Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy

G Model CULHER-3601; No. of Pages 9 ARTICLE IN PRESS Journal of Cultural Heritage xxx (2019) xxx–xxx Available online at ScienceDirect www.sciencedirect.com Original article Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy Amal Khedr a , Hamada Sadek b , Olodia Aied Nassef a , Mahmoud Abdelhamid a , Mohamed Abdel Harith a,∗ a b National Institute of Laser Enhanced Science, Cairo University, Cairo, Egypt Faculty of Archaeology, Conservation Department, Fayoum University, Al Fayoum, Egypt a r t i c l e i n f o Article history: Received 7 February 2019 Accepted 15 May 2019 Available online xxx Keywords: Ancient Egypt Ushabtis Genuine and fake LIBS SEM-EDX a b s t r a c t The potential of laser-induced breakdown spectroscopy (LIBS) in differentiating between types of genuine and fake Ushabtis was studied. For this goal, the present study was initiated by a comprehensive spectrochemical analytical study of four archeological Ushabti statues, dated back to the late Ptolemaic period (332–330 B.C), which have been excavated south of El-Shawaf pyramid in Saqqara, Giza, Egypt. This study was followed by performing the same investigations on several imitated Ushabtis, which were purchased from the touristic marketplace in the region of the pyramids. For the archeological samples, optimization of LIBS experimental parameters in terms of temporal detection conditions along with laser wavelengths and their corresponding ablation rate was performed. The spectrochemical analyses of the genuine archaeological samples were carried out both superficially and stratigraphically. Main surface elements such as Ca, Na, Al, Si, Cu, Fe, and Mn were detected while large amounts of Ca and Sr represent their bulk elements. The LIBS results were validated by results obtained by SEM-EDX technique, which provided complementary information such as surface morphological images and quantitative composition of the samples’ constituents. Under the optimum experimental conditions, LIBS spectral emission of the imitated samples was acquired and compared to those for true archeological ones for the sake of possible discrimination between them. Successful utilization of LIBS in examining the authenticity of suspect Ushabtis has been accomplished and presented. This study may suggest the need for archeologists to be equipped with portable LIBS systems in the excavation sites or museums, which facilitate the immediate discrimination process when LIBS data of genuine samples are initially available. © 2019 Elsevier Masson SAS. All rights reserved. 1. Introduction In ancient Egyptian royal tombs, numerous funerary figurines are typically found. Such figurines, called Ushabtis or Shabtis, are supposed to act as servants or minions for the deceased person [1]. The ancient Egyptians believed that Ushabtis would answer for their master and perform the daily life routine chores for the deceased, which gods required from him in the afterlife. One Ushabti has to serve for one day of the year; hence, usually, around 360 Ushabtis are found in the tomb. As Shabtis represent the most numerous of all ancient Egyptian antiquities to survive, they also represent the most commonly displayed ancient Egyptian collections in museums of cultural heritage worldwide. ∗ Corresponding author. Tel.:/Fax: +202 35675335. E-mail address: mharithm@niles.edu.eg (M.A. Harith). Colors play a substantial role in ancient Egyptian art and belief. Blue, for example, refer to life symbolized by the river Nile and heaven. On the other hand, the green could exemplify vegetation’s cycle, and consequently, revival and regeneration [2,3]. Certain sacredness is added to any figurine by the application of meaningful colors to it. Hence, most Shabtis were blue or green. One of the most available coloring agents for Shabtis, dating back to the beginning of the dynastic era, was copper compounds. Around the time of Thahtmus III (1479–1425 BC) in the new kingdom period, other colorants such as lead, cobalt, antimony, tin, iron, and manganese oxides were used for obtaining different colors such as white, green, purple, yellow, and black respectively [3,4]. An archaeological Ushabti usually is made of faience coated with a shiny colored or transparent thin glaze layer. Faience is produced in ancient Egypt by mixing up to 99% of high-grade silica, produced by crushing quartz pebbles, with alkali soda (Na2 O) 0.3 to 5%, and lime 1 to 5% at high temperature [5–9]. The shaping of the figurine is then performed via a process of melting and casting. In these old times, the production https://doi.org/10.1016/j.culher.2019.05.006 1296-2074/© 2019 Elsevier Masson SAS. All rights reserved. Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006 G Model ARTICLE IN PRESS CULHER-3601; No. of Pages 9 2 A. Khedr et al. / Journal of Cultural Heritage xxx (2019) xxx–xxx technology of the Egyptian faience has changed over time and from place to place in the country. Such changes are relevant to faience’s composition itself and the adopted production techniques [10]. For example, by the beginning of the first millennium BC, natron has been used instead of soda as a flux that helps in lowering the materials fusion temperature [11]. The lack of knowledge of the Ushabti origin and the absence of inscriptions makes it difficult to conclude whether or not a particular figurine is a copy or an original item. However, careful analysis of the materials used in manufacturing such Ushabtis can be decisive in this matter. The knowledge of the elemental composition of the surface and the core of the archaeological samples could help in constructing scenarios for the production technology utilized by the craftsmen and the row materials resources at such old eras. With such precious and unique objects, nondestructive analytical techniques are the most appropriate. Many works have been reported using Raman spectroscopy, Fourier transforms infrared spectroscopy (FTIR), and X-ray fluorescence (XRF) [12–16]. Though these techniques are nondestructive and, commercially available for several years ago, the first two provide only superficial information about the molecular, and not the elemental, composition, and cannot go beneath the surface like LIBS. On the other hand, though XRF can provide limited elemental stratigraphic details on the sample, it is unable in detecting elements lighter than magnesium. During the past two decades, laser-induced breakdown spectroscopy (LIBS) has emerged as a fast, simple and quasinondestructive spectrochemical analytical technique at a reasonable cost in the field of archaeology [17,18]. In LIBS, high power laser pulses are focused onto the surface of the target material. Concentrating such a tremendous amount of energy on a tiny volume causes the evaporation of few nanograms to micrograms of the target material. With continuous absorption of energy from the laser pulse, the atoms in the vapor are excited then ionized and at the end, a collection of ions and swirling electrons, at tremendously high temperature (6000–60,000 K), will compose the so-called the plasma plume. The laser-induced plasma cools down and gets rid of the absorbed energy in the form of electromagnetic radiation emission. The radiation emitted in the optical range (UV, Vis, and IR) is collected, dispersed spectroscopically, and the obtained spectra are displayed on a suitable PC for further processing and analysis. The LIBS spectra provide qualitative information about the elemental composition of the target material, where the identified spectral lines represent the spectral fingerprint of the elements existing in the plasma plume and consequently in the target material, in case of stoichiometric ablation. Quantitatively, the intensity of the elements spectral lines is proportional to the concentration of such elements in the target material, taking into consideration the matrix effect and the self-absorption. LIBS does not need sample preparation, where the sample is analyzed as is without dissolution, digestion or any other preparation procedures. On the other hand, LIBS is now available as commercial portable or mobile systems that can be used in excavation sites or museums for archaeological samples analysis. Since the laser parameters significantly affect the ablation process, plume and plasma characteristics, and more importantly the analytical performance of LIBS, the goal of the present paper was to: • find out the optimum detection conditions for a good LIBS spectroscopic practice; Table 1 List of experimental parameters used to produce the same irradiance. Laser wavelength (nm) Pulse energy (mJ) The lens focal length (cm) Spatial filter Irradiance (W/cm2 ) UV – 355 Green – 532 IR – 1064 25 65 225 10 10 20 With With Without 7.64 × 1011 Fig. 1. The archeological Ushabti samples, designated as S1–S4. Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006 G Model CULHER-3601; No. of Pages 9 ARTICLE IN PRESS A. Khedr et al. / Journal of Cultural Heritage xxx (2019) xxx–xxx • investigate the effect of laser wavelengths systematically on the investigated Ushabtis to obtain a good insight on their characterizing surface and bulk elements under fixed laser irradiance; • validate LIBS results with the SEM-EDX qualitative and quantitative analyses results; • examine the authenticity of the Ushabtis by comparing the LIBS spectra of the fake Ushabtis samples with that of the original ones under the same adopted experimental conditions. 