Documents

10 pages
189 views

Archaeometry 2017

of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Sulphur, fats and beeswax in the Iberian rites of the sanctuary of the oppidum of Puente Tablas (Jaén, Spain)
Transcript
  An integrated multianalytical approach to the reconstruction of dailyactivities at the Bronze Age settlement in Peñalosa (Jaén, Spain) E. Manzano a, ⁎ , A. García b , E. Alarcón b , S. Cantarero a , F. Contreras b , J.L. Vílchez a a Department of Analytical Chemistry, University of Granada, E-18071 Granada, Spain b Department of Prehistory and Archaeology, University of Granada, E-18071 Granada, Spain a b s t r a c ta r t i c l e i n f o  Article history: Received 4 April 2015Accepted 26 April 2015Available online 3 May 2015 Keywords: GC – MSGC – C – IRMSHRMSOrganic residueBronze AgePeñalosa site ThispaperseekstoreconstructeverydayhumanactivitiesinaBronzeAgeArgaricsiteinsoutheasternSpain.Weused an integrated multi-analytical approach to study residues extracted from 5 ceramic pieces and 3 soil sam-ples. The results of these analyses were complemented with archaeological interpretations to produce the  󿬁 rst 󿬁 ndings on domestic activities four millennia ago at the Peñalosa settlement in Jaén, Spain. The bands observedinFTIRandATR  – FTIRspectraofthearchaeologicalresiduesshowedapeakassignmentthatcon 󿬁 rmedthesilicatecomponents of the clay as raw material and the presence of organic matter. The amorphous organic materialswereanalyzedbyGC – MS,HRMSandGC – C – IRMS,andthemineralresiduesbynon-invasiveportableX-ray 󿬂 uo-rescence (pXRF). We identi 󿬁 ed  α , ω -dicarboxylic and  ω -(o-alkylphenyl) alkanoic acids produced by oxidativedegradationofunsaturatedfattyacids,aswellasmonocarboxylicacids,n-alkanes,diterpenoidacids,cholesterol,dyesandchemicalgrapemarkers.Assignmentswerebasedonlipidpro 󿬁 le,fattyacidratiosandthe δ 13 Cvaluesof C 16:0 andC 18:0 .Theresultsareconsistentwiththepresenceoftissuesfromruminant(ovineorbovine)andnon-ruminant(swineandequine)animals, vegetableoils, waxes,coniferresinandgrapeseeds.Thepresenceof  󿬁 sh-based fats was suggested in one of the residues. Our evidence also suggests that grape juice or wine may havebeen consumed at this Bronze Age site. As far as we know this is the  󿬁 rst investigation of the organic residuesabsorbed into an Argaric goblet used in a domestic situation.© 2015 Elsevier B.V. All rights reserved. 1. Introduction Food preparation and consumption are essential daily activities forthe survival of any social group. These timeless practices generate resi-dues and every remain artifact and ecofact in the archaeological recordof a site is not only a re 󿬂 ection of the decisions taken by a group of humansandtherelationsbetweenthem[1,2]butisalsooftentheprod-uct of a particular phase in the food production chain. Thanks to theseremains we know where (space) and in what (ceramic vessels) peoplestored, cooked and ate food four thousand years ago, but answeringquestions as to what they cooked or what they ate remains dif  󿬁 cult.Inorder to 󿬁 nd out which foods were cooked or consumed inan an-cient potsherd, it is necessary to analyze the organic residues absorbedbytheporesinthepotteryorthoseadheringtoitsinnersurface.Inaddi-tion to foods, these vessels may also have contained beverages, medi-cines, dyestuffs, essential oil, wine, etc. For archaeologists, residuescontaining lipids are perhapsof most interest, because the altered prod-ucts formed as a result of burial or other human activities (for exampleby heating) are relatively stable. They can therefore remain intact for along time in a ceramic matrix, although this depends on their ability towithstand biodegradation. The