The authors have declared that no competing interests exist. The commercial affiliations do not alter our adherence to PLOS ONE policies on sharing data and materials.
The Tomb of the Diver has been subject for many decades of fierce debate among archaeologists and classicists. Since its discovery in 1968, some scholars have considered it a unique example of the lost tradition of Greek painting, others have emphasized Etruscan or Italic parallels. More recently, a possible local production has been suggested. With the aim of trying to solve the archaeological question, an archaeometric comparison among this well-known artwork and several frescoed tombs coming from Hellenistic and Lucan necropolis was carried out. The multi-analytical study was focused on the identification of peculiar features of executive techniques and raw materials since the first period of the archaeological site. The analytical investigation has been preliminary based on a non-destructive approach, performed in-situ by portable equipment including imaging diagnostics and compositional spectroscopic techniques for identifying pigments and the conservation state of original painted surface; subsequently, a further deepening by using destructive techniques was performed in-lab for the mortar-based supports characterization. Archaeometric study suggested that technological choices slightly changed in a time span of about two centuries, highlighting important markers that allow clustering the contemporary artistic productions. Moreover, a comparison with mortars from temples decorations was provided to better understand the whole artistic context. The archaeometric data showed that the Tomb of the Diver could be traced back to a local artisanal tradition and therefore is neither Etruscan nor Greek, but the first and foremost an expression of the local elite culture of Paestum.
The Tomb of the Diver is an exceptional painted tomb from the Greek colony of Paestum in southern Italy (
(a) Cover slab (215×112×20 cm; TFC). (b) southern slab (225×80×11 cm, TFS). (c) northern slab (244×80×11 cm, TFN).
Since its discovery in 1968, in a small necropolis, 1.5 kilometres south of the ancient city of Paestum, it has been subject of fierce debate among archaeologists and classicists [
(a) Cover slab (241×118×20 cm, TPS). (b) Eastern short slab (110×92×10 cm, TPC). (c) Northern slab (240×92×15 cm; TPL).
The present paper presents new archaeometric analysis that shed light on the question of how the Tomb of the Diver relates to other frescoed tombs and great Doric temples from Paestum, starting from the identification of peculiar characteristics of executive techniques and raw materials used in the artworks belonging to the different chronological phases at the Paestum site. In addition to the pigment palettes identification, the present study has examined the whole pictorial stratigraphy and the minero-petrographic features of the mortars from tombs and temples, in order to highlight characteristics of raw materials and painting executive techniques for each archaeological investigated phase.
To date previous scientific studies have been focused on the identification of the used pigments for the pictorial representations on a considerable number of painted tombs. In 1997 H. Brecoulaki analysed different samples of pigments and plasters taken from some tombs, to carry out a scientific study on colours, binders and materials used in Paestum, and in Italy in general, in the pre-roman period [
Recently, the archaeometric campaign has been carried out by Italian Association of Archaeometry (AIAr) in collaboration with the Archaeological Park of Paestum. Following an interdisciplinary and multi-analytical approach, the research activity has been focused on:
The following scientific methodologies was involved in the present project and carried out by portable equipment: Visible-induced infrared luminescence (VIL); X-Ray Fluorescence analysis (XRF); Fiber Optic Reflectance spectroscopy (FORS); Fourier Transform Infrared Spectroscopy in External Reflectance mode (ER-FTIR); Raman Spectroscopy (RS). Moreover, to guarantee the significance of the results, a preliminary mapping of the surfaces by diagnostics imaging techniques (Infrared reflectography (IRR), UV Fluorescence imaging) was carried out to localise pictorial areas or preparation layers altered by past restorations and to know the conservative interventions that may have changed the original phase of the studied tombs.
In the second phase of the project, in order to acquire the analytical information on the sampled fragments of the mortars the following techniques were employed: Optical Microscopy (OM); Electron Microprobe Analysis coupled with Energy Dispersive Spectroscopy (EMPA-EDS); Attenuated Total Reflection Fourier Transform Infrared Spectroscopy (ATR-FTIR); Simultaneous Thermal Analyses (STA) coupled with FTIR for Evolved Gas Analyses (EGA). Moreover, to guarantee the significance of the results, a preliminary mapping of the surfaces by diagnostics imaging techniques (Infrared reflectography, IRR; UV Fluorescence imaging) was carried out to localise the pictorial areas or preparation layers altered by past restorations and to know the conservative interventions that may have changed the original phase of the studied tombs.
