 1557 cm−1 are bands associated with the D
and G normal modes in graphite respectively [14]. The D and G bands
intensities ratio can be used to determine amorphization stages, and
H.B. Lucas et al.
Fig. 5. Comparison of Raman signatures with red hematite pigment.
Fig. 6. Representative Mössbauer spectra collected at RT for samples C1 and
C2.
therefore, to deduce the origin of black pigment [15].
Red pigment – Raman spectroscopy on the red painted surface
revealed the likely substance that makes up the pigment. By comparing
the vibrational modes of the sample with iron oxide (hematite), it is
possible to emphasize the similarity of the spectral signatures (Fig. 5).
The use of red pigment based on iron oxide was used in rock art, ac-
cording to the literature [13]. The occurrence of iron oxide is very
common and is not inconsistent the fact find this substance in ancestral
143
Vibrational Spectroscopy 97 (2018) 140–145
paintings. Vibrational modes 227, 296, 411 and 612 cm−1 are con-
clusive to say that our standard samples are containing hematite [16].
3.2. Mössbauer spectroscopy
57
Fe Mössbauer spectroscopy is an excellent microscopic method in
the investigation of clay-based ceramics since practically all pottery
clays contain iron, usually in concentrations between 1 and 10 wt%,
enough to easily obtain 57Fe Mössbauer spectra. The changes which
pottery clays undergo during firing are reflected in the Mössbauer
spectra of the fired ceramics and hence can be used, even after mil-
lennia of burial, to obtain information on the original firing conditions
and on techniques and abilities of the ancient potters. Mössbauer
spectroscopy can also be applied in the classification of pottery on the
basis of their Mössbauer patterns. The different types of Mössbauer
spectra, which can be associated to different firing procedures, were
defined based on the presence and intensity or the absence of certain
components, mainly in the RT Mössbauer spectra. Typical Mössbauer
spectra at room temperature of the two representative samples are
shown in Fig. 6. The hyperfine parameters for each sample can be found
in Table 1.
Samples C1 and C2 are dominated by the presence of quadrupole
doublets associated with the presence of Fe3+. The fitting of the sam-
ples C1 was made with two different quadrupole doublets associated
with the presence of Fe2+, which are characterized by their isomer shift
and a relatively large quadrupole splitting, and a quadrupole doublet
associated with the presence of Fe3+, whose quadrupole splitting of
QS = 0.85 mm/s is characteristic of clay minerals or iron oxides in a
superparamagnetic state. Additionally one magnetic components asso-
ciated with hematite is observed. The simultaneous presence of both
Fe3+ and Fe2+ indicates a non-equilibrium state between oxidizing and
reducing firing conditions that may have arisen from a change in the
atmosphere. The strong Fe3+ presence indicates that the cooling oc-
curred in an oxidizing environment. The spectra of samples C2 are
fitting with one quadrupole doublet associated with the presence of
Fe3+ and do not show any quadrupole doublets attributable to Fe2+.
Additionally one magnetic components associated with hematite is
observed. The absence of Fe2+ suggests an oxidizing environment
during the firing of the ceramics. Previous work has shown that the
quadrupole splitting of the paramagnetic Fe3+ component at ambient
temperature can be used for determining the firing temperature by
comparing it with the values found in firing clays in the laboratory.
H.B. Lucas et al. Vibrational Spectroscopy 97 (2018) 140–145

Table 1 exhibits, besides quartz, feldspars, amphibole, layer silicates and the
Hyperfine parameters obtained at room temperature (IS: isomer shift; QS: presence of mica and the second one (samples C2) without mica. Mica
quadrupole splitting; Bhf: hyperfine field) and relative area of the different Fe usually disappears on firing at about 950 °C, suggesting that the max-
species. imum firing temperature was certainly below 950 °C for sample C1, and
Species Parameters Sample C1 Sample C2 higher for samples C2. This is in accordance with Raman analysis,
where we have observed the occurrence of anatase phase in rutile. The
3+
Fe ISO (mm/s) 0.28 0.26
XRD measurements are performed in the internal and external part of
QS (mm/s) 0.85 0.78
A (%) 61.0 50.0
the samples, in order to observe some difference in the mineralogical
composition as consequence of the manufacturing process. No differ-
Fe2+ ISO (mm/s) 0.97
QS (mm/s) 2.48 —
ences are observed in the measurements of internal and external parts
A (%) 14.0 of samples, which would indicate a uniform temperature in the man-
ufacturing process. In contrast to Raman spectroscopy, the presence of
Hematite ISO (mm/s) 0.21 0.26
QS (mm/s) −0.11 −0.17 Fe oxides was not observed by XRD, iron oxides often unnoticed by X-
Bhf (T) 50.0 49.5 ray diffraction due to their poor crystallinity and small particle size.
A (%) 25.0 50.0