chemical characterization of solid resi-dues is a dif  󿬁 cult task due to their complex composition and their de-graded state. In order to identify these residues, speci 󿬁 c methodologiesfor extraction, puri 󿬁 cation and further analysis of the total lipid extract(TLE)mustbedeveloped.Theanalyticaltechniquesusedmostcommon-ly for this purpose are Fourier transformed infrared spectroscopy (FTIR)[3], Raman spectroscopy (RS) [4], gas chromatography – mass spectrom-etry (GC – MS) [5 – 8], high-performance liquid chromatography – massspectrometry (HPLC – MS) [9,10] and pyrolysis – gas chromatography – mass spectrometry (Py – GC/MS) [11]. More recently the determinationof the ratios of the stable isotopes of carbon ( 12 C and  13 C) and nitrogen( 14 N and  15 N) using gas chromatography – combustion – isotope ratiomass spectrometry (GC – C – IRMS) [12] has provided insights into thetype of food consumed and the source of the srcinal lipids giving risetotheseresidues.Abetterunderstandingoftheanalyticaldatacanbeob-tained by establishing the relationship between the molecular constitu-ents that remain in the organic residues and the source from whichthey srcinate. To this end, we used various criteria, the simplest of which was the comparison of   “ chemical 󿬁 ngerprints ”  of molecules thathave remained in the ceramics for many millennia with the referencematerials.Theratiosofsomecommonfattyacids,thepresenceofspeci 󿬁 c Microchemical Journal 122 (2015) 127 – 136 ⁎  Corresponding author. Tel.: +34 958 243388; fax: +34 958 243328. E-mail address:  emanzano@ugr.es (E. Manzano).http://dx.doi.org/10.1016/j.microc.2015.04.0210026-265X/© 2015 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Microchemical Journal  journal homepage: www.elsevier.com/locate/microc  archaeologicalbiomarkersandthestablecarbonisotopevaluesofspecif-iclipidcomponentshavealsoprovedusefulforthischaracterization[13,14].This paper focuses on the analysis of residues discovered in  󿬁 ve ce-ramic vessels recovered from the Peñalosa site (Baños de la Encina, Jaén, Spain) and the associated soil residues found inside three of thesevessels.ThesevesselsdatefromtheBronzeAgeandinchronolog-ical and cultural terms are associated with the Argar people, an ancientculture that  󿬂 ourished in south-east Spain around the period 2250 – 1550 BC [15]. Since the 19th century, this ancient culture has been a subjectofgreatinterestamongthescienti 󿬁 ccommunity,asexempli 󿬁 edbytheabundantbibliography[16,17].Theanalysisandstudyoftheirfu-nerary rituals have been a recurrent theme, to the extent that it haseclipsedotherpossiblelinesofinvestigationsuchaseverydayactivities,which in many ways offer a broader view of life in ancient civilizations[2]. In fact, in research into the Bronze Age in Southeast Spain, onlytwo publications have analyzed the organic remains found in ceramicvesselsandbothofthesewerefromfunerary 󿬁 nds[18,19].Thenovelas-pect of our research is that it is the  󿬁 rst study in which domestic ce-ramics, rather than funerary objects, have been analyzed. To this end,we embarked on interdisciplinary research usinga combination of cur-rent analytical techniques (pXRF, FTIR, ATR  – FTIR, HRMS, GC/MS andGC – C – IRMS) with two ambitious goals: (1) to identify the compoundspresentintheorganicresiduesextractedfromthevessels,(2)todiscussthe possible biomolecular srcin of these organic residues in order tobetter understand what the people from the Argaric site in Peñalosacooked and ate. The results obtained using this combination of analyti-cal techniques showed unambiguous identi 󿬁 cations of constituents of chemically complex residues and offer a rigorous scienti 󿬁 c insight intoeverydayactivitiesinthispartofsoutheasternSpainfourmillenniaago. 