I. INTRODUCTION
Natural earths have been used as colour pigments since prehistoric times. They have been detected in works of art everywhere and in any historical period probably due to their availability, high colouring capacities, and stabilities to light and a variety of weather conditions. Ochres are natural earth pigments varying from yellow to red and brown shades. The colour shade of ochre depends on the type of the iron oxide chromophore. The red ochres contain mainly haematite (Fe2O3), while the yellowish ochres are rich in hydrated iron oxide (goethite, FeO·OH) (Cornell and Schwertmann, Reference Cornell and Schwertmann1996; Helwig, Reference Helwig1997, Reference Helwig and Berrie2007; Fiore et al., Reference Fiore, Maier, Parera, Orquera and Piana2008).
Most often the red ochre was detected as a main component of rock paintings and human funerals (Mortimore et al., Reference Mortimore, Marshall, Almond, Hollins and Matthews2004; Bikiaris et al., Reference Bikiaris, Daniilia, Sotiropoulou, Katsimbiri, Pavlidou, Moutsatsou and Chryssoulakis2000). Red colour plays an important role in human behavior. Studies and opinions on the mechanisms of colour preference can be found in such fields as anthropology, psychology, and linguistics (Leach, Reference Leach1976). The presence of other minerals, such as clay minerals or metal oxides, can also influence the colour of the ochres. It is well known that clay materials and iron oxides were adopted as mineral pigments along the history (Hradil et al., Reference Hradil, Grygar, Hradilová and Bezdička2003). They were used for ethnic or social marks expressed as rock art, burial ceremonies, paintings or engravings, and also body paintings or exhibition of adornments or sumptuary objects such as engraved plates. The marks with colour, generally red, were not hidden but, on the contrary, there are evidences they were intentionally shown (Gradín et al., Reference Gradín, Aguerre and Albornoz2003; Albornoz, Reference Albornoz and Otero1996). To a large extent, red pigments were selected for rock paintings (Hajduk et al., Reference Hajduk, Albornoz, Lezcano, Civalero, Fernández and Guráieb2004; Podestá, Reference Podestá, Bahn and Fossati2003; Wainwright et al., Reference Wainwright, Helwig, Podestá and Bellelli2000). Red pigments were also found colouring valves of molluscs (British) or mollusks (American) (Diplodon patagonicus) intentionally or unintentionally (Parada and Peredo, Reference Parada and Peredo2008; Trubitt, Reference Trubitt2003; Bar-Yosef Mayer, Reference Bar-Yosef Mayer and Gopher1997).
Mineral pigment colours varying from yellow to red are given by the presence of different iron oxyhydroxides and oxides, mainly goethite and haematite.
Figure 1. Map of the site location (by Mabel Fernández).
Figure 2. (Color online) Flat view of Carriqueo shelter (by Luis Teira).
Their structural and mineralogical compositions are related with their natural genesis and provenance. Compositional characterization helps archaeologists and anthropologists to reconstruct ancient society life style as well as to hypothesize the group’s peculiar mobility and to infer different uses.
The purposes of this research were to determine the mineralogical compositions of prehistoric pigments from excavated layers collected at the Carriqueo rock shelter archaeological site and to compare them with a suspected provenance source, La Oficina. Carriqueo shelter (S 40°37′27″; W 70°31′42″; W ) is located on the west side of La Oficina creek, a tributary of the Limay river, Pilcaniyeu area, in the Río Negro province (Figure 1) (Palacios and Ramos, Reference Palacios and Ramos2009; Crivelli Montero et al., Reference Crivelli Montero, Cordero, Palacios and Ramos2007).
From an archaeological point of view, Carriqueo is a small specialized site with a doubtless hunting activity and, to a less extent, a sedentary living site (Crivelli Montero et al., Reference Crivelli Montero, Cordero, Palacios and Ramos2007). This area is also archaeologically related with other sites such as La Divisoria, a chert lithic manufacturing area located 350 m away on the other side of the creek. Several surface sites that seem to have been active at the same time in the neighborhoods had been detected. Radiocarbons analysis [14C] reveals two dates: 2620±110 BPand 610±50 BP. From November to December 2006, the archaeological Carriqueo shelter site was investigated (Crivelli Montero et al., Reference Crivelli Montero, Cordero, Palacios and Ramos2007). A flat view of this site is shown in Figure 2. During the excavations, several pigment samples, deposited in layers in the sedimentary pile, were collected.