This new integrated analytical approach attempts to provide technical and chemical markers attributable to the artistic production of the first period in order to verify from the archaeometric point of view the belonging of Tomb of the Diver and the coeval Tomb of the Palmettes to a still unexplored local artisanal tradition at the Greek city in 500–475 BC.
A multidisciplinary analysis has been carried out on thirteen painted tombs, displayed at museum or stored in warehouses (Figs
Tomb | Necropolis | Chronological phase |
---|---|---|
Tomb of the Diver | Tempa del Prete | Phase I |
Tomb of the Palmettes | Arcioni | Phase I |
Tomb 210 | Gaudo | Phase I |
Tomb 314 | Gaudo | Phase I |
Tomb 76 | Andriuolo | Phase II |
Tomb 109 | Santa Venera | Phase II |
Tomb 110 | Santa Venera | Phase II |
Tomb 6 | Andriuolo | Phase III |
Tomb 11 | Andriuolo | Phase III |
Tomb 12 | Andriuolo | Phase III |
Tomb 20 | Andriuolo | Phase III |
Tomb 21 | Andriuolo | Phase III |
Tomb 23 | Andriuolo | Phase III |
Neptune temple | - | 560–520 BC–Phase I |
- | 460 BC–Phase I |
Phase I (ca. 500 BC); Phase II (ca. 400 B.C); Phase III (ca. 300 BC)
The archaeometric study of painted slabs in the first step was based on a non-destructive approach, performed
Moreover, for a deeper characterisation of painted slabs and temples decorations, a second step based on micro-samplings of mortars was conducted, allowing unveiling technological information on the underlying support. An integrated analytical protocol by destructive investigation has been carried out on 30 samples (
ID sample | Tomb/Necropolis | OM | ATR-FTIR | EPMA-EDS | STA |
---|---|---|---|---|---|
TUF1 | Tomb of the Diver/ Tempa del Prete | * | * | ||
TUF2 | * | * | |||
TUF3 | * | * | |||
TUF4 | * | * | |||
PAL1 | Tomb of the Palmettes/ Arcioni 781 | * | * | * | * |
PAL2 | * | * | * | ||
PAL2R | * | * | |||
PAL2N | * | * | |||
T11C | Tomb 11/Andriuolo | * | * | * | |
T11L | * | * | * | ||
T12 | Tomb 12/Andriuolo | * | * | * | * |
T20 | Tomb 20/Andriuolo | * | * | * | |
T20_1 | * | * | |||
T21 | Tomb 21/Andriuolo | * | * | * | * |
T23 | Tomb 23/Andriuolo | * | * | ||
T76C | Tomb 76/Andriuolo | * | * | * | * |
T76L | * | * | * | * | |
T109C | Tomb 109/Santa Venera | * | * | * | * |
T110C | Tomb 110/Santa Venera | * | * | * | |
T110C_1 | * | * | |||
T210 | Tomb 210/Gaudo | * | * | * | * |
T210L | * | * | * | ||
T314C | Tomb 314/Gaudo | * | * | * | |
T314L | * | * | * | ||
TN1 | * | * | |||
TN2 | * | * | * | * | |
TN3 | Neptune temple | * | * | * | |
TN4 | * | * | * | ||
TN5 | * | * | * | * | |
BS1 | * | * | * | ||
BS2 | * | * | * | ||
BS3 | * |
Aimed to obtain a preliminary original and restoration pictorial materials mapping of painted surfaces, multispectral imaging analyses have been carried out by using a high-definition scientific CCD camera characterised by the following technical features: Spectral Sensitivity 0.8–1.1 micron; Quantum Efficiency @ 450, 550, 650, 800, 900, 1000, 1100 nm: 40, 55, 64, 32, 20, 5, 1%; Spatial Resolution 3072×2048 pixel; Reflectogram grey levels: 12 bit/pixel; Large pixel 9x9μm; Full Well Capacity: 100 ke; Dark Current: 0.5 e-/pixel sec; Fill Factor: 100%; Peltier Cooled DT = 40°. The acquisition system makes possible to quantify, store without ageing problems, form data bases and process: calibrate, correct for stray light, form thematic maps, compare, highlight details, etc. In particular, Calibrated UltraViolet fluorescence analysis, with specially filtered UV sources and multispectral acquisition technique, helps to identify the possible existence of inhomogeneity on the surface due to non-original materials, to assess the artwork conservation state and to reveal faded substances or other traces of possible interventions at later stages, guiding us to identify the original materials employed. Moreover, Infrared Reflectography survey allows the study of the deeper layers and in visualising, if existing, the presence or absence of an underdrawing carried out with carbonaceous matter and possibly the technique employed for it, such as tracing of the cardboard, dusting or using a grid.