2. Experimental  2.1. Chemicals and reagents Dichloromethane and methanol (Analytical Grade) purchasedfrom Fluka (St. Louis, MO, USA) were used as extraction solvents. Tolu-ene (Analytical Grade) and Meth-Prep II (m-tri 󿬂 uoromethylphenyltrimethylammonium hydroxide) were respectively selected as the sol-ventandthereagentforthederivatizationprocesspriortogaschroma-tography analysis. Both were purchased from Sigma-Aldrich (St. Louis,MO, USA). Inorder to preventdegradation andobtain high levels of re-producibility, Meth-Prep II was stored at − 4 °C in the freezer. Further-more, LC-MS grade Formic Acid and Methanol (Sigma-Aldrich) wereused in High Resolution Mass Spectrometry assays.Analytical grade standards of fatty acids (C 10 , C 12 , C 13 , C 14 , C 15 , C 16 ,C 16:1,  C 18 , C 18:1 , C 18:2 , C 19 , C 20 , C 22 , C 26,  C 30 ) and Cinnamic acid, Azelaicacid, Suberic acid, Adipic acid, Tartaric acid, Syringic acid, Cholesteroland  β -Sitosterol were purchased from Sigma-Aldrich (St. Louis, MO,USA). Fatty acid C 13  was used as an Internal Standard. Individual stan-dardsolutionsofcompounds(1000mgmL  − 1 )werepreparedinmeth-anol and stored at 20 °C. These solutions were prepared fresh monthly.Working standard mixtures were prepared by diluting the individualstock solution in methanol. They were stored at 4 °C and preparedfresh weekly. All solutions were stored in dark glass bottles to preventphotodegradation.  2.2. Instrumentation 2.2.1. Attenuatedtotalre  󿬂 ectance – Fouriertransforminfrared spectroscopy(ATR – FT-IR) and Fourier transform infrared spectroscopy (FT-IR) The FT-IR spectra were collected using a JASCO Spectrometer 6200,working in transmission mode and diamond micro-ATR accessory. Theinstrument was connected to a Pentium 200 and the instrument soft-ware was SPECTRA MANAGER v2. The FT-IR spectra were registeredfrom 400 cm − 1 to 4000 cm − 1 and the ATR spectra from 600 to4000cm − 1 ,witharesolutionof2cm − 1 and200scan.Thespectrometerhas a TGS Detector. The spectrum of pure KBr was used as background.  2.2.2. Gas chromatography – mass spectrometry (GC  – MS) The chromatographic system consisted of an Agilent 6890 N series(Agilent Technologies, Palo Alto, CA, USA) equipped with an automaticinjector (model 7683) and automatic sample tray (model 7683). AnHP-5 (5% Phenylsiloxane 30 m × 0.25 mm, 0.25  μ  m of particle size)was used as a capillary column. In addition, an Agilent 5973N massspectrometry detector (Agilent Technologies, Palo Alto, CA, USA) wascoupled to the chromatograph. The detector consisted of an ionic im-pact (70 eV) and a quadrupole as an ionization source and analyzer re-spectively.Table1showstheparametersusedforgaschromatography – mass spectrometry analysis. Each parameter was optimized beforecommencing the analysis. In order to achieve the most ef  󿬁 cient chro-matographic separation,  󿬁 ve different temperature gradients wereassayed. As regards gas  󿬂 ow, several  󿬂 ows were tested with a  󿬂 ow of 1mLmin − 1 producingthebestresults.Fourdifferentinjectionvolumesweretestedandthebestchromatogramwasobtainedusingavolumeof 1.5 μ  L.Afterassayingfourvoltages,wefoundthatthebestresultwasob-tained with the 3 EV gain factor. Finally, we optimized the  m/z   range,selecting the50 – 520umarangeasthe  m/z  workingrange.Peakassign-mentswereperformedonthebasisoftheanalysisofavailablestandardcompounds and by comparing their mass spectra to those from theWiley mass spectra database.  