Figure 3. (Color online) La Oficina site, suspected source of red pigments.
Figure 4. (Color online) Pigment sample sizes.
The environs of the Carriqueo shelter were also prospected discovering, 200 m away, a probable source of pigments: two blocks of raw material for red pigment production. The site was denominated La Oficina. Figure 3 shows a partial view of 0.80 m3 of the La Oficina rock, the suspected main pigment source.
With the aim to look for chemical compositional analogies between Carriqueo and La Oficina sites, a set of 34 specimens was analyzed. Synchrotron radiation X-ray diffraction (SR-XRD) was selected as the instrumental technique used in this study (Bugoi et al., Reference Bugoi, Constantinescu, Pantos and Popovici2008; Welcomme et al., Reference Welcomme, Walter, Bleuet, Hodeau, Dooryhee, Martinetto and Menu2007; Calza et al. Reference Calza, Anjos, Mendonca de Souza, Bracaglion and Lopes2008; Sánchez del Río et al., Reference Sánchez del Río, Gutiérrez-León, Castro, Rubio-Zuazo, Solís, Sánchez-Hernández, Robles-Camacho and Rojas-Gaytán2008). Remarkable reasons are the higher flux which improves the signal to noise ratio and the polarization of the incident beam in relation to the sample, increasing in this way the sensitivity. The detection limit by conventional laboratory X-ray diffraction is in the range of a few weight percent, limiting the required amount of sample. SR-XRD improved the detection limit allowing the analyses of minute samples. Elemental analysis is widely applied for this kind of study. However, an appreciable amount of sample is required for most of the analytical techniques. In addition, phase identification is mandatory since the colour could depend not only on the element amount but also on the crystalline composition of the pigment (Dillmann et al., Reference Dillmann, Neff, Mazaudier, Hoerle, Chevallier and Beranger2002; Artioli, Reference Artioli2008; Simova et al., Reference Simova, Bezdicka, Hradilova, Hradil and Grygar2005). Then, the use of X-ray fluorescence, inductively coupled plasma, atomic emission spectrometry, and neutron activation analysis is limited. It is expected that SR-XRD would provide valuable information about the phase composition without compromising the integrity of the archaeological information.
II. EXPERIMENTAL
A. Sampling and sample preparation
Carriqueo samples were obtained during excavation at different levels. Figure 4 shows three examples of the analyzed samples.
Figure 5. (Color online) Rotating capillary containing the pigment sample in the X-ray path.
TABLE I. SRDRX results, site, stratigraphic level, stratum, and identified colour according to Munsell chart for the 34 analyzed samples.
As shown in this figure, the amount of samples is extremely small, less than 100 mg in all the cases. Two representative samples from La Oficina were taken for the comparison. An agate mortar and pestle were used to crush and grind each specimen into powders to ensure better homogeneity. The powders were passed through a plastic sieve with mesh of 0.074 mm and carefully introduced in a 0.3 mm diameter Mark-Rohrchen boron capillary sample holder; this procedure minimises preferential orientation (Piszora et al., Reference Piszora, Nawrocki, Darul, Nowicki and Evans2008). The use of the capillary method allows the use of small amounts of sample material and preserves the sample for further microanalysis. Finally, a set of 34 samples was measured.
B. XRD Instrument
The obtained samples were characterized in the D12A-XRD1 beamline at the Brazilian Synchrotron Light Laboratory (LNLS). The beamline is equipped with a three-circle Huber diffractometer (θ,2θ,φ), which permits the rotation of a sample in three axes (Figure 5) (Lima et al., Reference Lima, Barroso, Braz, Droppa, Oliveira and Lopes2007).
Figure 6. Synchrotron powder diffraction pattern of sample 6 corresponding to group I showing the presence of haematite.
The energy of the beam was 7.100 keV and the intensities were monitored with a detector perpendicularly located. The scan speed was 0.02° min−1 for the 2θ range of 10 to 70°.
C. Analysis
After the beamline was adjusted, the capillary containing the sample was mounted and located in the X-ray path. This operation was controlled using an external screen. Once the samples were measured, the collected data were analyzed using the CRYSTALLOGRAPHICA free software package (Oxford Cryosystems, Inc., 2007).