2. Experimental setup The LIBS experimental setup used in the present work has been explained in detail elsewhere [19,20]. A Q-switched Nd: YAG laser (Brilliant Eazy, Quantel, France) was used as an excitation source operated at its three harmonics of 1064 nm, 532 nm, and 355 nm and pulse duration of 5 ns. The pulse energy was measured by a power meter (SCIENTECH model AC5001, USA) at a repetition rate of 10 Hz. The LIBS spectral emission was produced by focusing the laser beam orthogonally onto the investigated sample, which is mounted on an X-Y-Z micrometric translation stage for fresh spot irradiation. A 1.5 m-fiber optic cable of 600 ␮m core diameter was used to collect the emission from the laser-induced plasma to be fed to an echelle spectrometer (Mechelle7500, Multichannel, Sweden), coupled to an ICCD camera, DiCAM-PRO (PCO-computer optics, Germany). The control of both acquisition time and delay with respect to the onset of laser pulse was performed using the echelle spectrometer software. LIBS spectra were acquired and stored for analysis using LIBS++ software [21]. 3 In order to carry out a comparison among samples’ characteristic spectral emission, the same irradiance for the three used laser wavelengths (1064, 532, and 355 nm) is allowed to be incident on all investigated samples. This has been performed by controlling the experimental parameters of each laser wavelength, including the laser pulse energy, laser beam diameter (via a spatial filter), and the focusing conditions. Table 1 displays a list of the controlled experimental parameters that were applied throughout the whole measurements. For LIBS surface study, single laser pulse operational mode was allowed on fresh spots and accumulated for five pulses while for bulk depth profiling, up to fifty successive pulses were applied to the same spot of the sample and the spectrum was saved after every single pulse. Scanning Electron Microscope (Joel JSM-5410, Japan) in conjunction with Energy Dispersive X-Ray unit (OXFORD INCA Penta FETx3, England) was used to investigate the Egyptian archeological samples. SEM-EDX not only has been used to study the morphological structure but also utilized for the elemental analysis of the samples; qualitatively and quantitatively, which was used for validation of the LIBS results. 3. Samples Two sets of samples have been studied via LIBS. The first set consists of four genuine archeological Ushabti samples dated back to the Ptolemaic period collected from excavation sites in Saqqara, Giza, Egypt and are designated as S1–S5 as shown in Fig. 1. The second set, including six selected fake samples, was purchased from the tourists’ marketplace in the region of the pyramids. Their shapes and colors will be displayed in a following section. All the samples’ height ranges from approximately, 4 cm to 6 cm. 4. Results and discussion This section is divided into two main sections starting with comprehensive study of the archeological Ushabtis samples through an optimization of the LIBS experimental conditions in terms of detection parameters and the applied wavelengths [The three harmonics of Q-switched Nd-YAG laser; the fundamental IR wavelength (␭ = 1064 nm), the visible green (␭ = 532 nm) and the UV wavelength (␭ = 355 nm)]. This includes the LIBS qualitative analysis of the samples by identifying atomic and ionic emission lines in the spectral range between 200 nm and 900 nm and studying the craters formed by the application of each of the three wavelengths. This was followed by validating LIBS qualitative results by results obtained via SEM-EDX technique, which provide complementary images of the morphology of the investigated samples and their quantitative compositions. The second section investigates fake Ushabtis, using LIBS under the same previously obtained optimum experimental conditions. 4.1. Study of archeological Ushabtis Fig. 2. Laser-induced breakdown spectroscopy (LIBS) spectra at the three applied laser wavelengths representing common elements such as Ca, Al, Na, Si, Mg, and Fe (top) and zoomed spectra parts in two wavelength ranges. At the early stage of the interaction of a Q-switched nanosecondlaser pulse with a solid sample, an initial intense continuum is promptly emitted close to the target surface, which corresponds to the emission of blackbody radiation from the dense hot plasma. As the plasma cools down and expands away from the target, the process of recombination of electrons and ions eventually results in the production of spectral lines emission. Temporally, the radiation processes usually tend to be dominated with the highly ionized lines being emitted close to the target (superimposed on the continuum) and, then, the atomic lines appear at a later time of the plume lifetime. Line emission prolongs for a few microseconds compared to that of the continuum emission, which lasts for ns time range. Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006 G Model CULHER-3601; No. of Pages 9 4 ARTICLE IN PRESS A. Khedr et al. / Journal of Cultural Heritage xxx (2019) xxx–xxx Therefore, the exploitation of time-resolved LIBS (TR-LIBS) allows obtaining a temporal variation of LIBS signal which enables the choice of best detection time for the best signal to background (S/B) ratios. Moreover, the analytical signal intensity of an element in the plasma, which is related to the ejected mass of the investigated sample mainly depends on laser radiation parameters (laser energy and focusing). Any random changes of these parameters can strongly affect the analytical precision and accuracy. Additionally, the ablation rate may vary as a result of the laser pulse-to-pulse energy fluctuations and the relevant laser irradiance changes. Therefore, the laser irradiance incident on the solid surface has been adjusted and kept the same under the conditions listed in Table 1. Various approaches to analytical signal correction and characterization improvement have been reviewed [22]. A straightforward method is to consider the ratio of the intensity of the analytical spectral line to the intensity of a spectral line of another component of the material instead of considering the absolute intensity of the spectral line. Despite the change in measured values, the intensity ratio remained almost the same. Two types of internal standard lines can be used [23]: • non-sensitive spectral line of the major component of the sample; • sensitive line of an element that is not initially present, but added to the sample. In the present work, the intensity of the spectral emission lines was analyzed either as it was measured under the specified experimental conditions (laser pulse energy, laser wavelength) or normalized with respect to an internal reference selected among those major elements characteristic of the matrix of the Ushabtis’ material, as it will be mentioned. Based on the former reasoning, under the adopted experimental setup, Ushabtis’ LIBS spectral emissions were stored as a function of the time window (gate width G) during which the spectral emission is acquired, and its delay from the onset of the laser pulse (delay time D) under the application of the three adopted laser wavelengths. For data analysis and manipulation, the line ratio of the spectral line of Cu I at 327.3 nm was normalized to that of Ca I at 428.9 nm (one major component of the sample) and plotted at different acquisition and delay time values for the three laser wavelengths. Nearly the same values of the optimum gate width (G) and delay time (D) at 1064 nm, 532 nm, and 355 nm were obtained. Under the adopted experimental conditions, the values of (G) and (D) were set at fixed values; 2500 ns and 1500 ns, respectively, throughout all measurements. 4.1.1. LIBS signal dependence on laser wavelength The laser wavelength is an essential factor for the initiation of plasma, and its choice is dependent on its analytical capability. Elemental composition of Ushabti samples has been investigated via LIBS under the application of three different laser wavelengths, namely, the fundamental wavelength at 1064 nm, Green at 532 nm and UV at 355 nm. Under the adopted optimum D/G time values, spectral emission was recorded for the four archeological samples. For all samples, surface elements such as Ca, Al, Na, and Fe were easily detected, and a noticeable amount of Mn and Cu could be observed for black and greenish blue pigmented regions in the samples, respectively. Moreover, Si and Mg spectral lines indicate a high content of such elements, which may be attributed to their presence in the ancient ground quartz or quartz sand. One selected example of a LIBS spectrum covering a wide spectral range from ∼ 240 nm to 650 nm under the applied three laser wavelengths is shown in Fig. 2 (top). Spectral lines of Mn and Cu characterizing respective black and blue pigmented regions are displayed in the zoomed sections in the wavelength ranges from 474 nm to 484 nm in Fig. 2 Table 2 A list of LIBS detected elements and their corresponding spectral lines. Elements Detected LIBS spectral lines Ca 315.88 nm, 317.93 nm, 428.3 nm, 393.3 nm, 396.8 nm 308 nm, 309 nm 263.12 nm, 269.25 nm, 273.96 nm, 274.67 nm, 275.57 nm, 426.04 nm, 427.17 nm 324.75 nm, 327.39 nm, 510.58 nm, 515.32 nm, 521.80 nm 288.15 nm 470.11 nm, 472.74 nm, 473.91 nm, 475.20 nm, 476.27 nm, 476.67 nm, 478.38 nm, 482.43 nm 588.99 nm, 589.60 nm 279.55 nm, 280.27 nm, 285.16 nm, 516.74 nm, 517.31 nm, 518.39 nm Al Fe Cu Si Mn Na Mg LIBS: laser-induced breakdown spectroscopy. (middle) and from 510 nm to 524 nm in Fig. 2 (bottom). For clarification, detected elements and their corresponding spectral lines are listed in Table 2. To avoid the inherent fluctuations of LIBS spectral data; the intensity was normalized to the line intensity of Ca (I) at 428.3 nm. The effect of the laser wavelength on the intensity of some selected elemental lines was studied for samples S1, S2, S3 and S4 as shown in Fig. 3. It is clear that, for all samples, the normalized intensity for most elements is the highest when the UV laser at 355 nm was applied. This effect may be explained given the higher absorption of the UV in such type of targets and that among the applied wavelengths, the high photon energy of the UV laser induces plasma characterized by higher emissions, less continuum, short lifetime and narrow emission lines. Additionally, the signal intensity of the spectral emission lines is highly dependent on the optical penetration depth, which is high for IR laser compared to that for UV laser. In the case of a UV laser, energy is rapidly absorbed in the surface of the samples, which has been previously reported for substrates of high evaporation threshold such as ceramics, stones and faience [24]. For all used wavelengths, the appearance of elements such as Ca, Si and Al in all samples with higher abundance supports publications reporting that major components of Ushabti samples are quartz and limestone. The Ca is derived from lime (CaO), which comprises between 1–5% of the composition of faience and can be extracted from natural resources such as limestone or chalk. It is assumed that the ancient Egyptians may have had lime in archaeological faience as an impurity in the used sand or it was added deliberately from chalk and/or limestone. On the other hand, analysis of the Egyptian sand has revealed that it contains 2–6% calcareous material [25]. These findings may also explain the unconscious existence of the lime on old manufactured faience. Moreover, when applying each of the three laser wavelengths, the spectral emission lines of silicon suggest that it represents a major constituent of the samples. It might be claimed that the source of silicon may be the quartz used in producing the bulk body of the Ushabti. As the quartz bulk object is finished, it may disintegrate. Therefore, the addition of alkali, such as P, Na, and K as well as the lime can help in binding the quartz grains together during the drying phase, which gives the produced faience higher mechanical strength [26,27]. We believe that sodium might have been obtained in the form of Natron (Na2 O) mined by ancient Egyptians from Wadi El-Natrun, a valley northwest of Egypt and the Elkab area, eighty kilometers south of Luxor, Upper Egypt [4]. Bearing in mind that the ablation rate is dependent on many factors that include the laser parameters, surface conditions, and laser absorption in the plasma [28], the cross-sectional profiles of the craters produced by the three applied laser wavelengths are Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006 G Model CULHER-3601; No. of Pages 9 ARTICLE IN PRESS A. Khedr et al. / Journal of Cultural Heritage xxx (2019) xxx–xxx 5 Fig. 3. Normalized laser-induced breakdown spectroscopy (LIBS) intensity using three laser wavelengths; 1064 nm, 532 nm, and 355 nm on the Ushabti samples S1 (a), S2 (b), S3 (c) and S4 (d). Fig. 4. The cross-sectional profiles of the craters obtained by the laser wavelengths at 1064 nm (a), 532 nm (b), and 355 nm (c). obtained, as shown in Fig. 4. It shows typical features of ns-laser surface interactions, a central crater with specific area and depth, which change with the applied laser wavelength. For the three used laser wavelengths, a list of depths and diameters for the produced crates are given in Table 3. It is observed that the crater’s profile produced by the UV laser is more symmetrical compared to those produced by both the visible and IR lasers whose craters seem to be irregular. At fixed laser irradiance, the difference in the craters’ Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006 G Model ARTICLE IN PRESS CULHER-3601; No. of Pages 9 6 A. Khedr et al. / Journal of Cultural Heritage xxx (2019) xxx–xxx Table 3 A list of the values of both depths and diameters for the produced craters when applying the three mentioned laser wavelengths. Wavelength (nm) 1064 532 355 Depth (␮m) Diameter (␮m) 65 45 46 62 24 19 profile can be attributed to the laser wavelengths and their corresponding optical penetration depths. 4.1.2. Surface LIBS versus stratigraphic LIBS This section focuses on both the identification of Ushabti’s surface pigments, which can be inferred from their color in addition to their obtained LIBS spectroscopic data (their characteristic spectral lines), and the LIBS stratigraphic measurements which provide information about the composition of the glaze layer and the below substrate (bulk) of the Ushabti sample. A comparison of LIBS spectra for the surface and the bulk of Ushabti’s head for the sample (S3) is illustrated in Fig. 5. It is observed that different elements are enabling the differentiation between the surface and the bulk. While surface spectral analysis shows emission lines of Mn that is responsible for the black color of the head, the bulk is characterized by the large amount of Sr, and Ca. It is most probable that Mn comes from Pyrolusite (MnO2 ) while the presence of Sr and Ca presumably exist via the use of lime (CaO) and Celestine (SrSO4 ) that may be originated from the clay co-existing with the faience components and the higher Sr signal on the surface could be due to contaminants from the burial environment. The high intensity of Cu spectral lines was observed for the surface layers of the rest of the samples which is supposed to be responsible for the blue/green color. The source of Cu is the Egyptian blue pigment (CaCuSi4 O10 ), which was made by heating a mixture of sand, Natron (Na2 CO3 •10H2 O) and a copper compound such as malachite. This scenario was explained by the fact