2.2.3. Gas chromatography – combustion – isotope ratio mass spectrometry(GC  – C  – IRMS) The GC – C – IRMS equipment was based on a Thermo Trace GC ultraand Thermo Delta V Advantage as an IRMS detector (Thermo FisherScienti 󿬁 c, Waltham, MA). Con 󿬂 o III was selected as an interface andreactor temperature (Cu – Ni – Pt) was set at 940 °C. The mass spectrom-eter source pressure was 1.9 × 10 − 06 mBar. HP-1 (30 m × 0.25 mmID × 0.25 μ  m) was used as a GC column for assessing chromatographicconditions. The carrier gas was helium with the GC oven programmedfrom60(hold1min)to320°Cat6°Cmin − 1 for20min.CarbonisotoperatiosarereportedinthestandarddeltanotationrelativetothePeeDeeBelemnite (PDB) standard. The results were expressed as  δ 13 C (%) =[(R  sample − R  standard ) / R  standard ], where R is  13 C/ 12 C in per mil.  2.2.4. High resolution mass spectrometry (HRMS) HRMS analysis was performed to identify other residues apart fromfatty acids and lipids, which cannot be detected by gas chromatogra-phy – mass spectrometry, focusing speci 󿬁 cally on the identi 󿬁 cation of peptides and proteins.An LCT Premier (Waters, Manchester, UK) spectrometer was usedfor the HRMS analysis and was operated with electrospray ionization(ESI)in positive mode.Mass spectrometer parameters were asfollows:capillaryvoltage,2.60kV;sourcetemperature,100°C;desolvationtem-perature, 400 °C; cone gas  󿬂 ow, 40 L h − 1 ; desolvation gas  󿬂 ow,400 L h − 1 ; nitrogen ( ≥ 99.995%) was used as cone and desolvation gasand cone gas  󿬂 ow. It is important to mention that before injection thesamples were diluted with a mixture of methanol (0.1% acid formic)to ionize the compounds, so enabling them to be detected by HRMS.  2.2.5. Portable X-ray  󿬂 uorescence (pXRF) Portable X-ray  󿬂 uorescence analyzer, with a 40 kV X-ray tube withAganodetargetexcitationsource,andasiliconPIN-diodewithaPeltiercooled detector. The analyzer was initially calibrated using the silverand tungsten shielding on the inside of the shutter, and the sourcecount time for analysis was 󿬁 xed in 90s.Direct measurements can be performed with portable instrumentsthat do not affect the integrity of the sample. 128  E. Manzano et al. / Microchemical Journal 122 (2015) 127  – 136    2.3. Archaeological samples. Criteria for selecting the artefacts Archaeological samples were taken from ceramics found in situ atthe occupation level of House 25 of the Peñalosa site. This domesticspace is located on the western side of the oriental acropolis of the set-tlement, neara further three domestic and productivespaces. Togetherthey present a complex constructive and urban system, located in thehighest part of the settlement. Archaeological research has de 󿬁 nedHouse 25 as a large social space, measuring approximately 20 m 2 . It iscompletely open and has no subdivisions. Its rectangular layout has anE – W orientation, as do most of the documented houses in Peñalosa.We studied 󿬁 ve ceramic vessels (Fig. 1) and the associated soil resi-dues preserved inside them. These vessels were selected in accordancewith a series of archaeological criteria, including their good state of   Table 1 Gas chromatography-mass spectrometry parameters.Gas chromatography Mass spectrometryInjector temperature 250 °C Interface temperature 230 °CInitial column temperature 70 °C (2 minute hold)  m/z   range 50 – 520 umaGas  󿬂 ow 1.0 mL min − 1 (helium) Voltage 2235 V Mode Splitless Quadrupole temperature 150 °CColumn HP-5 Mode ScanTemperature gradient 12 °C min − 1 until 250 °C Scan velocity 2.12 scans/s20 °C min − 1 until 290 °CInjection volume 1.5  μ  L Gain factor 3 EV  Fig. 1.  