III. RESULTS AND DISCUSSION
Table I shows the SR-XRD results as well as the site, stratigraphic level, stratum, and identified colour according to the Munsell chart (Munsell Soil Colour Charts, 1994) for the 34 analyzed samples. In order to group the samples, the presence of haematite (Htt, α-Fe2O3) was used as one of the references. Following this criterion, four groups were formed:
Figure 7. Synchrotron powder diffraction pattern of La Oficina creek sample corresponding to group I showing the presence of haematite.
Group II shows the presence of goethite (JPCDS, ICDD, No. 29-0713) associated with kaolinite (K) and quartz. The samples exhibit high crystallinity, similar to those of group I (Figure 8).
Figure 8. Synchrotron powder diffraction pattern of sample 8 corresponding to group II showing the presence of goethite.
I. Presence of Htt corresponding to samples 2 to 6, 10 to 17, 20 to 21, and 24 to 34;
II. Presence of goethite (Gtt, FeOOH) and no Htt corresponding to samples 8, 9, 18, and 19;
III. Nonreddish samples corresponding to samples 7 and 23; and
IV. Reddish samples but no chromophores detected corresponding to samples 1 and 22 (Q, quartz).
In order to assess the validity of the classification based on the presence or absence of haematite, the following remarks based on the SR-XRD results can be made.
Group I shows similar diffraction patterns: the peaks of quartz (JPCDS, ICDD, Card No. 46-1045) and haematite (JPCDS, ICDD, Card No. 33-0664) are intense and well defined indicating good crystallinity. The presence of haematite is associated in some samples also with a zeolite. Natural zeolites could come from volcanic ashes and crystallize in postdepositional environments over large periods of time in marine basins, which are the case of Los Andes mountains (Bouza et al. Reference Bouza, Simón, Aguilar, del Valle and Rostagno2007) (Figure 6).
Figure 9. Synchrotron powder diffraction pattern of sample 23 corresponding to group III with nonreddish evidences.
Figure 10. Synchrotron powder diffraction pattern of sample 22 corresponding to group IV with no evidences of chromophore substances.
It must be noted that La Oficina samples belong to group I, as the majority of the red coloured samples, and have similar diffraction patterns regarding the phases and crystallinity (see Figures 6 and 7).
Only two green samples correspond to group III. The phase found in both samples was mica (Mi). The green-grey colour leads to associate samples 7 and 23 with green earths. There are several green earth mica minerals such as celadonite, glauconite, etc., which have been used as green pigments since ancient periods (Murray, Reference Murray2000). Identification of green earth mica minerals is complicated because of the similarities in their powder patterns (Tamburini et al., Reference Tamburini, Adatte, Föllmi, Prell, Wang, Blum, Rea and Clemens2003). In this case the results show the presence of mica but they are inconclusive about the nature of the source of the green colour (Figure 9). Finally, group IV corresponding to reddish samples presents the sharp peaks of quartz but no evidence of red pigments. This could be the result of the presence of a noncrystalline pigment or of its low concentration (Figure 10).
IV. CONCLUSIONS
It has been demonstrated the potential of SR-XRD in the archaeometry field and its suitability as an analytical technique for provenance studies. In this work, it was successfully employed for explaining chemical compositional analogies between samples coming from Carriqueo and La Oficina sites. In this way, it could be concluded that La Oficina site seems to be the provenance source for red pigment samples found at Carriqueo archaeological site. Although four different groups have been identified, the provenance of them would still need to be investigated using complementary techniques such as microanalysis by total-reflection X-ray fluorescence spectrometry. In terms of archaeology, the importance of this work is to confirm that the human groups in the surroundings of Limay river had a strategy of supplying of raw material based on the use of the closest resource. This research helps us to increase the understanding of the human group mobility in the North Patagonia area of Argentina using pigment information as an archaeological evidence. Data published in this research contribute to interested archaeologists and archaeometrists as a reference for further studies.
ACKNOWLEDGMENTS
This work has been supported by the Laboratório Nacional de Luz Síncrotron (LNLS), Campinas, Brazil, under Proposal Nos. D12A XRD1 6558/08 and IAEA 13864, and the following Argentinean Projects, under Grant Nos. PIP 5344, UBACYT F059 y I809, and PICT14171. The authors also thank the Council for Scientific, Technological and Humanistic Research of the Universidad Centro Occidental Lisandro Alvarado, Barquisimeto, Venezuela.