Master curves show that up to 400 °C there is no change in the QS.
Between 400 °C and 650 °C it can be observed that QS increases from
0.5 mm/s to values that may be as high as 1.4 mm/s. In the temperature
range of 650 °C to 800 °C there is a further small variation in the
splitting. Above 800 °C QS decreases again. The average value of the
quadrupole splitting (QS) of Fe3+ species obtained in the fitting of our
samples was compared to the master curves reported in the literature
[17,18] for clays fired in laboratory under oxidizing conditions (firing
in air). The estimated firing temperature for samples C1 and C2 is ap-
proximately between 850 °C and 950 °C. These results are in agreement
with the X-ray results shown below.
3.3. X-ray diffraction
X-ray diffraction (XRD) is an important tool in mineralogy for
identifying, quantifying and characterizing minerals in complex mi-
neral assemblages. Its application to ancient ceramics, which are a
mixture of clay minerals, additive minerals and their transformation
products, yields information on the mineral composition of objects.
Details of production processes, like firing conditions as well as appli-
cations of slips or glazes may thus become transparent. The XRD dif-
fractograms, shown in Fig. 7, exhibit two main patterns. Sample C1
4. Conclusions
The Raman spectroscopy allowed to determine the composition of
the pigments used in the manufacture of Nasca ceramics. The white
pigments exhibit in their composition rutile and anatase, however black
pigments show the presence of amorphous carbon possibly obtained
carbonized vegetable material, and red pigments are associated with
the presence of hematite. According to the type of Mössbauer patterns it
was possible obtain information to the firing conditions. Thus, asso-
ciating these information with XRD results we can suggest that the
firing occurred under reducing conditions and that the ceramics were
partially (sample C1) or completely re-oxidized (sample C2) later in the
firing cycle. This is in agreement with the color (red) observed in the
samples. The firing temperatures is estimated were lower than 950 °C
for sample C1 and higher for sample C2 which suggests different
manufacturing processes in these samples. This firing temperature is in
agreement with the presence of Anatase observed by Raman spectro-
scopy. In the future, more samples will be analyzed in order to obtain
additional information about the firing technique used in making Nasca
pottery. Measurements of Mössbauer spectroscopy will be performed in
appropriate fired clay and with ceramic samples refired under con-
trolled conditions in the laboratory.

Fig. 7. X-ray diffraction for samples C1 and C2.

144
H.B. Lucas et al.
Acknowledgements
We acknowledge “Laboratório Multiusuário de Técnicas Analíticas”
(LAMUTA/ UFMT–Cuiabá-MT–Brazil) and Project FINEP #
01.04.0121.00 for the XRD measurements. H.B. Lucas is grateful to
Fundação de Amparo à Pesquisa do Estado de Mato Grosso (FAPEMAT)
and Propeq/UFMT for his fellowship.
References
[1] L. Stievano, M. Bertelle, S. Calogero, Application of 57Fe Mössbauer spectroscopy
for the characterisation of materials of archaeological interest: the work performed
in Italy, Hyperfine Interact. 150 (1–4) (2003) 13–31.
[2] J. Van Pevenage, P. Vandenabeele, Raman spectroscopy and the study of ceramic
manufactures. Possibilities, results and challenges, Archaeological Ceramic
Analysis, Oxford University Press, 2017.
[3] D.A. Proulx, A Source Book of Nasca Iconography, University of Iowa Press, Iowa
City, Iowa, USA, 2006.
[4] H. Silverman, D.A. Proulx, The Nazca. Peoples of the America Series, Blackwell
Publishers, Oxford, 2002.
[5] P. Munayco, R. Scorzelli, Characterization of the firing conditions of archaeological
Marajoara pottery by Mössbauer spectroscopy and X-ray diffraction, Hyperfine
Interact. 222 (1) (2013) 69–75.
[6] R. Scorzelli, L. Bertolino, A. Luz, M. Duttine, F. Silva, P. Munayco, Spectroscopic
studies of kaolin from different Brazilian regions, Clay Miner. 43 (1) (2008)
145
Vibrational Spectroscopy 97 (2018) 140–145
129–135.
[7] R. Brand, Normos mössbauer fitting program, Universität Duisburg, Duisburg
(Germany), 2007.
[8] L. Lutterotti, Total pattern fitting for the combined size–strain–stress–texture de-
termination in thin film diffraction, Nucl. Instrum. Methods Phys. Res. Sect. B:
Beam Interact. Mater. Atoms 268 (3–4) (2010) 334–340.
[9] J. Winkler, Titanium Dioxide, European Coatings Literature, Vincentz Network,
Hannover, 2003.
[10] E. Murad, H. Köster, Determination of the Ti speciation in commercial kaolins by
Raman spectroscopy, Clay Miner. 34 (3) (1999) 479–485.
[11] H. Edwards, IR and Raman Spectroscopies, The Study of Art Works, Elsevier, 2017.
[12] B. Berenblut, P. Dawson, G. Wilkinson, The Raman spectrum of gypsum,
Spectrochim. Acta Part A: Mol. Spectrosc. 27 (9) (1971) 1849–1863.
[13] Á.V. Vialou, Pré-história do Mato Grosso: Santa Elina vol. 1, EdUSP, 2005.
[14] A.C. Ferrari, Raman spectroscopy of graphene and graphite: disorder, electron-
phonon coupling, doping and nonadiabatic effects, Solid State Commun. 143 (1-2)
(2007) 47–57.
[15] A.C. Ferrari, J. Robertson, Raman spectroscopy of amorphous, nanostructured,
diamond-like carbon, and nanodiamond, Philos. Trans. R. Soc. Lond. A: Math. Phys.
Eng. Sci. 362 (1824) (2004) 2477–2512.
[16] D. De Faria, S. Venâncio Silva, M. De Oliveira, Raman microspectroscopy of some
iron oxides and oxyhydroxides, J. Raman Spectrosc. 28 (11) (1997) 873–878.
[17] R. Gebhard, R. Bott, N. Distler, J. Michálek, J. Riederer, F. Wagner, U. Wagner,
Ceramics from the Celtic Oppidum of Manching and its influence in central Europe,
Hyperfine Interact. 154 (1–4) (2004) 199–214.
[18] I. Shimada, W. Häusler, T. Hutzelmann, U. Wagner, Early pottery making in
northern coastal Peru. Part I: Mössbauer study of clays, Hyperfine Interact. 150
(1–4) (2003) 73–89.