that the addition of sodium (in Natron) as a fluxing agent, reduces the melting point of pure silica from 1700 ◦ C to less than 1000 ◦ C and the inclusion of limestone (calcium oxide) works as a stabilizer [27]. Ultimately, with the aid of the pigments’ apparent color in addition to their characteristic atomic and ionic LIBS spectral emission, one can successfully discriminate among different pigments and their composition. By applying UV laser at 355 nm on the Ushabti samples S3 and S4, the depth profiling of two key elements, which are responsible for the color pigments are investigated in Fig. 6. This figure shows the depth profiling plot of spectral emission lines of Mn and Cu which are responsible for the black and blue/green colors in samples S3 and S4, respectively. It is clear that the intensity of the spectral emission line, Mn I 475.2 nm decreases with increasing the depth. In other words, as the laser beam goes more in-depth in the sample, the black pigment on the surface is gradually decreasing until it is nearly removed after about ten micrometers. On the other hand, inspection of depth profiling of the emission line Cu I 324.7 nm for sample S4 revealed that the color layer is almost removed after about five micrometers. This observation indicates that monitoring the characterizing elements for each pigment allows the estimation of its thickness, which seems to be different for each surface pigmented layer. 4.1.3. SEM-EDX measurements Scanning electron microscopy and energy dispersive X-ray (SEM-EDX) were used to validate LIBS results and to provide quantitative analyses for a comprehensive overview of the archeological samples. Due to its finely focused high energetic e-beam, SEM can scan around even rough surfaces producing clear morphological images [29,30]. Fig. 7 shows the local surface elements for samples S1 and S4 at a particular position with their corresponding weight percentage relative to the predominant concentration of the expected and characteristic pigment’s elements. The insets in the same figure depict the micrographs of the same analyzed parts of the sample surface. Most of the results agree with LIBS analysis, especially in accordance with the major constituent elements such as Si, Al, and Ca. It should be noted that the emission lines of K, P, Cl, and S were out of the spectral range considered in the present LIBS study, but EDX has observed them. The source of such effective alkali elements may be found as rock salts or from ashes. For Fig. 5. Laser-induced breakdown spectroscopy (LIBS) spectra of the surface and the bulk for sample (S3) at two wavelength ranges. Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006 G Model CULHER-3601; No. of Pages 9 ARTICLE IN PRESS A. Khedr et al. / Journal of Cultural Heritage xxx (2019) xxx–xxx 7 Fig. 6. The intensity change against the depth in microns for Mn I 475.2 nm for sample S3 and Cu I 324.7 nm for sample S4. Fig. 7. The elements weight percentage of the energy dispersive X-ray (EDX) analysis for both samples S1 (left), and S4 (right). The inset is the scanning electron microscopy (SEM) micrograph of the analyzed surface. example, P generally comes from bone ashes while K is believed to come from the ashes of halophytic plants particularly Salicornia which was the major source of alkali until the Ptolemaic period onwards [31]. The existence of relatively high weight percent of Br in both samples may be originated from the raw materials found in several types of clay, especially, carbonate-rich clays. It is worth mentioning here that it is challenging to detect bromine, as well as most halogens, via LIBS due to its high excitation energy. Bromine is usually accumulated in the seawater and because of its high reactivity; it volatilizes into the atmosphere, and precipitate to the land surface where it is usually combined with other elements at various firing temperature. Since Br sustains its original formation in burned objects up to 800 ◦ C, it is considered as firing temperature indicator [32]. Therefore, Ushabti figurines were assumed to be burnt below 800 C. To that end, complementary results have been offered by SEMEDX technique, which, along with LIBS, provide more reliable information about the surface pigments and the bulk elemental composition of the archeological Egyptian Ushabti. 4.2. Examining the fake Ushabtis Given the obtained significant results for the identification of the pigments of the archaeological Ushabtis, the capability of the LIBS technique in examining fake faience Ushabtis collected from tourist bazaar in the vicinity of the Giza pyramids have been investigated. Under the optimum experimental conditions mentioned above, the obtained LIBS spectra for the investigated fake Ushabtis, mainly, their blue pigmented regions, have been recorded. Such spectral data has been compared to those previously acquired for the genuine archeological samples. Fig. 8 illustrates an example of the obtained LIBS spectral emission of the blue pigment for the fake Ushabtis. When performing a comparison between such spectral data and those previously acquired for the ancient Ushabtis, some elements such as Mg and Si coexist in both. However, they exist in lower content in the fake samples. Also, the appearance of the different elemental composition such as Co and Ti represents significant differentiating elements. The presence of cobalt, known to be currently used for obtaining blue color faience, may explain the appearance of the stable colorant giving the blue color on the surface of these imitated samples. According to the required selected colors, cobalt compounds are recently used to produce two colors; the light purple of the cobalt carbonate (CoCO3 ) and the black of the cobalt oxide (CoO). By adding very low quantity (one-fourth percent) in glaze results in satisfactorily colored blue. However, the blue color produced by cobalt oxide can be changed by the addition of other coloring oxides. Moreover, the presence of Fe may justify the yellowish brown color observed in the fake samples [33]. Furthermore, the presence of several recognizable Ti spectral lines in the fake samples is believed to be commonly present as titanium dioxide or titania, which is used to opacify the glaze by creating small crystals within it, so it is well suited for macro-crystalline glazes [34]. Titania may accentuate the color on the clay body, and its high viscosity may stiffen the glaze. Insoluble sources of titania that currently used are rutile and ilmenite (iron titanate). As a result, one can conclude that fake samples constituted different elemental composition than those contained in the archaeological Ushabtis when the same color pigment is examined for both types of samples. The LIBS elemental differentiation capability performed successful examination of the authenticity of the suspect samples, and an indication of their fake entity was revealed. Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006 G Model CULHER-3601; No. of Pages 9 8 ARTICLE IN PRESS A. Khedr et al. / Journal of Cultural Heritage xxx (2019) xxx–xxx Fig. 8. Zoomed laser-induced breakdown spectroscopy (LIBS) spectral emission for the fake Ushabtis showing some common elements like those found in the ancient samples, but with lower content, such as Mg and Si and some differentiating elements such as Co and Ti. 5. Conclusion As a quasi-nondestructive and fast spectrochemical analytical technique, LIBS was used to identify the elemental composition of the surface and bulk of four Ushabti samples dating back to the Ptolemaic period. Three laser wavelengths (355, 532, and 1064 nm) were exploited to study the effect of the wavelength on the analytical results. It has been demonstrated that the UV laser wavelength at 355 nm is the most appropriate as for the dynamic spectral range and the signal to noise (S/N) ratio in the obtained LIBS spectra. The superficial colors could be justified by the presence of specific element appearing in the relevant spectra such as Mn and black color, as well as Cu and green color. Stratigraphic measurements depicted the elemental structure of the glaze layer and the beneath bulk layer too. The LIBS results were validated successfully via EDX-SEM measurements, which also provided a complementary image that may help infer the old techniques adopted by the Egyptians back then. Moreover, to make use of these measurements, a comparison between the LIBS spectra of the archaeological Ushabti and that of two fake Ushabti samples was performed. An apparant discrepancy in the elemental composition was observed that indicates the possibility of using LIBS in the authentication of the archaeological samples. Future work is planned for in situ measurements, in museums and excavation sites using portable LIBS systems, to save time and preserve the precious ancient samples. References [1] R.O. Faulkner, The Ancient Egyptian Book of the Dead: the Book of Going Forth by Day, Ed. Carol Andrews, British Museum Press, USA, 1990. [2] K. Yamahana, Synchrotron radiation analysis on ancient Egyptian vitreous materials (446), in: Proceedings of the 25th Linear Accelerator Meeting, Himeji, Japan, 2000, pp. 92–96. [3] A. Sparavigna, Ancient technologies: the Egyptian sintered-quartz ceramics, philica.com 426 (2014) 1–12. [4] A. Kaczmarczyk, R.E.M. Hedges, Ancient Egyptian Faience: An Analytical Survey of Egyptian Faience from Pre-Dynastic to Roman Times, Aris and Phillips, Warminster, 1983. [5] M. Bimoson, I.C. Freestone, Early faience recent studies, origins and relations with glass, in: Early Vitreous Materials, British Museum Occasional paper (BMP), London, 1987 (56[5]). [6] P. Vandiver, W. Kingery, Egyptian faience: the first high-tech ceramic, In ceramics and civilization, in: The 88th Annual Meeting of the ACS, Chicago, Illinois, 1986, pp. 19–34. [7] H.M. Stewart, Egyptian Shabtis, Shire - Egyptology, London, 1995. [8] M. Panagiotaki, A Study of Vitreous Materials from Minoan Crete, Ancient Greek Technology, in: Proceedings from the 1st International Conference on Ancient Greek Technology, Thessaloniki, Greece, 1998. [9] P.B. Vandiver, in: J.S. Olin, A.D. Franklin (Eds.), Technological change in Egyptian faience in archaeological ceramics, Smithsonian Institution Press, Washington DC, 1982, pp. 167–179. [10] M.S. Tite, A. Shortland, Y. Maniatis, D. Kavoussanaki, S.A. Harris, The composition of the soda-rich and mixed alkali plant ashes used in the production of glass, J. Archaeol. Sci. 33 (2006) 1284–1292. [11] W. Noll, R. Holm, L. Born, Painting of ancient ceramic, Angew. Chem. Int. Ed. 14 (1975) 602–613. [12] S. Sotiropoulou, V. Perdikatsis, K. Birtacha, C. Apostolaki, A. Devetzi, Physicochemical characterization and provenance of colouring materials from Akrotiri-Thera in relation to their archaeological context and application, Archaeol. Anthropol. Sci. 4 (2012) 263–275. [13] L. Burgio, R.J.H. Clark, T. Stratoudaki, M. Doulgeridis, D. Anglos, Pigment identification: a dual analytical approach employing laser-induced breakdown spectroscopy and Raman microscopy, Appl. Spectrosc. 54 (2000) 463–469. [14] M. Mantler, M. Schreiner, X-ray fluorescence spectrometry in art and archaeology, X-Ray Spectrom. 29 (2000) 3–17. [15] G. Banik, M. Schreiner, F. Mairinger, H. Stachelberger, Scanning electron microscopic investigation of cross sections taken from paint layer specimen, Prakt Metallogr. 19 (1982) 104–108. [16] A. Loon, J. Boon, Characterization of the deterioration of bone black in the 17th century Oranjezaal paintings using electron microscopic and microspectroscopic imaging techniques, Spectrochim. Acta B 59 (2004) 1601–1609. [17] R. Gaudiuso, M. Dell’Aglio, O. De Pascale, G.S. Senesi, A. De Giacomo, Laserinduced breakdown spectroscopy for elemental analysis in environmental, cultural heritage and space applications: a review of methods and results, Sensors 10 (2010) 7434–7468. [18] D.W. Hahn, N. Omenetto, Laser-induced breakdown spectroscopy (LIBS), part I: review of basic diagnostics and plasma-particle interactions: still-challenging issues within the analytical plasma community, Appl. Spectrosc. 64 (2010) 335–366. Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006 G Model CULHER-3601; No. of Pages 9 ARTICLE IN PRESS A. Khedr et al. / Journal of Cultural Heritage xxx (2019) xxx–xxx [19] O. Abdel-Kareem, M.A. Harith, Evaluating the use of laser radiation in cleaning of copper embroidery threads on archaeological Egyptian textiles, Appl. Surf. Sci. 254 (2008) 5854–5860. [20] O.A. Nassef, H. Ahmed, M.A. Harith, Surface and stratigraphic elemental analysis of an ancient Egyptian cartonnage using laser-induced breakdown spectroscopy (LIBS), Anal. Methods 8 (2016) 7096–7106. [21] M. Corsi, G. Cristoforetti, V. Palleschi, A. Salvetti, E. Tognoni, Fast and accurate method for the determination of precious alloys caratage by laser-induced plasma spectroscopy, Eur. Phys. J. D 13 (2001) 373–377. [22] N.B. Zorov, A.A. Gorbatenko, T.A. Labutin, A.M. Popov, A review of normalization techniques in analytical atomic spectrometry with laser sampling: from single to multivariate correction, Spectrochim. Acta B 65 (2011) 642–657. [23] V. Thomsen, W. Gerlach, The foundations of modern spectrochemical analysis, Spectroscopy 17 (2002) 117–120. [24] B. Di Bartolo, O. Forte, Advances in Spectroscopy for Lasers and Sensing, Springer, The Netherlands, 2006. [25] C. Kiefer, A. Allibert, Pharaonic blue ceramics: the process of self-glazing, Archeology 24 (1971) 107–117. [26] A. Lucas, J.R. Harris, Ancient Egyptian Materials and Industries, 4th ed., Arnold, London, 1962. 9 [27] P.T. Nicholson, Egyptian Faience and Glass, Shire - Egyptology, Aylesbury, 1993, pp. 137. [28] T. Efthimiopoulos, E. Kritsotakis, H. Kiagias, C. Savvakis, Y. Bertachas, Laser ablation rate of materials using the generated acoustic waves, J. Phys. D: Appl. Phys. 31 (19) (1998) 2648–2658. [29] R. Ravisankar, G. Raja Annamalai, A. Naseerutheen, Application of a multianalytical toolset to characterize the ancient pottery from Kottapalayam, Tamilnadu, India Using spectroscopic techniques, Int. Res. J. Pure Appl. Chem. 3 (3) (2013) 210–219. [30] L. Moraru, F. Szendrei, Ancient pottery analysis using SEM image processing, Eur. J. Sci. Theol. 6 (2010) 69–78. [31] P.T. Nicholson, I. Shaw, Ancient Egyptian Materials and Technology, Cambridge University Press, Cambridge, UK, 2000. [32] M.J. Trindade, M.I. Dias, F. Rocha, M.I. Prudêncio, J. Coroado, Bromine volatilization during firing of calcareous and non-calcareous clays: archaeometric implications, Appl. Clay Sci. 53 (3) (2011) 489–499. [33] D. Rhodes, Clay and Glazes for the Potter, Chilton, Philadelphia, 1957. [34] J. Britt, The Complete Guide to High-Fire Glazes, Lark Books, Sterling Publishing Co., New York, 2007. Please cite this article in press as: A. Khedr, et al., Discrimination between the authentic and fake Egyptian funerary figurines “Ushabtis” via laser-induced breakdown spectroscopy, Journal of Cultural Heritage (2019), https://doi.org/10.1016/j.culher.2019.05.006