Two flashes Quantum T5D mounted with B+W 486 UV/IR blocking filter were used for irradiating with visible light the surfaces of painted slabs, in order to explore the peculiar characteristics of some pigments to be luminescent in the infrared region when excited with visible light. The infrared emission was collected with a modified (built–in filter for IR removed) Canon EOS 400D (10.1 Mpixel, CMOS sensor) with Canon lens EFS 28 mm fitted with B+W 093 IR830 infrared filter to cut all stray radiation from visible spectrum and thus collecting only infrared luminescence emission. A white plate Spectralon® (WS-1S-L Labsphere certified standard) and a self-made mock up with Egyptian Blue were used as reference. In this context, such a technique represented a useful tool for the identification and localization of Egyptian Blue pigment [
Chemical investigations were performed by using a portable spectrometer consisting of a miniature X-ray tube system, which includes the X-ray tube (max voltage of 40 kV, max current of 0.2 mA, target Rh, collimator 1 or 2 mm), the power supply, the control electronics and the USB communication for remote control; a Silicon Drift Detector (SDD) with a 125 to 140 eV FWHM @ 5.9 keV Mn Kα line Energy Resolution (depends on peaking time and temperature); 1 keV to 40 keV Detection range of energy; max rate of counts to 5.6×105 cps; software for acquiring and processing the XRF spectra. Primary beam and detector axis form an angle of 0 and 40 degrees respectively with the perpendicular to the sample surface.
Tube voltage 35 kV, current 80 μA, acquisition time of 60 s, no filter was applied between the X-Ray tube and the sample, distance between sample and detector around 1 cm are the measurement parameters adopted for this study. The setup parameters were selected to have a good spectral signal and to optimize the signal to noise ratio (SNR).
Statistical treatment of XRF spectra has been carried out for identifying the main compositional clusters and evaluating eventual differences within the three chronological phases.
Raw data (namely the acquired XRF spectra) were calibrated in order to precisely assign the characteristic 956 emission lines (in KeV) of each chemical element. After the calibration, the data were treated for enhancing the most relevant peaks and removing the noise. We log-transformed each sample xj with j = 1,2,3…956 and got yj = logxj. The yj were smoothed with a moving average based on 10-konts with Gaussian weighting system. Then, we considered the residuals zj = xj-yj that are clean from the background noise.
To enhance the peaks, we considered the
Such transformation from original counts to maximum of the function was performed for each sample. The peaks in each sample where used to recognize the chemical elements. We restricted the analysis to those elements appearing in at least the 5% of the samples, and no more than 95%.
The final data-matrix (105 samples and 34 chemical elements) contain 0s and 1s, where the 1s means that the element is observed in the samples, while the 0 is the absence of the substance. From the data-matrix we compute the distance matrix with the simple-match distance, and then we performed a hierarchical clustering with the ward’s linkage.
Single spot analyses for each pigment were performed using Fibre Optics Reflectance Spectroscopy (FORS), a spectroscopic technique which analyses the light reflected from a surface illuminated with visible light.
FORS measurements were carried out in the spectral range 400–800 nm by using a tungsten lamp (20W) as source and the grating Ocean Optics (model HR2000) as detector. Optical fibre bundles were used both to drive the light on the surface under analysis and to collect the reflected radiation. The measuring head geometry was 45°/0°/45°. The probe, in contact with the surface, was a homemade fibre holder, which, at the same time, guarantees a soft contact and permits to fix the best distance from the surface. This allow to maximize the signal and to maintain the measuring area shielded from undesired external light. The analysed area was 2 mm
BRAVO Handheld Raman Spectrometer by Duo LASER was used to collect Raman spectra of the tombs in situ. BRAVO uses a patented technology (SSE™, Sequentially Shifted Excitation, patent number US8570507B1) [
In combination with Raman analysis, SERS spectroscopy has been used for better investigating the possible presence of protective films. This technique exploits the amplification of Raman scattering by molecules adsorbed on a surface of a noble metal.
Surfaces of painted slabs have been analysed at room temperature by means of a Bruker Optics Alpha-R portable spectrometer with an External Reflectance (ER) module for contactless and non-destructive analyses, covering a circular area of about 3 mm of diameter [
The instrument is equipped with a ROCKSOLID
Mortars were, instead, analysed by means of Attenuated Total Reflectance module (ATR), equipped with a diamond crystal, at same resolutions and spectral range, using 64 scans for each run.