The  󿬁 ve ceramic vessels studied (  25567  ,  25617  ,  25638 ,  25705  and  25789 ).129 E. Manzano et al. / Microchemical Journal 122 (2015) 127  – 136   preservation(completevessels),thecontexttheywerefoundin(alltheceramics were found in primary position on the paved  󿬂 oor of theroom), their possible uses and functions and their position in the foodproduction chain (relations morphology-use/function). These vesselsincluded a medium-sized spherical pot with a convex base, threebowls (one spherical, one semi-spherical and one with open sides),andagobletwithatall,thinstem,themostcharacteristic andemblem-atic element of the El Argar Culture. All these pieces belong to differentphasesinthefoodpreparationprocessandprovideafascinatinginsightinto daily life 4000 years ago.Thegobletisperhapsthemostinterestingofallthesepieces.Archae-ologists believe that in the Argaric Culture, these kinds of goblets wereassociatedwithhighsocialstatusandhierarchicalorganization,becauseof their design and technical complexity. This goblet is especially inter-estingbecausealthoughsimilarvesselshavebeenfoundamongArgaricgravegoods,thisisalmosttheonlygobletthoughttohavebeenusedforeveryday domestic purposes. Chemical residue analysis on this gobletcan provide an insight into the daily life of this culture from the recentprehistory of southern Spain, since it would be the  󿬁 rst Argaric gobletto be analyzed from a domestic context from the recent prehistory of southern Spain.  2.4. Sampling and sample handling  In-situ measurements were carried out using a portable X-ray  󿬂 uo-rescence (pXRF) spectrometer for trace elements from samples  25567  and  25638 . XRF analysis does not require sampling.Small amounts of   󿬁 ne powder were collected for FTIR, ATR  – FTIR,GC – MS, HRMS and GC – C – IRMS analysis by:- scraping the inner surface and the bottom of the 5 pieces of pottery(samples  25567  ,  25617  ,  25638 ,  25705  and  25789 ),- taking the associated soil residues (samples  25568 ,  25706   and  25791 ) found inside 3 of the above pieces —  25567  ,  25705  and  25789 , respectively.The samples analyzed using the attenuated total re 󿬂 ectance (ATR)samplingaccessorywastaken 󿬁 nely groundand did not require prepa-ration. For Fourier transform infrared spectroscopy (FTIR), in transmis-sion mode, the samples were mixed with KBr (sample:KBr, 1:100) andpressedintodiscs.ForGC – MSasmallamountof  󿬁 nepowderfromeachofthe8sampleswasweighedandanalyzed(samples  25567  (0.9264g),  25617  (0.3675g),  25638 (0.2771g),  25705 (0.5235g),  25789 (0.3678g),  25568 (0.9772g),  25706  (0.9961g)and  25791 (0.9919g)).Eachsamplewascrushedandgroundinanagatemortarforchromatographicanaly-sis. For GC – C – IRMS similar amounts of samples were processed. Thistechnique was only applied to samples  25706  and  25791 duetotheun-availability of samples. To avoid contamination, researchers wore kidgloves and a mask when handling the vessels and the samples.  2.5. Sample treatment  2.5.1. Extraction and derivatization The samples were examined for lipids and proteins. Organic resi-dueswereextractedfromthepotteryusingtheprocedure 󿬁 rstpresent-ed by Evershed et al. [5], which we optimized for these purposes,assayingvarioussolvents,mixedindifferentproportions,namelymeth-anol, dichloromethane, chloroform and toluene. The best results wereachieved with mixtures of methanol:dichloromethane (1:2) andmethanol:chloroform (1:2). Due to the high toxicity of chloroform,the methanol:dichloromethane mixture was selected as our extractionsolvent. Proteins were extracted using a procedure based on Chertovet al. [20]. The extracts we obtained were  󿬁 ltered through 0.22  μ  mnylon Millipore  󿬁 lters before injection into the high resolution massspectrometer (100  μ  L).So as to analyze lipid matter and fatty acids by gas chromatographycoupledmassspectrometry,aderivatizationreactionwasrequiredpriorto injection into the chromatograph. Meth-Prep II was selected as thederivatization reagent because it can derivatize both fatty acids andlipid materials simply in a single step without further collateral reac-tions. The derivatization procedure was based on a procedure for char-acterizing drying oil in paintings that we developed and successfullytested in previous research [21].  