The transmitted light optical microscopy investigations (OM), performed by using a Zeiss Axiolab associated with AxioCam MR for digital image acquisition, were aimed to identify the mineral-petrographic characteristics of the mortars, in terms of both binder and aggregate typology; to highlight the stratigraphic relationships between the layers; determine the conservation state of original materials and any contribute due to previous restoration.
Electron microprobe analysis (EMPA) coupled with energy-dispersive spectroscopy (EDS) were performed on 14 micro samples (
The EMPA images were acquired according to the following instrumental parameters: HV: 15 KeV; probe current: 10 nA; working distance: 11mm; Image: BSE–SE signal; detector image: Solid State detector (SSD), Everhart Thornley detector (SE); Image size: 2560 x 1920 pixel. Moreover, the EDS analyses were carried out according to the following measurement parameters: 15 keV HV; 10 nA probe current; 11mm working distance; 40° take off; and 30 seconds live time.
The measurements were performed after coating the sample with a thin and highly conductive film ultra-pure graphite (± 5 nm thickness, applied by Sputter—Carbon Coater QUORUM Q150T-ES, 70 A pulse current and 2.5 sec pulse time).
Thermogravimetric (TG) and Differential Scanning Calorimetric (DSC) analyses were simultaneously performed using a Netzsch STA 449 F3 Jupiter thermal analyser coupled with a Bruker Tensor 27 for Evolved Gas Analysis (EGA) in FTIR. Powdered samples (20–30 mg) were placed in alumina crucibles and heated from room temperature up to 1050°C (10°C/min heating rate) in ultra-pure air atmosphere. The FTIR spectra were acquired using 8 cm
In order to guarantee a correct identification of original pictorial materials, vibrational spectroscopic analyses (ER-FTIR, SERS) highlighted the spectral responses of the surface treatments due to previous restoration works and weathering products affecting the painted slabs.
ER-FTIR spectra, in fact, revealed the typical signatures of organic materials at ca. 2982, 2953, 2926, 2870 (C-H stretching vibration) and 1740 cm
T11: Tomb 11, TFN: Tomb of the Diver, northern slab, T23: Tomb 23. Infrared bands of organic compounds are reported in italics.
(a, c) Tomb of the Palmettes. (b) Tomb of the Diver.
It should be noted that on the Tomb T11 and on the Tomb of the Diver sharper peaks at ca. 2960, 2930 and 2850 cm
As alteration products, ER-FTIR and Raman spectroscopies detected the typical features of sulphates. Actually, on the tomb T11 infrared broad bands at ca. 2230 (2ν
a) XRF spectra of black degradation layers acquired on the pictorial surface of the painted tomb T110 enriched in manganese, compared with a representative spectrum of black pictorial layers (tomb T11); digital microscope images acquired on black layer of T110 (b) and T11 (c) respectively.
For each pigment hue, XRF, Raman, FORS and ER-FTIR analyses was carried out on original pictorial layers. Below the results obtained from the whole set of analytical techniques used for
The obtained results are supported by several studies carried out on the frescoes of other Campania archaeological sites, which confirm the use of calcite as the most widespread white pigment [
(a) VIL image of a northern slab detail of the Tomb of the Diver. (b) Corresponding visible image of the northern slab detail of the Tomb of the Diver. (c) VIL image of the green decoration of the Tomb 21. (d) Corresponding visible image of the green decoration of the Tomb 21.
In Campania region, attestations of Egyptian blue production have been found in Cuma and Pozzuoli [
(a) XRF spectra of green pigments identified as green earths. (b) XRF spectra of green pigments identified as a mixture of Egyptian blue and iron-rich earths. Digital microscope images of both typologies of green pigments are also reported.
The second typology of green pigments (
Moreover, a mixture of Egyptian blue with black pigments (carbon- and/or iron-based pigments) and iron-rich earths was already detected during previous analyses on painted slabs form Paestum [
Fe-As-Pb XRF intensity plot for red pigments in which three compositional groups can be distinguished (Group I: red pigments composed of iron oxides -Tombs 6, 11, 12, 20, 21, 23 and 110-; Group II: iron-based pigments containing arsenic traces -Tombs 76 and 210-; Group III: iron-based pigments containing arsenic and lead traces -Tomb 314, Tomb of the Palmettes and Tomb of the Diver-). (b) Representative XRF spectra (Tomb 23 for Group I, Tomb 210 for Group II, Tomb of the Diver for Group III).
Iron(III) oxide is the principal colouring matter in red ochre, the most widespread and used red pigment in the past [
A second compositional group was constituted by red pigment of the tombs T76 and T210; here, red hues show, along with iron and calcium, the presence of arsenic traces that suggests the use of different red pigments (
By contrast, the yellow layers analysed in the other tombs (Andriuolo T12 and T20 Tombs, Phase III) are instead made of iron-based pigments. XRF analyses showed the presence of iron along with calcium (
(a) XRF spectrum. (b) Raman spectrum. Tomb T12, short slab.
Interestingly, the same results have been obtained using a statistical approach that automatically compares the whole XRF spectrum of each measurement along with the age and the necropolis of provenance. The resulting dendrogram (
The investigation of technological changes in the considered time-span passed through the study of mortar-based supports, analysed by micro-destructive mineralogical and geochemical techniques. The sampling also involved coating mortars from the temples preserved in the archaeological site of Paestum, dated back to the Phase I, with the aim of evaluating eventual similarities with the decoration supports of painted tombs and that could suggest a standardization of production techniques.
From a macroscopic point of view, the analysed samples show high variability. With the exception of sample T21, all mortars show a multi-layered structure (at least two layers, namely
OM analyses showed that the samples of mortar-based supports of painted tombs are constituted of micritic and/or cryptocrystalline binders (
(a) Sample PAL2R, microscopic features of mortar that allowing recognizing two different layers, plane polarized light, 1x. (b) Sample T12, microcrystalline binder containing fragments of fossils shells, crossed polars, 1x. (c) Sample TN4, calcite-bearing aggregate, crossed polars, 5x. (d) Sample BS2, aggregate-bearing mortar, crossed polars, 1x. (e) Sample TN3, aggregate composed by marble dust, crossed polars, 5x. (f) Sample T109C, pictorial layer, crossed polarized light, 10x. (g) Sample T21, aggregate in the inner layer of the mortar, crossed polars, 5x.(h) Sample T23, aggregate in the inner layer of the mortar, crossed polars, 1x. (i) Sample T76, aggregate in the inner layer of the mortar, crossed polars, 1x.
(a) Fractured binder of the outer layer of sample PAL2. (b) Microcrystalline binder of sample T109. (c) Binder of sample T76L. (d) Outer area of gypsum accumulation in the aggregate-bearing mortar TN1. (e) External surface of sample TN2. (f) Gypsum crystals in a pore of sample TN5. Yellow boxes indicate the areas of EDS analysis.
In detail, two distinguishable layers of plaster were identified in different samples (TUF4, BS1, PAL1, PAL2R, PAL2N) thanks to the appearance of the binder, its compactness or fracturing degree (
Moreover in many samples (TUF1, TUF2, TUF3, TUF4, PAL2R, PAL2N, T11C, T11L, T12, T20; T109, T110, T210) traces of the overlying pictorial layer are visible; its thickness is variable from 12 to 60
The lime-rich nature of the samples is also confirmed by spectroscopic and thermal analyses. ATR-FTIR spectra (
The aggregate was observed only in the samples T21, T23 and T76L, in which it was added to the binder in the innermost layers of the mortar-based supports of the painting layer.
In T21 the aggregate is composed of a few crystals of quartz and plagioclase (
Thus, the presence of the aggregate fraction, as well as the typology of the fragments inside, permitted a distinction between the plasters that, however, lies outside from the chronological subdivision in the three phases but likely depends on the different functions that samples had. For samples T21, T23 (Phase III) and T76L (Phase II), in fact, a multi-layer technology was adopted, preparing at least two layers (an inner layer containing rounded aggregate and an outer layer only composed of binder); the addition of aggregate in the innermost layer, likely in contact with slab surfaces in travertine, was probably due the creation of a workable mixture that permitted a planar correction of the slab roughness whereas the outermost layer only composed of binder was likely needed for creating a white supporting medium for the painting [
The plasters of the slabs placed as decoration of the tombs and the coating mortars of the temples show similar characteristics with no clear changes relative at the three chronological phases. The only criterion that can be partially applied is the presence of the aggregate fraction in some samples and its absence in others, which could suggest both a different origin of the raw materials or another kind of executive technique.
EMPA-EDS microanalysis provided morphological and compositional information on the binder. The first suggests how the binder is strongly recrystallized and degraded in many areas; moreover, common phenomena of secondary calcite formation were observed in the pores and in the fractures as previously in OM (
The chemical data are reported in Tables
Hydraulicity Index HI = (Al2O3 + SiO2) / (CaO+MgO). Legend type: (A) aerial lime; (AH) average hydraulic lime.
Period | Tomb | ID | Na2O | MgO | Al2O3 | SiO2 | ClO | CaO | Total | Al2O3 + SiO2 | CaO + MgO | HI | Type |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
TUF1_01 | - | 0.07 | 0.25 | 1.02 | 0.78 | 97.89 | 100.00 | 1.27 | 97.96 | 0.01 | A | ||
TUF1_02 | - | 0.28 | 0.37 | 0.78 | 0.75 | 97.82 | 100.00 | 1.15 | 98.10 | 0.01 | A | ||
TUF2_01 | - | 0.58 | 0.63 | 1.14 | - | 97.65 | 100.00 | 1.77 | 98.23 | 0.02 | A | ||
TUF2_02 | - | 0.62 | - | 1.22 | - | 98.17 | 100.00 | 1.22 | 98.79 | 0.01 | A | ||
TUF3_01 | - | 0.25 | 0.31 | 0.51 | 0.77 | 98.15 | 100.00 | 0.82 | 98.40 | 0.01 | A | ||
TUF3_02 | - | 0.23 | 0.35 | 0.57 | 1.04 | 97.81 | 100.00 | 0.92 | 98.04 | 0.01 | A | ||
TUF4_01 | - | 0.52 | 0.33 | 1.15 | 1.63 | 96.37 | 100.00 | 1.48 | 96.89 | 0.02 | A | ||
TUF4_02 | - | 0.78 | 0.19 | 1.43 | 1.36 | 96.25 | 100.00 | 1.62 | 97.03 | 0.02 | A | ||
PAL1_1 | - | 0.26 | 0.36 | 0.41 | 0.50 | 98.47 | 100.00 | 0.77 | 98.73 | 0.01 | A | ||
PAL1_2 | - | 0.33 | 0.22 | 0.36 | 0.51 | 98.58 | 100.00 | 0.58 | 98.91 | 0.01 | A | ||
PAL2_1 | - | 0.91 | - | 1.00 | 0.88 | 97.21 | 100.00 | 1.00 | 98.12 | 0.01 | A | ||
PAL2_2 | 0.09 | 0.97 | - | 1.73 | 0.88 | 96.33 | 100.00 | 1.73 | 97.30 | 0.02 | A | ||
PAL2N_1 | 0.72 | 1.33 | 0.19 | 0.97 | 0.79 | 96.00 | 100.00 | 1.16 | 97.33 | 0.01 | A | ||
PAL2N_2 | 0.53 | 1.03 | 0.30 | 0.96 | 1.64 | 95.54 | 100.00 | 1.26 | 96.57 | 0.01 | A | ||
PAL2R_1 | 0.54 | 0.15 | 0.56 | 0.55 | 3.24 | 94.96 | 100.00 | 1.11 | 95.11 | 0.01 | A | ||
PAL2R_2 | - | 0.36 | 0.69 | 1.28 | 2.25 | 95.42 | 100.00 | 1.97 | 95.78 | 0.02 | A | ||
T210_1 | 0.24 | 0.45 | 2.05 | 12.42 | 0.99 | 83.85 | 100.00 | 14.47 | 84.30 | 0.17 | AH | ||
T210_2 | 0.12 | 0.57 | 2.84 | 12.16 | 0.67 | 83.64 | 100.00 | 15.00 | 84.21 | 0.18 | AH | ||
T76C_1 | 0.05 | 1.58 | 0.95 | 5.83 | 0.45 | 91.14 | 100.00 | 6.78 | 92.72 | 0.07 | A | ||
T76C_2 | 0.05 | 1.65 | 0.64 | 4.39 | 0.47 | 92.81 | 100.00 | 5.03 | 94.46 | 0.05 | A | ||
T76L_1 | 0.43 | 1.08 | 0.47 | 3.48 | 2.26 | 92.28 | 100.00 | 3.95 | 93.36 | 0.04 | A | ||
T76L_2 | 0.49 | 1.31 | 0.81 | 4.81 | 2.21 | 90.38 | 100.00 | 5.62 | 91.69 | 0.06 | A | ||
T109_1 | - | 1.36 | 0.54 | 4.41 | - | 93.70 | 100.00 | 4.95 | 95.06 | 0.05 | A | ||
T109_2 | - | 1.71 | 0.83 | 5.03 | - | 92.43 | 100.00 | 5.86 | 94.14 | 0.06 | A | ||
T12_1 | - | 0.76 | 1.27 | 18.34 | 1.45 | 78.18 | 100.00 | 19.61 | 78.94 | 0.25 | AH | ||
T12_2 | - | 0.66 | 1.22 | 19.65 | 1.41 | 77.06 | 100.00 | 20.87 | 77.72 | 0.27 | AH | ||
T21_1 | 0.23 | 1.10 | 4.09 | 11.69 | 1.57 | 81.32 | 100.00 | 15.78 | 82.42 | 0.19 | AH | ||
T21_2 | 0.25 | 1.27 | 4.07 | 11.70 | 1.83 | 80.88 | 100.00 | 15.77 | 82.15 | 0.19 | AH | ||
T23_1 | 0.22 | 0.40 | 0.21 | 16.16 | 0.50 | 82.51 | 100.00 | 16.37 | 82.91 | 0.20 | AH | ||
T23_2 | 0.35 | 0.60 | 1.23 | 12.61 | 1.10 | 84.11 | 100.00 | 13.84 | 84.71 | 0.16 | AH |
Hydraulicity Index HI = (Al2O3 + SiO2) / (CaO+MgO). Legend type: (A) aerial lime; (AH) average hydraulic lime.
Period | Tomb | ID | Na2O | MgO | Al2O3 | SiO2 | P2O5 | SO3 | ClO | K2O | CaO | Total | Al2O3 + SiO2 | CaO + MgO | HI | Type |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
TN1_02 | - | 2.59 | - | 1.37 | 3.27 | 0.56 | 2.60 | - | 89.61 | 100.00 | 1.37 | 92.20 | 0.01 | A | ||
TN1_03 | 0.90 | 1.92 | - | 0.85 | 1.77 | 0.64 | 3.49 | - | 90.43 | 100.00 | 0.85 | 92.35 | 0.01 | A | ||
TN2_01 | 0.56 | 2.18 | 0.16 | 1.06 | 0.19 | 0.36 | 1.78 | 0.28 | 93.43 | 100.00 | 1.22 | 95.61 | 0.01 | A | ||
TN2_02 | 1.11 | 1.55 | 0.34 | 1.24 | 0.37 | 0.79 | 1.93 | 0.58 | 92.09 | 100.00 | 1.58 | 93.64 | 0.02 | A | ||
TN2_03 | 0.98 | 1.58 | 0.30 | 1.52 | 0.19 | 0.63 | 3.60 | 0.38 | 90.82 | 100.00 | 1.82 | 92.40 | 0.02 | A | ||
TN2_04 | 0.66 | 1.65 | 0.21 | 1.67 | 0.25 | 0.61 | 4.37 | 0.31 | 90.27 | 100.00 | 1.88 | 91.92 | 0.02 | A | ||
TN2_05 | 0.96 | 1.38 | 0.29 | 1.17 | 0.38 | 0.33 | 6.13 | 0.49 | 88.87 | 100.00 | 1.46 | 90.25 | 0.02 | A | ||
TN5_01 | - | 1.14 | - | 0.61 | - | 0.34 | 3.91 | - | 94.00 | 100.00 | 0.61 | 95.14 | 0.01 | A |
Compositional data suggest how the binder consists almost exclusively of calcite (CaCO
The coating mortars from the Neptune Temple, also dated back to Phase I, show a certain compositional affinity with the binders of the tombs of the same phase with an average amount of CaO around 90%, and although higher contents of MgO occur (until 2.59 wt%;
Moreover, it should be noted that mortars from temples are characterized by the presence of ClO, P
The Hydraulicity Index (HI) (Tables
The chemical analysis and the HI of the binders suggest the probable use of two type of limestone for the preparation of the samples, namely pure and marly limestone. Even though it is important to consider the advanced process of recrystallization of the binder in the samples and their scarce state of conservation that could influence the original composition.
Conversely, compositional data collected by the analysis on the binders and, especially the determination of HI, suggested a certain variability over the time: the plasters of Phases I (Tomb of the Diver, Tomb of the Palmettes, Neptune Temple) and Phase II (Tomb 76 and Tomb 109) were made with aerial lime, whereas Tomb 210 and the tombs of the Phase III (Tomb 12, Tomb 21, Tomb 23) with average hydraulic lime. This could probably suggest a difference in the raw materials used, respectively a purer and a marly limestone.
As detected by EMPA-EDS analyses, coating mortars from temples are characterised by the presence of ClO P
The multi-analytical techniques used for the characterisation of the painted tomb slabs and temples from the archaeological site of Paestum revealed significant information on raw materials and technological features. Red and green identified pictorial mixtures have proven to be important markers for a systematic comparative analysis used pigments in the three different chronological phases. Moreover, the typologies of mortars, allowed differentiating the executive techniques, highlighting the change in choices and technological skills of artisans since the first period of the frescoed tombs tradition at Greek colony.
The results shed light on the decorative techniques and pigments adopted from the Hellenistic to Lucan period. In particular, interesting similarities in red and grey-green layers chemical composition and mortars technical features have been highlighted between the Tomb of the Diver and the Tomb of the Palmettes, which represented one of the main archaeological issues asked by the archaeologists supporting the belonging of these two coeval tombs to a still unexplored local artisanal tradition at the Greek city in 500–475 BC.
Valuable information has been obtained by the compositional data of pigments that permitted the identification of the palette used for decorating the slabs. The use of calcite, red and yellow ochres, green earths, Egyptian blue and black carbon was revealed by
The use of the Cu-based pigment in the green pictorial layers of the third period (post 300 BC) suggests a greater availability of this precious synthetic pigment starting from this time. The main evidence is related to the high similarity, in terms of pigment palette, between the Tomb of Diver and Tomb of the Palmettes, which shown similar chemical composition for each analysed colour. However, archaeometric evidences highlighted that they are quite different from the painted tombs of the Lucan period, supporting the archaeological hypothesis based on a stylistic point of view.
In order to better understand if such similarities can be also observed in the manufacturing of underlying plasters, analyses on micro-samplings (e.g. OM, EMPA-EDS, TG/DSC-FTIR(EGA), ATR-FTIR) have been carried out. Their features, as well as those of other public buildings (e.g.
Results of analyses on micro-sampling of the tombs shed light on the stratigraphic structure (from pictorial to preparation layers and supports) and mineralogical composition of samples, suggesting different technological choices used for the execution of the painted slabs, in particular during the Phase III.
The plasters of the tombs and of the temples show similar characteristics, highlighting no clear changes relative during the three chronological phases. The only criterion that can be partially applied is the presence of the aggregate fraction in some samples and its absence in others, which could suggest both a different origin of the raw materials or another kind a different of executive technique, depending on the different functions. Finally, a changing in the production technology of mortar-based supports has been verified thank to the hydraulicity index evaluation. The plasters of Phases I and Phase II were made with aerial lime, whereas the tombs of the Phase III (and the mortars of the tomb in addition to the T210 mortars, Phase II) with average hydraulic lime. This could probably suggest a difference in the raw materials used, respectively a purer and a marly limestone.
The archaeometric data suggest that the Tomb of the Diver could belong to a local artisanal tradition and therefore is neither Etruscan nor Greek, but first and foremost an expression of the local elite culture of Paestum. Furthermore, the data suggests that the expertise and technology used for the frescoed tombs originated in the field of temple architecture. Moreover, the analysis of plaster samples from the temples and the tombs suggests that the workshops were closely connected, or even the same.
Images of the tombs from Andriuolo, Gaudo and Santa Venera necropolis analysed within the research project.
(PDF)
Dendrogram obtained by statistical treatment of intensity XRF signal values, showing at least seven groups. Groups 1 and 2 mainly cluster white pigments (likely composed by predominant Ca and lower Al, Si, Fe and Sr) and black hues (likely here clustered since the impossibility of detecting carbon). The other colours here observed (red, yellow and green) thus contain the same elements (i.e. iron-rich earth pigments). The Group 3 is constituted of pigments of different tombs from Gaudo, Andriuolo and Santa Venera necropolis, characterised by the presence of external alteration black patinas containing manganese, which likely allowed isolating such a group. The Group 4 is composed of the Tomb of the Diver and the Tomb of the Palmette, confirming the similarities in the compositional features of pigments used for decorating them. Groups 5a and 6a are likely formed by pigments containing copper whereas the for the remaining groups (5b, 6b and 7a) a unique relation with the compositional Groups cannot be identified since the ubiquitous presence of chemical elements contained in the restoration products that likely influenced more than the constitutive elements of matrices and/or pigments.
(PDF)
Representative ATR-FTIR spectra and TG-DSC-DTG curves of samples PAL2 (a,b), T76L (c,d) and TN2 (e,f).
(PDF)
Infrared peaks obtained by ATR-FTIR analyses and vibrational assignments.
(PDF)
Weight-losses and enthalpy changes of analysed samples by STA analyses. Negative peaks observed on derivative thermogravimetric curves (DTG) are also reported.
(PDF)
The authors would like to thank the whole staff of the Archaeological Museum of Paestum for their collaboration during