2.5.2. Analytical procedure The analytical procedure for GC comprises the following steps:1) 0.3 – 1 g of sample were weighed and placed in an ultrasound vessel;2) 15 mL of the dichloromethane and methanol (2:1  v/v ) mixture wasadded; 3) lipids, fatty acids and other compounds were extractedby immersing the vessel in an ultrasonic bath (Mod 5133 JP Selecta,Barcelona, Spain) for 15 min (twice); 4) the extracts were collectedand centrifuged at 3500 rpm for 5 min; 5) the two extracted liquidswere collected and dried in a nitrogen atmosphere at 50 – 60 °C; 6)500  μ  L of Toluene and 37.5  μ  L of Meth-Prep II were added to carry outthe derivatization reaction in the ultrasonic bath, which took 30 minat ambient temperature; and 7)  󿬁 nally, the derivatized samples wereinjected into the chromatograph (1 μ  L).The procedure for analyzing the proteins by HRMS was based onChertovetal.[20].The 󿬁 nalextractsweredilutedinamixtureofmeth-anolwith0.1%offormicacidandweredirectlyinjected(100 μ  L)intothehigh resolution mass spectrometer. 3. Results and discussion Since the srcin of the organic residues was unknown, we began byanalyzing the samples using infrared spectroscopy. The preliminaryFTIRandFTIR-ATRanalysisprovideda 󿬁 ngerprintwhichsuggestedpos-sibleinitialhypothesesaboutthesubstancespresentintheresiduesandpottery characterization. Due to the complex nature of the organic ma-terials,a combinationofanalyticalapproaches(GC – MS,HRMSandGC – C – IRMS) wasused to ensure comprehensive analysis.Sincefewstudieshave focused on HRMS, in this paper we support the lipid residuesobtained from GC – MS with residues of peptide chains provided byHRMS. Finally, the mineral residue was analyzed by portable non-invasive X-ray 󿬂 uorescence (pXRF).  3.1. FTIR and ATR analysis The FTIR and ATR  – FTIR spectra for the eight residues analyzed arevery similar. Fig. 2 shows the FT-IR spectrum for one of the samples(sample  25789 ) with the main wavenumbers and their correspondingintensities.Allsamplesanalyzedarerichinclaymineralsshowingchar-acteristic bands in the region 3700 – 3600 cm − 1 (O – H stretching bandsat3694andmoreintenseat3619cm − 1 ),1100 – 1000cm − 1 (asymmet-ric Si – O – Si stretching bands) and 910 – 830 cm − 1 (Si – O stretchingbands)[22].Thisbandcontainscontributionsfromvarioussilicatemin-eralstypicallyfoundinclay.Quartzisidenti 󿬁 edbyacharacteristicdou-bletaround779and797cm − 1 andtheSi – Ostretcharound1000cm − 1 .The weak broad band around 3440 cm − 1 and less intensive absorptionat1635cm − 1 inallFTIRspectracanbeattributedtotheO – Hstretchingmode and H – O – H bendingmode respectively from water from crystal-lizationand/orhydroxylgroupsfromtheclay.ACO 32 − bandisobservedat 1420 – 1440 cm − 1 in the IR spectra of sample  25789  and its innersediment  25791  due to calcite remains probably srcinated from soildepositions. The weak band at 1538 cm − 1 in samples  25789  and  25791  attributed to  δ (N – H) and supported by a mild band (N – H) at3127 cm − 1 may be attributed to traces of proteinaceous material.Organic materials (wax traces) could be inferred from the featuresaround 2900 cm − 1 characteristic of C – H stretching in  25789 ,  25638 ,  25705  and  25791 , a  󿬁 nding later con 󿬁 rmed by GC – MS. 130  E. Manzano et al. / Microchemical Journal 122 (2015) 127  – 136 
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks