98
1 tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg Glyptotek Preliminary Report 1, 2009 Preliminary Report 1, 2009 Tracking Colour The polychromy of Greek and Roman sculpture in the Ny Carlsberg Glyptotek — the Copenhagen Polychromy Network

Tracking Colour

  • Upload
    devikke

  • View
    129

  • Download
    2

Embed Size (px)

Citation preview

Page 1: Tracking Colour

1

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

Preliminary Report 1, 2009

Tracking ColourThe polychromy of Greek and Roman sculpture in the Ny Carlsberg Glyptotek

— the Copenhagen Polychromy Network

Page 2: Tracking Colour

Preliminary Report 1, 2009

Tracking ColourThe polychromy of Greek and Roman sculpture in the Ny Carlsberg Glyptotek

— the Copenhagen Polychromy Network

Page 3: Tracking Colour

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

Ny Carlsberg Glyptotek & the Copenhagen Polychromy Network

Tracking ColourThe polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

The Copenhagen Polychromy Network is an interdisciplinary body formed in 2004 on the in-itiative of the Ny Carlsberg Glyptotek to conduct research on ancient sculptural polychromy in the collections of the Glyptotek.

Participants in the Copenhagen Polychromy Network

• NyCarlsbergGlyptotek Jan Stubbe Østergaard, MA, research curator, project coordinator Maria Louise Sargent, project conservator, B.Sc. • TheDepartmentsofPaintingandMonumentalArt,TheSchoolofConservationofTheRoyal

AcademyofFineArts,Copenhagen Mikkel Scharff, M.Sc., conservator, Head of Departments • NordicCenterforEarthEvolution,NaturalHistoryMuseumofDenmark Professor, Dr. Minik T. Rosing • TheInstituteofChemistry,TechnicalUniversityofDenmark,Copenhagen Rolf W. Berg, Ph.D., associate professor

Acknowledgments

We gratefully acknowledge the grant for microscopes and camera given in 2008 byTheKirstenandFreddyJohansenFoundation

Valuable advice and support has generously been provided by

StatensMuseumforKunstTheNationalMuseumofDenmarkLeicaMicrosystemsDenmarkA/S

Editor:JanStubbeØstergaardandtheCopenhagenPolychromyNetworkGraphicdesign:JakobHelmer,MAA,Copenhagen(www.jakobhelmer.com)

Published by the Ny Carlsberg Glyptotek, 2009ISSN:1904-1888Copyright:NyCarlsbergGlyptotekandtheauthors

Copyrightandlicencestatement:www.creativecommons.org/licenses/by-nc-nd/3.0/

Page 4: Tracking Colour

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

Contents

Contents 4

Preface 5

IntroducingtheCopenhagenPolychromyNetwork 6 Jan Stubbe Østergaard

The Copenhagen Polychromy Network pilot project 2004–2008

The CPNPilotProject:abriefintroductionandevaluation 11 Jan Stubbe Østergaard

InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830 13 Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

Preliminary results from geochemical analysis of 41 pigments on ancient Greek & Roman marble sculptures Minik T. Rosing and Jan Stubbe Østergaard

Raman spectroscopy characterization of colored pigments in archaeological materials 48 Rolf W. Berg

The Copenhagen Polychromy Network main project 2008–2011

The CPNMainProject2008–2011:anoutline 68 Jan Stubbe Østergaard

DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCG IN1203 74 Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

AuthorContact 90

Other CPN activities 91

Contents

Page 5: Tracking Colour

5

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

Preface

From2004to2008,theNyCarlsbergGlyptotekandtheCopenhagenPolychromyNetwork(CPN)carriedoutaPilotProjectinthemuseum’scollectionofGreekandRomansculpture.InJune2008westartedtheso-calledMainProjectwhichwillrununtil2011.ThefirstphaseofvisualexaminationbeganinJanuary2009andlasteduntilthissummer.

SomeprovisionalresultsoftheseactivitiesarepublishedinthisfirstPreliminaryReport.The Report also contains an introduction to the CPNandoutlinesofboththePilotProjectandtheMainProject.Theprotocolfollowedinvisualexaminationanddocumentationisnotthesubjectofaseparatearticle,butispresentedinsteadasappliedinpractise.

On a practical note, the system of citation and abbreviation in this Report is not homogenous.

This first Preliminary Reportwill be followed by (at least) twomore, to be published in2010 and 2011. The format will be electronic and the publications will be freely available for downloadfromtheGlyptotekwebsiteatwww.glyptoteket.dk.Atpresent,financialrestric-tions allow only a quite unsophisticated pdf-solution.

IthasbeenourideaforsometimethatthefollowingReportsshouldhavetwosections:Sec-tionIwithpreliminarypublicationoftheactualprojectresults;andaSectionIIcontainingother contributions on ancient sculptural polychromy, whether related to our results or not.

Wefullyrealizethattheadventofbibliometricalmethodsofacademic‘rating’willmakeitevenlessattractivetopublishinaforumsuchasours.ItisthereforeourambitiontoseethefollowingReportsappearase-publicationsundertheauspicesofanaccreditedDanishacademicpublisher.Suchapublisherhasalreadybeenfoundandinitialdiscussionshavetakenplacewith positive results.Anumber of colleagues on our international advisorypanelhaveagreedtoundertakepeer-reviewingofarticlesappearinginSectionII.Itremainsto be seen whether the necessary funding can be found.

Asitis,weinvitecontributionstosectionIIofthePreliminaryReport2,2010,tobereceivednolaterthanMay1,2010.EditorialguidelineswillbemadeavailableonrequesttotheEditor.

ItgoeswithoutsayingthatanycriticalcommentsonthisfirstReportandsuggestionsforimprovement will be much appreciated.

Our thanks go at this point to all those within and outside the Ny Carlsberg Glyptotek who havecontributedtobringingourprojectthusfar–especiallytotheKirstenandFreddyJo-hansenFoundationforprovidinguswithmicroscopesandcamera,toLeicaMicrosystemsDenmarkA/SandLarsHvalsumforuntiringsupportandadonationtotheproject.Moreo-ver,totheStatensMuseumforKunstwhichhasgenerouslylentusglasspartitionsfortheprojectworkspace.Finally,weexpressourappreciationoftheinterestandsupportshownbythemembersofourInternationalAdvisoryPanel.

JanStubbeØstergaardEditor

On behalf of the Ny Carlsberg Glyptotek and the Copenhagen Polychromy Network

Page 6: Tracking Colour

6

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

IntroducingtheCopenhagenPolychromy Network

Jan Stubbe Østergaard1

TheCopenhagenPolychromyNetwork(CPN)wasformedontheinitiativeoftheNyCarlsbergGlyptotek,Copenhagen,inthewakeoftheexhibition‘ClassiColor–thecolourofancientsculpture.’TheexhibitionwasshownatthemuseumfromMarchtoMay,2004.2

Factors determining the establishment of the network

The decision to form the network was taken in the light of a number of different, but inter-relatedfactors:

• Theexhibitionprojectmobilizedawarenessoftheimportanceofpolychromystudieswith-in the museum.

• Theexhibitionwassuccessfulinreachingawideaudience,intermsofbothvisitornumbersandmediacoverage.Amongotherthings,thisservedtogenerateinterestandunderstand-ingamongavarietyofimportantaudiences,suchasinstitutionswithscientificresourcesrelevanttothisfieldofarchaeologicalandarthistoricalresearch.

• ThecollectionofancientsculptureintheNyCarlsbergGlyptotekhas,since1991,beenthesubjectofsystematicdocumentationresultinginadigitallyproduced,printedcataloguese-riesavailablealsoinanEnglishversion.3Relevantdataonthecollection,intheformoftextsand high quality photography, are therefore now accessible to international scholarship. Thecataloguecontentsareintheprocessofbeingmadeavailableviaanobjectdatabaseonthemuseum’swebsite,www.glyptoteket.dk.

• Theonlyaspectofthecollectionnotyetsystematicallyexaminedanddocumentedisthatof traces of ancient polychromy.4

• Thecollectionshavegoodpotentialforprovidinginformationonancientsculpturalpoly-chromy.ManyworkswereacquiredontheartmarketinRomefromthemiddleofthe1880sto the early years of the 20thcentury.InthesameperiodacquisitionsweremadeontheartmarketinMunichandinAthens.Afairproportionoftheseacquisitionswererecentfinds.Dealerswereawareof the importanceplacedbytheGlyptotekonacquiringworks inaspristine a condition as possible.

Overtheyears,subsequenttreatmentintheGlyptotek’sdepartmentofconservationhasseen the application to many pieces of methods of cleaning which would not be acceptable todayandcastshavebeentakenofanumberofimportantworks.Itislikelythatthishascausedsomedamage.Apartfromworksacquiredfromoldcollections,ancientsculptures

1 Researchcurator,ancientart,NyCarlsbergGlyptotek,DantesPlads7,DK-1556Kø[email protected].

2 ThiswastheCopenhagenversionoftheexhibitionshownfirstintheStaatlicheAntikensammlungenundGlyptothek,Munich,in2003(BunteGötter.DieFarbigkeitantikerSkulptur)andlaterintheMuseiVaticani,in2004/5.UndertheauspicesoftheStiftungArchäologie,Munich,andthroughtheeffortsofVinzenzBrinkmannandUlrikeKoch-Brink-mann,continuallyrevisedandexpandedversionsoftheexhibitionhavesincebeenpresentedatanumberofothervenues.Thecontributionthusmadetogeneratingbothpublicandprofessionalinterestinthesubjectcanhardlybeoverestimated.

3 ThecatalogueseriesisavailablefromthemuseumeShopatwww.glyptoteket.dk

4 AsisthecaseforallothermajorcollectionsofGreekandRomansculpture.

Page 7: Tracking Colour

7Introducing the Copenhagen Polychromy NetworkJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

havehowevergenerallynotbeensubjectedto‘rigorous’cleaning.Manypiecesretaintracesof polychromy.5

• Themuseumhousescollectionsofpolychromeworks inotherart formsandfromothercultures whose relevance as comparative material is becoming increasingly clear. We have Greek,EtruscanandRomanterracottas;Etruscanstonesculpture,Egyptianstonesculptureof all periods aswell as Egyptian terracottas of theHellenistic and Romanperiods andpaintedmummyportraitsfromRomanEgypt.

• Themuseumhadbecomeawareofthefactthattracesofancientpolychromyonthestonesculptures in its care were deteriorating and have in some cases disappeared. The processes atworkinthedeteriorationofancientpigmentsarenotyetbeenpreciselyunderstood.Ac-tion to document traces of polychromy in the collection was therefore agreed to have a high priority.

• Thedisciplinesneededtoconductresearchinancientpolychromyinamuseumcollectionwas available in Copenhagen.

TheabovementionedfactorsdeterminedtheNetwork’saimsandstructure.

TheNetwork’saims

The CPN’s aim is to contribute to an increase of data available to the study of ancient sculp-tural polychromy through systematic documentation of traces of polychromy in the collec-tionofGreekandRomansculptureintheGlyptotek,in-depthconservationscientificandnaturalscientificstudyofsculpturesidentifiedashavingspecialimportance,andpublica-tion of the acquired data. Itwasdecidedthattheprojectpublicationsshouldappearinopenaccesselectronicfor-matswhosesophisticationwoulddependonavailablefunding.Ahighdegreeofinteractiv-ity with readers was aimed for.

TheNetwork’sStructure: Participating institutions and competences

Theparticipatinginstitutionsare: • TheNyCarlsbergGlyptotek • TheDepartmentsof Painting andMonumentalArt at theSchool ofConservationof the

RoyalDanishAcademyofFineArts,Copenhagen • TheGeologicalMuseumandInstituteofGeology,UniversityofCopenhagen(NaturalHistory

MuseumofDenmark) • TheInstituteofChemistry,TechnicalUniversityofDenmark,Copenhagen.

5 Visibleremainsofpigmenthavebeenobservedonc.120GreekandRomansculptures,fromminiaturetocolossalformats.Sinceexperiencehasshownthatcloseexaminationrevealsremainsnotseenbythenakedeye,thenumberofsculpturesinthecollectionthatarerelevanttothisfieldofresearchiscertainlyhigher,butdifficulttoestimate.

Page 8: Tracking Colour

8Introducing the Copenhagen Polychromy NetworkJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Together, these institutions provide the competences needed in this field of research,namely:6

• Classicalarchaeological,curatorialcompetence(1).AresearchcuratorofancientartattheNyCarlsbergGlyptotekdirects theproject,and is responsibleprojectmanagement, strategyandprioritiesincollaborationwiththemuseum’scuratorialdepartmentanditsDirector.Heis in charge of archaeological documentation, contact with museum conservation, network partners and an international collegial network.

• Museumconservationcompetence(2). This member of the team is in charge of a variety of con-servationtasks:Visualexaminationanddocumentation(ambientlight,rakinglight,IRandUVphotography;microscopy;cross-sectionpreparationandinvestigation;documentationofsamples;organisationofdigitaldata),anddialoguewiththecuratorialdepartmentandexternalpartnersinconservationscience.

• Conservationandnaturalscientificcompetences(externalnetworkpartners)representedby(3)anexperiencedconservator,contributinginthefieldofscientificconservationphotography;instrumental pigment analyses (SEM/EDX; FT-IR), chemical spot tests and pigment andbinding media analysis. The network partners in natural science offer SEM/EDX,SEM/XRD,XRF and ICP-MS (InductivelyCoupledPlasmaMassSpectrometry)forpigmentidentificationand pigment localisation through isotopic trace element analysis and comparative analyses ofprovenancedspecimens(4),andchemicalanalysesusingRamanLaserSpectroscopy(5).

Theseexternalnetworkpartnersoffertheirservicefreeofcharge.Theyalsoonoccasionactivatetheircontactsinotherfields,inDenmarkandabroad.

TheCPN’soverallstrategy

Asimpleoverallstrategyconsistingofthreestageswasestablishedattheoutset.First,tocarryoutaPilotProjectstudy,proceedingastheresourcesoftheCPN might allow. This was thentoprovideaplatformforstagetwo,amoreambitiousMainProject,plannedaccordingto the resources that might be allotted by the museum or raised by other means. The last stagewastocompriseafinalpublicationoftheaccumulatedresultsandthecommunica-tionoftheseresultstothepublicintheformofanexhibitionattheNyCarlsbergGlyptotek.

Progress hitherto

Asitturnedout,byearly2008theresultsachievedbythePilotProjectledtoadecisionbythemuseumtogoaheadwiththeMainProject.ThePilotProjectandsomeofitsresultsaredealt with elsewhere in this Report. Themuseum’smaincontributiontotheprojectwasafull-timepositionasresearchcu-ratorforthethreeyears2008–2011,andahalf-timepositionasprojectconservatorforthesameperiod.Applications foradditional fundinghavenotso farmetwithsuccess– theexceptionbeingagenerousdonationmadebytheKirstenandFreddyJohansenFoundationfor the acquisition of instruments.

6 In the following, thepersons involvedare referred tobynumbers inbrackets.Theyare (1) JanStubbeØstergaard,research curator,MA; (2) Conservator RebeccaHast and subsequentlyMaria Louise Sargent BA conservation andclassicalarchaeology,postgraduatestudent inconservation, (3)MikkelScharff,HeadofthePaintingsDepartmentandoftheMonumentalArtDepartment.SchoolofConservation,TheRoyalAcademyofFineArts,Copenhagen(withattachedadvancedstudents),(4)Prof.dr.MinikRosing,NaturalHistoryMuseumofDenmark,(5)AssociateprofessorRolfW.Berg,Phd,InstituteofChemistry,TechnicalUniversityofDenmark.

Page 9: Tracking Colour

9Introducing the Copenhagen Polychromy NetworkJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

TheMainProjectisoutlinedinaseparatearticle,togetherwithareportonthefirstsculp-turetohavebeenexamined. Workonthefinalpublicationistotakeplace2011–2012andtheexhibitionisscheduledforthespringof2013.

Challenges

Of the challenges facing the CPN’s activities, the greatest is that of ensuring the contin-uation of coherent and systematic research on ancient sculptural polychromy in the Ny CarlsbergGlyptotekafterthecloseoftheMainProject. IthasbynowbecomeclearthatwithintheframeworkoftheMainProject,itwillonlybepossible to investigate a limited number of the sculptures which can supply us with evi-dence of the polychromy of Greek and Roman sculpture – furthermore, the polychromy of themuseum’srichholdingsofEgyptianandEtruscansculptureremainsunexplored. Ifwedonotsucceedinmeetingthischallenge,theresultsachievedbytheCPN will thus standasatorso;theexperienceandexpertiseacquiredbytheCPN will dissipate, and, worst ofall,wewillbeunabletoofferyoungerDanisharchaeologistsandconservatorsaplatformallowingthemtotakeoverfromthoseleavingthefieldintheforeseeablefuture.Tonourishresearch‘Nachwuchs’isessential,asalways.

Despite the seminal contributionmade by such scholars asVinzenz Brinkmann, UlrikeKoch-Brinkmann,PhilippeJockey,BrigitteBourgeoisandmanyothers,researchinancientsculpturalpolychromystillfindsitselfinaformativephase.Itsinterdisciplinarycharacterandtheconsiderableresourcesitrequires,makesitthemoredifficulttofindafootingofanypermanenceathumanisticacademicinstitutions.Itisindicativeofthisthat,asfarasIknow, the only organization wholly dedicated to sculptural polychromy research is a foun-dation,namelytheStiftungArchäologieinMunich.7

Visions

The CPN should therefore meet the challenge described above by developing visions of how theresearchithasinitiatedmaybecontinuedinthefuture.Suchvisionsmustbebasedonsomeestablished,keyfactors:

• ThesizeandresearchpotentialofthecollectionsofancientsculptureintheNyCarlsbergGlyptotek.

• Themuseum’slong-standingpolicyofwelcomingresearchonitsholdingsbyscholarsfromoutside the institution and allowing open access to works in the collection – combined with an open-minded attitude to the taking of samples for analysis elsewhere.

• Theavailabilityofrelevantresearchfacilitiesatthemuseum,includingafirst-ratelibrary,officespaceforvisitingscholars,amini-auditoriumforscholarlyresearchmeetings,aswellasworkspaceandequipmentforvisualexaminationanddocumentation.

• Thenot inconsiderable international, interdisciplinarynetworkofcontactsdevelopedbythe CPN.

• Theexistenceofup-comingDanishconservatorsandarchaeologistswiththenecessarytal-ent and enthusiasm.

7 Foundedin2005ontheinitiativeofVinzenzBrinkmannandUlrikeKoch-Brinkmann(www.stiftung-archaeologie.de).

Page 10: Tracking Colour

10Introducing the Copenhagen Polychromy NetworkJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

An international research centre in Copenhagen

Taken together, these key factors allow one to envision the creation in Copenhagen of a re-searchcentreforancientsculpturalpolychromy.Inoutline,suchacentrecouldbedescribedasfollows:

• ThecentrehasitsbaseintheNyCarlsbergGlyptotek.Itisinterdisciplinaryincharacterandinternationalinoutlook.Itfocusesitsresearchonancientsculpturalpolychromyasrepre-sented in the collections of the Ny Carlsberg Glyptotek. The centre conducts a research pro-gramme, including research training of young scholars, and a variety of activities connected with ancient sculptural polychromy. This involves collaboration with other museums and relevant universities. The centre issues relevant publications and organises scholarly meet-ings. The possibility of offering courses on aspects of ancient sculptural polychromy on rel-evantacademiclevelsshouldbeexplored.

• Thecentreisdirectedbyaclassicalarchaeologist,amuseumstoneconservatorandacon-servation scientist.

• Themissionofthecentreistopromote,expandandconsolidateresearchonancientsculp-tural polychromy in the academic world and communicate research results to a wider pub-lic.Thecentreaimsto:

1 Establishavisibleandexplicitresearchprofilewithnationalandinternationalreferences,setting standards for future research on ancient sculptural polychromy.

2 Exploreandconsolidateinternationalknowledgeofancientsculpturalpolychromybyinvit-ing colleagues from abroad to conduct research in the collections and by establishing col-laborationwithresearchprojectsinothermuseums.

3 Achievenewresultsbyconductingresearchprogrammes,andbyinspiringyoungscholarsand scientists to include ancient sculptural polychromy in their research.

• Incarryingoutthismission,thecentrecollaboratescloselywiththeCPN and an international advisory board.

Someof theelementsof thisvisionwerepresentonamodestscaleatan international,interdisciplinary Round Table on ancient sculptural polychromy held in the Glyptotek in September2009.Acoupleofparticipantsarrivedsomedayspriortothemeetingtoexam-ine sculptures in our collection. The sculptures were brought to work spaces set up for the purpose.Inconnectionwithasecondroundtablemeetingnextyear,onemightimaginethisactivityexpanded.Atthemeetingitself,scholarsdirectlyinvolvedinresearchprojectspresentedanddiscussedresults.Participation,byinvitationonly,includedDanisharchae-ologist and conservators.

Todiscuss,develop,adjustandrefinesuchvisionsforthefutureisavitalpartoftheworknow being done by the CPN and the Ny Carlsberg Glyptotek.

Page 11: Tracking Colour

11

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

The CPNPilotProject:abriefintroduction and evaluation

Jan Stubbe Østergaard

TheprovisionalresultsofthePilotProjectarereportedoninthearticlesbelow.Here,Ire-strictmyselftosomegeneralremarksontheaimsoftheprojectandsomecommentsinthe way of evaluation.

Threesculpturesfromourcollectionswerechosenforstudyfollowingtwocriteria:thepres-ence of traces of ancient polychromy, and the potential for throwing light on sculptural polychromyinpoorlydocumentedperiods–namelyClassicalandHellenisticGreekoriginalsculptureandRomanidealsculpture.Throughthestudyoftheseworks,wehopedto:

• Acquireexperienceinstructuringandoperatinganinterdisciplinarynetwork. • Developaprotocolforvisualexamination. • Acquiredataontheexpenditureinvolvedinvisualexaminationanddocumentationand

in-depthconservationalandnaturalscientificstudy,intermsoftime,personnelresourcesandinstrumentalexaminationcosts.

• ProduceprimaryresearchdataontheobjectschosenforthePilotProject. • Establishcontactwithothersworkinginthefield. • Provideaplatformforamoreambitious,longtermresearchprogram.

Lookingback,thePilotProjectserveditspurposeinmostrespects.

However, inthematterofoperatingthenetwork,examinationofthesculptureswasnotcarriedoutinaworkspacespecificallydesignedforthepurpose,butatvariouslocations–in situ in the galleries, in the museum conservation workshop, in our photo studio and in workspacesattheSchoolofConservation.Asaresultwedidnotdevelopaclearideaofhowtoestablishandequipadedicatedworkspaceinthegalleries,fortheMainProject.Thiswaseventually to cost us time in preparing that facility.

Asfarastheplanningofnon-invasiveorsampleanalysestobecarriedoutby,orusingtheinstrumentationof,ourexternalpartnersinconservationandnaturalscience,wedidnotcome to fully recognize a basic condition. Namely that such analyses could only take place totheextentallowedbythe internalworkprogrammesofthe institutions involved,andalreadyhardpressedinthematterofresources.Forperfectlyunderstandablereasons,theywerenot–andarenot–inapositiontoallotagreedresourcestoaprojectsuchasourswiththe regularity needed for any kind of planning. This in no way affects our gratitude for the unstintingexternalnetworksupportgenerouslygiventotheproject. WithintheframeworkoftheMainProject,appropriatemeasureshavenowbeentakentomakethebestpossibleuseoftheresourcesofexternalpartners.Inpractise,thismeansthebatchingoftheanalysesweneed,ratherthanwaitingforanalyseswhenexamininganindividual sculpture.

Furthermore,wedidnotgainanyproperideaoftheamountoftimetheexaminationofanindividualsculpturewouldinvolve.DuringthePilotProject,workwasdoneadhoc,ascon-ditionspermitted,andnologwaskept;thedevelopmentofaprotocolwentaheadashoped,but in a way which did not result in an understanding of the time required to implement theprotocol.ThishasinfluencedtheplanningoftheMainProject,whichhasturnedouttobeunrealisticallyoptimisticasregardsthenumberofobjectswewouldbeabletoexamine.

Page 12: Tracking Colour

12TheCPNPilotProject:abriefintroductionandevaluationJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Asaresult,ashorter,so-calledSurveyProtocol,hasnowbeendevelopedto increasethenumberofsculpturesexaminedduringtheMainProject.

TheexperiencegainedduringthePilotProjecthasservedtostrengthentheconvictionthatthe way forward in the study of ancient sculptural polychromy is to establish committed interdisciplinarynetworksformedaroundmajorcollectionsofancientsculpture, intheirpluralityforminga‘networkofnetworks’insynergeticcollaboration.Ithasalsoservedtoclarify that such networks will be of at least two different types. The CPNisanexampleofanetworkmadeupofsomecompetencesexistinginthemu-seum,whileothersareprovidedbyexternalpartnerinstitutions–withthelimitationsthisentails. Other collections are in the keeping of museums which have the prerequisite inter-disciplinary organisation in-house and are therefore able to allot resources in a more organ-ized manner once the decision has been taken to give polychromy research the necessary priority.Networksofthistypearelikelytobecometheleadersinthefield. Somemuseumswill have to establish networks like that of theCPN, created through decisions taken by the keepers of their collections and their conservators, and by those in charge of institutions of conservation science and natural science. Inbothtypesofnetwork,decisionmakersarefacedwithmanydifferentwell-reasoned,urgentapplications forallotmentof resources toavarietyofprojects. It comesdowntoestablishingpriorities. Inthissituation,thecasefor interdisciplinaryresearchinancientsculptural polychromy must be forcefully made.

ThePilotProjectwassuccessfulincementingnetworkcollaborationandproducingresearchdata.Withoutthis,itwouldnothavebeenpossibletofulfiltheaimoflaunchingtheMainProject.

Page 13: Tracking Colour

13

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830

Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard 1

One of three sculptures selected for the CPNPilotProjectwasafragmentarymarbleheadofa woman2(Fig.1),chosenasarepresentativeoftheGreekClassicalPeriod.Thesuggesteddate of c. 425 BCandproposedAtticoriginisbasedpartlyonstylisticanalysis,partlyontheAthenianprovenanceofthehead.Aspreserved,theheadis23cmhigh,16cmwideand21cmdeep;ifitcomesfromastatue,theheightofthiscanbecalculatedtoc.2m,orslightlyover life size.3Themarblehasbeenidentifiedbyisotopicanalysisas‘probablyParianLych-nites;Ephesianapossibility.’4

1 MikkelScharff,conservator,HeadoftheDepartmentsofPaintingandofMonumentalArt, RoyalDanishAcademyofFineArts,TheSchoolofConservation

JanStubbeØstergaard,researchcurator,Ancientart,NyCarlsbergGlyptotek,Copenhagen. JanStubbeØstergaard’scontributionislimitedtothearchaeologicalcomment.

RebeccaHast,stoneconservator,MScinConservation,Copenhagen NicolineKalsbeek,chemist,ass.prof.,Dr.,formerlyTheSchoolofConservation,

nowTeamleaderNovozymes,Bagsværd,Denmark.

2 iN2830.Poulsen1941,pp.111–118,pl.13;Poulsen1951,p.219,no.298a;Moltesenetal.1995,p.70,no.16.

3 Poulsen1941,p.112;Berger1956,p.171,no.11foradditionalmeasurements.

4 Sampletaken1993forisotopicanalysis.Result:d13C,5,13–d18O,-3,26;Poulsen1941,p.112‘Pentelic’

Fig.1:NyCarlsbergGlyptotek IN2830.

Page 14: Tracking Colour

14InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Introduction – technical examination, pilot study

Prior to selecting relevant examination techniques for the Pilot Project, literature on ex-amination of the polychromy of Greek and Roman sculpture was studied.5 The technical examinationofancientpolychromyhasconcentratedonstudyingthecompositionofpaintfragments and the traces they left of the stone surface. The authors have summarized their findings of pigments, establishedhypotheses concerning the original appearance of thesculptures and tried to verify these hypotheses by reconstructing the polychromy of the sculptures.6

The published studies of ancient polychromy7 left us with the impression that most painted surfaces were made with paints composed of a selection of a total of 15–20 basic pigments (withvariations)andafeworganiccolorants8madeintolakes/dyesandallmixedwithbind-ing media of a rather limited variation but including animal glue.

Apparentlythesamesetofpigments,bindingmediaandstratigraphywerefoundintheancientGreekandRomansculptures, inotherancientcultures (e.g.Xi’an9)aswellas inEuropeanmedievalpolychromy.10 This led us to assume that we may trace the origin of medieval painting technique in ancient sculpture. This view is supported when studying the materials and the development of some of the recipes in medieval treatises from the 8thcenturyMappaClaviculato12th century Theophilus.11

ThetypicalstratigraphyinEuropeanmedievalpolychromyonwoodwouldbeaso-calledground structure, covered with one or two layers of paint. Occasionally metal leaves and somemorespecializedmaterials (e.g. semi-preciousstone)wereapplied to thesurfaces.Thisisalsotheruleformedievalpaintingordecorationofstonesupports.Scharffhaspre-viously suggested that the origin of certain kinds of ground structures (lead-containinggroundswithdryingoilasbindingmedium)inEuropeanmedievalwoodenpolychromyhasbeen developed from using the same composition on stone surfaces.12

Sinceliteratureontheexaminationofancientpolychromyisrelativelylimiteditwasdecid-edthattheexaminationtechniqueschosenforthepilotstudyweretobebasedonanexist-ingtraditionofstudyingpolychromyonEuropeanmedievalpolychromewoodensculptureand panel paintings,13aswellasthegeneraltechniquesdevelopedforexaminingpaintlay-

5 Bourgeois – Jockey 2005; Brinkmann 2003; Brinkmann –Wünsche 2003 (version 2008 from Liebieghaus Skulptu-rensammlung, Frankfurt 2008); Chrissoulakis – Queyrel – Perdikatsis 1989; Kakoulli – Kottaridou – Minos 2001; Graeve–Preusser1981;Nielsen–Østergaard2004;Santamaria–Morresi2004;Santamaria–Morresi–DelleRose2004;Wallert 1995.

6 Brinkmann–Wünsche2003passim.

7 Cf.note11;Thieme–Emmerling2001,pp.335–369;Wu–Zhang–Petzet–Emmerling–Blänsdorf2001.

8 Grzywacz–Su–Fan–Wouters2008,pp.534–541.

9 Cf.note13.

10 Plahter 2002, pp. 446–454.

11 Hawthorne–Smith1974;Hawthorne–Smith1979.

12 Scharff1999,pp.47–52.

13 Nadolny2006;Tångeberg1986;Kühn–Straub1984,pp.7–54,pp.125–260;Scharff1995,pp.83–91.

Page 15: Tracking Colour

15InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

ers on various supports.14Publishedliteratureofpaintingsonstoneorplaster(howevernottraditionalwallpaintingsinlimeorfrescotechniques)wasalsostudied.15

Whiletherewereexpectationsaboutthepigmentstobefoundinanexamination,theusesof binders for the pigments are less well established and we had fewer prior ideas about what binders might be found during analyses. Proteinaceous glue and egg tempera have beenanalyticallyidentifiedinsomecases,whilePliniustheElderdescribesatechniqueus-ingwax(encaustictechnique).16InEuropeanmedievalsculpturalpolychromyavarietyofbinders were used17andacertainsystemseemstohavebeenapplied:ifapaintedsculpturewastobeleftoutsideitwaspaintedinasomewhatweatherresistanttechnique(e.g.adry-ingoilbinder)whilewatersolubletechniques(e.g.proteinaceousglue)wassometimesusedandacceptedforindoorsculpture;othertechniquesincludedeggtemperaoroil.ThesamemayormaynothavebeenthecaseinGreekandRomanpolychromytechnique:however,itwouldmakesensetouseaweatherresistanttechniqueoutofdoors.Ifso,itmightbecomepossible – by determining the nature of the binding medium – to be able to suggest the originalpositionsofasculpture(outsideorinside:weatherproofpaintingtechniqueornot)whenthisisnotknownaboutaspecificobject.

Other features of antique paint layers not very well understood are the surface appear-ances:mattorglossy,opaqueortranslucent,orvariationsonthesameobjectasitissome-times seen on medieval sculpture? The state of most of the surviving paint fragments or painted areas on ancient sculptures are probably not representative or reliable concerning thesurfaceappearance.Itisassumedthatthesurfaceshavebeendegradedorchangedinnumerous ways and for a number of reasons.

Someproblemsarerelatedtoagenerallackofprovenanceintheolderpartofmanycollec-tions. Greek and Roman sculptures in old collections have been acquired from dealers, the sculptures are generally without provenance and thus the condition of the sculptures when foundisnotknown.Whenpainttraces/samplesarelocatedonthesurfaceitisnotpossibleto ascertain whether they were present at the time when the sculpture was found or if paint orothersurfacetreatmentshavebeenadded(deliberatelyornot)aftertheobjecthadbeenfound. Neither is it known whether the surfaces of the paint traces are as when found or they have been tampered with. Asaworkinghypothesisforthepilotstudyweassumedthatpaintlayers(compositionand stratigraphy) on polychrome sculpture does not differ widely between various geo-graphicallocations,culturesandtimelines.Ifso,weshouldbeabletousethesameexami-nationtechniquesforancientpolychromyasforEuropeanmedievalpolychromyandweassume that stratigraphic features on Greek and Roman sculpture could be interpreted and comparedwithEuropeanmedievalpractice.WhenwestartedthePilotProjectexaminationwethusexpectedtofindpigments,bindingmedia,stratigraphy,andperhapsevenmetalleaveswhichappearinpublishedliteratureonancientaswellasEuropeanmedievalpoly-chromy.Ifsuchsimilaritiesweretobefound,wemightgainanideaoftheoriginalsurfaceappearance of ancient polychromies.

14 Khandekar2003,pp.52–64;Autenrieth1990,pp.215–233;Walmsley1993,pp.57–62.

15 Katz1998,pp.27–33;Sauerberg2003,pp.189–199.

16 NaturalisHistoriaXXXV,149.

17 Bindingmediumsidentified–andreferredinEupopeanmedievalpaintingtreatises(seee.g.note11)–includepro-teinaceousglue,starch,waxes,(drying)oils,egg,gums,perhapsresins,resinouscompoundsorbalsams–ormixturessuch as in various types of tempera.

Page 16: Tracking Colour

16InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Wehope–throughfurtherexaminationsandanalyses–togain insight intotheoriginalappearance of the ancient polychrome sculptures, and through this to better understand the present state and deterioration phenomena. This would enable us to assess the risk of further deterioration and establish a strategy for future preventive means of protection.

ThemedievalpolychromyexaminationtechniquesappliedinthePilotProjectaredescribedin the following.

Methods & materials

TheexaminationofthefemaleheadintheNyCarlsbergGlyptotek(IN2830)isdividedintovarioussteps:visualexamination/analysis,photographicdocumentation,paintsampling,light and UV-fluorescence (UV-FL)microscopy, scanning electronmicroscopy (SEM) withEDX and SEM imaging, Fourier transform-infrared spectroscopy (FT-IR), and gas chroma-tography-massspectrometry(GC-MS).

visualexamination/analysis

Thevisualexamination/analysisconsistedofavisualexaminationwiththenakedeyeofthesurfaceoftheheadbeginningwiththefaceandsubsequentlyturningaroundandex-aminingthesculptureclockwise360degrees.Thevisualexaminationwassupportedbyamagnificationglassandtheuseofanoperationmicroscope18toexamineindetailvariousphenomena.Thelightandradiationtypesusedatthisstagewere:tungstenlightinsym-metricsetup(inordertodistinguishvisiblecolourdifferencesonthesurface);rakinglightoverthesurfaceinordertoexaminesurfacephenomenasuchastoolmarks,incisionsandtracesofpaint;UV-radiation causing UV-FLtoexaminethesurfaceforsurfacefluorescencephenomena.19

Photographicdocumentation(digital)andexamination

Systematicdocumentationwasmadeofthesculptureandfurthermorephotographyofthedetail phenomena observed during the visual examination, supplemented by examina-tionanddocumentationbymeansofinfraredreflectography(IRR)andUV-FL. Photographic documentation in colour20wasmadeofthesculpturein6stages(beginningwiththeface)and subsequently in 60 degree steps. One image using a colour reference standard21 for sub-sequent colour balancing was made for each new set-up, and grey background used on all images.DigitalphotographsweresavedinthecameraasRAW-files,post-processed,colourcorrected and saved as 16 and 8 bit tiffimages.Inallcasesthephotographsweremadeusingfixedexposuretimeandaperture.Exposuredetailsofthephotographsareembeddedas EXIF-data with the digital tiff images.

18 Operationsmicroscope,CarlZeissJena.

19 The UV-radiationsystembuildsontheprinciplesdescribedinthepaper:Autenrieth1990,pp.215–233.

20 ANikonD100digitalcamera(3008×2000pixelsCCD-array),60mmMacro-Nikkorlenswasusedforcapturingphoto-graphs, symmetrically positioned and balanced halogen light provided form and shadows.

21 GretagMcBethColorChecker® chart, a 24-patch reference standard to reproduce colours.

Page 17: Tracking Colour

17InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Ultravioletfluorescence(UV-FL)photography

The same 6 images were taken under UV radiation22 as well as some details. The photo-graphs were saved in the camera as RAW-files,post-processedandcolour-correctedusingagrey card and saved as 16 and 8 bit tiff images.

IR-Reflectography(IRR)

AnIRR-camera system23wasusedtoexamineareasonthesurfacewherecolour/paintwasfound in order to subsequently document whether the paint or surrounding areas provided information otherwise invisible.

Paint sampling

Foursampleseachwithasurfaceareaofcirca0.25–1mm2 were taken, one from each of four surfaceswithpigmentremains.Approximatelyhalfofeachsamplewasembeddedinpoly-ester resin and prepared for paint cross sections.24 The remaining parts of the four samples were left for subsequent analysis.

Microscopy

The fourembeddedpaintsamplesascrosssectionswerestudied indarkfield lightandUV-fl microscopy25usingtwosetsoffilters26inordertoexaminethestratigraphyandob-tainthefirstideaofthepigmentsandfluorescencephenomenafrompigmentsand/orbind-ing media. The cross sections were subsequently photographed.27

SEM/EDAX

To determine the chemical elements present in the various layers in the samples, a JEOL 5310 LV SEM with a Link EDXunitwasusedfortheanalyses.Lowvacuum(LV)wasappliedwithapressureofapprox.20Pa,theaccelerationvoltagewas15kV.Backscatter(BSE)pic-tures were taken of all samples. Contrast differences in BSE pictures are caused by relative differencesinatomicnumberZoftheelementsconstitutingthesample.Heavierelements(higherZ)appearlighterandbrighterinthepicturethanelementswithlowerZ.DuringEDXanalysis single elements are mapped or monitored across the sample surface, the lighter

22 Camera,seenote27;UV-radiationsystem,seenote25;filtersinfrontoflens:Wratten2A(cuttingaround400nm),Wratten 2A + Wratten 85B(cuttingoffsomeofthebluepartofthespectrum).

23 IRRcamera(InframetricsInfraCamSWIR®)witha256x256pixelPtSi-detectorarray,lowerthreshold1100nm,equippedwithabandpassfilter1500nmto1800nmandassourceofIRradiationatungstenbulbwereusedinthestudy.Viaa frame grabber, the 8-bit signals were transferred to a computer, subsequently processed in the image processing software VIPS/nip2(GNU)andsavedas8bittiff images.

24 PaintsampleswereembeddedinSerifix® polyester resin with MEKPhardener,curedat50°C,preparedwithSi-basedwet-grinding paper 240, 800, 1000, 1200mesh anddry-polishedusing 4000mesh; seeChristensen – Scharff 1986,pp. 10–12.

25 ZeissAxiotech 100HDmicroscope, ×20/0.40 EpiplanHD lens, used indarkfield technique,halogen light andUV-radiation.

26 3SP395+LWP420,FT510+LP520.

27 Olympus OM-1camera,Ectachrome64ISO and 200 ISOfilm(EPY for tungsten light, EPD for UV-FL),autoexposure;thecolourslidefilmsweredevelopedandafterwardsscanned.

Page 18: Tracking Colour

18InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

the colour of an area of the sample the higher is the concentration of the investigated ele-mentinthatspecificarea.

FT-IR with ATR microscopy28

To determine the chemical composition of selected elements, individual layers on the four samples were analysed using the FT-IR microscope with ATR directly on the cross section of theembeddedsamples.Typically,6–10areasorspecificgrainsoneachlayerwereexamined.The results were determined immediately or compared with references, mainly the IRUG database.29,30

GC-MS

Todeterminepossibleorganicbinders.Derivatisationwasdonebymakinganalkalinehy-drolysis.Next thesamplewasacidifiedandextractedwithether, transferred toanauto-samplervialandthesolventevaporated.Finallythesamplewasre-dissolvedint-butylme-thylethercontainingdiazomethane(1)andinjectedon-columnat64°C.Thetemperatureisprogrammedtoinitially190°C@10°C/min,andfromthereto320°C@4°C/min.Dataarecollectedin‘fullscan’onaSaturnIonTrap.31

SEM imaging

Thiswasdoneinordertoexaminecalcium-containinglayersonsample1andSample3forpossiblepresenceofcoccoliths.SpecimenswerepreparedforSEM imaging by coating with Platinum-Palladium in a JEOL JFC-2300HRhighresolutioncoaterpriortoscanningat7kVina JEOL JSM-6335FFieldEmissionElectronMicroscope.32

Results and discussion

ThefollowingresultsanddiscussionsarebasedonthePilotProjectexaminationsandana-lysesdescribedabove.Acertainmarginoferrororuncertaintyisduepartlytotheminutesample size and partly to lack of parallel samples that might – through repetitions – have made some analyses more reliable.

28 FT-IRPerkinElmerSpectrumOne,AutoimageFT-IRMicroscope,ATR.

29 InfraredandRamanUsers’Group(IRUG);forfurtherinformationandtheIRUGdatabase,see:www.irug.org

30 SpecialthankstoDr.JanJørnHansen,SchoolofConservation,forperformingandinterpretingtheFT-IR spectra.

31 Glastrup1998,p.133;SpecialthankstoDr.JensGlastrup,theNationalMuseumofDenmark,forperformingandinter-preting the GC-MS analyses.

32 SpecialthanksforperformingSEMimagingtoDr.NikolajScharff,theZoologicalMuseumattheNaturalHistoryMu-seumofDenmark(TheUniversityofCopenhagen).ThanksaswelltoProf.Dr.MinikRosing,theGeologicalMuseumattheNaturalHistoryMuseumofDenmarkforhelpininterpretingtheimages.

Page 19: Tracking Colour

19InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Visual examination and photographic documentation

Asaresultofthevisualexamination–whichconfirmedthepresenceofpaintfragments–fiveareaswithpaintfragmentswereselectedforfurtherstudy:

1 A few fragments of presumed skin coloured paint on the right cheek inminute hollowgrateswereidentified,onewasextracted(Fig.2,sample1);

2 Afewfragmentsofskincolouredpaintintheareaaroundthelefteyewereidentified,onewasextracted(Fig.3,sample3);

3 Anareaaroundandbelowtheleftearwithskincolouredpaintwasidentified,nosamplewasextracted;

4 Anumberofred-brownpaintfragmentswereidentifiedinthecavitiesofthehairabovetheleftearandinotherpartsofthehair.Onesamplewasextractedfromtheareaabovetheleftear(Fig.3,sample4);

5 Anareawithbrownpaintwasidentifiedontopofaputty-likematerialcoveringwhatwasassumed to be an old damage and subsequent repair at the top of the forehead at the edge ofthehairline(Fig.2,sample2).

Thegeneralconditionofthesculpturesurfacecould–basedpurelyonthevisualexamina-tion–bedividedintofivefairlydistinctlevels:

1 A level of theassumedoriginal, relatively smooth surfaceof the facewhere skin colorsmighthavebeenapplied–‘warm’greyishappearance;

2 Alevelofsmoothsurfaceofthefacearoundthelefteyewherethemarblesurfaceisconsid-erablymorewhite.Suchsurfacecharacteristicsareusuallyinterpretedbyarchaeologistsasalaterre-workingofthesurfaceinordertomakedamagedareaslessconspicuous;

3 Areasof relativelywhitebut fairly roughsurfaceapparentlyatprotrudingareas like thehair, eyebrows and nose, interpreted as the wear and tear of the surface rather than proper breaks;

4 Obviouslydamagedareasexhibitaroughbreaksurfaceandacoloursomewhatdarkerthanthe supposedly original cheek area. This is the case on most of the hair-area as well as most of the left lower half of the cheek, and the bottom of the head where the break is located betweentheexistingpartoftheheadandthenowlostpartsofthehead,i.emostofthemouth,chinandneck;

5 Anareaabovetheforeheadandatthehairline,seemstohavebeendamagedandhasbeenfilledwithaputty-likesubstancethathasbeencoveredwithadarkred-brownlayerofpaintof which fragments are still seen.

UltravioletFluorescence(UV-FL) enhanced the presence of some of the paint fragments (Fig.4),andshowedthreeotherphenomena:

1 The area of the right cheek, the forehead, the area around the right eye and the nose as-sumed to be the original surface showed a discreet fluorescence while the white area of the righteyeexhibitedastrongyellowish-whitishfluorescencewhiletheirisshowedthesamediscreetfluorescencease.g.thecheek(Fig.5).

2 The discreet fluorescence of the face was specked with scattered dark red-violet fluorescing spots at places where in normal light brown spots were seen. This was also evident in the re-worked or damaged areas.

3 Aprominentdifference influorescencewasseenbetweentheabovementioneddiscreetfluorescence in areas considered to be the original surface and in areas considered to be

Page 20: Tracking Colour

20InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.2:IN2830.Paintsamples1+2,topcenter:headwithlocationofsamples(bottomleftandrightaredetails);topandcentreleft:Sample1(fleshcolour)intungstenlightandUV-FL;topandcentreright:Sample2(damagedareaatforeheadandhairline)intungstenlightandUV-FL.

Fig.3:IN2830.Paintsamples3+4,topcenter:headwithlocationofsamples(bottomleftandrightaredetails);topandcentreleft:Sample3(fleshcolour)intungstenlightandUV-FL;topandcentreright:Sample4(haircolour)intungsten light and UV-FL.

Page 21: Tracking Colour

21InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.4:UV-fluorescence (UV-FL)offemalehead,in 2830.

Fig.5:UV-FL of white in the right eye and iris of female head, IN2830,detail.

Page 22: Tracking Colour

22InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

laterre-workedareas–i.e.mostofthelefteyeandcheek.Inthere-workedareasthefluo-rescence was brighter, bluer.33

Infrared Reflectography (IRR) imageswerecapturedof relevantareas (asdocumenta-tion)butonthissculpturethetechniquedidnotrevealinformationthatwasnotvisibleintungsten light or in UV-FL.

Cross sections

Thefourcrosssectionsofpaintexhibitedthefollowingfeaturesundermicroscopy: Sample 1(fromareaofpresumedskincolouronrightcheek):Athree-layerstructurewithawhitelayeratthetopandbottomandalightochrelayerasthemiddlelayer.Eachlayerisapprox.100micrometerthick(withsomevariation),thetopandbottomlayersbeingfinergrainedthanthemiddlelayer.Attherightsideofthecrosssectionarearelativelylargeorange-redgrainandsomefairlysmallgrainsscatteredovertheentireochrelayer(Fig.2,leftpart). InUV-FL the middle ochre layer absorbs more UV than the top and bottom lay-ersandnoneexhibitastrongfluorescence.Theorange-redgrainabsorbswhileafewothergrains fluoresce vividly. Sample2 (fromdamagedareaabove forehead):Consistsmainlyof two layerswith themajorityof thesamplebeinga thickwhite layerandat the topof thewhite layer largered-brown grains. Some of these grains appear semi-transparent, thewhite layer appar-ently shining through. The white layers fluoresce vividly with a yellowish fluorescence. The largered-browngrainsexhibitanearlyfullUV-absorbancethusappearingdark(Fig.2,rightpart). Sample 3 (fromareaofpresumedskincolourunderlefteye):AsinSample1,athree-layerstructurewithwhitelayersatthetopandbottomofthecrosssections.ContrarytoSample1,onlythewhitetop-layerappearsfinely-grainedwhilethemiddleochrelayerandthebot-tomgroundlayerappearcomparativelycoarse;eachlayerisapprox.100micrometerwithvariations.Whilethetwolowerlayersshowadiscreetfluorescencetheupperlayerexhibita fairly strong yellow UV-FL(Fig.3,leftpart). Sample 4 (from area of presumedhair colour):A relatively coarse layer structurewithmany largeandsmallwhitegrainsmixedwithavarietyof smallochre,brown, redandorangegrains.Mixedintothisaresomesmallblackgrainsandonelargeblackgrain.Whilemost of the layer has a discreet UV-FL an area on top of the structure and partly downwards throughthecenterexhibitayellowishUV-FL (Fig.3,rightpart).

Visual examination, UV-FL, IRR and cross sections provided a considerable amount of information as well as some yet unsolved problems. No tool marks or surface phenom-enathatcouldberelatedtolocalcolourswerefoundwhenexaminingthesurface.PaintfragmentswereconfirmedvisuallyinnormallightandinUV-FL.Afewweresampledandmadeintocrosssections.Summinguptheresultsofthevisualanalysisandcrosssectionsitappearedthattheskincolourswerethree-layerstructures,thehairexhibitedaone-ortwo-layer structure and the putty-like material with red-brown paint on the forehead had a two layer structure. Atthebottomofthetwoskincoloursamplesawhitelayerwaspresent,assumedtobeagroundlayer.ComparedwithatypicalEuropeanmedievalgroundlayeroftenexceeding1mmthissculpturehadathingroundlayer.Thismaybeexplainedbytheverycarefulfinishof the surface of the face – on medieval sculpture the wood carver usually left the surface

33 Thisisinaccordancewithsimilarobservations,seeRorimer1931,p.15ff.

Page 23: Tracking Colour

23InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

fairly coarse, relying on the persons producing the thick ground and paint layers to level out theunevenness.Fromatechnicalpointofviewthisisremarkable:itmusthavebeenmorelabour intensive for the sculptors to have produced a smooth surface e.g. on the skin area ofthemarblethanifapainterweretoproduceasimilarsmoothsurface(ontopofamorecoarsemarblesurface)usingcalciumcarbonateinabindingmedia. The two samples of skin colour – very alike – had a fairly thin layer of light ochre on top of the ground, and in both cases a white layer on top. The top white layer could either be a sec-ondary addition or it could be a thin white layer to regulate the effect of the ochre - perhaps make the ochre appear more light and lifelike? One might also consider the light refracting impact of the crystalline structure of the marble through the very thin paint layers. No obvi-ouscomparisonwithmedievalEuropeanpaintingtechniquecouldbeidentifiedwithregardto a white layer on top of a skin colour. UV-FL made the difference between the apparent original surface with the discrete UV-FL and other parts with a more white-blue UV-FL stand out – the latter interpreted as second-arysurfaces.Whileitcanbeassumedtobeacharacteristicforanoriginalsurfacetoex-hibitdiscrete‘warm’greyUV-FLthereisnosimpleexplanationforthereason.Deteriorationphenomena of the stone itself as a possibility has been suggested by Rorimer34 but remains fromthepaintlayers(e.g.microparticlesandbindingmedia,ormicrobiologicaldeteriora-tion)mayalsoplayapart.UV-FL revealed important information when studying the white of the eye in the right eye in UV-FL. The white of the eye showed a prominent fluorescence while no pigments particles were visible leading to the assumption that the white part of the eye was painted with a pigment and binding media layer that has left a fluorescing sub-stance in the stone surface.

SEM/EDAX

Sample1:The BSEpictureofsample1isgiveninFig.6(topleft).Themagnificationis×1000,meaningthatthesampleisapprox.0.15mmincrosssection.FromFig.6itisevidentthatacoarsethickmiddlelayerisboundedbythinfinegrainedlayersatthetopandthebottom.MapswiththedistributionofCa,Mg,Fe,S,Cl,Na,SiandPareshownintheremainingpartofFig.6. Allthreelayersinsample1consistofcalciumcarbonate(CaCO3)withanumberofgrainsinthemiddlecoarselayerthatareveryrichinCa(thewhitegrainsintheCamap). FromtheFedistributionitisevidentthattheconcentrationofFeishigherinthecoarsemiddlelayerespeciallyintherightsideofthesample.ThepresenceofFeindicatesthatanochre paint was used for the colour of the skin. Sioriginatingfromsilicateaggregatelikequartz(SiO2)isconcentratedinthelowerlayerofthesample,butSiisalsoobservedoutsidethesampleintheembeddingmaterial,henceoriginatingfromaSicontainingabrasiveusedinthemanufactureofthesample.Alargesaltgrain(NaCl)isobservedintheupperrightpartofthesample. AninterestingfeatureisobservedintheSmap.Asulfurcontainingmaterial,probablygypsum(CaSO4 ·2H2O)isobservedinhighconcentrationsontopoftheupperlayerandrun-ningthroughthecoarselayer.ThissecondScontaininglayer,thatisparallelwiththefirst,revealsthatthecoarsemiddlelayerconsistsofmorethanonesinglelayer.Agypsumlayerlike the one found indicates that this layer for a period of time was the surface of the sam-ple.BetweenthetwohighlyconcentratedSlayers,theconcentrationofSisslightlyelevatedcompared to the rest of the sample.

34 Cf. note 24.

Page 24: Tracking Colour

24InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

FromtheEDXresultsthereisnoexplanationfortheverybrightorangecolourfoundinthetungstenlightphotographofsample1(Fig.2,upperleft).

Sample 2:The red-brown colour of the damaged area at the forehead presumed to be a secondary surface appears as very bright red large grains in the tungsten light photograph (Fig.2,upperright).BSEpicturesofthetwohalvesofsample2areshownupperleftinFig.7aand7b.Themagnificationsare×350and×200,meaningthatthesampleisapprox.0.5mmincrosssection.ItisevidentfromthepicturesthatintherightareaofthesampletheredgrainslookverywhiteinFig.7a(upperleft);thismeansthataheavyelementisobservedinrelationtotheredcolour.Intheleftpartofthesample(Fig.7b,upperleft),theredgrainsap-pearmuchdarkerindicatinglighterelements.Herethedarkergrainsarecoatedbyheavierelementalcompounds.Asaconsequencethetwopartsofthesamplewereexaminedsepa-rately. ABSEpictureandelementmapsofCa,Ba,S,Fe,K,Al,Si,NaandClintherightpartofthesampleareshowninFig.7a.FromthemapsofBaandSitisevidentthatthesetwoelementsare responsible for the compound found in connectionwith the red layer. Ba andS arefoundasbarite(BaSO4),whichiswhiteandnotred.FromthemapofFeitseemsthataFecompoundiscoatingthebaritegrains.IfthisFecompoundisresponsiblefortheredcolour,itisahematiticredochrecompound(Fe2O3).TheCamaprevealsthatthewhitematrixiscalciumcarbonate,andmapsofNa,K,AlandSishowthatlargegrainsofpotassiumfeld-spar(KAlSi3O8)areusedasaggregateinthechalk.Theotheralkalifeldsparmineralalbite(NaAlSi3O8)isnotpresent. ABSEpictureandelementmapsofCa,Ba,S,Fe,K,Al,Si,NaandClintheleftpartofthesampleareshowninFig.7b. Inthispartofthesampleonlyonegrainofbarite isfound,implying that in the left part of the sample, the red colour has a different origin. The map ofFeshowsthataFecompound,againredochre(hematiteispresumed),isresponsiblefortheredcolour.TheFecompoundiscoatinglargegrains,asisevidentfromtheBSE picture oftheleftsideofthesampleandfromtheFemap.Thisiscontrastedbythetungstenlightpictureofsample2inFig.2(toprightimage),whereentirelyredgrainsareobserved.Theochreiscoatinggrainsofquarts(SiO2)andpotassiumfeldspar(KAlSi3O8)andagainnoalbite(NaAlSi3O8)isobserved,seethemapsofNa,K,AlandSiinFig.7b.TheCamapconfirmsthepresence of calcium carbonate. Wearenotfamiliarwithearlierfindingsofredochreonbaritegrainsoronsilicatesre-sponsibleforredpaintlayers.ExaminationoftheredcolourofthegrainsinFig.2(toprightimage)revealsthattheydonothaveanordinaryochrecolour,theredlooksratherasanorganiclacquerorcinnabar(HgS)orsomeleadred.HowevernotracesofeitherHgorPbarefound in the sample by EDXanalyses.AredlacquerprecipitatedonbariteorsilicatewouldbeexpectedtofluoresceinUVpictures;thisisnotthecase,asevidentfromFig.2(centerright UV-FL image).

Page 25: Tracking Colour

25InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.6:Sample1fleshcolouroncheek;SEM/EDX,×1000,20kV.Fromtoplefttobottomright:BSE picture (topleft)andelementmapswiththedistributionof[none],Na,Mg,P,Si,S,Cl,CaandFe(bottomright).

Page 26: Tracking Colour

26InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.7a:Sample2rightpart,dam-agedareaatforehead;SEM/EDX, ×350,20kV.Fromtoplefttobottomright:BSEpicture(topleft)andele-ment maps with the distribution ofNa,K,Ca,Ba,Fe,Al,Si,SandCl(bottomright).

Page 27: Tracking Colour

27InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.7b:Sample2leftpart,damagedareaatforehead;SEM/EDX,×200,20kV.Fromtoplefttobottomright:BSE picture(topleft)andelementmapswith the distribution of Na, K, Ca, Ba,Fe,Al,Si,SandCl(bottomright).

Page 28: Tracking Colour

28InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.8:Sample3fleshcolournearleft eye, SEM/EDX,×750,20kV.Fromtoplefttobottomright:BSE picture (topleft)andelementmapswiththedistributionofNa,K,Cl,Ca,Fe,Mg,Al,SandP(bottomright).

Page 29: Tracking Colour

29InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.9:Sample4hairaboveleftear(detailofsample),SEM/EDX,×500,20kV.Fromtoplefttobottomright:BSEpicture(topleft,withminiatureBSE picture of the whole sample in-serted)andelementmapswiththedistributionofNa,Mg,Al,Si,P,S,Cl,K,Ca,FeandC(bottomright).

Page 30: Tracking Colour

30InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Sample3:The BSEpictureofsample3isgiveninFig.8,themagnificationis×750,meaningthatthesampleisapprox.0.25mmincrosssection.FromFig.8(topleft)itisevidentthatacoarsethicklayerisoverlainbyathinnerfinergrainedlayer.MapswiththedistributionofCa,Mg,Fe,S,Cl,Na,Si,AlandPareshownintheremainingpartofFig.8. Bothlayersinsample3consistofcalciumcarbonatewithanumberofgrainsinthelowercoarselayerthatareveryrichinCa(thewhitegrainsintheCamap).Alisdistributedinbothlayersprobablyoriginatingfromclayminerals.AfewlargegrainsarerichinNa,AlandSiindicatingthefeldsparmineralalbite(NaAlSi3O8)incontrasttothepotassiumfeldsparfoundinsample1and2,andquartsgrainsarealsoobserved.ApeculiarityisPfoundincon-nectiontothesilicates.Fe,probablyasayellowochre,isresponsiblefortheskincolouredappearanceofthesample.Sulfur(S)observedinsample3isconcentratedinspotsonthetopofthesampleandintwothinlayersthatareangledapprox.45degreestoeachother;onelayerisinacrack.Webelievethattheselayershavebeenexposedassurfaces,whereSfrom the environment is able to react with the calcium carbonate resulting in the formation of gypsum on the surface.

Sample 4: BSEpicturesofsample4aregiventopleftinFig.9,withthewholesampleinminiature and a detail of the paint layer. The detail is from the left in the sample.35 The mag-nificationofthedetailis×500,meaningthatthesampleisapprox.0.45mmincrosssection.The BSEpictureofthewholesamplerevealsaratherhomogeneoussurface(withrespecttoelementnumber)withapartinthemiddleofthesamplethatismuchdarker,implyingsignificantlylighterelements.MapswiththedistributionofCa,Mg,Fe,S,Si,Al,K,Na,Cl,PandCareshownforthedetailofthesampleintheremainingpartofFig.9. ThetungstenlightpictureinFig.3(topright)showsawhitishcoarsegrainedsamplewithorange pigment grains and one very large black grain in the middle of the sample. TheCmapinFig.9identifiestheblackgrainascharcoal.Visualidentificationoftheor-angegrainsinthesamplewouldindicateapigmentlikeminimum(Pb3O4),howevernolead(Pb)wasdiscoveredinthesample.TheorangecolourisattributedtoFepresentinthesam-ple(FemapinFig.9)whereFerichparticlesareobserved. ThewhitematrixisCarich(Fig.9)andconsistsofcalciumcarbonate.Theleftpartofthesample(Fig.9)isrelativelyricherinMgthantherestofthematrix,indicatingtheuseofdolomiticlimestone,i.e.thechalkcontainingmineraldolomiteCaMg(CO3)2. QuartzgrainsSiO2 (Simap)andpotassiumfeldspargrainsKAlSi3O8 (K,AlandSimaps)havebeenusedasaggregate(Fig.9).SaltNaCl(Fig.9,NaandClmaps)isobservedinthelower right part of the sample. AgrainwithanelevatedcontentofSandPisobservedinFig.9.Wehavenoobviousex-planation for this.

FT-IR + ATR

Themainpigments/fillersinall layersinallfoursamplesweredeterminedbycarbonatebandstobesomeformofcalciumcarbonate(CaCO3). While the pigment composition appeared the same in the white top and bottom layer in Sample 1(skincolour)acertainphysicaldifferencecouldbeseenbetweenthesetwolayers.ThetoplayerofSample1appearedalittlelessdensejudgedbyaslightlyhigheramountofestersfromthe(polyester)matrixhadpenetratedthetoplayerthanthebottomlayerandthe mid ochre layer. The mid ochre layer in Sample 1 and Sample 3 (skincolour)didnotshowtracesofironoxidesorsilicatesasexpected.

35 The SEM/EDX analysis was done with the sample positioned upside down.

Page 31: Tracking Colour

31InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Thewhite(main)partofSample2(damagedpartatedgeofforehead)againcomprisedofcalciumcarbonate(verymuchlikeinSample 1 and Sample 3 and as these with no traces of silicates).Thevariousredgrainswereexaminedandgaveslightlydifferentresults.Attheright side of the sample the main part of the grains was calcium carbonate while on the left sideitappearedmorelikeamixtureofcalciumcarbonateandsilicatessuchasfeldspar.Inthreegrainstotherightnocalciumcarbonatewasdetectedatall.Bandssuggestingsul-phates were detected. Comparisons with IRUG tables like IMP00194 anglesite and IMP00197anorthitesupportedtheseobservations.BariumwasnotdetectedinthegrainsbyFT-IR. The redcolorofthegrainsappearedtobelinkedtoironoxides. Sample 4 (hair)consistedmainlyofcalciumcarbonatesaswellbutwasdistinguishedfromSamples 1–3 by the presence of silicates with the calcium carbonate. The brownish color of thesamplewasduetoironoxides,silicates(ochre)aswellassmallblackparticles(carbonblack)andafairlylargeblackgrain.

FT-IR wasusedtoexaminethefoursamplesfortracesoforganicmaterialslikebinders.Ingeneral itwasnecessarytodistinguishbetweentracesthatmaybe linkedto thematrixpolyesterandpossibleoriginalorganicmaterials.Inmostcases–examininglayersofthefour samples – there were no traces of binding media. Neither proteins, carbon hydrates nor resinscouldbefoundbutinafewcasesahintofapossiblewaxor(morelikely)oilwerefound.Oilwaspartlyjudgedbytopsaroundfrom1700to1715–20cm-1 and some CH groups around3000thatappearedalittlestrongerthanthegroupsinthematrixester.Thatapos-sibleoilwaspartlydeterioratedwasdeterminedbytheshiftoftopsfromaround1730–40toalowerlevel.TheseobservationsweremadeonSamples1,2and3whilehardlyanysignifi-canttopsaround1700werefoundinSample4.

The FT-IR analysisofthesamplesconfirmedthecalciumcompoundtobecalciumcarbon-ate(CaCO3).AlthoughanumberofanalysesweredonewiththeATRonsmallareas(approx.35×35micrometer)onthecrosssectionsinordertoverifyorexcludethepresenceofpos-sibleorganicbindersonlyweaksignalswerefoundandthusnodefiniteconclusionscouldbedrawn.However,inafewcases–onafewspotsontheuppermostlayerofsample1andsample3–someindicationsofeithertracesofoilorawaxcouldbefound.Itmightbecor-related with the fact that the uppermost white layer appeared more porous – a fact that caused some discussion as it might also have allowed the embedding media to be absorbed inthesamelayer.Repeatedly,theembeddingmedium(polyester)seemedtocreatefalsesignals of an organic binder near edges of the embedded sample which led to considera-tions about alternative embedding strategies and materials.

GC-MS

GC-MSextendedtheanalysisoforganicmaterialsinthesamplesandindeedshowedthepresenceofthefattyacidsP(Palmitine)andS(Stearine)thusindicatingoil.Thismayalsocomplement the weak signal of possible oil in some of the FT-IR analyses.HowevernoA(Azlaine)wasfoundandthustheoilappearsnottohavebeenadryingoil. Therearesomepossibleexplanationsforthis.Oneisthatanon-dryingoilmayhavebeenapplied in antiquity, another that the non-drying oil has been applied in recent time, either as a part of the preparation of the sculpture for sale or a treatment done by the two known owners in recent times.36Itislikelythatatreatmentwithanon-dryingoil(oracomposi-

36 ThepersonwhoboughtthesculptureinAthensatthebeginningofthe20th century or staff at the Ny Carlsberg Glyp-totek.

Page 32: Tracking Colour

32InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

tionwheresuchoilwereincluded)hasbeenappliedtosaturatearecentlycleanedoraleanor dusty surface.37Ithasnotbeenverifiedifsuchoilispresentinnon-paintedareasofthesculpture.Thebackgroundanalysisexhibitedafairamountofpollutionandforthatreasonit is included in the chromatogram.

SEM imaging

26imagesofcalcareousfragmentsfrompaintsample1(17imagescaptured38)andsample3(9imagescaptured39).Ononefragmentapossibleorganicformwasidentified,lengthap-prox.15micrometer,widthapprox.10micrometer.40

Nococcolithswereidentifiedonthesurfaceofthe26imagesofcalcareousfragmentsfrompaintsample1andsample3.Ononefragmentanorganicformwasidentified.Thepossibleorganicformhasnotbeenfurtheridentified;itmightbeafragmentofafungusorspore?41 Thewell-definedcalcium-crystals in the fragmentsand the lackof coccoliths led to theassumption that the calcium containing grounds and pigments derived from marble – or rather marble-dust.42

Whencombininganddiscussingtheresultsfromthevisualexaminations,UV-FL, IRR and cross sections with the instrumental analysis it appeared that the stratigraphy of the origi-nalsculpturemayhavehadonlytwolayerstructures.Intheskincolourareasagroundanda paint layer – light ochre – while the white top layer might be secondary as a thin line or spotsofgypsumbetweenthelayersindicate.Itis,however,notpossibletodeterminetheagedifference between the original and the possible secondary layer. The chemical composition ofthelayerswasmostlydetermined–intermsofpigment/filler–tobecalciumcarbonate,dolomite, some clay minerals, ochre and black. The calcium carbonate probably is marble powderasnococcolithswerepresent.Nothingdefinitewasfoundconcerningbindingme-diainthefoursamples.Onlytracesofanon-dryingoilwereidentified:suchoilhaswithallprobability not been used as the paint layer would not have dried and thus may have been a lateaddition.Althoughnothingwasidentifiedthepaintmayhavecontainedoneormoreofanumberofpossiblebindingmediaincludinganimalglue,starch,egg,waxes,(drying)oils,gums,perhapsresins,resinouscompoundsorbalsams–ormixtures.However,mostsuchorganic compounds may have deteriorated by now to a point beyond analytical detection. One reason for deterioration could be the possibility that NCG in2830–asisthecasewithotherantiquestatues–mayhavebeenexposedtoe.g.bacteriawhileburiedunderground.Furthermoreitcanprobablynotberuledoutthatan‘alfresco’techniquehasbeenusedtobind the pigments.

37 Forasurveyofpossibletraditionalsurfacetreatmentstomarblesculptures,seePlenderleith–Werner1971,p.311ff.

38 ImagesNCG2830sample1a–NCG2830sample1q.

39 ImagesNCG2830sample3a–NCG2830sample3h.

40 ImagesNCG2830sample1m,1n(overview)+NCG2830sample1o(detail).

41 OralcommunicationfromProf.Dr.MinikRosing.

42 Marbledustisknowntohavebeenusedinsomecasesforgroundmaterialsandisreferredtoinpaintingtreatises,seeforexample‘MappaClavicula’(note11),chapter122B with a description of powdered marble and glue.

Page 33: Tracking Colour

33InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Conclusion

Tracesofcolourwasconfirmedasbeingpaint,appliedtoagroundlayer.Themostpromi-nent paint layer is the skin colour. The general stratigraphy and pigment composition has beenaffirmedwhilethebindingmediaisyetnotconfirmed.ThetechniquesusedduringthePilotProjecthasledtoadjustmentstotheprotocoltobeappliedintheexaminationofsculptureswithintheframeworkoftheMainProject.Someuncertaintiesmayberelatedtothe fact that most sculptures in older collections have been acquired without information aboutexcavationandpriortreatment.Furthermore,fewcollectionshaverecordofearlierin-housetreatments(ifany)makinganassessmentofpreviousconditioncomplicatedandmakingitdifficulttoevaluatematerialspresent(ornotpresent)onthesurface.HopefullytheresultsoftheMainProjectaswellasresultsfromsimilarinvestigationcampaignsinothercollectionsorbycontemporaryfieldarchaeologywillcreateabetterknowledgebasisconcerning the original state and visual appearance of the sculptures as well as a better understanding of the deterioration process – the latter in order to enhance the protection and preservation of the remains of colour on the ancient sculptures.

Archaeological comment

Thisfragmentofafemalehead(Fig.1)wasboughtc.1910atAthensbytheDanisharchitectSvenRisomandacquired for theGlyptotek in 1940.Themarble isprobablyParianLych-nites.43 On stylistic grounds, it is unanimously held to be a Classical Greek original, made at Athensc.425BCE. TheheadisrelatedbyitsstyletoagroupofotherAtticfemalemarbleheadsandfrag-ments associated with the Parthenon sculptures and sculptures of the immediate post-Parthenonic period.44 Like the Glyptotek head, a number of these pieces have drill holes in the hair for the attachment of metal ornaments such as diadems and wreaths. This feature and the slightly over life size format suggest that the head is that of a goddess. ComparedtothatoftheGreekArchaicandEarlyClassicalperiods,thesculpturalpoly-chromy of the Classical and Late Classical period remains poorly documented. The sum-maryofferedbyReuterswärdin196045 has not been substantially challenged by subsequent research. The results related above have special bearing on one of the most contested issues, namely how the nude parts of marble sculptures were dealt with. Was a skin colour applied or not? FortheArchaicperiod,theevidenceavailableremainsmeagreanditsinterpretationde-bated.46 Inthelightofthepresentstateofourknowledge,itwouldseemadvisabletobevery open-minded and avoid generalizations. The decision to apply a skin colour or not may

43 Cf.note4.TheuseofParianLychnitesmarblebyAtticsculptorswaswidespread.

44 Berger1956.SeealsotwootherfragmentsinAthens:NationalArchaeologicalMuseuminv.4491(referenceinMoltesen1995)and3739(Kaltsas2002,p.123,no.228w.bibl.).

45 Reuterswärd1960,pp.85–88.

46 Reuterswärd1960,p.68,liststheArchaic(andClassical)instancesofaskincolourknowntohim,withpagereferencestohistreatmentofthesculpturesconcerned;thepolychromyoftheperiodissummarizedpp.69–70and70–74(onsurfacetreatmentwithwax,ganosis),andpp.74–80.Reuterswärd’sdiscussionofskincolourlacksfocusandstructure.Brinkmann2003,pp.43–45assertsthataskincolourwastherule.ThispositionisstronglychallengedbySchmaltz2005,pp.24–30.Tomyknowledge,thequestionofskincolourhasnotbeenaddressedasaseparatesubjectsincethestillimportantstudiesbyRichter1928/29and1944.Aconcisepresentationanddiscussionoftheevidencenowavail-able is needed.

Page 34: Tracking Colour

34InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

havebeen influencedby factorssuchasmaterial,monumenttype,context, formatsandregional traditions. As for theClassicalperiod,evidence iseventhinner.Thenumberofsculpturesknownto have or have had a skin colour is very small and none have been investigated in depth. To my knowledge – which may well be too limited – the monuments in question are the following:47

• Hephaisteion,eastandwestfrieze,Athens,c.450BCE.Tracesofredonmalefigures.48 • Funerary relief of Philis, from Thasos, c. 450–44 BCE.Skincolouronthecheeks.49

• Reliefs from Myra, Lycia. Grave 81, entrance room. 4th century BCE.Skincolouronadultmale,woman and boy.50(Fig.10)

• Mausoleum at Halicarnassos.Mid4th century BCE.SkincolouronAmazonFriezeandcofferre-liefs.51

• Sarcophagi from Sidon. 4th century BCE.MaleskincolouronAlexanderSarcophagusandfemaleskincolouronlidpedimentreliefsonMourningWomanSarcophagus.52

• Attic funeraryreliefwithhorseandblackgroom. 4th century BCE.Bluish-blackskincolourongroom.53

47 Cf.Reuterswärd1960asinnote52.Onthecolourofskinin4th century BCEwrittensourcesseeVillard2006,pp.43–54,focusingontheword‘andreikelon’meaningbothastatueofamanandthecolourofskin.Cf.Theophrastos,Delapidi-bus,51,6onthepaintersuseofochre(‘miltos’)inpreparingskincolours,‘taandreikela’,Villard2006,p.48.

48 Athens,insituinthepronaosofthetemple.Reuterswärd1960,p.52,cf.Koch1954,p.101,withhisownobservationofreddishskincolouronbothfriezes(nocolourreproduction).Idonotknowofanyvisualdocumentationincolour.

49 Paris,LouvreMa766.Reuterswärd1960,p.58(notLouvreMa706asinhisnote129);Hamiaux1992,p.108,no.97,‘thecolourmentionedinearlierliteraturehasdisappeared.’Nocolourreproductionofearlierstateexists.

50 Reuterswärd1960,p.58;Borchhardt1975,pp.135–146,FarbtafelI,1–2(colourphotographsofReliefIIandIVinTomb81).Borchhardt’sdescriptionofpreservedcolourtracesdoesnotincludeskincolour.Cf.thenowlost(seeBorchhardt1975,p.135)paintedcastsoftheBritishMuseumBMCat.954ofthetomb81reliefsBorchhardt1975,81,II,IIIandIVasdescribedinReuterswärd1960,loc.cit.,n.130.TheonlyexistingvisualdocumentationofthepolychromyofthesethreereliefsisnowthecolourplatereproductionofwatercolourrenderingsbyG.ScharfinFellows1841,p.198ff.

51 BodrumandLondon,BritishMuseum.Reuterswärd1960,p.59;Jenkinsetal.1997,especiallyp.38,rightcolumn,withmentionofskincolourontheAmazonfrieze,butnocolourreproduction;Cook2005,p.30.SkincolouronthereliefsculpturesoftheMausoleumremainstobevisuallydocumented.Forpolychromyonthefree-standingsculpturesoftheMausoleumseeWaywell1978andHiggs2006,pp.190–193(ColourontheAmazonFriezeandfree-standingsculp-ture).

52 Istanbul,ArchaeologicalMuseum, inv. no. 368 (Mourning) and 370 (Alexander). Reuterswärd 1960, pp. 60–62, p. 60w. references toearlycolour reproductions innote133. For theAlexanderSarcophaguscf.V.Brinkmann inBrink-mann–Wünsche2007,pp.150–162andH.Piening,ibid.,pp.168–171,withdocumentationincolour.FortheMourningWomensarcophaguscf.nowFleischer1975,p.60reportingskincolourstillvisible(nocolourillustrations).Forcolourphotographsofnear-presentstate:Pasinli1989,pp.14–17(Mourning),pp.18–33(Alexander).

53 Athens,NationalArchaeologicalMuseum,inv.no.4464.Reuterswärd1960,pp.62–64;Kaltsas2002,p.206,no.415.Nocolour reproduction.

Page 35: Tracking Colour

35InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

These monuments have in common the fact that they are all reliefs, rather than free-stand-ing sculptures in the round. From this,no conclusions canbedrawnat thispoint as towhetherthepolychromyofclassicalreliefsdifferedfromthatofsculpturesintheround:fartoofewsculpturesintheroundoftheClassicalperiodhavesofarbeenexaminedin-depthwith the technology now available.54

To date, we have therefore no published study of samples taken from skin colour remaining on Greek marble sculpture of the Classical period. The cross sections of the skin colour of ngc IN2830andthedataobtainedfromthemthereforehavenoparallelswithwhichtheymight be compared. Consequently, this skin colour may in principle have been applied in post-classicalorevenpost-antiquetimes.Furthercomparativedatamustbeaccumulatedinordertoreachfirmerground. Inthisregard,coordinated,systematicandnotleast,interdisciplinarystudyofpolychro-my in museum collections of Greek and Roman sculpture, conducted to some agreed stand-ards,isneeded.ApilotstudyofsculpturesintheBritishMuseum,launchedinthesummerof 2009, is the more welcome. One among many important issues to be faced is that of the taking of samples. The use of invasive methods55 needs to be discussed in the light of present day technology, in a forum composed of curators and conservators. Insuchadiscussion,itneedstoberecognizedthatthesizeofasampleasneededforcrosssectionanalysisandfurtherin-depthinstrumentalexaminationisminutetothepointofbeinginvisibletothenakedeye,i.e.onanaveragec.0.3–1mm2. One should obviously not remove the last remaining shred of ancient pigment, but have some agreed standard to de-termine when the taking of a sample is permissible.

54 ItisinstructivetoreadBourgeois–Jockey2005reportingontheirstudyofLateHellenisticsculpturesintheroundfromDelos.Examinationunderhighmagnificationrevealsevidencenotvisibletothenakedeye–thoughnotofaskincolour.Thehypothesesputforwardinexistingliteratureonthepossiblydifferentiatedpolychromyofthevariousclasses of Classical Greek sculpture will be discussed in forthcoming publications by the authors of this article.

55 Tobedistinguishedfrom‘destructivemethods,’atermwhichrelatestotheimpactonsamplesoftheanalyticalmeth-ods employed.

Fig.10:Tomb81,Myra.Reliefsintheentrance room. 4th century BCE. Water colourbyG.ScharfasreproducedinCharles Fellows’AnAccount ofDis-coveries in Lycia, 1841.

Page 36: Tracking Colour

36InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

If,onethicalorothergrounds,thetakingofsamplesisdisallowed, itmustberealizedthat we are thereby prevented from gaining data pertaining to essential aspects of ancient sculptural polychromy. AsfarasourworkattheNyCarlsbergGlyptotekisconcerned,thetakingofsamplesisnotruled out a priori. Conditions allowing, it will be part of our standard protocol.

Justasimportantisanincreasedawarenessamongarchaeologistsandconservatorswork-inginthefield,ofpossible,usuallyminuteandveryfragileremainsofpolychromyonfindsof sculpture and architectural members.56

56 DiscoveriesofancientsculptureandarchitecturewithremainsofpolychromyareregularlyreportedfromsitesaroundtheMediterranean(i.e.Aphrodisias,BetShean,Corinth,Herculaneum,Gortyn,Narona,Oudna,Rome,Tarragona).Buttomyknowledgeexcavationdirectors,fieldarchaeologists,fieldconservatorsandspecialistsinancientsculpturalpolychromyhavesofarnotmettodiscussandformulateguidelinesforbestpractisewhendealingwithsuchfinds.Ameeting should be organized as soon as possible.

Page 37: Tracking Colour

37InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

references

Autenrieth 1990:H.P.Autenriethetal.,DiePraxisderUV-Fluoreszenzfotografie,ZeitschriftfürTechnologieundKonservierung,4,1990,215–233.

Berger 1956:E.Berger,EinneuerKopfausdenParthenongiebeln?,AM71,1956,153–172.

Borchhardt 1975:J.Borchhardt(ed.),Myra.EinelykischeMetropole(1975).

Bourgeois – Jockey 2005:B.Bourgeois–Ph.Jockey,Ladoruredesmarbresgrecques.NouvelleenquêtesurlasculpturehellénistiquedeDélos,JournaldesSavants2005,253–315.

Brinkmann 2003: V. Brinkmann, Die Polychromie der archaischen und frühklassischenSkulptur(2003).

Brinkmann – Wünsche 2003:V.Brinkmann–R.Wünsche (eds.),BunteGötter.DieFarbig-keitantikerSkulptur.GlyptothekMünchen(2003)(withlaterrevisededitionsfromBasel,Skulpturhalle2005andHamburg,MuseumfürKunstundGewerbe2007;editioninTurkishfromtheNationalArchaeologicalMuseum,Istanbul2006andinGreekfromtheNationalArchaeologicalMuseum,Athens2007;versioninEnglish,‘GodsinColor.PaintedSculp-tureofClassicalAntiquity,’ArthurM.SacklerMuseum,HarvardUniversityArtMuseums(2007).ThemostrecenteditionisfromtheLiebieghausSkulpturensammlung(2008).

Christensen – Scharff 1986:M.C.Christensen–M.Scharff,Cross-sectionsintheexamina-tionofPictorialArt,inStructure10,StruerMetallographicNews,April1986,10–12.

Cook 2005:B.F.Cook,ReliefSculptureoftheMausoleumofHalicarnassus(2005),esp.29–30‘Colourandsurfacetreatment.’

Fellows 1841:CharlesFellows,AnAccountofDiscoveriesinLycia(London1841).

Fleischer 1975:R.Fleischer,DerKlagenfrauensarkophagausSidon.IstForsch34(1983).Esp.p.60‘Polychromie’(skincolouronfemalefiguresinlidpediment).

v. Graeve – Preusser 1981:V.v.Graeve–F.Preusser,ZurTechnikgriechischerMalereiaufMarmor,JdI96,1981,120–156.

Grzywacz – Su – Fan – Wouters 2008:C.Grzywacz–B.Su–Y.Fan–J.Wouters,DevelopmentofidentificationstrategiesforAsianorganiccolorantsonhistoricChinesewallpaintings,inJ.Bridgland(ed.),ICOM-CC 15thtriennialconference,NewDelhi,22–26September2008:preprints/ICOMCommitteeforConservation,(2008)534-541.

Glastrup 1998: J. Glastrup, Diazomethane preparation for gas chromatographic analysis,J.Chromatogr.A827,1998,133.

Hamiaux 1992:M.Hamiaux,LessculpturesgrecquesI.DesoriginesàlafinduIVesiècleavantJ.-C.Louvre(Paris1992).

Page 38: Tracking Colour

38InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Hawthorne – Smith 1974:J.G.Hawthorne–C.S.Smith,MappaeClavicula–alittlekeytotheworldofMedievaltechniques, J.G.Hawthorne&C.S.Smith(eds.),TheAmericanPhilo-sophicalSociety(NewYork1974).

Hawthorne – Smith 1979: J.G. Hawthorne – C.S. Smith,Theophilus on Divers Arts: the

foremostmedievaltreatiseonpainting,glassmakingandmetal-work,(J.G.Hawthorne–C.S.Smith(eds.)(Secondedition,NewYork1979).

Higgs 2006:P.Higgs,LateClassicalAsiaMinor:DynastsandtheirTombs,inO.Palagia(ed.),GreekSculpture.Function,MaterialsandTechniquesintheArchaicandClassicalPeriods(NewYork2006)163–207.

Jenkins et al. 1997:I.Jenkins–C.Gratziu–A.Middleton,ThepolychromyoftheMausoleum,inI.Jenkins–G.B.Waywell(eds.),SculptureandSculptorsofCariaandtheDodecanese(London1997)35–41.

Kaltsas 2002:N.Kaltsas,SculptureintheNationalArchaeologicalMuseum,Athens(Athens2002).

Katz 1998:M.R.Katz,ThemedievalpolychromyofthemajesticwestportalofToro,Spain:InsightintoworkshopActivitiesoflatemedievalpaintersandpolychromers,inPaintingTechniques–History,MaterialsandStudioPractice,IICDublinCongress7–11September1998,27–33.

Khandekar 2003:N.Khandekar,Preparationofcross-sectionsfromeaselpaintings,ReviewsinConservation,No.4,2003,52–64.

Koch 1954:H.Koch,StudienzumTheseustempelinAthen(Berlin1954).

Kühn 1984:H.Kühn,Farbmaterialien,inReclamsHandbuchderkünstlerischenTechniken,Bd.1,(Stuttgart1984)7–54.

Liverani (ed.) 2004:P.Liverani(ed.),Icoloridelbianco.Policromianellasculturaantica.Mu-seiVaticani.CollanadiStudieDocumentazione(CittàdelVaticano2004).

Moltesen et al. 1995:M.Moltesenetal.,Catalogue.NyCarlsbergGlyptotek.GreeceintheClassicalPeriod(Copenhagen1995).

Moltesen – Nielsen 1996:M.Moltesen–M.Nielsen,Catalogue.NyCarlsbergGlyptotek.Etru-riaandCentralItaly450–30B.C.(Copenhagen1996).

Moltesen 2002:M.Moltesen,Catalogue.NyCarlsbergGlyptotek.ImperialRomeII(Copen-hagen2002).

Nadolny 2006:J.Nadolnyetal,MedievalPaintinginNorthernEurope(London2006).

Nielsen – Østergaard 2004:A.M.Nielsen–J.S.Østergaard(eds.),ClassiColor–farveniantikskulptur.MeddelelserfraNyCarlsbergGlyptotekN.S.6(Copenhagen2004).

Pasinli 1989:AlpayPasinli,IstanbulArchaeologicalMuseums(Istanbul1989).

Page 39: Tracking Colour

39InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Poulsen 1951:F.Poulsen,CatalogueofancientsculptureintheNyCarlsbergGlyptotek(Co-penhagen1951).

Poulsen 1941:V.H.Poulsen,Etattiskskulpturfragment fraFidias’ tid, inArkæologiskeogKunsthistoriskeAfhandlingertilegnedeFrederikPoulsen.FestskrifttilFrederikPoulsen(Copenhagen1941).

Panzanelli 2008:R.Panzanelli(ed.),TheColorofLife.PolychromyinSculpturefromAntiq-uitytothepresent.Exhibitioncatalogue,J.PaulGettyMuseum2008(2008).

Plahter 2002:U.Plahter,Asurveyonpigmentsandpigmentapplicationon31Norwegianmedieval oil paintings dated to 1250–1350, in R.Vontobel (ed), ICOM-CC 13th triennial meeting,RiodeJaneiro,preprints(London2002)446–454.

Plenderleith – Werner 1971:H.J.Plenderleith–A.E.A.Werner,Theconservationofantiqui-tiesandworksofart(2nded.,London1971).

Price 2007: M. Price Preparation and application of mineral pigments, inV. Brinkmann,R.Wünsche(eds.),GodsinColor.PaintedsculptureinclassicalAntiquity.Exhibitioncata-logue,ArthurM.SacklerMuseum,Harvard(2007)217–218.

Reuterswärd 1960:PatrikReuterswärd,StudienzurPolychromiederPlastik.Griechenlandund Rom.Untersuchungen über die Farbwirkung derMarmor- und Bronzeskulpturen(Uppsala1960).

Richter 1928/29:G.M.A.Richter,WeretheNudePartsinGreekMarbleSculpturePainted?,

MetropolitanMuseumStudiesI,1928/29,25–31.

Richter 1944:G.M.A.Richter,PolychromyinGreekSculpture,withSpecialReferencetotheArchaicGravestonesintheMetropolitanMuseum,AJA48,1944,327–336.

Rorimer 1931:J.J.Rorimer,Ultra-violetraysandtheiruseintheexaminationofworksofart,MetropolitanMuseumofArt(NewYork,1931).

Santamaria – Morresi 2004:U.Santamaria–F.Morresi,Leindaginiscientificheperlostudiodellecromiadell’AugustodiPrimaPorta,inLiverani2004,241–248.

Santamaria – Morresi – Delle Rose 2004:U.Santamaria–F.Morresi–M.DelleRose,Indaginiscientifichedeipigmentielegantidellelastremarmoreedipintedell’AuladelColossodelForodiAugusto,inLiverani2004,281–289.

Sauerberg et al. 2003:M.L.Sauerbergetal,Thewallpaintingsofc.1300intheAnte-ReliquaryChapel,NorwichCathedral and theThornhamParvaRetable: a technical comparison,inAnnMassing(ed.),Paintingandpractice–TheThornhamParvaRetable–Technique,conservationandcontextofanEnglishmedievalpainting.TheHamiltonKerrInstitute,UniversityofCambridge(2003)189–199.

Scharff 1995:M.Scharff,ThealtarfrontalfromLøgum,inNorwegianMedievalAltarFron-talsandrelatedmaterials (papers fromtheConference inOslo16–19December1989)publishedinActaadarchaeologiametartiumhistoriampertinentia,XI,1995,83–91.

Page 40: Tracking Colour

40InvestigatingthepolychromyofaClassicalAtticGreekmarblefemaleheadNCGIN2830Mikkel Scharff, Rebecca Hast, Nicoline Kalsbeek, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Scharff 1999:M.Scharff,EarlymedievalpaintingtechniquesinNorthernEurope:adiscus-sionoftheuseofcoloredgroundsandothernotabletechniques,inPolychromeSkulpturinEuropa,Tagungsbeiträge,Symposium,Dresden(1999)47–52.

Schmaltz 2005:B.Schmalz,GöttingischeGelehrteAnzeigen257,2005,9–33(reviewofBrink-mann2003).

Straub 1984:R.E.Straub,Tafel-undtuchleinmalereidesMittelalters,inReclamsHandbuchderkünstlerischenTechniken,Bd.1,(Stuttgart,1984)125–260.

Tångeberg 1986:P.Tångeberg,MittelalterlicheHolzskulpturenundAltarschreineinSchwe-den(Stockholm1986).

Thieme – Emmerling 2001:C.Thieme–E.Emmerling,ZurFarbigkeitderTerracottaarmee/OnthePolychromyoftheTerracottaArmy,inC.Blänsdorf–E.Emmerling–M.Petzet(eds.),TheTerracottaArmy of the First Chinese Emperor Qin Shihuang, ICOMOS (München2001)335–369.

Villard 2008:L.Villard,L’essorduchroamtismeauIVesiécle:quelquestemoignagescontem-porains,inA.Rouveret–S.Dubel–V.Naas(eds.),Couleursetmatiéresdansl’Antiquité:textes,techniquesetpratiques(Paris2006)43–54.

Wallert 1995:A.Wallert,UnusualpigmentsonaGreekmarblebasin,inStudiesinConserva-tion40,1995,177–188.

Walmsley 1993:E.Walmsleyetal.,EvaluationofPlatinumSilicideCamerasforUseinInfra-red Reflectography, in ICOM Committee for Conservation, 10thTriennialMeetingPreprints,WashingtonD.C.(1993),57–62.

Waywell 1978:G.B.Waywell,TheFree-StandingSculpturesoftheMausoleumatHalicarnas-sosintheBritishMuseum(London1978).

Wu – Zhang – Petzet – Emmerling – Blänsdorf 2001:WuYongqi–ZhangTinghao–M.Petzet–E.Emmerling–C.Blänsdorf (eds.),ThePolychromyofAntiqueSculpturesand theTer-racottaArmyof theFirstChineseEmperor. StudiesonMaterials, PaintingTechniquesandConservation.ArbeitsheftedesBayerischenLandesamt fürDenkmalpflege, Bd. 11(München2001).

Page 41: Tracking Colour

41

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

Preliminary results from geochemical analysis of pigments on ancient Greek & Roman marble sculptures

Minik T. Rosing1 and Jan Stubbe Østergaard2

Introduction

This is the report on analyses carried out within the framework of the Copenhagen Poly-chromyNetworkPilotProject.3 The aim was to use geochemical methods and geological reference materials to determine the mineralogical identity of ancient pigments, and to testwhetherspecificregionsofprovenanceofspecificpigmentscanbeidentified.Thelat-tercouldbeaddressedthroughmajorandtraceelementandisotopiccompositionsorbydifferences in the mineralogy of impurities in the pigments. When classifying ancient pig-ments,isimportanttokeepinmindthatclassificationofmineralsinthepresentmeaningofthewordisarathernewconcept,whichprobablyhadnomeaninginClassicalAntiquity.Likewise, the understanding of chemical components was also sketchy at best, at the time ofmanufacturingofthepigments.Forthesereasons,pigmentswouldprobablyhavebeenclassifiedaccordingtocoloursandtechnicalpropertiesandperhapsaccordingtothesiteof their origin.4Wemay therefore introduceaschemeofclassification that transgressesancientclassificationschemesandwhichonlyreflectspoorlythefactorsuponwhichan-cientartistswouldhavebasedtheirchoiceofpigmentsforagivenproject.5However,aclas-sificationbasedonmoderngeochemicalandmineralogicalprinciplesmayprovideinsightintotheoriginsandprocessesofmanufacturingofthepigments.InthisPilotProject,twosculptures in the Ny Carlsberg Glyptotek with red pigmentation were investigated by two different approaches.

Identification of red pigments

ThefirstwasacolossalmarbleheadofZeus/Jupiter(Fig.1–5),6 which carries a red pigment onthebeard,cheeksandeyebrows.Asampleofthepigmentanditssubstratewasscrapedfrom the sculpture. The powder from the beard was investigated under the microscope and by qualitative analysis by EDS (energydispersiveX-rayspectrometry) inaSEM (scanningelectronmicroscope).Thepowderwasfixedonstickytape,goldcoatedandinvestigatedun-der the SEM.Thisshoweddiscretepigmentgrainsmixedwithsmallparticlesofcarbonatepresumablyderivedfromthesubstrate.Quantitativeanalysisofthepigmentgrainscould

1 Responsibleforanalysis.NordicCentreforEarthEvolution, NaturalHistoryMuseumofDenmark.ØsterVoldgade5–7,DK-1350KøbenhavnK,Denmark.Email:[email protected]

2 Responsibleforarchaeologicalcomment.Researchcurator,Ancientart, NyCarlsbergGlyptotek,DantesPlads7,DK-1556KøbenhavnV.Email:[email protected]

3 SeetheintroductiontothePilotProject.

4 Onancientpigmentsandterminology:Koch2000,37–50andAnhangII(DiePigmentedergriechischenMalerei);DerNeuePaulyIV(1998)s.v.Farben(N.Hoesch).

5 SeeforexamplePollitt2002.

6 Inv.no.1664.Moltesen1996,227–229,no.101.ThereportedfindspotisalocalityinsouthernLatium.Thesuggesteddate is c. 100 BCE.Seearchaeologicalcommentbelow.

Page 42: Tracking Colour

42Preliminary results from geochemical analysis of pigments on ancient Greek & Roman marble sculpturesMinik T. Rosing and Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.5:AsFig.1.Detailofredinthearea of the left eye. Photo courtesy M.Abbe.

Fig.1:HeadofZeus/Jupiter,NyCarlsbergGlyptotekIN1664.Marble.Height54cm,width45cm.Fromanacrolithstatue.Frontview. Museumphoto.

Fig.2:AsFig.1.Leftprofile. Museumphoto.

Fig.3:AsFig.1.Rightprofile. Museumphoto.

Fig.4:AsFig.1.Detailoftheleftsideof the face with red on cheek and beard.PhotocourtesyM.Abbe.

Page 43: Tracking Colour

43Preliminary results from geochemical analysis of pigments on ancient Greek & Roman marble sculpturesMinik T. Rosing and Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

not be carried out by this method. The qualitative analysis showed that the pigment grains havehighconcentrationsofPb(lead),mostlikelyintheformofleadoxide.Chlorineandtracesofbarium(Ba)andsulphur(S)werefoundinvaryingproportionsinassociationwiththePb.ThissuggeststhatthepigmentcouldbeanimpurePb-oxidemineralpowdercon-taining other minerals derived either from an impure source, or formed by reaction of the primaryPb-oxidewithagentsfromtheenvironmentduringthemillenniapassedsincetheapplicationofthepaint.ItislikelythatBaandSrepresentthemineralbarite(BaSO4),whichmay have been used as a white pigment or may be a constituent of the marble substrate. InadditiontothePb-richpigmentgrains,themineralchromitewasalsoidentifiedinthepowder. This may possibly be a contaminant from the preparation of the pigment, as this mineral does not commonly occur with Pb-ores. Chromite is a common mineral in serpen-tinerocksthatarefoundthroughouttheMediterraneanregion,butcouldalsobederivedfrom gabbro or other dense massive rocks hypothetically used as mortars during crushing of the pigment.

Interpretation

Thesimplest interpretationisthattheredpigmentistheleadoxidemineralredleadorminium(Pb3O4).ThetermminiumhaspreviouslybeenappliedtobothPboxideandtothemercury sulphide mineral cinnabar. Pliny distinguishes between true scarlet minium, which was probably cinnabar, and an inferior product called minium secondarium derived from silver and lead foundries.7Thelatterwasprobablytheleadoxidemineralwenowcallmin-ium. The high content of Cl in some grains might hypothetically be due to the presence of theminerallaurionite(PbClOH),whichmayhavebeenpresentinthesourceofthepigment,orperhapshaveformedbyreactionofprimaryminiumasaconsequenceofexposuretosaltwater.ThishasbeenthecaseintheslagpilesaroundtheminesofLaurionSEofAthenswhere from a number of secondary Pb minerals including Laurionite have been described. The provenance of the Pb in the pigment has been tested by Pb isotopic analysis by ICP-MS(inductivelycoupledplasmamassspectrometry8).ThePbisotopicdataunambiguouslyshowsthatthePbcannotbederivedfromLaurionortheAegeanregionandrecentcontami-nationbymodernindustrialleadproductscanalsoberuledoutasasourceofthelead.InaPb-isotope discrimination diagram by Kylander et al.,9 the isotopic composition of the lead inthepigmentplotliesoutsidethefieldsofanyRomanleadoresminedatthetime.Thismightsuggestthattheminiumwasindeedpreparedasaby-productofsilverproduction.Inantiquity silver was produced by a process called cupellation. Theprocessinvolvestwosteps.Firstly,thesilveroreisheatedwithmetalliclead,whichdissolvesthesilver,andformsanalloythatcanbeseparatedfromtheslag.Secondlythelead–silveralloyisheatedinafurnace.Duringthisprocesstheleadisoxidizedtothewhiteleadoxidelitharge(lithargyrus=silverstone)whichseparatesasafroththatfloatsontopofthe silver metal. The litharge is subsequently converted to minium by heating in air to c. 450° C.Thelithargeandtheminiumproducedfromitthereforecontainamixtureofleadfromboththeoreandtheleadaddedduringthefirststepoftheprocess.Followingtherationale

7 Plinius theElder,NaturalisHistoria 33, 111–119 (trans.H.Rackman, LoebClassical Library,Vol. IX,Cambridge,MA, 1952);WaltonandTrentelman2009,846–847.

8 AnalysedataVGElementalAxiuminstrumentattheUniversityofCopenhagen,andthedatacorrectedformassbiasusing a Ta spike. T. Kokfelt personal communication 2004.

9 Kylander2005,467–485.

Page 44: Tracking Colour

44Preliminary results from geochemical analysis of pigments on ancient Greek & Roman marble sculpturesMinik T. Rosing and Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

of Walton and Trentelman, 2009,10theisotopiccompositionofthepigmentonZeus’sbeardhasbeencomparedtoPbfromtheRioTintosilveroreandleadfromtheleadmineatMurciainSouthernSpain.Leadingotsstampedwith‘NovaCarthago’havebeenfoundinexcavatedsilversmeltingplantsneartheancientRomansilverminesofRioTintoinsouthernSpain.11 ThisleadwasprobablyextractedfromtheMurciaminebyCartagena.Theisotopiccomposi-tionoftheredleadpigmentusedtocolourZeus’sbeardplotsonamixinglinebetweentheMurciaandtheRioTintoleadcompositions(Fig.6),andisthereforemostlikelymanufac-tured from litharge from the silver mines at Rio Tinto. The mass balance suggests that about 1/3oftheleadusedintheprocesswasleadmetalfromCartagena,and2/3wasleadcarriedby the lead – silver ore.

The second sculpture investigated in this Pilot Project was anArchaic Greek limestoneSphinxfromAttica12withtracesofredcolour.Hereweusedahand-heldXRF(X-rayfluo-rescence spectrometer13).Thisinstrumenthasaca.1cm2 area of analysis. The analysis of afaintredpigmentationoftheSphinxshowedhigherlevelsofFeintheseareascomparedto background, and there were no traces of Pb or other heavy elements in the fluorescence spectra from the pigmented areas, which lead to the conclusion that the red pigment on thissculpturewasaFe-oxidebasedpigmentsuchasredochre,whichisabundantlypresentthroughouttheMediterraneanregion.

10 Walton – Trentelman 2008.

11 Domergueetal.2006,14.

12 Inv.no.1203.Johansenetal.1994,40,no.4.Thepreliminaryresultsoftheexaminationofthissculpturearepresentedin an article further below.

13 Innov-XAlphaSeries8000LZX/R.

Fig. 6: The isotopic composition ofPb from the red pigment of Zeus’sbeard (open square). Also shownarecompositionalfieldsforPbfromRio Tinto and Cartagena together with analysis of red pigments from Egyptianmummies shown as opencircles from Walton – Trentelman, 2009.ThePbfromtheZeuspigmentsplots onamixing linebetween theCartagenian and the Rio Tinto lead, and is dominated by lead from the Rio Tinto source.

Page 45: Tracking Colour

45Preliminary results from geochemical analysis of pigments on ancient Greek & Roman marble sculpturesMinik T. Rosing and Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Conclusions

Ithasprovenpossibletoidentifythemineralogicalcompositionofpigmentsusedonan-tiquemarbles,and,inthisexample,distinguishbetweentwodifferentredpigments,whichsuperficiallyresembleeachother.TheanalysisleaveslittledoubtthatthebeardontheZeushead was deliberately painted with a red lead pigment. The pigment was most likely manu-factured on a large scale during Roman times by secondary heating of litharge which was a by-product from silver production by cuppelation, and probably derived from Rio Tinto in SouthernSpain.ThelithargecarriedasignificantamountofleadfromMurcia/Cartagenaaddedasmetalduringsilverextractionbycupellation.

Itseemsthatthehand-heldXRFspectrometerisaveryefficientexploratorytool,toguidemore specific analytical strategies, such as isotopic analysis, XRD (X-ray diffraction) andEDS or WDS(wavelengthdispersiveX-rayspectrometry)microanalysis.TheXRF instrument has the clear advantage that it is easy to operate, fast and non-destructive, while the other methods require micro-sampling of the artefacts.

ArchaeologicalcommentontheZeus/JupiterIN1664

Thisoverlife-sizeheadinGreek(?)marblecomesfromanacrolithicstatue(onlyheadandnudepartsinstone;dressinothermaterialsmountedonawoodencore).14ItwasacquiredfortheNyCarlsbergGlyptotekbyWolfgangHelbigin1898,ontheartmarketinRome.ThefindspotwasreportedtoHelbigasbeingtheruinsofaRomanvillasouthofCeprano,nearS.GiovanniIncarico,intheinteriorofsouthernmostLatium.15 There is no reason to doubt thisinformation:thesitewasnotofanyconsequence. AcoupleofkilometresnorthofthevillageofS.GiovanniIncaricolayFabrateriaNova,16 a coloniaofRome’s,foundedin124BCEafterthedestructionoftherebellioustownofFregellaein 125 BCE.ThislatterimportanttownwassituatedanothertwokilometresnorthofFabra-teriaNova,onthebanksoftheriverLiriandastridetheViaLatina.17 TheheadhasgenerallybeenidentifiedasZeus/Jupiter,18thoughPoseidon/Neptuneisalsoapossibility.IthasbeensuggestedthatitisfromacultstatueatFabrateriaNovaorFregel-lae.19 Stylistically, theheadbelongs togetherwithagroupof fragmentaryabove life-sizesculpturesfromCentralItalyinGreekmarble,byGreeksculptors,ofthemid2nd to mid 1st century BCE.Themajorityareacrolithsandtheyaregenerallyregardedascomingfromcultstatues.20

14 Themarblehasnotbeenidentified.TothebibliographygivenbyMoltesen1996,227–229,no.101add:Reuterswärd1960,200;Reusser1993,104n.61;Dörig1995,301–302;LIMC1997;Coarelli–Monti1998,108no.101;Ghisellini2003–2004, 514 no. 22.

15 LetterstoCarlJacobsen11September.,17September.and15October,1898,intheNyCarlsbergGlyptotekarchives.

16 Coarelli1984,207–208.

17 Cf.Coarelli–Monti1993.

18 Cf. LIMCVIII(1997)342s.v.Zeusno.219b(I.Leventi–V.Mackaira).

19 Reusser1993,104n.61;Moltesenetal.1996,227–228(fromtheCapitoliumatFabrateriaNova?).NoCapitoliumorothersanctuariesrelevanttoJupiter/Neptunehavebeenidentifiedinthetwotowns.

20 Forthegroup,seeMartin1987,207–248(catalogue);Reusser1993,104n.104(additionstoMartin’scatalogue);Ghisellini2003–2004,510–519(updatingofMartinandReusser,includespiecesfromelsewhereinIntaly).

Page 46: Tracking Colour

46Preliminary results from geochemical analysis of pigments on ancient Greek & Roman marble sculpturesMinik T. Rosing and Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

ThesesculpturesmaybroadlyspeakingbedescribedasLateRepublicanorLateHellenis-ticinstyle,andaretobeseeninthecontextofthehellenizationofLatiumandCampaniainthewakeoftheincreasedRomancontactwiththeGreekworldoftheEasternMediterra-nean from the early 2ndcenturyonwards.A‘Hellenistic’anda‘Classicizing’stylisticcurrentrunningsidebysidemayberecognized.Theidentificationanddatingofindividualpiecesisoftendifficult.21 Our head is usually dated to ‘1st century BCE’andseenasavariantoftheZeusOtricoliintheMuseiVaticani,aworkwhichinitsturnisoftenconnectedwithastatueofSerapisinAlexandria,byBryaxisandbelongingtothelate4th century BCE.

The remains of colour on IN 1664 have so far been interpreted either as evidence of the ritual paintingwithredofcultimagesofJupiterrecordedbyancientauthors,orasamordantforgilding.TracesofgildingwerenotobservedwhentheheadwasexaminedunderbinocularmicroscopeinSeptember2009.22Indirectevidencethatthehairoftheheadwasgivensomesort of surface covering is found in the raw treatment and ancient piecing of the back. The fact that we are dealing with an acrolithic statue also entails the element of colour provided bythematerialusedforthesurfacesofthenon-marbleparts(drapery). Inthegroupofsculptureswithwhichthispieceisrelated,weareawareofonlytwoin-stancesofpreservedpolychromy:theheadofthecultstatueofDianaNemorensisintheGlyptotek has faint traces of the pupil in its left eye, and red is preserved on the sandal sole ofafootfragmentoftheFidesfromhertempleontheCapitolineHillinRome.23Indirectevi-dence of polychrome elements on Late Republican cult statues are to be found in the form of inlaid eyes and piercing of ears for earrings.24Inlaideyesaretoourknowledgenotrepre-sentedamongthemalesculpturestowhichourheadisrelated,butarefoundinHellenisticmalemarblecultstatuesfromGreece–acaseinpointbeingtheheadofAnytosfromthegroupofcultstatuesintheDespoinasanctuaryatLykosourabyDamophonfromMessene,ofthelate3rd century BCE25ItshouldbenotedthataredpigmentwasobservedonthefootofDespoinaandthecuirassofAnytos;itisapparentlynolongervisible.Thequalityofthelocal stone used at Lykosoura would seem to recommend some sort of surface treatment.26

This does not affect the fact that the valuable evidence of painted sculptural polychromy provided by IN 1664 has limitations because it stands completely isolated. Comparative data from similar monuments are not available. The results of the analysis carried out would seemtoconfirmthatthepigmentwasappliedinancienttimes,butwhetheritistheorigi-nal polychromy or a secondary one, we cannot tell. Tomakeanyprogress,closeexaminationisneededoftheother,similarsculpturesfromCentralItaly–andofLateHellenisticmarblecultstatuesfromtheGreekworld.

21 FortheseaspectsseeReusser1993,104–112andGhisellini2003–2004,479–509,bothwithextensivereferencestorel-evant literature.

22 Reuterswärd1960,198–200w.n.560(ritual);Moltesen1996,227(forgilding?).MicroscopybyMr.MarkAbbe,LaRentaFellowattheFairchildCentreofObjectsConservationattheMetropolitanMuseumofArt,NewYork.

23 DianaNemorensis:IN1517.Martin1987,236–237no.15;Moltesen1996,205–206no.90(colournotmentioned).Fides:Reusser1993,218–219no.1d.

24 Inlaideyes:Femalehead,MuseiCapitoliniinv.no.253,Ghisellini2003–2004,511no.2.Earspierced:FeroniafromTer-racina,Terracina,MuseoCivicoinv.no.16,Martin1987,232,no.13.

25 MostrecentlyH.SchraudolphinBol2007,190–197.397(bibliography).figs.174k-m.Cf.Ghisellini2003–2004,490–493.

26 Faulstich1997,163w.n.682

Page 47: Tracking Colour

47Preliminary results from geochemical analysis of pigments on ancient Greek & Roman marble sculpturesMinik T. Rosing and Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

references

Bol 2007:P.C.Bol(ed.),DieGeschichtederantikenBildhauerkunst.IIIHellenistischePlastik(FrankfurtamMain2007).

Bilde 1996: P. Guldager Bilde,The Sanctuary of DianaNemorensis.The Late RepublicanAcrolithicStatues,ActaArch66,1995,191–217.

Coarelli 1984:F.Coarelli,Lazio(Rome1984).

Domergue et al. 2006: C.Domergue – P.Quarati –A.Nesta – P.R.Trincherini, Retour surles ingots de plomb de Commachio (Ferrara, Italie) en passant par l’archeométrie etl’épigraphie.Online:http://arxiv.org/ftp/physics/papers/0605/0605044.pdf

Faulstich 1997:E.I.Faulstich,HellenistischenKultstatuenundihreVorbilder(FrankfurtamMain1997).

Johansen et al. 1994:F. Johansenetal.,Catalogue.NyCarlsbergGlyptotek.Greece intheArchaicPeriod(1994).

Koch 2000: N.J. Koch,Techne und Erfindung in der klassischenMalerei. Eine terminolo-gischeUntersuchung(2000).

Kylander et al. 2005:M.E.Kylanderetal.(2005),Refiningthepre-industrialatmosphericPbisotopeevolutioncurveinEuropeusingan8000yearoldpeatcorefromNWSpain.EarthPlanet.Sci.Letters240,2005,467–485.

Martin 1987:HansGüntherMartin,RömischeTempelkultbilder.EinearchäologischeUnter-suchungzurspätenRepublik(Rome1987).

Moltesen 1996:M.Moltesen,Catalogue.NyCarlsbergGlyptotek.EtruriaandCentral Italy(1996).

Plinius the Elder:NaturalisHistoria33,111–119(trans.H.Rackman,LoebClassicalLibrary,Vol.IX,Cambridge,MA,1952).

Pollitt 2002:J.J.Pollitt,PeriChromaton:WhatAncientGreekPaintersThoughtaboutColor,inM.A.TiveriosandD.S.Tsiafakis(eds),ColorinAncientGreece.TheRoleofColorinAn-cientGreekArtandArchitecture700–31BC(Thessaloniki2002)1–8.

Reusser 1993: Ch. Reusser, Das Fidestempel auf demKapitol In Rom. BullCom Suppl. 2(Rome1993).

Reuterswärd 1960:P.Reuterswärd,StudienzurPolychromiederPlastik.GriechenlandundRom(Uppsala1960).

Walton – Trentelman 2009:M.S.Walton–K.Trentelman,2008.Romano–Egyptianredleadpig-ment:AsubsidiarycommodityofSpanishsilverminingandrefinement.Archaeometry51,2009, 845–860.

Page 48: Tracking Colour

48

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

Raman spectroscopy characterization of colored pigments in archaeological materials

Rolf W. Berg

INTRODUCTION

Archaeologicalartifactsandartworkswithcolorsarebeingincreasinglyanalyzedintheseyears. The pigment materials used for their creation are much studied after the development ofmodernnon-destructivemicrosamplinganalyticaltechniques.Arthistorians,museumconservators, and archaeological scientists are now much aware of the importance of physi-cochemical characterization for the attribution of the historical period and genuineness of anitem.Ancienttechnologicalmethodsusedintheconstructionoftheitemsmaybechar-acterized by spectroscopists with a minimal disturbance to the artifacts or artworks.

InthisconnectiontheRamanspectroscopytechniquemustbeconsideredamostelegantmethod for pigment and materials analysis of relevant museum and archaeological ma-terials. This is done by correlating some bands in the studied pigments with those of well characterizedreferences.TheuseofRamanspectroscopycanbetakentoillustratethis:Itprovides e.g. information of importance to art restorers and museum conservation scien-tistsinpreservingmaterialsandtheunderstandingofdeteriorationprocesses.Itdoessobyidentificationofkeycomponents,asshowninFig.1.

PriortotheadventofmodernRamanspectroscopyitwasdifficulttoanalyzearchaeologi-calmaterials,duetodifficultiesarisingprincipallyfromthegenerationoffluorescencebyshorter-wavelength visible radiation. There was also the real possibility of sample degrada-tion occurring from the use of the relatively high laser powers that years ago were necessary forsufficientsampleexcitation.ThemajoradvantagesofRamanspectroscopyoverinfraredspectroscopy,namelytheweakRamanscatteringofwater,couldnotbefullyexploited.Thissituation has changed recently, as shown below in this review.

ThenumberofresearchpapersonthesubjectofRamanspectroscopyappliedtopigmentsand art has been growing very fast during the last years. To get a comprehensive overview werefertothreerecentthemenumbersof JournalofRamanSpectroscopy1,2,3 and other dedicatedtextssuchase.g.Edwardsetal.4, 5

To help the reader we start by a short presentation of some technical details in Raman spectroscopy.

1 P.Vandenabeele,Ramanspectroscopyinartandarchaeology,J.RamanSpectrosc.2004;35nos.8–9:607–609,andtherestofthespecialissuecontainingfurther31dedicatedpapers,pp.606–818.

2 W.Kiefer,Ramanspectroscopyinartandarchaeology,J.RamanSpectrosc.2006;37:no.10,961–1237.

3 P.BaraldiandA.Tinti,Ramanspectroscopyinartandarchaeology,J.RamanSpectrosc.2008;39:963–965andthefol-lowing 22 papers.

4 H.G.M.Edwards.RamanspectroscopyinthecharacterizationofArchaeologicalMaterials,chpt.26(p.1111–1044)inHandbookofRamanSpectroscopy.FromtheResearchtotheProcessLine.EditedbyI.R.LewisandH.G.M.Edwards,M.Dekker,N.Y.2001.

5 Edwards,H.G.M.,Farwell,D.W.,Theconservationalheritageofwallpaintingsandbuildings:anFT-Raman spectro-scopicstudyofprehistoric,Roman,mediaevalandRenaissancelimesubstratesandmortarsJournalofRamanSpec-troscopy,Vol.39,Issue8,987–992,2008.

Page 49: Tracking Colour

49Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.2:Therelationshipsbetween infrared absorp-tion, Rayleigh and Raman scattering;seethetextfordetails. The wavenumber σ is the number of waves per cm and a measure of the energy, the frequency or the reciprocal wave-length.

Fig.3:Ramanspectrumofsulfurcrystals;a‘notch’filtertakesawaythelight between ca. 50 and -150 cm-1, so that neither the Rayleigh scattering at ca. 0 cm-1 nor the Raman bands in the range from ca.70to-100cm-1 are seen.Someauthorsflipthe direction of their wavenumbershiftaxis.

Fig.1:TheroleofRamanspectroscopy in the char-acterization of art and archaeological materials.

Page 50: Tracking Colour

50Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

BRIEFINTRODUCTIONTORAMANSPECTROSCOPY

The Raman scattering effectwasdiscoveredin1928byIndianphysicistsRamanandKrishnan.6 Ramanscatteringmaybedefinedasinstantaneousinelasticscatteringoflight(electromag-neticradiation).Whenalight quantum (akindofparticleinthelight,calledaphoton)collideswith a sample, the photonmaybescattered,eitherelastically(calledRayleigh scattering)orinelastically (called Raman scattering). In the Ramanprocess an amount of energy is ex-changed with the sample as shown schematically in the quantum energy level diagram in Fig.2.Accordingly,theoutgoingphoton has less or moreenergythantheincomingone(Stokesoranti-StokesRamanprocesses).Theenergyismeasuredbythefrequencyv or wavenum-ber σaccordingtotherelationE=h v=h c σ(hisPlanck’sconstantandcisspeedoflight).ARamanprocesscorrespondstoa(fundamental)transitionamongcertaingroupvibrationalstatesasdescribedinthetheoryofquantummechanics.Forapeak(aRamanband)tooccurintheRamanspectrumwithasignificantintensity,themolecularbondstretchingorangledeformationvibrationmustcauseachangeintheso-called‘polarizability’ofthemolecule(thepolarizabilityistheabilityofamoleculetobecomepolarwhensubjecttoanelectricfield).TheensembleoflightscatteringbandsconstitutestheRamanspectrum.MostoftenonlyStokes-shiftRamanspectraaremeasuredandshown;i.e.thescatteredphotonshavelowerfrequencythantheincidentradiation.TheStokesandanti-Stokesspectraofatypicalsample,yellowsulfurcrystals,isshowninFig.3.

Duringaninfrared(IR)absorptionprocess,aquantumofIRradiationisabsorbed(aphoton ofaparticularenergyEandfrequencyv or wavenumber σ,againbecauseE=hv=h c σ).Themolecularsystemundergoesatransitionfromthegroundstate(quantumnumberv =0)toanexcitedstate(v=1),inthepresentcasecorrespondingtoe.g.abondstretchinginaCH3 group and a wavenumber shift of ca. 2900 cm-1. Incontrast to this IR absorption, during Rayleigh and Ramanscattering,anexcitingphotonofmuchhigherenergy(visiblelight)hitsthemolecularsystemandraisesittoavirtualstate,fromwhereit‘immediately’fallsback.Therearetwopossibilities,illustratedinFig.2withgreenAr+ light of 514.5 nm wavelength, correspondingtoawavenumberof19435cm-1. In theso-calledStokesRamanscattering(notsolikely),thesystemfallsbacktothev=1state(emittinga16535cm-1photon),orinRayleigh scattering (more likely) to thev=0 ground state (emitting light at ~19435 cm-1),producingtheso-calledRayleighwing.Ifthesystemisstartingfromthev=1state(notsolikelyatroomtemperaturebecauseoftheBoltzmanndistribution),similartransitionscanhappen.Nowalsoaso-calledanti-StokesRamanprocessispossibleproducingphotonsat22335cm-1.MostRamanspectroscopystudiesreportdatacorrespondingtoStokesRamantransitions.Samplesorimpuritiesthereinhavingenergystatesnearthe‘virtual’ones(hereate.g.~19435cm-1)mayabsorbphotonsfromtheincidentlightandlaterre-emitthelightas a broad intensive background called fluorescence.

Dramaticimprovementsininstrumentation(lasers,detectors,optics,computers,etc.)dur-ing recent years have raised the Raman spectroscopy technique to a level where it can beusedfor‘speciesspecific’quantitativechemicalanalysis.Althoughmanytimesnotassensitive as e.g. infrared absorption, the Raman technique has the advantage that it can non-destructively measure samples directly without touching or any sample preparation. Furthermore,byuseofpolarized lighttechniques,onecanderivemolecular information

6 Raman,C.V.andKrishnan,K.S.,AnewTypeofSecondaryRadiation,Nature,121,501,1928.

Page 51: Tracking Colour

51Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

that cannot be obtained from infrared spectra. Good starting references dealing with Raman instruments and lasers are perhaps.7,8,9,10,11

RamanandInfraredtechniquesarecloselyinterrelatedinthattheybothsamplemolecules.Thespectralbandsdependoncharacteristic‘group’motionsinthemoleculespresentinthesample.Onemaysaythatthevibrationalbandsinthespectragiverisetoa‘fingerprint’ofthemolecules.Asanexample,bandsoccurringnear2950cm-1 often arise from transitions foraliphaticC–Hstretchingvibrations(althoughsometimesperturbedby‘Fermi-resonance’withovertonesandothernonfundamentaltransitions).So-calledempiricalgroupfrequen-cychartsareavailable,specifyingthe‘fingerprint’bands,thatmaybeusedtoidentifypurematerialsorthepresenceofaparticularcomponentinamixture,seee.g.12,13,14,15

Although similar transitional energy ranges occur in IR and Raman spectroscopy, differ-ent selection rules govern the intensities in Raman scattering and IR absorption spectra. Hencebothtypesofspectramayberequiredtofullycharacteriseasubstance:Anecessaryrequirementforamolecularmotion(suchasavibration,rotation,rotation/vibration,orlat-ticenormalmode)tobemeasurableinIR spectra it is needed that an oscillating dipole mo-mentisproducedduringthevibration(inRamanthemotionwithinthemolecularsystemshouldvarythepolarizability).Combinations,differencesorovertonesofthesetransitionscan occur, but normally only weakly. The selection rules of the transitions are described in quantum mechanics and group theory, see e.g.16,17,18,19,20

7 Diem,M.,IntroductiontoModernVibrationalSpectroscopy,J.Wiley&Sons,NewYork,USA,1993.

8 Laserna,J.J.(Ed.),ModernTechniquesinRamanSpectroscopy,J.Wiley&Sons,ChichesterUK,1996.

9 Weber,W.H.,Weber,W.H.&Merlin,R.,RamanScatteringinMaterialsScience,Springer,1–492,2000.

10 McCreery,R.L.,RamanSpectroscopyforChemicalAnalysis,J.Wiley&Sons,Chichester,1–415,2001.

11 Smith,E.andDent,G.,ModernRamanSpectroscopy–ApracticalApproach,J.Wiley&Sons,Ltd.,ISBN0-471-49668-5,1–210, 2005.

12 Lin-Vien,D.,Colthup,N.B.,Fateley,W.G.,Grasselli,J.G.,HandbookofInfraredandRamanCharacteristicFrequenciesofOrganicMolecules,AcademicPress,NewYork,1991.

13 Nyquist, R.A., Kagel, R.O., Putzig, C.L, Leugers,M.A.,Handbook of Infrared andRaman Spectra of Inorganic Com-poundsandOrganicSalts(4vols.),AcademicPress,NewYorkandLondon,1996.

14 Nakamoto,K.,InfraredandRamanSpectraofInorganicandCoordinationCompounds,5thed.,PartA:TheoryandAp-plicationsinInorganicChemistry,andPartB,ApplicationsinCoordination,OrganometallicandBioinorganicChemis-try,JohnWiley&Sons,NewYork,1997.

15 Socrates,G.,InfraredandRamanCharacteristicGroupFrequencies,TablesandCharts,3rded.,J.Wiley&Sons,Chiches-terUK,1–249,2001.

16 Herzberg,G.,MolecularSpectraandMolecularStructure,Vol.II,InfraredandRamanSpectraofPolyatomicMolecules,vanNostrand,NewYork,1945.

17 Wilson,E.B.Jr.,Decius,J.C.andCross,P.C.,MolecularVibrations,TheTheoryofInfraredandRamanVibrationalSpectra,McGraw-HillBookCo.,NewYork,1955.

18 Long,D.A.,TheRamanEffect:AUnifiedTreatmentoftheTheoryofRamanScatteringbyMolecules,JohnWileyandSons,ISBN0471490288,2002.

19 Carter,R.L.MolecularSymmetryandGroupTheory,JohnWiley&Sons,NewYork,1998.

20 Atkins,P.W.andFriedman,R.S.,MolecularQuantumMechanics,4thedition,OxfordUniversityPress,ISBN13978-0-19-927498-7,1–608,2004.

Page 52: Tracking Colour

52Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

FLUORESCENCE, INSTRUMENTATION ANDEXPERIMENTALDETAILS

Applications of Raman spectroscopy have been limited by the presence of fluorescencefromthesamples(orfromimpurities inthesamples). Incaseofstrongfluorescencetheuse of less-energetic near-IRlasersfortheexcitationisoftennecessary.Fourier-transform(FT-)Ramaninstrumentsareavailablethattypicallyapply~1064nmlaserexcitation(fromsolid state Nd-YAG or Nd-YVO4lasers)toavoidthefluorescence.21 The advantage of Raman spectroscopy over IRandotheranalyticaltechniques(ifthefluorescenceproblemscanbecircumvented)stemsfromtheabilityofRamantoidentifydiscretespeciesinsitu.Ramanspectracanbeobtaineddirectlyfrom‘samplesonthewalls.’

The polarization properties of the Raman scattered light may be employed to select only the isotropic intensity of the symmetric vibrational modes, thereby helping conclusive assign-ment of the spectra.

The Raman effect is weak, perhaps only 10-8 of the photons hitting the sample are scattered inRaman.Theuseofhighpowerlasers(tocircumventthelowscatteringefficiency)mayoften result in sample decomposition, and fluorescence interference from impurities must be considered a likely problem when using visible light, at least for samples that do emit muchfluorescence.Thedevelopmentofchargecoupleddetectors(CCDs)andnotchfiltershave revolutionized theRamantechnique.Sampling throughamicroscope–underhighmagnification–isaneffectivewaytocollectRamanlightoveralargesolidangle,andthusonly minute sample quantities are necessary.

MuseumartefactsandartpieceshavebeentheobjectofseveralRamanspectroscopystud-ies as discussed in the following. Raman microscopy is, in principle, an appropriate tech-nique,sinceitallowsfocusingonverysmallgrainsfromthesamples.Howeversuchsam-plesoftenemitintensivebroadfluorescence.Inourownexperimentswithvisible(green514.5orred784nm)laserlight,alsostrongfluorescencewasobserved.Similarobservationshavebeenreportedbymanyotherresearchers.Thereforetogetgoodexperimentalspectraithasbecomecommonpracticetousethe1064nmnear-infraredexcitinglasers(e.g.aNd3+-YAGlaserat100mWofpower)andscanningFT-Ramaninstruments,e.g.theBrukertypeRaman FT spectrometer equipped with a liquid-N2cooledGe-diodedetectorshowninFig.4.Insuchaninstrumentalsetup,thescatteredlight(excitedbythelaserhittingthesample)iscollected–eventuallyatadistancebyafibreopticsystem,seeFig.5andFig.6–filteredfor the laser light itself and sent into the spectrometer for spectral analysis. The spectra arecalculatedafteraveraging~200to500scansfollowedbyapodizationandfast-Fourier-transformationtoobtainaspectral resolutionof~2cm-1 within a stable precision better than 1 cm-1.Thespectramaybecorrectedfor(small)intensitychangesindetectorresponseversus wavelength but this is many times not needed.

BecauseefficientinfraredspectrometersweredevelopedatamuchearlierdatethanRamaninstrumentation, infrared databases for natural or synthetic materials and minerals are morecomprehensivethanforRamanspectra.Ancientspecimensoftengivehighlyfluores-centRamanspectraexcitedwithvisiblelightduetoimpuritiesincoatingsorpreviouscon-servation techniques. These factors have favoured the use of infrared spectroscopy, because the fluorescence emission may swamp the lower-intensity Raman signals.

21 Chase,D.B.,andRabolt,J.F.,FourierTransformRamanSpectroscopy,AcademicPress,NewYork,1994.

Page 53: Tracking Colour

53Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.4:SimplifiedopticaldiagramofaFT-Raman instrument operating with a 1064nminfraredlaser(beamshowngoingdowntotherightinthepicture);BS=beamsplitter. The sample sits at the focus in the lower right corner of the picture or at theendofabundleofopticalfibres(notshown).Seereferencesandtextfordetails.

Fig.5:FibreopticalRamanprobeandtheopticalprinciples.

Fig.6:TheirisoftheheadofZeusintheCopenhagenNyCarlsbergGlyptotekisbeingexaminedwithlightfromafiber.Thepracticalvalueoftheprobeisthatitis easy to get the laser light to the sample, collect the Raman scattered light and send it to the spectrometer, asshowninFig.5.

Fig.7:AnexampleofMycenaeanpaintedplasterartfromBronzeageGreece(theperiodofroughly1450–1200 BC):DolphinscenefromtheQueen’sMegaronatKnossos,Crete.Figurecitedfrom26.

Page 54: Tracking Colour

54Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

On the other hand, in the area of paintings and manuscripts, Raman spectroscopy is used extensivelybecauseof theaccessibilityof the low-wavenumberregionofthevibrationalspectra (<500 cm-1),which aremost important for the characterization of inorganic pig-ments. This region is not easily observed in the infrared.

Manymuseumspecimensdonotallowmechanicalorchemicaldestructivesampling–thetakingofsmallsamplesbydrilling,scraping,orexcisionisprohibitedorstronglydiscour-agedbyarchaeologicalconservators.Theuseoffiber-opticprobesfor‘remote’analysisofartobjectsisnowadvocatedformanyinfraredandRamanapplications.Theobviousadvan-tage of in situ recordingRamanspectrainmuseumenvironmentsorarchaeologicalexcava-tionfieldsisnowaidedbytheappearanceofportableunits,seeFig.6.

With modern FT-Raman spectrometers or dispersive Raman spectrometers using confocal microscopy with CCD detection – characterization it is now possible via Raman spectra of sampleswithdimensionsassmallasafewμm,usingvisibleornear-infraredexcitation.

PIGMENTSTUDIES

Paintiscoveringmanymuseumobjects.Pigmentscanbefoundonsuchitemsasstatues,ceramics,frescoes,paintings,andmanuscripts. Incasesofancientstatues,rockart,wallpaintings and mural frescoes, often only a very small part of the pigments remains, perhaps onlyseveralgrains,becausetheitemshavebeensubjectedtoenvironmentaldegradationdue to wear, time, local climate and restoration.

Inthefollowingwereviewsometypicalinvestigationsmadebymeansofthemicro-Ram-an spectroscopy method. We start with some recent mural pigment studies, because they show most of the essential features of the state of the modern micro-Raman spectrometric method.Otherexamplesnotdiscussedherecanbefoundintheliterature,e.g.analysisofdated authentic wall paintings from archaeological sites representing some 6000 years of human occupation.22,23,24,25

Mural paintings in Mycenaean Greece

ThefirstexampletobediscussedisapilotstudyonBronzeagepaintedplasterinMycenaeanGreece discovered around the end of the 19th century.26Anexampleonthesefragmentarypaintings is shown in Fig. 7.These kind ofwall paintings are claimed to be among thefirst ones to have been executed in the buon fresco technique. This has serious implica-tions for our general understanding and appreciation of the materials used and such early

22 H.G.M.Edwards,L.Drummond,J.Russ.,Fourier-transformRamanspectroscopicstudyofprehistoricrockpaintingsfromtheBigBendRegion,Texas.J.RamanSpectrosc.30:421–428,1999.

23 H.G.M.Edwards,D.W.Farwell,S.Rozenberg.,RamanspectroscopicstudyofredpigmentandfrescopigmentsfromKingHerod’spalaceatJericho.J.RamanSpectrosc.30:361–363,1999.

24 H.G.M.Edwards,D.W.Farwell,F.RullPerez,S.JorgeVillar.,SpanishmediaevalfrescoesatBasconillosdelTozo:AnFT RamanSpectroscopicStudy.J.RamanSpectrosc30:307-312,1999.

25 H.G.M.Edwards,P.Vandenabeele,E.M.Newton,F.RullPerez,S. JorgeVillar.,FT Raman spectroscopic studies of the wall-paintingsatthemonasteryofSanBaudelio,Spain.Appl.Spectrosc,2001.

26 A.BrysbaertandP.Vandenabeele,BronzeAgepaintedplasterinMycenaeanGreece:apilotstudyonthetestingandapplicationofmicro-Ramanspectroscopy,J.RamanSpectrosc.2004;35:686–693.

Page 55: Tracking Colour

55Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.8.Ramanspectraofanindigo sample from Greek painted plaster and a reference of pure indigo. Cited from 26.

Fig.9.Ramanspectraofalimonite sample from Greek painted plaster and a corresponding reference. Cited from 26.

Fig.10.Schematicrepresentationof the organic molecule of indigo or indigotin thatcanbeextractedfromplants. The size of the molecule and interatomic distances are given in Å. Cited from 27.

Fig.11.Ramanspectraof:(1)syntheticindigo,(2)apalygorskite(99%)andindigo(1%),finelygroundunheatedmixture,(3)apalygorskite(99%)andindigo(1%)heated(5h190°C)pigment(syntheticMayablue),and(4)archaeologicalMayablue pigment from a mural paint fromCacaxtla.Indigoisclearlypresent in all samples. Cited from 27.

Page 56: Tracking Colour

56Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

developed technologies applied in prehistoric societies. Non-destructive sampling methods are required for analysis because very little original material remains. Consequently, micro-Raman spectroscopy was applied to this early fragmentary material.26

Interestingresultswereobtained,althoughnotwithoutproblems.Highfluorescencepre-ventedtheidentificationbyRamanofe.g.theEgyptian Blue,knowntobepresent.Egyptianblueisthemostcommonblueinantiquity,anartificialpigmentmadefromcoppersalts,cu-prorivaite,CaCuSi4O10orsimilarsalts.Othercompoundswereidentified,bothorganic(indigo,Fig.8)andnon-crystallinematerials(limonite, an amorphous yellow form of goethite,FeOOH,incombinationwithFe2O3, pointing to ochre and also of goethite used as pigments originally, Fig.9).Suchresultsareimportant,complementingourtraditionalknowledgeknownfromancient sources and obtained from other modern techniques. One should however be care-ful,becausee.g.thered-purplepigmentsidentifiedbyothermeansashaematite α-Fe2O3, did not give Raman signals whereas calcite, CaCO3,wasidentifiedprobablyfromthelimeplastersubstrate.Aclear identificationof indigo, applied in blue painting materials in decorated plastersurfaces(Fig.8)mustbeconsideredofhighinterest.Indigoisknownasaverycolor-fastmaterialandhasbeenmentionedinEgyptiancontextsfromthe5thDynastyonwards(ca.2400BC).Ithasbeenusedasapigmentforwallpaintings,executedintheal fresco tech-niqueinlaterperiods(seethewritingsofPlinyandVitruviusforitsuseinClassicaltimes)bothinEuropeandintheNewWorld.TheidentificationofthismaterialonpaintedplasterfromtheAegeanBronzeAgeisanewknowledgethathasapotentialofshowingtrendsintechnologicaltransfer,innovationandcontinuityfeaturedamongtheAegeanandeasternMediterraneansocieties.

AsthesamplingmethodtinycottonwoolQ-tipswereusedforremovalof1microgramofpigmentfromthesubstrate.TheQ-tipwasthentappedonthesurfaceofacleanedmicro-scopeglassslidetoreleasesomepigmentparticles.Byfocusingthelaserontoeachparticleusinga×50objective,spectrawereobtained.ThemeasurementsweredoneusingaRen-ishawSystem-1000spectrometerwitha785nmdiodelaserandacooledcharge-coupleddevice(CCD)detector.Grainsfromatotalof87samplesweremeasuredandanalyzed.

Mesoamerican Maya Blue Pigments

Mostorganicmoleculescannotbeusedascolorants inartworksbecausetheyfadewithlight,age,andreactwithotherchemicals(oil,resins,substrates,etc.)oftheartworkitselfor present pollutants. On the other hand pigments made from minerals are very stable and are therefore preferred for artistic techniques, whether it is in mural paintings, oil paintings, or polychromatic pottery, etc.

MesoamericanMayapeoplesinpre-Hispanictimes,ataroundtheVII–VIIIcentury,inventedthemarvelousMayaBlueartificialpigment.Thisextraordinarypigmentwasapparentlyob-tained by encapsulating the natural organic compound indigo intoaninorganicmatrixofpalygorskite. Indigo,seeFig.10, isextractedfromthe Indigofera suffruticosa plants. Palygor-skite isaveryparticularfibrousclaymineralpresentingchannels in itsmolecularstruc-ture. Palygorskite (almost free of impurities)may be obtained e.g. from amine close toTicul (Yucatan,Mexico).Theindigo-palygorskitemixtureacquiresaconsiderablestabilitywhenamoderatethermaltreatment isapplied,constitutinganartificial‘mineralization’.

Page 57: Tracking Colour

57Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Thechemicalreasonsoftheunusualstabilityofthepigmentandtheexactmechanismofinteraction between the indigo and the clay are not entirely understood.27,28

TheMayabluetechnologyhasbeenexaminedbyRamanspectroscopy.27Spectraofdiffer-entauthenticsampleswerecomparedwithpreparationsofsyntheticMayaBlueandothermixturesofindigowithotherinorganicmaterials,seee.g.Fig.11.

UnheatedorheatedmixturesofindigowithpalygorskitegiveRamanspectrasimilartothatofMayablue,asdoalsoindigomixedwithotherclays,likesepioliteormontmorillonite,in-dicating that the indigo spectrum is not much affected during the thermal treatment mak-ing the pigment robust.27,28ForRamanstudiesonpalygorskitealone,aninfraredNd:YAG laser and an FT-instrument were used, because a high fluorescence background was hiding thebandswhentheexcitinglaserwasinthevisiblerange.Onthecontrary,indigoproduceswell-definedbandsthatcanidentifythiscolorant,whenoneisusingaconfocalJobin-Yvondispersive instrument with a CCDdetectorandanotchfilter,excitedbyagreenlaser(wave-lengthof532nm,intensityfrom50μWto2mW,producedbyafrequency-doubledNd:YAG laser).FortheidentificationofMayablueonemusttakeintoaccounttheobservedintensityincrease in some indigo bands and the presence of new bands in the Raman spectrum of the archaeological pigment with respect to that of modern indigo, possibly because of inter-actions with palygorskite or loss of planarity of the indigo molecule introduced when the heatingprocesstookplace.Itmaybeaddedthatalsogreenpigmentsmaycontainindigo.28

Nepalese Temple Mural Paintings

To studymaterials and techniques used by ancientAsiatic artists, pigments present inpainted mural decorations in a monastery temple from the 15thcenturylocatedinLoMan-thang,upperMustang,Nepal,havebeenscientificallyexaminedbymicro-Ramanmethods.29 Thetemplehasbeencompletedin1472,andwasbuiltinrammedmudandwood.Cross-sectionsof14collectedauthenticpigmentsampleswereanalyzed.Forthemicro-Ramananalyses paint fragments and cross-sectioned samples were placed on the microscope stage andilluminatedwithlaserlight(632.8nmatapowerrangingfrom0.5to5mW)througha×50objective.ThescatteredlightwasanalyzedwithaLabramdipersiveRamaninstrumentequipped with a CCD detector cooled to 200° K.

The results showed that for the blue colors, azurite (2CuCO3·Cu(OH)2) was used, some-times in combination with lazurite particles. Lazurite is a sodium aluminosilicate mineral, Na8[Al6Si6O24]Sn, (nvaries),withaverydeepbluecolorcomingfromsulfurradicalanionsresidinginholesinthecrystallattice.Itoccursnaturallyinsemi-preciousstonesoflazurite(Lapis Lasulae)andprobablywasimportedtoNepalfromknownsourcesinBadakhshaninnortheasternAfghanistan.TheRamanspectraofthesebluepigmentsareshowninFig.12.

Malachite, a green, basic copper carbonate Cu2CO3(OH)2, that also occurs as a natural mineral, was similarly found to have been used as a pigment to paint green temple decorations, and

27 M.SánchezdelRío,M.Picquart,E.Haro-Poniatowski,E.vanElslandeandV.HugoUc,OntheRamanspectrumofMayablue,J.RamanSpectrosc.2006;37:1046–1053.

28 R.G.Moreno,D.StrivayandB.Gilbert,Mayablue–greenpigmentsfoundinCalakmul,Mexico:astudybyRamanandUV-visiblespectroscopy,J.RamanSpectrosc.2008;39:1050–1056.

29 R.Mazzeo,P.Baraldi,R.LujànandC.Fagnano,CharacterizationofmuralpaintingpigmentsfromtheThubchenLa-khangtempleinLoManthang,Nepal,J.RamanSpectrosc.,2004;35:678–685.

Page 58: Tracking Colour

58Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.12.Ramanspectraobtainedfromthebluepig-mentslayersamplesfromaNepalesetemple:(a)azurite(2CuCO3·Cu(OH)2)and(b)lazurite(Na8[Al6Si6O24]Sn, (nvaries)).Citedfrom29.

Fig.13.RamanspectraobtainedfrompigmentsfoundinorangeandredpaintlayersfromaNepalesetemple:(a)vermilionand(b)haematite. Cited from 29.

Fig.14.RamanspectraobtainedfromtheyellowlayerunderneathgoldleafofBuddha’sfaceinaNepalesetemple:(a)para-realgarand(b)orpiment. Cited from 29.

Page 59: Tracking Colour

59Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

also the copper sulfate hydrate brochantite was sometimes present, perhaps representing an alteration product of malachite.29Similarlyredandorangepaintedlayerswerefoundtobe constituted of orpiment(ayellowarsenicsulfide,As2S3)andvermilion(HgS,knownfromChinasinceprehistorictimes),bothaloneandincombination(seeFig.13).Averyinterest-inggildingtechniquehasbeenusedtoproduceBuddha’sface:goldleafwasaffixedtoanunderlyingmixturecontainingorange-redparticlesofpara-realgar, a polymorph of arsenic sulfide, realgar (As4S4),with traces of orpiment and vermilion. Red ochre (haematite)waspresent in the brown color decorations.29SomerepresentativespectraarereproducedhereinFig.13andFig.14.

Roman Frescoes

Certain‘greenearth’pigments,mainlyconstitutedoftheclaymicamineralsceladonite or glauconite, have been used since antiquity in the creation of artworks.30 The green color comes from small celadonite micaceous scales, about 1–10 μm in length. The chemical com-positionofceladoniteisapproximatelyK[(Al,Fe),(Fe,Mg)](AlSi3,Si4)O10(OH)2 with substitu-tions varying inside the round parentheses. The micro-Raman procedure has been applied tosomefragmentsofRomanfrescoesfromtwoarchaeologicalsitesincentralItalyatSuasa(Ancona)andPisaurum(Pesaro-Urbino)totestthegreenpigments.Ramanspectraofcela-donite(seeanexampleinFig.15)showsfeaturesintherange100–1200cm-1. The spectral regionbelow300cm-1 is considered toshowbands related to internalvibrationsofMO6 octahedra,whereMstandsforSi,Mg,Al,orFeatoms.Inthespectralrange300–800cm-1, bandsaremainlyduetovibrationalmodesofSiO4 tetrahedra.

MostinterestingfortheanalysisofceladonitepigmentsbyRamanistheobservationthatthe use of different lasers revealed resonance effects in the Raman spectra. With visible ar-gonionlaserlightexcitationthebandsat459and960cm-1 were found to be stronger, while forexcitationwithredandinfraredlongerwavelengthsthebandat394cm-1 is more evident. Whenthewavelengthwasincreased,from514.5to785nm,anapparentshifttowardslowerwavenumbers of the band at about 545 cm-1wasobserved(from548to538cm-1),maybedueto intensity changes in an unresolved doublet.30

Thanks to the quality and spatial resolution of Raman microscopy it was possible to identify celadonite from glauconite,alsoemployedintheItalianfrescoes(seeFig.16).Finegreengrainsof glauconitehaveanapproximatechemicalcompositionof(K,Na)(Fe,Al,Mg)2(Si,Al)4O10(OH)2, somewhatlikeceladonite,butwithalargercontentofAlduetoapartialsubstitutionofAlforSi(relativeatomic%Si=5.28,Al=1.25;Mg=0.70;Fe=1.60).Commonimpurities,crystalsof pyriteFeS2, haematite α-Fe2O3, goethiteFeOOH,andFexOyalsooccur.InSuasaandPisaurum,the glauconite and celadonite pigments were found in different frescoes but never found mixedtogether,andceladonitewasmorediffusethanglauconiteinbothlocations.30

Pigmentslabelledas‘greenearth’arecommerciallyavailablenowadays,butaccordingtoRamanexaminations30onlyinsomecasestheyarereallybasedonceladoniteorglauconite;oftentheycontainmixturesofdifferentminerals,fromaegirine to epidote, from lizardite to illite,andinsomecasesanorganicsyntheticdye(pigmosol green)hasbeenaddedaspigmentto a combination of minerals, leading to a product with characteristics very different from that of the green earths.30

30 F.Ospitali,D.Bersani,G.DiLonardoandP.PaoloLottici,‘Greenearths’:Vibrationalandelementalcharacterizationofglauconites,celadonitesandhistoricalpigments,J.RamanSpectrosc.2008;39:1066–1073.

Page 60: Tracking Colour

60Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.15:Ramanspectraofaceladonitesample(relativeatomic%Si=4.54;Mg=0.80;Al=0.24;Fe=1.68)showingresonanceRamaneffectsatshorterexcitationwave-lengths. Cited from 30.

Fig.16:Ramanspectraofglauconite and celadonite found in Roman frescoe fragments coming from archaeological sitesandshowingextrabandsathighwavenumbers.Cited from 30.

Fig.17:RamanspectraofthecomponentsofthewhitecoatingrenderofthewallsoftheTournaiNotre-DameCathedral.Dehydrationofthegypsumseemstohaveoccurred with time resulting in spectra characteristic of the hemi hydrated or anhydrous forms of calcium sulfate.Itisnotoriginatingfromatoohighlaserpower.Cited from 31.

Fig.18:Ramanspectraoftheyellow,orangeandredpigmentsofthewallsoftheTournaiNotre-DameCathedral, cited from 31.

Fig.19:Ramanspectraofyellow,orangetoredironoxidepigmentsfromthewallsoftheTournaiNotre-DameCathedral, cited from 31.

Page 61: Tracking Colour

61Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

European Cathedral Mural Paintings

AnotherstudyexamplehasbeenconcernedwithpigmentsextractedfrommuralpaintingsfromtheNotre-DameCathedralofTournai(Belgium).Samplesfromherehavee.g.beenan-alyzedbymicro-Ramanspectroscopyusingvisiblelightofwavelengths488,514.5and752.6nm and low intensity, 0.1–10 mW.31 The Raman method was preferred because of its non- or micro-destructivecharacter(withsomecaution).Amajorproblemencounteredwas,asinmany other studies, local fluorescence originating from organic binders or fatty pollutants, whichoftenmaskedtheRamanbandstosuchanextentthatusefulspectracouldnotbeobtained.However,variouspigmentsoftheRomanesqueandGothicpalettes,insomein-stances dating back to perhaps 1250, could be conclusively identified fromhundreds ofmicro-samples taken from walls.

The wall linings of the Tournai Cathedral have originally been covered with a render layer of limestone(CaCO3),gypsum(CaSO4·2H2O)andquartzsand(SiO2);thecorrespondingspectraarepresentedinFig.17.31Therenderhasbeenappliedtothemasonryinordertofillinthedepressionsandthesurfaceporosityofthewallstones.Itwasmeanttoplaytwodifferentroles:protectingthenaturalstonefromdegradationasasacrificiallayer,andpreparingwallsurfaces before the mural paintings.

AccordingtothestudyoftheRamanspectraofthewallsinthecathedral,thetypicalMid-dleAgepalettepigmentsare shown inFig. 18andFig. 19.31Oxidesof lead, tinand ironwereclearly identifiedbymeansof their characteristic spectra.Pigmentsused toobtainyellow, orange and red colors comprised massicot (yelloworthorhombic PbO), lead tin yel-low(Pb2SnO4),yellowtoredochres(Fe(OH)3–Fe2O3),haematite(redFe2O3),minium(orangeredPb3O4)andcinnabar(deeplyredHgS),suitablymixedtoobtainthewantedhue,colorsandbrightness.Themost commonpigments seem tohavebeen the redcolored ironoxides.The yellow pigments might have an origin from old glass making techniques, according to newevidenceontheexistenceof‘non-standardized’yellowPb–Sn–Sbtripleoxidepigments,verycloseincompositionandstructuretothePb–SnandPb–Sbtypeoxides.32

Apparently,wellcrystallizedα-Fe2O3, haematite, has been used for wall paintings since an-cienttimes.Asanexampleofthis,onemaycomparetheMedievalsamplespectrum,Fig.19, with e.g. results from prehistoric cave wall-paintings.33 The Raman spectrum of the red haematite pigments always present a typical narrow peak at ca. 294 cm-1(sometimesasanasymmetric doublet at 291–299cm-1)combinedwithweaker-intensitybandsate.g.226,411and 611 cm-1,seee.g.Fig.20.Inparticular,the411cm-1bandissufficientlyhighinwavenum-bertoconfirmthepresenceofwell-crystallizedhaematite, since this band is known to occur at400–407cm-1instructuralstatesintermediatebetweengoethite,FeOOH,andhaematite,Fe2O3.Alsothetypicalverywidebandofhaematiteatca.1320cm-1 is evident in the prehis-toric sample.33

31 L.Lepot,S.DenoëlandB.Gilbert,ThetechniqueofthemuralpaintingsoftheTournaiCathedral,J.RamanSpectrosc.2006,37,1098–1103.

32 C.Sandalinas,S.Ruiz-Moreno,A.López-GilandJ.Miralles,ExperimentalconfirmationbyRamanspectroscopyofaPb–Sn–Sbtripleoxideyellowpigmentinsixteenth-centuryItalianpottery,J.RamanSpectrosc.2006;37:1146–1153.

33 F.Ospitali,D.C.SmithandM.Lorblanchet,PreliminaryinvestigationsbyRamanmicroscopyofprehistoricpigmentsinthewall-paintedcaveatRoucadour,Quercy,France,J.RamanSpectrosc.2006;37:1063–1071.

Page 62: Tracking Colour

62Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Identifiedblue and greenpigmentsspectra fromthewallsof theTournaiNotre-Dameca-thedralarereproducedinFig.21.PulverizedLapis lazuli and azurite (2CuCO3·Cu(OH)2)pig-ments were apparently much used in the blue zones from the 12th century. Lapis lazuli, (Na8[Al6Si6O24]Sn, also called lazurite or ultramarine blue)wasintroducedinEuropefromAf-ghanistan. Itwas extensively used in theMiddleAges andduring theRenaissance.Thepigmentwasextremelyexpensiveandunaffordableformanyartists.BecauseofthepriceinoldtimeMiddleAgeEurope,largebluesurfacesofmuralandotherpaintingscontainla-zuriteonlyinexceptionalcases.IntheTournaicathedraltheratherextensiveuseoflazurite is an indication of the economical power of the church.31

Green areas were often found to contain atacamite and posnjakite. Atacamite is a pigment mentioned in old treatises such as the De Diversis Artibus. It consistsofagreenmineral(CuCl2·Cu(OH)2).Posjnakite isacoppersulfate (CuSO4·3Cu(OH)2·H2O),although itmaynotoriginally have been present and could be formed during the time spent on the walls.

With respect to white and blackcolor,respectively,leadwhite(2PbCO3·Pb(OH)2)andcarbon black (sootetc.)wereidentifiedincertainareasintheTournaiCathedral.31Somepaintlayerswere found to show signs of egg yolk as the binding media on top of thin layers made of gyp-sumandchalkcoveringthestone,seeFig.22.Theapplicationofeggyolkcanbeconsideredas a common way to apply pigments to a surface layer of certain mural paintings, or could also have been used as a way to grind the pigments before use of another binding medium. Theeggyolkdurabilitywasfoundremarkableincombinationwithmostpigments,exceptmaybethegreencopperones.Furthermore,signsofbeeswaxwereidentified.Beeswaxmaybe regarded as a protecting agent, applied in the past or by a modern conservation campaign toconferwaterrepellencetotheexternalsurface–toprotectfromanywaterinfiltrationmany times responsible of gypsum solubilization–crystallization processes that give rise to wallcracking.Sincehumidityisoneofthemaincausesforthedecayofmuralpaintings,itwasnotsurprisingfortheexpertpeopletofindsuchbeeswaxorganictraces.31 Clear spectra ofeggyolkandbeeswaxareshowninFig.22.

Finally,thepresenceofoxalateswasseen(Fig.22),relatedtoagingduetodegradationofchalk and organic substances applied during past restorations. Calciumoxalatecan be attrib-uted to metabolic processes by surface micro-organisms such as aggressive lichens. They produce oxalic acidwhich reactswith theCaCO3, chalk, to forma thin calciumoxalatemembrane.InthePalazzoFarnese,theFrescoespaintedin1560byZuccarihavebeenfoundtobeverysignificantlydamagedbytheinvasionofaggressivelichencolonies.Theseorgan-ismsmayproduceupto50%oftheirownbiomassashydratedcalciumoxalate, especially near green copper based pigments and lead white, thereby destroying the platform on which the artwork is based.34

Pigments resting on painted wood

Therearemanyexamplesofstudiesonpolychromepigmentsonwoodenhistoricalitems.AsarecentexampleweshallonlymentionthecaseofaRamanspectroscopicstudyonacrucifix(fromabout1400AD)nowintheNationalMuseuminGdansk,Poland.35 Typically, characteristic spectral bands found were associated with different pigments on painted

34 H.G.M.Edwards,ProbinghistorywithRamanspectroscopy,Analyst2004,129,870–879.

35 A.Kaminska,M.Sawczak,M.Oujja,C.Domingo,M.CastillejoandG.Sliwinski,PigmentidentificationofaXIV/XVc.woodencrucifixbymeansoftheRamanspectroscopictechnique,J.RamanSpectrosc.2006;37:1125–1130.

Page 63: Tracking Colour

63Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.20:Ramanspectrumofadarkredcrystalinaredpigment from a prehistoric wall-painted sample. Well-crystallised haematite, α-Fe2O3, is clearly recognized. The Ramanspectrumobtaineddirectlyinthecave,excitedbyuseofadiodelaser,workingat780.0nmwithanintensity of 1.50 mW. Cited after 33.

Fig.21:Ramanspectraofblue and green pigments from thewallsoftheTournaiNotre-DameCathedral,citedfrom 31.

Fig.23:Ramanspectrumofaredpigmentareawitharust-coloredshade,depictingthesidewoundofJesus,onawoodencrucifixfromtheNationalMuseuminGdansk Poland. Cited from 35.

Fig.22:Ramanspectraofdetectedorganiccompounds:eggyolk,beeswaxandoxalatefromthewallsoftheTournaiNotre-DameCathedral,citedfrom31.

Page 64: Tracking Colour

64Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

areasoftheobjectandalsoincrosssections.AsseeninFig.23,amixtureofpigmentsmusthavebeenusedbytheartisttoacquirethedesiredcolorofe.g.thesidewoundofJesus:Bandscorresponding to redpigmentswere identified:vermilion (253,284,343cm-1) and red lead minium (313,390,548cm-1).Inotherplacesnumerousbandswereascribedtoe.g.malachiteand azurite. Bands corresponding to the pigments prussian blue, (Fe[Fe3+Fe2+(CN)6]3) (282,538cm-1),andchromeyellow,(PbCrO4)(338,360,403cm-1),werealsoobserved,butbecausethesepigmentsarerathernewinventions(knownonlysince1704and1803,respectively),partial retouching of some areas of the statue must have taken place.35

Modern pigments

To complete this discussion it was considered of interest also to mention modern samples of art.Alsoherearchitecturalpaintanalysistechniquehavebecomeimportantforestablish-ing the palette of historical paint colors, and reconstructing how a room or a facade might have looked at an earlier time.36Asan illustrativeexamplewemayrefer tomuralpaint-ingsbytheinnovativeItalianminiaturistNapoleoneVerga(1833–1918).Hisarthasrecentlybeen studied by a number of techniques including in situ non-destructive micro-Raman investigation.37Theexperimentalresultsallowedidentificationofnineteenthcenturyblueandgreenpigments,givingdetailedinformationnotonlyonVerga’spaletteandpaintingtechniques but also on undocumented retouches. On the blue and green areas of the wall painting,13differentpigmentswereidentifiedbyRaman-microscopy,viz.ultramarine blue, cobalt blue (CoOAl2O3),prussian blue, emerald green (3Cu(AsO2)2Cu(CH3COO)2),copper phthalocy-anine, chromeyellow, massicot, cinnabar, carbon, leadwhite(2PbCO3·Pb(OH)2),whiteSanGiovanni(CaCO3),gypsum (CaSO4·2H2O)andbarite (BaSO4).Amongthese,onlycopper phthalocyanine, anorganicdyeinventedin1936,isthoughttobenotoriginal.37 The spectra were obtained using a JASCOVentunoRamanspectrophotometercoupledtoamicroscopeandequippedwith a CCDdetector,cooledto-50°C.ThespectrawereexcitedusinggreenradiationfromadoubledNd:YAG laser. The laser power at the sample was always kept below 4 mW, and the exposuretimeswentupto400s.37 Other modern pigments are discussed in38.

AnalternativetoRaman:Theverylowfrequencyinfraredrange

Providedthatenoughquantityofthesampleisavailable,farinfrared(FIR)spectroscopymayrepresent a useful analytical method, especially in the case of inorganic compounds some of which are not active in the mid infrared region and often, because of the fluorescent effect produced by organic media, not detectable by Raman spectroscopy.39Studies fromtheearly1970shavereporteduseofFIRspectroscopyinthefieldofcharacterizationofin-organicpigments,usingapolyethylenesamplingmethod.Analysescanbeandhavebeenmadewithamoderatequantityofsample(0.5–1.5mg).39 The data obtained compare well withRamanresults.AnexampleofsuchresultsisshowninFig.24.

36 E.Kendix,O.F.NielsenandM.C.Christensen,Theuseofmicro-Ramanspectroscopyinarchitecturalpaintanalysis,J.RamanSpectrosc.2004;35:796–799.

37 F.Rosi,C.Miliani,I.Borgia,B.BrunettiandA.Sgamellotti,IdentificationofnineteenthcenturyblueandgreenpigmentsbyinsituX-rayfluorescenceandmicro-Ramanspectroscopy,J.RamanSpectrosc.2004;35:610–615.

38 C.L.Aibéo,S.Goffin,O.Schalm,G.vanderSnickt,N.Laquière,P.EyskensandK.Janssens,Micro-Ramananalysisfortheidentificationofpigmentsfrom19th and 20thcenturypaintings,RamanSpectrosc.2008;39:1091–1098.

39 E.Kendix,G.Moscardi,R.Mazzeo,P.Baraldi,S.Prati,E.Joseph,S.Capelli,FarinfraredandRamanspectroscopyanalysisofinorganicpigmentsJ.RamanSpectrosc.2008;39:1104–1112.

Page 65: Tracking Colour

65Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

UV as a means to avoid fluorescence

Asmentioned,fluorescenceduetothebindingmediaetc.isamainprobleminRamanspec-troscopy analysis of artistic pigments.40Ithasbeenproposedthatonecoulduseultraviolet(UV)lighttoavoidthefluorescence.41Thefluorescenceprobleminpigmentidentificationcanbe brought down and minimized by application of local irradiation with controlled levels of pulsed UVlaserlightontheanalyzedpaintsample(pigment+bindingmedia),followedbyrather conventional Raman analysis with an IR source.40Fromaquantitativepointofview,insomecases(chromiumyellowandultramarineblue)asignaltonoiseratioimprovementof16dBcanbeachievedbyusingpulsedUV irradiation and IR laser Raman analysis instead of direct Raman analysis with a visible laser.40

WehavetriedinanexampletousecontinuousdeepUVlaserlight(229nm,withapowerofafewmW)fromadoubledLexelArgonLaser.Thesampleconsistedofonlytwomicroscopicfragments of a statue, IN2830,fromtheNyCarlsbergGlyptotekinCopenhagen,seeFig.25.The light was analyzed with a Renishaw UV 2000 microscope spectrometer.42 The samples were also analyzed with red light and green visible light in our laboratory, but in these cases the fluorescence was too strong to give usable spectra. Not even the strongest calcite band at 1086 cm-1 could be seen.

AsisobviousfromFig.25,thefluorescencehasgoneawaywhenUVlightisusedforexci-tation.Unfortunatelythereisnotanysignoftheredpigmentthatwasdestroyed.Soitisobvious that great care must be taken, at least when deep UV light is used. Weak light power levels must be used.

40 A.López-Gil,S.Ruiz-MorenoandJ.Miralles,OptimumacquisitionofRamanspectrainpigmentanalysiswithIRlaserdiodeandpulsedUVirradiation,J.RamanSpectrosc.2006;37:966–973.

41 S.A.Asher&C.R.Johnson,Ramanspectroscopyofacoalliquidshowsthatfluorescenceinterferenceisminimizedwithultravioletexcitation,ScienceN.S.vol.225,No.4659,311–313,1984.

42 B.Sharma,R.W.BergandS.A.Asher,unpublishedwork.

Page 66: Tracking Colour

66Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.24:Comparisonof(A)aFarInfraRed(FIR)absorptionspectrum,(B)aMiddleInfraRed(MIR)absorp-tionspectrumand(C)aRamanscattering spectrum of malachite, Cu2CO3(OH)2. The sample was groundanddispersedinamatrixsubstance and pressed into a disc. The FIR spectrometer was a Thermo Nicolet5700withaParker/BalstonFT-IR PurgeGas Generator, prevent-ingwaterandcarbondioxideintheatmosphere from interacting with thepolyethylenematrixofthepelletand avoiding interference bands in the spectrum. The MIR spectrometer wasaPerkinElmer1700equippedwith an MCTdetector.InthiscasethesamplewasrecordedinaKBrmatrixpellet.Citedafter39.

Fig.25:Woman’shead,statueIN2830fromtheNyCarlsbergGlyptotekinCopenhagen.Tothelefttheheadisshown,to the right spectra of the fragments are shown and for reference the deep UV Raman spectrum of clear calcite.42 ComparealsowiththecalcitespectruminFig.17.

Page 67: Tracking Colour

67Raman spectroscopy characterization of colored pigments in archaeological materialsRolf W. Berg

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

CONCLUSIONS

Fromtheabovesummaryofresultsobtainedbymanyresearchersonmanyinstruments,itcan be stated that the use of Raman microscopy has been proven to be successful in many instances.UsingRamanspectroscopydirectlyonspallatedorscalpel-excisedfragmentsofart and wall paintings, it has many times been possible to deduce information about the paint layers and thereby to provide information on the methodology used in the construc-tionoftheartwork.Amajoradvantageofthetechniqueisitssemi-nondestructivenatureand the ease of sampling, which requires little specimen preparation or treatment.

Asoneexample, indigoblue,acolor-fastpigmentknown inantiquity,has recentlybeenidentifiedonpaintedplasterfrombeforetheBronzeAge.26 Raman worked for a range of pig-mentsbutnotforall.Manyblue,greenandpurplepigmentscouldnotbeidentified,possiblybecause the wavelength of the laser was not appropriate for these pigments, often causing an overload or fluorescence in the spectra.

Althoughtherearenowseveraldatabasespublishedwhichassisttheidentificationofartpigments, an integral part of the interpretation of the Raman spectra is the obtaining of related reference spectra. There have in the past been several attempts on the construction ofdatabasesofpigmentmaterial(includingplastersandmortars),andonbindingagentsof organic origin used to increase the adhesion of the pigment to the substratum, but there is still a lack of complete data bases that are easy and cheap to use. This is so because in-formation about inorganic and organic components in specimens is accessible only from commercial data bases or specialist reports.

ItisthehopethatthissummarymayservetoshowthattheuseofRamanmicroscopyforthe nondestructive analysis of archaeological specimens has a future.

ACKNOWLEDGMENTS

The author wishes to thank his collaborators in the area of Raman spectroscopy, archaeology andart;inparticular,heisgratefultoJanStubbeØstergaard,SusanneBrunsgaardHansen,Ane Sælland, Kim Pilkjær Simonsen, Conny Hansen, Ole Faurskov Nielsen, and LykkeRyelund.Withrespect toultravioletRamanspectroscopy,BhavyaSharmaandSanfordA.AsherfromPittsburghUniversity,Pennsylvania,USA,arethanked.TheDanishAgencyforScience,TechnologyandInnovationprovidedfundingforthisproject.

Page 68: Tracking Colour

68

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

The CPNMainProject2008–2011:anoutline

Jan Stubbe Østergaard1

ThefollowingisbasedonatextconceivedattheoutsetoftheMainProjectin2008.Wheresignificantchangesofplanhavesincetakenplace,acommentisadded.

Motivation

InlaunchingtheMainProject,theNyCarlsbergGlyptotek(NCG)wasmotivatedbytherec-ognitionthat:

• itisabasicmuseumtasktoacquireknowledgeofallaspectsoftheworksinitscareandto make this knowledge available to the general public as well as to scholarship. This task hasnotbeenfulfilledasregardstracesofpolychromyinitscollectionofGreekandRomansculpture.

• informationontheGreekandRomansculpturecollectionisaccessibletointernationalschol-arship.Adocumentationand researchproject focusingon tracesof ancientpolychromymaythereforecontributesignificantlytothisdynamicfieldofstudy.

• polychromystudiesarenotonlyinternationalincharacterbutalsonecessarilyinterdisci-plinary.ThemuseumhasestablishedtheCopenhagenPolychromyNetwork(CPN)andhasinternationalcontactswithscholarsandconservationscientistworkinginthefield.

• tomaintainandincreasethepresentmomentumofthestudyofancientsculpturalpoly-chromy, one of the essential prerequisites is an increase in basic data acquired through interdisciplinary research activity and the publication thereof.

The international dimension

Statusinancientsculpturalpolychromystudiesischaracterizedontheonehandbyincreas-ing activity, on the other by a lack of close collaboration and coordination. The aims and meansofthefieldhavenotbeenthesubjectofanyprofessional,internationaldiscussion;2 ageneralwrittenintroductiontothefielddoesnotexist;researchandpublicationarenotsubjecttoanycommonstrategy,nortoanyagreedsetsofstandardssuchaswouldseemnecessary to ensure comparative studies. ItisthereforeanessentialelementoftheMainProjecttokeeprelevantcolleaguesandinstitutionsabroadinformedaboutourwork.Asafirststep,anInternationalAdvisoryPanelwasestablishedtocommentonourprojectprogram.

1 Research curator, Ny Carlsberg Glyptotek.

2 Onlyfewmeetingsdevotedtoancient(andmedieval)sculpturalpolychromyhavesofarbeenheld:‘DieFarbenderAntike–NeueBeobachtungenzurPolychromieundderenWahrnehmung.’GlyptothekMünchen,4–5June,2005(pro-ceedingsnotpublished);‘Lapolicromiasupietranaturaleeceramicadall’AntichitàalMedioevo.’GiornatadiStudio,UniversitàdegliStudidellaTuscia,Viterbo,26.10.2007. (publicationofproceedingsplanned);‘CIRCUMLITIO. Interna-tionalesColloquiumzurPolychromiederantikenundmittelalterlichenSkulptur.’LiebieghausSkulpturensammlung,FrankfurtamMain,10–12December,2008(proceedingstobepublished);‘Lapolicromiadimarmiantichi:L’AraPacisel’etàaugustea’.Giornatadistudio11.3.2009,L’AuditoriumdelMuseodell’AraPacis(proceedingstobepublished);‘InternationalRoundTableonancientsculpturalpolychromy,’NyCarlsbergGlyptotek,Copenhagen,10–11September,2009(proceedingsnottobepublished).

Page 69: Tracking Colour

69TheCPNMainProject2008–2011:anoutlineJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

The critical comments and suggestions received from the panel have been incorporated intheprojectplanning.Thepanelmembersaswellasothersinvolvedinthefieldinterna-tionallywillbekeptfullyinformedandconsultedadhocastheprojectproceeds. Our aim is to contribute to a discussion of such matters of common interests as docu-mentationstandards,researchstrategiesandthedesirability/possibilityofformingprofes-sionalforaforinformationsharing,publicationanddiscussion(suchasawebsiteorotherdigitalforum;electronicpublication/e-journal;internationalmeetings). The issue of conservation documentation in digital form will be faced in the light of on-goinginternationalinitiativesonthesubject.3

General project structure

Theprojectwilldevelopalongthreeparallellinesofinvestigation: 1 Visualexaminationanddocumentationofworks followingapredeterminedstrategy for

their selection and sequence.The protocol includesmacroscopic examination, tungstenlight photography, UV-fluorescence and IR reflectography and microscopy. This is being con-ducted by the NCGprojectconservatorandattachedinterns.

Comment:Thepredeterminedsequenceofsculpturesforexaminationislikelytobere-vised. Rather than following a chronological order, it may be necessary to follow a strategy determined by what is of interest to potential funding parties. The protocol we developed remainsrelevant,butashortened‘Surveyprotocol’hasbeenintroducedtoincreasevolumeofexaminedsculptures.Effectivelythen,visualexaminationwillinitiallyproceedaccordingtotheSurveyprotocol.Thefullprotocolwillbeusedwhendeemedprofitable.4

2 In-depthinstrumentalanalysisofworksidentifiedthroughvisualexaminationashavingparticularpotential.Thisinvolvesconservationscientificandnaturalscientificanalysesasrelevant, by the range of instrumentation available to the CPNexternalnetworkpartners.

3 Ordering,evaluation,discussionandpublicationofacquireddata,beingconductedbyNCG, CPNexternalpartnersandinternationalnetwork.

Project strategies for selection of sculptures for examination

The collection in the Glyptotek comprises works in both stone, terracotta and bronze and allareinprincipleofinteresttothisproject.Inpractise,priorityisgiventostonesculptures(limestoneandmarble),makingupas theydothemainbodyof thecollection. It isalsotowards sculpture in this material that research in sculptural polychromy has mainly been directed.Byfocusingonstonesculpturetheprojectisthereforemostlikelytomakethebestpossiblecontributiontothefield. The collection consists of more than 1000 stone sculptures, dating from c. 600 BCE to c. 400 CE. The whole range of sculptural types, formats and genres are represented. The follow-ingmajorgroupingsmaybeidentified:ArchaicGreekoriginals(18);ClassicalGreekoriginals(80);HellenisticGreekoriginals(71);Greekportraits(Romancopies55,originals12);Romanportraits(363);Romanidealsculpture(353);Romanfunerarymonuments,altars,votivere-liefsandvaria(61).

3 TheAndrewW.MellonFoundationinitiative,see:http://mac.mellon.org/issues-in-conservation-documentation

4 InthisReport,theprotocolisdemonstratedasusedinthearticlebelowonthesphinxNyCarlsbergGlyptotekIN1203.TheSurveyprotocolwillbepresentedinour2010report.

Page 70: Tracking Colour

70TheCPNMainProject2008–2011:anoutlineJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Of these sculptures, c. 120 have traces of polychromy visible to the naked eye. Closer examinationislikelytorevealtracesonothersculptures. Facedwith this large andheterogeneous bodyofworks, theprojecthas established astrategyfortheselectionofsculpturestobeexaminedbasedonthefollowingcriteria,inorderofimportance:relevancetostateofresearch;potentialforacquisitionofdata;relationtoparticularstrengthsofthecollectionandmuseumresearchtradition;relationtoresearchinterests of CPNexternalpartners Theselectionofworksforthefirstphaseofvisualexamination(Spring2009)wasdeter-minedbytheaddedcriterionofrepresentativityofformats(fromstatuettestolargescalestatuary),sinceamainobjectiveofthisphasewastheassessmentofthetimeexpenditureconnectedwiththevisualexaminationprocedure. Comment:Examinationprogresshasbeenfarslowerthanoptimisticallyplanned.Assess-mentofevenapproximatetimeexpenditureisthereforenotyetpossible.

Project phases

Toprovideaframeworkforobservationofprogress,evaluationandadjustmentsinthelightofexperienceswon,theprojecthasbeendividedintoanumberofphases.Thephasesarealsoatoolforbudgetcontrol,aswellasbeingconceivedas‘milestones’tomarkandreportonprojectprogress intheformof internalmeetingsandpreliminarypublications.Sincethereisnoprecedentforaprojectofthiskind,itishopedthattheproposedphasingwilloffer a point of departure when organizing similar investigations elsewhere. Thephasecontentandobjectivesaresubjecttoaconsiderabledegreeofuncertaintystem-minginthemainfromthreecircumstances.Firstly,itisdifficulttoestimatethetimeandresourcesneededtoreachthestatedobjectives.Existingpublicationsontheinvestigationoftracesofsculpturalpolychromydonotmentionthetimeexpenditureinvolved.Secondly,incontinuationofthis,itisforobviousreasonsimpossibletoknowtheextentandcharacterofthetracesofpolychromywhichtheprojectwillrevealandhowthiswillimpactoneachphase. Thirdly, at the time of writing, funding remains to be found for a number of key ele-mentsoftheproject–especiallyasconcernsmuseumconservationstaff,publications,web-siteanddatabaseprogrammingaswellasplannedscholarlymeetings.Theprojectprogramwill be continually updated in the light of developments in this regard. Inasimilarvein,itmustalsoberememberedthattheexternalpartnerswillbemakingresourcesforinstrumentalanalysisavailabletotheprojectonlytoanextentthatiscompat-ible with the work programme at the various institutions. Comment:ItwasoriginallyplannedthatthefirstphaseofvisualexaminationshouldbecompletedbyJune1,2009.Asitis,thisphasehasnotyetbeencompleted.Progresshasbeenfar slower than anticipated. The most important reason is undoubtedly that conservation staff resources are half of what was foreseen.

Work spaces

To facilitate the logistics involved in moving sculptures, and to improve communication of theproject to themuseumvisitors,aworkspace forvisualexaminationhasbeenes-tablishedinthecornerofoneofthesculpturegalleries(Fig.1–4).Havingnoprecedentstofollow,developingthespacecalledforcreativeimprovisationanda‘learning-by-doing’at-titude.

Page 71: Tracking Colour

71TheCPNMainProject2008–2011:anoutlineJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Thespacemeasures5×3metersandissurroundedbya2meterhighmodularglassparti-tionandalockabledoor.Informationmaterialforvisitorsismountedontheoutsideofthepartition. Onehalfoftheinterior(Fig.3)istakenupbyaworkdeskandanequipmentstowagearea.Theotherhalfisdesignedforvisualexamination.Here,sculpturesupto300kgmaybesetup on a turntable mounted on a wheeled platform whose height may be hydraulically regu-lateduptoaheightof90cm.Forfullscalestatues,aheavydutysteelturntableforloadsupto1000kghasbeendevised(Fig.4).Bothturntableshavedegreemarkingforcontrolofphotography angles. The equipment needed for photography, including UV-FL and IR, using a Canon EOS 5D MarkIIdigitalcamera,isavailable.Acommerciallyavailablecollapsiblepavilionhasitsblackoutroofingpermanentlymountedtoneutralizenaturallightfromthegalleryskylights.ForUV-photography, black out side panels are mounted. The four legs of the pavilion have been adaptedtoaheightallowingstatuesupto2.5metresintheworkspace(Fig.1–2). Magnification tools for visual examination include a variety of magnifying glasses, aDinoLiteProvideomicroscopeandaLeicaM651operationsmicroscope. Ideally, theworkspaceshouldalsobeequippedwithahighendvideomicroscope(KeyenceVHX-600 or simi-lar),butthisisbeyondthepresentfinancialreachoftheproject. Theconservatorsworkingontheprojecthaveeasyaccesstoan(albeitimprovised)officeadjacenttothegallerywhereaLeicaDM2500darkfieldmicroscopeisavailableforcrosssectionmicroscopy.Theofficealsoprovideswhatisnecessaryforprocessingandstoringdigital images.

Visual examination protocol

ThemodelfortheprotocolistheSchoolofConservationvisualexaminationreportformforpolychromesculpture.ThereportgivenaboveonthePilotProjectstudyofIN2830andbelowontheMainProjectstudyoftheArchaicGreeksphinxIN1203mustfornowserveasanindicationoftheprotocolformat.ThenewSurveyprotocolwillbepresentedinourPreliminary Report 2.

ForthenextReport: completed and on-going visual examination

The second report will also contain the results of the work that has already been completed, but too late tobe includedhere.ExaminationhasbeencarriedoutofaLateHellenistic/Roman Late Republican portrait, IN1583(Fig.5),oftheLateArchaicGreekheadofakouros(‘TheRayetHead’)IN418(Fig.6)andofaportraitoftheemperorTiberiusIN1750(Fig.7).Atthetimeofpublicationofthispreliminaryreport,thefirstfull-scalesculpturehasbeenmovedtotheworkspace:theLateHellenistic/RomanEarlyImperialgroupofArtemisandIphigeneia, IN 481–482a (Fig. 8). Our Preliminary Report 2will containmuch else beside. Publication is planned for the autumn of 2010.

Page 72: Tracking Colour

72TheCPNMainProject2008–2011:anoutlineJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.1:TheCPNworkspaceforvisualexaminationand documentation in the Ny Carlsberg Glyptotek. Informationforvisitorsismountedontheoutside.Museumphoto.

Fig.2:AsFig.1,butwithincreasedheightto accommodate full-size statuary. Museumphoto.

Fig.3:TheinterioroftheCPN workspace. Museumphoto.

Fig.4:Heavydutysteelturntablewithdegreemarkings.Height50cm.Museumphoto.

Page 73: Tracking Colour

73TheCPNMainProject2008–2011:anoutlineJan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.8:ArtemisandIphigeneia,marble,1st century BCE–1st century CE. Ny Carlsberg Glyptotek IN 481–482a. Museumphoto.

Fig.5:LateHellenistic/LateRepublicanmarbleportrait,c. 100–50 BCE. Ny Carlsberg Glyptotek IN1583. Museumphoto.

Fig.6:HeadofanAtticKourosstatue(‘RayetHead’),marble, c. 525 BCE. Ny Carlsberg Glyptotek IN 418. Museumphoto.

Fig.7:MarbleportraitoftheemperorTiberius, c.20–30CE. Ny Carlsberg Glyptotek IN1750. Museumphoto.

Page 74: Tracking Colour

74

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCG IN1203

Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard1

Abstract

TheexaminationoftracesofpolychromyonanArchaicsphinxintheNyCarlsbergGlyp-totek was undertaken using a protocol developed for the purpose of co-ordinating the docu-mentationandinformationconcerningancientpolychromesculptures.TheArchaicsphinxwasselectedasthefirstofanumberofsculpturesfromthecollectionofGreekandRomansculpture,whichwillbeexaminedforremainsoftheiroforiginalcolours.Thesphinxhascleartracesofredonlargeareas,whichcanbeobservedwiththenakedeye.Asampleofthered was studied by cross section analysis, micro-chemical analysis, scanning electron mi-croscopy(SEM)combinedwithEnergydispersiveX-ray(EDX)andX-rayfluorescenceanaly-sis(XRF).Allthestudiespointinthesamedirection:theredcolourishematite(Fe2O3).

Keywords

Polychromy.Archaic.Sphinx.limestone.Red.Hematite(Fe2O3).NyCarlsbergGlyptotek.

Introduction

TheArchaicsphinx2 isthefirstofanumberofsculpturesfromtheGlyptotek’scollectionofGreekandRomansculpturewhichhasbeenexaminedanddocumented for tracesofcolour within the framework the CPNMainProject.Theexaminationanddocumentationwas undertaken following a protocol developed for the purpose of co-ordinating the documentation and information concerning ancient polychrome sculpture. The protocol providesstandards for themethodswhichcanbeused for theexaminationof tracesofcolour on ancient sculpture. ExaminationofthesphinxstartedinJanuary2009incollaborationwithconservatorLinRosaSpaabækandwascarriedout inaworkspaceup inoneof thegalleriesofancientsculpture in the Glyptotek. While facilitating the logistics of moving sculpture to and from the work space, the location simultaneously gives the public an opportunity to follow the progressoftheconservator’swork.

ArchaicsphinxIN1203

ThefollowingsectiondetailstheexaminationsoftheArchaicsphinxIN1203andtheresultsobtained.

1 MariaLouiseSargent,Projectconservator,NyCarlsbergGlyptotek;LinRosaSpaabæk,Conservator,Copenhagen;Mik-kelScharff,conservator,HeadoftheDepartmentsofPaintingandofMonumentalArt,RoyalDanishAcademyofFineArts,TheSchoolofConservation;JanStubbeØstergaard,researchcurator,Ancientart,NyCarlsbergGlyptotek,Copen-hagen(archaeologicalcomment).

2 Johansen1994,40no.4(withearlierliterature,towhichadd:Floren1987,283;Boardman1991,Fig.225w.text);Kreiken-bom2002,159–160;Brinkmann2003,no.234,Fig.234.1–5.

Page 75: Tracking Colour

75DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

BasicData

IN1203Motif:SphinxDimensions:84×70cmDate:c.570BCE3

Typeofwork:Sculptureintheround,tombmarkerPlaceofmanufacture:Attica,Sparta(?)Material:PoroslimestoneAcquisition:1895Period:Archaic,700–500BCE

Methodology

Examinationofthesphinxwasdividedupintothefollowingcategories:Visualexamination,photographic documentation, ultraviolet fluorescence (UV-FL), raking light examination,videomicroscopy,microscopy,thetakingofsamplesandexaminationofcrosssections,mi-cro-chemicalanalysis,ScanningElectronMicroscopy(SEM)combinedwithEnergyDisper-siveX-ray(EDX),X-RayFluorescenceanalysis(XRF),FT-IRandGasChromatography–MassSpectrometry(GC-MS).

Visual examination

Initiallythesphinxwassystematicallyexaminedwiththenakedeyeandwithabinocularmagnifierwith×3magnification.Theexaminationstookplaceinthelightfromtwotung-sten light sources which were located at a 45 degree angle to the centre of the sculpture.

Photographic documentation and examination

The photographic documentation was made with a digital camera Canon Eos5DMarkII.Thesculpturewasrotatedaboutitsaxiswithanintervalof45degrees,andthusphotographedwitheightexposures.Thetopandbottomofthesculpturewerelikewisephotographed(Fig.1–8)Inaddition,detailedphotosweretakenofvariousselectedareasofthesculpture.Inorder to subsequently make possible orientation in relation to the taking of samples and microscope images an alphanumerical grid was imposed on each of the detailed pictures with given coordinates.

Ultravioletfluorescence(UV-FL)

Inadditiontobeingphotographedinthetungstenlight,thesphinxwasalsoexaminedandphotographed under ultraviolet radiation, by means of which pigments, dyes and binding mediainsomecasesfluoresce:inthiswaytheyareemphasised,appearmoreclearlyandexhibitgreatercontrast(Fig.9&10).4 The precondition for UV-FL photography is the use of long wave UV-radiationandafilterwhichpreventstheUV-radiation from being reproduced in the photograph. UV-fluorescence which, on the other hand is transmitted in the range ofvisible light,passesunhinderedthroughthefilterandcanthereforebereproduced in

3 Johansen1994,loc.cit.;Richter1961,10–11,no.3datesthesphinxtoc.590BCE;Floren1987,283‘580–570’;Boardman1991,Fig.225‘c.580’,Kreikenbom2002,159–160‘580–570’.

4 Bjerre1980,22.

Page 76: Tracking Colour

76DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.1:Sphinx.ArchaicGreekoriginalc.570BCE, Poros limestone NyCarlsbergGlyptotek(NCG)IN1203,tungstenlight.

Fig.2–8:Photographicdocumen-tation with 45 degree rotation of SphinxNCG IN1203,tungstenlight.

Page 77: Tracking Colour

77DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.9:NCG IN1203, UV fluorescence photo, right side.

Fig.10:NCG IN1203, UV fluorescence photo, left side.

Fig.11:NCG IN 1203,detailphoto of chest, location of sample 1.

Fig.12:NCG IN1203,detailphoto of right shoulder, location of sample 6 and sample 2–2b.

Fig.13:NCG IN 1203,detailphoto of back of wings, locationofsample3–3band sample 4 and 4b.

Fig.14:NCG IN 1203,detailphoto of left wing, loca-tion of sample 9.

Page 78: Tracking Colour

78DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.15:NCG IN 1203Microscopeimageinsitu.26×Red above left eye.

Fig.16:NCG IN 1203,Sample5Microscopeimageinsitu.26×Modernbluefromleftshoulder.

Fig.17:NCG IN 1203Microscopeimageinsitu.26×Yellowochrefromthebackhair.

Fig.18:NCG IN 1203,Sample3a,crosssection,100×tungstenlight.

Fig.19:NCG IN 1203,Sample3a,crosssection,100×UV light.

Page 79: Tracking Colour

79DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.21:NCG IN 1203 location of XRF-08(yellowarea).

Fig.22:NCG IN1203location of XRF-03,07.

Fig.23:NCG IN 1203 location of XRF 01,02 and 04–06.

Fig.20:NCG IN 1203,Sample3a SEM/EDXbackscatterpicture.200×.

Page 80: Tracking Colour

80DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

thephotograph.ForthispurposeaTiffen2Afilterwasused(cuttingataroundapprox.400mm).

Raking tungsten light

Inaddition,thesphinxwasexaminedandphotographedinrakinglight,whichimpliestheuse of a mono-directional light source, the purpose of which is the revealing and emphasis-ing of the differences in the level in the surface, e.g. tool marks, incised preliminary work andmarkedguidelines.Raking lightexaminationwasparticularlyemployedtoshowupany contour incisions of scale-shaped feathers on the chest and the shoulder. The earrings wereexaminedforanypossiblepattern,suchasrosettes.Inadditiontheflatbackhairwasexaminedwithregardtofindingincisionsforpaintedlocksofhair,whichareknownfromcontemporarysphinxes(Sphinx,AthensAcropolis632). BothUV-FL and the raking light exposuresdemanda total blackoutof the immediatesurroundings. Therefore the public will also notice a blackout tent being raised in the work spacewhenthisformofexaminationanddocumentationistobecarriedout.

Microscopic examinations

Formicroscopicexaminationsthreedifferenttypesofmicroscopewereused: • LeicaM651operationsmicroscope,whichmakesitpossibletoexaminethesculptureinsitu

atupto×26magnificationwitha150mmobjective(lens)ataworkingdistanceof15cm. • DinoLiteProvideomicroscope,asmallhand-heldmicroscopewhichcanbeconnectedtoa

computer which makes it possible to record both still and video images. • LeicaDM2500 Mstereomicroscopewhichisusedtoexaminecrosssections.

Theentiresphinxwassystematicallyexaminedwiththeaidoftheoperationsmicroscope,supportedbythevideomicroscope.Inadditionmicroscopicpicturesweretakenofselectedareas by connecting a camera attached to the operations microscope.

The taking of samples

Itwasnecessarytotakesamplesofvariouspigments.Thesamplescanbeusedforthecrosssectionexaminations,micro-chemicalanalysesaswellaselementanalyseswithSEM/DEX. The samples do not need to be larger than a very few pigment grains, i.e. with a surface area as small as 0.25–1 mm2.Thefollowingsamplesweretaken:

1 Redfromthechest(Fig.11) 2 Redfromthedamagedpatchonthechest(Fig.11) 3 Redfromtheareabetweentheflightfeathersonthebackoftheleftwing(Fig.13) 4 Blackfromtheflightfeathersonthebackoftheleftwing(Fig.13) 5 Light blue from the left shoulder 6 Bluefromtherightshoulder(Fig.12) 7 Blackdotsfromtherightwing 8 Blackfromtheleftshoulder 9 Yellow/brownfromtheflightfeathersonthebackoftherightwing(Fig.14) 10 Yellowfromtheleftshoulder

Cross section examination

ForthepreparationofcrosssectionsthecoloursamplesandindividualgrainsofpigmentwereplacedinaSerifex®polyesterresinwithstyreneassolventandhardener(methylethyl

Page 81: Tracking Colour

81DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

ketoneperoxide)fromStruers.Thecrosssectionwassandedwithwet-grindingpaper240,400,800,1000,1200anddry-polishedpaper4000withodour-freeparaffin. Acrosssectionmayrevealsomethingaboutthegrainsizeofthepigment.Thecapacityofacolourlayertocovercanbedefinedfrome.g.whetherthepigmentsofthepaintlayerarefineorcoarsegrained.Atthesametimeonecanalsotellfromacrosssectionwhetherthereis a build-up of layers of pigments. Cross sectionof sample1–4wereprepared. Eachof thevarious cross sectionswasas-signed an inventory number and a sample number as well as being photographed with a camera connected to the Leica DM2500M stereo microscope, in both halogen light and un-der UV-radiation.Finallythephotographsweregivenascale(Fig.18–19).

Micro-chemical analysis

Micro-chemicalanalysisisarapid,straightforwardandeconomicalmethodofidentifyingpigments, in which chemical reactions are used to identify which elements the pigments consistof.Themethodshouldbecomparedwithotherexaminations, since it isneitherquantitative, nor can it provide information on the composition of the individual pigment. In addition themicro-chemical analysismethod is regarded as a destructive analyticalmethod–evenifonaminutescale:thesampleusedisdestroyedintheanalysis.Micro-chemicalanalyseswerecarriedoutonSample1,takenfromthechestofthesphinx.Itwastestedforcinnabar(HgS),redlead(Pb3O4)andhematite(Fe2O3).

SEM/EDX(ScanningElectronMicroscopy/EnergyDispersiveX-Ray)

AllsamplesweretestedwiththeaidofSEM/EDX.Forthis,theapparatus(JeolJMS5310-LV lowvacuum)oftheRoyalDanishAcademyofFineArts,SchoolofConservation,wasusedwith assistance from Jettie van Lanschot (conservator, cand.scient.cons) and JørnBredalJørgensen (geologist, cand.scient.). In order to observe how the various elements weredistributed,amapping(X-raymap)wasmadewhichindicatestheX-ray intensityoftheselectedelementsandtheirmutualdispositioninthesurfaceofthecrosssection(Fig.20).

X-RayFluorescenceAnalysis(XRF)

Severaldifferentareasofthesphinxwereexamined,andresultsinterpreted,byprofessorMinikRosingoftheMuseumofGeology/NaturalHistoryMuseumofDenmarkusingahand-heldX-rayfluorescencedevice(XRF)(Fig.21–23). Inordertodeterminewhichelementswerepresentinthestoneitself,withoutthepig-mentremains,abreaksurfacewasexamined.Inaddition,theredandbluepigmentsfromtherightshoulderwereexamined.Alsoblackfromthebackoftheleftwingwasexaminedinordertoestablishwhetheritwasremainsofapigment.Ontheexaminationoftheredpigment,MinikRosingwrites: ‘ThesecondsculptureinvestigatedforitsredpigmentwasanArchaicGreeklimestoneSphinx fromAtticawith tracesof redcolour.Hereweusedahand-heldXRF (X-rayfluo-rescencespectrometer).5 This instrument has a c. 1 cm2 area of analysis. The analysis of a faintredpigmentationoftheSphinxshowedhigherlevelsofFeintheseareascomparedto background, and there were no traces of Pb or other heavy elements in the fluorescence spectra from the pigmented areas, which lead to the conclusion that the red pigment on

5 Innov-XAlphaSeries8000LZX/R

Page 82: Tracking Colour

82DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

thissculpturewasaFe-oxidebasedpigmentsuchasredochre,whichisabundantlypresentthroughouttheMediterraneanregion.’

Binding medium analysis

InAntiquity,usewasoftenmadeoforganicbindingmediasuchasanimalglue,wax,eggordryingoil,whichisoftennot–oronlytoalimitedextent–preserved. Sample1oftheredpigmentfromthechestwasexaminedforbindingmediumbymeansof FT-IR spectroscopy, the results of which may indicate the presence of organic or inorganic compounds.ThesamplehasalsobeensenttotheDepartmentofChemistryandIndustrialChemistryattheUniversityofPisawhereDrIlariaBonaducewillundertakeaGC-MS analy-sis for binding medium.

Results

ThefollowingresultsanddiscussionarebasedontheexaminationsandanalysesofthesphinxIN 1203asdescribedabove.Samplesremovedfromthesculpturewereverysmall,whichmadesomeoftheanalysesverydifficulttoperform.Acertainmarginoferrororun-certainty is partly due to this fact.

Visual examinations

Thevisualexaminations,includingthemicroscopicexaminationscarriedoutinsitu,indi-catethatthereisredpaintinmanyareasofthesphinx:

1 Onalargeareaofthechest(Fig.11). 2 On the front edge of both wings. 3 Alongtheentireundersideofbothwings. 4 Manyareasalongtheedgebetweentheflightfeathers(Fig.14). 5 Alongtheheadbandaroundthehead. 6 On the right area of the hair. 7 Onfourofthetriangularornamentsdecoratingthe(dress?)borderattheneck,corresponding

toeveryotheroftheseornaments(Fig.12). 8 Alongthebottomedgeoftheborderwhichmarksoffthechestfromtheplumage. 9 Inanindividualtriangularornamentwhichextendsasfarasthelocksofhairontheleft

side(Fig.12). 10 Betweeneachofthelocksofhair(Fig.12). 11 On the right side, under the wing. 12 On the underside of the body between the forelegs.

Otherpigments: 1 Ayellowochrehasbeendetectedonthehairattheback(Fig.17). 2 Ochre has also been found on one of the flight feathers on the back of the right wing. 3 Severalsmalltracesofochreontheleftsideoftheabdomen. 4 Tracesofochreaboveandbelowtherighteye(Fig.15). 5 Minutetracesofblueontherightshoulder. 6 Lightblueontheleftshoulderswhichisbelievedtobemodern(Fig.16). 7 Black,correspondingtoeveryotherwingfeatheronthebackofbothwings,leavingtheques-

tionopenwhetheritispigmentornot(Fig.14).

Page 83: Tracking Colour

83DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

UltravioletFluorescence(UV-FL)

Thesphinxfluorescedbrightorange,inunevenlayersovermostofthesculpture,mostpow-erfully on the wing feathers and under the eyes. To what this fluorescence is due has yet to be determined. The result can be due both to originally applied material, possibly a combi-nationofapigmentordyeandabindingmediumwhich,togetherfluoresceorange,and/oralatersurfacetreatmentofthesculpture.Thesepossibilitiesshouldbefurtherexamined.AreaswithredpaintvisibletothenakedeyeabsorbedtheUV-radiation and appeared dark.

Raking light

Rakinglightexaminationwasparticularlyusedtoexplorewhetherthereactuallywerein-cisionsofshell-shapedcontourfeathersonthechestandshoulder.Theexaminationpro-ducednoresult.Inaddition,theflatbackhairwasexaminedinthehopeoffindingincisedlines indicating the contours of locks of hair, but without result. On the other hand raking light highlighted the grooves outlining the musculature on the right and left hind legs and this was documented photographically.

Cross section examination

Examinationof thecrosssectionwith the redpigment fromSamples1–3 indicatesonlyonelayer,asnotracehasbeenfoundofa‘primer’undertheredpigment.Thepigmentwasexceedinglyfine-grainedwhichindicatescarefulpreparationandahighdegreeofcoveringcapacity. Common to all three samples is the additional characteristic that the red pigment grains are found located betweenwhite, semi-transparent grains from the stone. Smallblack grains, which may be the result of pollution, lie sporadically between the pigment and thewhitegrains.ExaminationwithUV-radiation revealed that the red pigment absorbed more than the surrounding white grains.

Micro-chemical analysis

Micro-chemicalanalysiswascarriedoutonsamplesfromthelayerofredpigmentonthechest.Theanalysisofcinnabar(HgS)andredlead(Pb3O4)wasnegative,whiletheanalysisofhematite was positive, which indicates that the red pigment contains iron.

SEM/EDX(ScanningElectronMicroscopy/EnergyDispersiveX-Ray)

Commontoall thesamples:LargequantitiesofCa (calcium)coveringmostof thesamples,whichmaybeassumedtobepartofthestone.SporadicpresenceofSi(silicate)whichmayeither come from clay particles from the ground or from the abrasive used in the manufac-ture of the sample. Sporadicappearanceofsulphurinallthesamples,andmaypossiblybeacomponentofplaster, which, again, may be a component of the stone. Samples 1–3:ExaminationofSamples1–3oftheredpigmentfailedtoindicatethepres-ence of mercury or lead, as a component of cinnabar and red lead respectively. On the other hand it was apparent that iron was present, which supported the result of the micro-chem-icalanalyses.Ironhasaweakcontrastandthereforeonlyappearsrarelywithanyclarityonan SEManalysis.BylettingtheSEM run for a longer time a clearer picture is obtained. The faint appearance of iron in the sample indicates that there has only been relatively little pig-ment,thatthishasbeenveryfine-grainedandpossiblymixedinanextender,aconclusionthatwouldbesupportedbythecrosssectionexaminations(Fig.20).

Page 84: Tracking Colour

84DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Sample 4:ExaminationofSample4(theblackfromthebackoftheleftwing)producednoclearresult. InAntiquitycarbonwasoftenusedfor thepreparationofblackpigment.CarbonisaverylightelementanditspresenceisthereforedifficulttoestablishwithaSEM/EDXanalysis.However,noelementswerefoundwhichwerecontainedintheotherblackpigmentsusedinAntiquity,suchasphosphorus(boneblack?)ormanganese(manganeseblack?).Ontheotherhand,ahighcontentoftitanium(Ti)indicatesthattheblackmightbepollution. Sample 5:TheresultsoftheexaminationofSample5showedthatthebluepigmentfromtheleftshoulderwasmodern,basedonthehighcontentoftitanium(Ti).Theresultwasreinforcedbymicroscopicexaminationinsitu,whichshowsauniformorientationofthepigment. The colour was added to the sculpture by accident and is, in all probability, the samepaintwhichformerlycoveredthewallsinRoom6wherethesphinxhasbeenexhib-ited since the early 1950s. Sample 6:Bluefromtherightshoulder.Hadafairlysmallcontentofcopper(Cu),thoughthiswasnotpersuasive.Thesampleshouldbemorecloselyexamined toachievea lessambiguous result. Sample 7 & 8:Sample7hadacharacterofanumberofuniformroundballslocalisedwithinalimitedarea.Theresultsoftheexaminationsuggestacontaminationfromcinders.TheresultsfromSample8showahighcontentofsulphurandindicatepollutionembeddedinplaster. Sample 9:Containediron,whichindicatesayellowochre. Sample 10:TheresultsofSample10(yellowfromtheleftshoulder)arestillawaited.

X-RayFluorescenceAnalysis(XRF)

Examinationof thedamagedarea revealeda largequantityof sulphurandcalcium,butalsoasmallamountofiron(3000ppm),whichmustberegardedasanaturaloccurrenceinthistypeofstone.Areaswithredpigmentexhibitedatwiceashighironcontent,which,additionally,testifiestothepresenceofochre.Itwasnotpossibletodefinethebluefromtherightshoulder.Theexplanationmaybethelimitedpresenceofpigment.Theblackformtheback of the left wing did not answer the question whether the black was pigment or not.

Binding medium analysis

The results of the FT-IR did not show the presence of any binding medium containing pro-tein(egg).However,itdidproduceapictureofanotherorganiccompoundwhichmightturnouttobeawaxoranoil,probablyofrecentcharacter,whichwasjudgedbythetopsaround2923and2854cm-1.Similarresultswereobservedfromtheexaminationofthewoman’shead IN2830intheNyCarlsbergGlyptotek(seethePilotProjectreportonthissculpture).The results of an additional binding medium analysis with GC-MS are still awaited.

Conclusion

ExaminationanddocumentationofthesphinxIN1203werecarriedoutusingaprotocoldeveloped for the purpose of coordinating documentation and information about antique polychromesculpture.Theprotocolmadeiteasytobegintheexaminationofthesphinx.Hopefullytheresultofthesphinxaswellassimilarexaminationswillcreateabetterknowl-edge basis concerning the original state and visual appearance of the sculptures as well as abetterunderstandingofthedeterioration.Someuncertaintiesmayberelatedtothefact

Page 85: Tracking Colour

85DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

thatmostsculpturesinoldercollectionsarewithoutanyinformationaboutexcavationandtreatment until acquisition by the museum. The result of the UV-FLexaminationshowedabrightorangefluorescence,dividedintoanevenlayerovermostofthesculpture:itwasatitsstrongestonthewingfeathersandunderthe eyes. The reason for the fluorescence has yet to be clearly established but can be due bothtooriginallyappliedmaterialand/oralatersurfacetreatmentofthesculpture. Rakinglightexaminationfailedtoindicateanytoolmarks,neitherontheneck-hair,theshoulders or the chest. Clear,visibletracesofredpigmentwerefoundonlargeareasofthesphinx.Theredpaintwas analysed with the aid of a microscope, cross section analysis, micro-chemical analysis, SEM/EDX and XRF.Theresultsofalltheanalysispointedinthesamedirection:theredpaintisaniron-containingcompoundandthusanochrecolour.Examinationalsoshowedtherewasaveryfine-grainedpigmentwithahighcoveringcapacityandasingle-layerstructure. Thegeneralpigmentcompositionhasbeenaffirmedofmostofthesampleswhilethebindingmediaisyetnotconfirmed.

Our thanks to

JettievanLanschot,Conservator,cand.scient.cons.,TheDanishSchoolofConservationJørnBredal-Jørgensen,Geologist,cand.scient.,TheDanishSchoolofConservationMogensS.Koch,PhotographerandConservator,TheDanishSchoolofConservationProf.MinikRosing,dr.phil,theMuseumofGeology/NaturalHistoryMuseumofDenmarkDr.IlariaBonaduce,DepartmentofChemistryandIndustrialChemistry,UniversityofPisa.

Archaeological comment6

AmongthestonesculpturesoftheGreekArchaicperiodthathavebychancecomedowntous,sphinxesformasmallbutdistinctandthoroughlyinterestinggroup.Itisthemoresurprising that to my knowledge no in-depth study of them has so far been published – nor ofarchaicrepresentationsofsphinxesassuch,forthatmatter. Stonesphinxesweresetupasvotivesinsanctuaries7 and as gravestones8. The latter, the funerarysphinxes,arealmostallfromAtticaandgenerallyhavetheheadturnedatrightangles to the body, towards passers-by. IN1203clearlybelongstothisclass. What is known of the polychromy of this particular group of monuments is found only as disiecta membra. The membra may however be assembled without too much trouble thanks tomonographsdevotedontheonehandtoAtticfunerarymonumentsoftheArchaicperiod,andontheother,toArchaicsculpturalpolychromy,respectivelybyG.M.A.Richterin1961andV.Brinkmannin2003,thelattersupplementingRichterbecauseitisnotrestrictedtofunerary monuments.9 ToprovideanuptodatearchaeologicalcommentonthepolychromyofoursphinxIN1203wouldthereforerequireasapointofdepartureacombineduseofRichterandBrinkmann,with the addition of monuments and literature that are either more recent or have been

6 By Jan Stubbe Østergaard. Abbreviations according to the guidelines of the Deutsches Archäologisches Institut (www.dainst.org)

7 Billot1977

8 Holzmann1991

9 Richter1961;Brinkmann2003.

Page 86: Tracking Colour

86DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

missed by both of them. The whole range of iconographically relevant winged parallels on monumentsoftheArchaicperiodmustofcoursealsobetakenintoaccount:Gorgons,Si-rens, Griffons, Pegasi and the like.10SuchastudycannothoweverbeundertakenwithinthecompassofaPreliminaryReportonourproject.

Inavailableliterature,pigmentsonArchaicsphinxeshavetodatenotbeenidentified,butrathermacroscopicallydescribed.Theonlyexceptionknowntomeisthesampleofredcol-ourtakenfromtheheadofafunerarysphinxfoundinAthensin1994intheareaoftheAch-arnianGateonOdosAiolou.(Fig.24)Itisdatedca.540BCE and is now kept in the magazines ofthe3rdEphoriainAthens.11 The sample was analyzed in the Laboratory of the National ArchaeologicalMuseumbyDr.EleniMagou;thepigmentwasidentifiedashaematite.12 The redcolorantonoursomewhatearliersphinxisofthesamecharacter.

Theproposeddatingofoursphinx13placesinthecompanyofthreeothersphinxesofthefirstquarterofthe6th century BCE. Colours remain on all of them, but no published visual documentationexists.Thesculpturesinquestionare:

1 Amarble sphinx in theMetropolitanMuseum ofArt in NewYork,MMA 24.97.87, datedc. 600 BCE.Tracesofredwasobservedonthefilletandonfeathers,blackisseenbetweenthe wings.14

2 Alimestonesphinxfragment,alsointheMetropolitan,MMA26.13,datedc.600BCE. White wasreportedontheface,theneck,theleftforeleg,thebellyandontheraisedborders;blackonthehairandontheleftwing;redonthebreast,theleftwingcovertandsomefeatherson the right wing.15

3 Finally,thereisthelimestonesphinxfromVari,intheNationalArchaeologicalMuseuminAthensNAM4476,16 dated c. 590 BCE. There are remains of the white stucco which coated the piece and some red colour on the stucco17(Fig.25).

Inthisgroup,2and3haveastylizationoftheflightfeathersintworegistersandauseofraisedborderssimilartoours.WealsofindparallelstotheuseofaredcolourfoundontheCopenhagensphinx:onthefillet,onthechest,onthewingcoverts,andontheflightfeath-ers.Theblackfoundonthewingsof1and2wouldatleastsupporttheidentificationoftheblack on IN1203asbeingancient.Butwefoundnotraceofastuccocoatingason3.

10 Cf. Richter 1961, 50–51 on the feathers of birds and creatures of myth.

11 Inv.no.M4518(head)and4519(wingfragment).Karagiorga-Stathakopoulou1996/1997(2000)2–3.pl.1a–c,4a–c,5a–b

12 Karagiorga-Stathakopulou1996/1997(2000)3no.9.Themethodofanalysisusedisnotdescribed.

13 Cf.note3above.Thedatingofthesesculpturesisdifficult,cf.Kreikenbom2002,159.

14 Richter 1961, 10 no. 1

15 Richter 1961, 10 no. 2

16 TheinventorygivenbyRichterseemstobeincorrect,cf.Kaltsas2003,322.no.676.inv.no.4476‘Youthwithchlamys’.TheVarisphinxisnotincludedinthecatalogue.FollowingthetworeferencesgivenbyKaltsasshowsthatthenumberhegivesisnotamisprint.Ihavenotbeenabletoestablishthecorrectinv.no.forthesphinx.

17 Richter 1961, 11 no. 4

Page 87: Tracking Colour

87DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Fig.29:Reconstructionofanarchaicmarblesphinx,c.550BCE.Athens,AcropolisMuseuminv.no.Acr.632.FromSchuchhardt–Neusser1940,colourplateafterp.76.

Fig.24:Headofanarchaicmar-blesphinx,c.540BCE.Athens,3Ephorate,inv.no.M4518.FromADelt51–52,1996/1997(2000)2–3.pl. 1 a–b.

Fig.25:Fragmentarylimestonesphinx,c.600–575BCE.Athens,Na-tionalArchaeologicalMuseuminv.no.4476?(cf.note16).FromRichter1961,fig.19.

Fig.26:Reconstructioninb/wofanarchaiclimestonesphinxfragment,c. 600 BCE.NewYork,MetropolitanMuseumofArt,inv.no.26.13.FromHall1944,332figs.9and10.

Fig.27:Reconstructionofanar-chaiclimestonesphinxfragment,c.575–550BCE.NewYork,Metropoli-tanMuseumofArt,inv.no.43.11.16.FromHall1944,pl.X.

Fig.28:Reconstructionofanarchaicmarblesphinx,c.540BCE.NewYork,MetropolitanMuseumofArt,inv.no.11.185.FromHall1944,pl.VII.

Page 88: Tracking Colour

88DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

FourpublishedreconstructionsofthepolychromyofarchaicAtticsphinxesareknowntomeandareshownhere(Figs.26–29).18 Closest in style to IN1203istheNewYorklimestonesphinxfragment,Fig.23,butthehairstyleisdifferentinhavinghorizontalratherthanverti-calstylizationofthelocks–afeaturefoundpreservedon3(Fig.25).

IN1203waschosenasourworking-up,trialpieceinvisualexaminationbecauseitisoneofthe oldest sculptures in the collection, because colour was clearly visible and because it is quitecomplexinform,despitethehighdegreeofstylization.Itserveditspurposewell.Asregards its polychromy, the results gained cannot claim to be spectacular. Their importance liesinthefactthattheyarethefirstwonfromanarchaicGreeksculpturetobepublishedin such a completely documented manner. Our hope is that this will encourage future close examinationofsimilarmonumentsalreadyidentified–notleastbyRichterandBrinkmann– as being rich in preserved remains of their original polychromy.

18 NewYork,MMA26.13:Hall1944,332figs.9and10inb/w.336.NewYork,MMA43.11.16:Hall1944,335pl.X;NewYork,MMA11.185:Hall1944,334pl.VII.Athens,AkropolisMuseum632:Schuchhardt–Neusser1940,colourplateafterp.76;Brinkmann2003,no.70,fig.70.7.

Page 89: Tracking Colour

89DocumentationandinvestigationoftracesofcolourontheArchaicSphinxNCGIN1203Maria Louise Sargent, Lin Rosa Spaabek, Mikkel Scharff, Jan Stubbe Østergaard

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

References

Billot 1977:M.Fr.Billot,RecherchessurlesphinxduLouvreCA637, BCH101,1977,385–387,419–420.

Bjerre 1980:H. Bjerre,Analytisk fotografering.Konservatorskolen.DetKongeligeDanskeKunstakademi.København(København1980).

Boardman 1991:J.Bordman,GreekSculpture:TheArchaicPeriod(2.ed.,1991).

Brinkmann 2003: V. Brinkmann, Die Polychromie der archaischen und frühklassischenSkulptur(2003).

Floren 1987:J.Floren,DiegeometrischeundarchaischePlastik(1987).

Hall 1944:L.F.Hall,NotesonthecolorspreservedonthearchaicAtticgravestonesintheMetropolitanMuseum,AJA48,1944,334–336B.

Holzmann 1991:Holzmann,UnesphingearchaïquedeThasos,BCH113(1991)125–165(withextensivereferencestoearlierliteratureinnotes).

Johansen 1994:F.Johansen,Catalogue.NyCarlsbergGlyptotek.GreeceintheArchaicPeriod(Copenhagen1994).

Kaltsas 2003:NikolaosKaltsas,AncientGreekandRomanSculptureintheNationalArca-heologicalMuseum,Athens(2003).

Karagiorga-Stathakopoulou 1996/1997: Th. Karagiorga-Stathakopulou, Athenaïka ergaorimesarchaïkesglyptikes,ADelt51–52,1996/97(2000),A,1–34.

Kreikenbom 2002:D.Kreikenbom,ReifarchaischePlastik,inP.Bol(ed.),DieGeschichtederantikenBildhauerkunstI(2002),133–169.

Richter 1961:G.M.A.Richter,TheArchaicGravestonesofAttica(1961).

Schuchhardt – Neusser 1940:W.-H.Schuchhardt–K.Neusser,BemaltearchaischeSphinx,DieAntike16,1940,1ff.

Page 90: Tracking Colour

90

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

AuthorContact

Emailaddressesofmainauthors

Rolf W. Berg [email protected]

Rebecca Hast [email protected]

Minik T. Rosing [email protected]

Maria Louise Sargent [email protected]

Mikkel Scharff [email protected]

Lin Rosa Spaabæk [email protected]

Jan Stubbe Østergaard [email protected]

Page 91: Tracking Colour

91

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009

Other CPN activities

Inthecourseoftheseprojects,theCPN has presented papers at a number of scholarly meet-ings and has organized a couple of meetings itself.

Papers presented at scholarly meetings

J.S.Østergaard“Primary researchonancientGreekandRomansculpturalpolychromy;aPilotProject inCopenhagen”At: Die Farben derAntike – Neue Beobachtungen zur Polychromie und derenWahrneh-mung“.StaatlicheAntikensammlungenundGlyptothek,München,4–5June,2005(proceed-ingsnotpublished).

M.Scharff–R.Hast–J.S.Østergaard“Remains of Polychromy on Ancient Greek and Roman Sculpture” (poster; published inD.Saunders–J.H.Townsend–S.Woodcock(eds.),TheObjectinContext:CrossingConserva-tionBoundaries.ContributionstotheMunichCongress,329).At:TheObjectinContext:CrossingConservationBoundaries.IICcongress,Munich,28Au-gust–1September,2006.

J.S.Østergaard“TheCopenhagenPolychromyNetwork:AprimaryresearchprojectonancientGreekandRoman sculptural polychromy in the Ny Carlsberg Glyptotek”At:Lapolicromiasupietranaturaleeceramicadall’AntichitàalMedioevo.GiornatadiStu-dio,Universitàdegli StudidellaTuscia,Viterbo, 26October, 2007. (proceedings tobepub-lished);

And,updated,at:“CIRCUMLITIO”.InternationalesColloquiumzurPolychromiederantikenundmittelalterlichenSkulptur.LiebieghausSkulpturensammlung,Frankfurta/M,10–12De-cember,2008(proceedingstobepublished).

M.Scharff–J.S.Østergaard“InvestigatingthePolychromyofGreekandRomanstonesculpture:TheCopenhagenPoly-chromyNetworkMainProject”At:The2.InternationalSymposiumofConservationandResearchontheTerracottaArmyandPolychromeCulturalRelics,Xi’an,P.R.China,23–25March,2009(proceedingstobepub-lished).

J.S.Østergaard“Researchonancient sculptural polychromy in theNyCarlsbergGlyptotek,Copenhagen:introductionandfirstresults”At:LesArtsdelacouleurenGrèceancienne–etailleurs”.ÉcoleFrançaised’Athènes,23–26April2009(proceedingstobepublished).

Page 92: Tracking Colour

92Other CPN activities

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Scholarly meetings

NyCarlsbergGlyptotek,Copenhagen,10–11September,2009(proceedingsnottobepublished).

Speakers

MarkBenfordAbbe(AdvancedCertificateinConservation,MA,Ph.D.Candidate,InstituteofFineArts,NewYorkUniversity;ResearchScholar,MetropolitanMuseumofArt,NewYork)Project:ThepolychromyofRomansculpture;Aphrodisias.“SomeRecentInvestigationsintothePolychromyofRomanMarbleStatuary”

ClarissaBlume(MPhil,PhDstudent,UniversityofBochum/UniversityofHeidelberg)Project:ThePolychromyofHellenisticSculpture.“PortraitsofthePtolemies–Greek-styleportraitsandtheirpolychromefinish”

Dr.VinzenzBrinkmann(Headof thecollectionofancientart,LiebieghausSkulpturensammlung;professor,Ruhr-UniversitätBochum)Project:Ancientsculpturalpolychromy.“LatestResearchonthePolychromyofEarlyGreekMarbleSculpture”(withU.Koch-Brink-mann)

Dr.UlrikeKoch-Brinkmann(Classicalarchaeologistandreconstructionexpert,München)Project:Thetechniquesofancientsculpturalpolychromy;experimentalreconstruction.“LatestResearchonthePolychromyofEarlyGreekMarbleSculpture”(withV.Brinkmann)

Prof.dr.PhilippeJockey(Professeurd’HistoireetCivilisationgrecquesUniversitédeProvence(Aix-Marseille1))Project:ThesurfacetreatmentofHellenisticsculpturefromDelos.“Theoretical and methodological issues raised by the study of the surface treatment of the HellenisticSculpturefromDelos”

Prof. dr. Paolo Liverani(ProfessoreAssociato,DipartimentodiScienzedell’Antichità,FacoltàdiLettereeFilosofia,UniversitàdiFirenze)Project:SculpturalpolychromyoftheRomanimperialage.“ObservationsonthecolourinRomanimperialsculpture:techniqueandmeaning”

AlexanderNagel(MA, PhD candidate,University ofMichigan,AnnArbor, KelseyMuseumofArchaeology;FreerandSacklerGalleries,SmithsonianInstitution,WashingtonDC)Project: The Persepolis Polychromy Project; Polychromy of Achaemenid Sculpture; NearEasternsculpturalpolychromy.“ApproachingthePolychromiesoftheAncientNearEast: ThePersepolisPolychromyProject”

Page 93: Tracking Colour

93Other CPN activities

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Dr.ThorstenOpper(Curator,DepartmentofGreeceandRome,TheBritishMuseum)Project:PolychromyofclassicalsculptureintheBritishMuseum(pilotstudy).“ThePolychromyofAncientSculpture: CurrentworkandfutureplansattheBritishMuseum”

FromtheCopenhagenPolychromyNetwork:

JanStubbeØstergaard(Researchcurator,ancientart,NCG/CPN)“MainProjectSummary”

MariaLouiseSargent(ConservatorNCG/CPN)“The CPN protocol in practise”

MikkelScharff(Conservator,HeadofDepartment,SchoolofConservation)“EstablishingtheCPNprotocolforvisualexamination”

Prof.dr.MinikRosing(Geologist,NaturalHistoryMuseumofDenmark)“Geochemical analyses of ancient pigments”

Page 94: Tracking Colour

94Other CPN activities

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Colour on form, form under colour – the aesthetic dimension in European polychrome sculpture

SeminarattheCarlsbergAcademy,Copenhagen,30October,2009

Speakers and abstracts

• JanStubbeØstergaard(MA,researchcurator,ancientart,NyCarlsbergGlyptotek)“Introductiontothematterathand”

AbstractIwillintroducethesubjectofancientsculpturalpolychromyandoutlinestatusinthisfieldof research. The aesthetic dimensions of ancient polychrome sculpture are as of self evi-dentresearchinterestastheyaredifficulttoexploreonthebasisofthemeagreevidenceprovidedby the traces remainingon theoriginals.Ausefulspin-offof theexperimentalarchaeological effort put into research based reconstructions of the polychromy of ancient sculptures has been the at times vehement criticism of their aesthetic and technical quality. Though based on a misunderstanding of the rationale of the reconstructions, such criticism has served to highlight the necessity of a dialogue with art historians and conservators con-versant with the techniques and aesthetics of the closest analogies to ancient polychrome sculpturethatwehaveavailable,namelythefarbetterpreservedearlymodernEuropeanpolychromesculpture.Fromthepointofviewofthestudyofancientsculpturalpolychromy,such a dialogue would serve as platform for hypotheses about the aesthetic dimension of colour on form in ancient times. Theparadoxisthatforreasonsrootedinthehistoriographyofarthistoryandtherecep-tionofGreekandRomansculpture,theaestheticdimensionofEuropeanpolychromesculp-turehasreceivedonlymodestscholarlyattention.Thisaspectwillbebrieflyexplored.

• Dr.SusieNash(SeniorResearcher,theCourtauldInstituteofArt,London)“TheGreatCrossatChartreusedeChampmol: ClausSluter,sculptor,andJeanMalouel,painter”

AbstractTheconstructionanddecorationoftheChartreusedeChampmolforPhiliptheBoldinthelate 14th century is remarkably well documented. We have detailed payment accounts for materials and labour, including pigments and gold leaf, and with them we can follow the processes of creating and polychroming large scale stone monuments like the Great Cross made for the central cloister of themonastery, andknown today as the ‘Well ofMoses’(1395–1404).Analysis of these documents in combinationwith recent technical analysisallowsusinsightsintohowasculpturalworkshop,headedbyClausSluter,andapaintingworkshop,headedbyJeanMalouel,co-ordinatedworkandcollaborated,ontheconceptionof the colour on this work as well as on its practical undertaking. What is also important to consider,andhardertodefine,iswhythisworkwaspaintedatall,givenitsexposedloca-tion(whichrequiredremedialmeasuresearlyon)andgiventhatdifferentchoicesconcern-ing the surface treatment of imagery in stone were made for other monuments within the complex.Thesedifferencesallowustoexplorehowsculpturalcolourmightholdmeaning.

Page 95: Tracking Colour

95Other CPN activities

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

• Dr.MariaFabriciusHansen(Arthistorian,Assistantprofessor,DepartmentofArtHistory,UniversityofÅrhus)“Colourblind:TheopticsofClassicismasaprobleminarthistoricalaccountsoftheMediaevalPeriodandtheRenaissance”

AbstractA reluctance to recognize the importanceof colour for sculptureandarchitecture isnotlimited to accounts dealing with classical antiquity. The wide spread tendency to overlook theroleofcolour inrelationtoformhasalsoaffectedarthistoricalpresentationsofMe-dievalandRenaissancesculptureandarchitecture.InhisbookFraAngelico(1990),thearthistorianGeorgesDidi-Hubermanwritesvividlyaboutourabilitynottosee;aboutthearthistorianspropensityfornotseeing,ortodisregardthatwhichdoesnotfitintotheschemeofthingsinthewayweexpect. The pronounced colour blindness found among classical archaeologists and historians of arthastodowiththecultofclassicismwhichhasdominatedthosesubjectsfromWinck-elmann onwards.This also goes to explain the lack of interest in the question ofwhatthe effect of a sculpture is when polychrome and when seen in its monochrome material. DrawingonexamplesfromMedievalandRenaissancesculptureandarchitecture,mypaperdiscusses the apparent conflict between classical, monochrome form, and the un-classical, coloured surface. We will be looking at colour as found in the reused Roman architecture of Medievaltimes,aswellinthesculptureandarchitectureintheItalianrenaissanceoftheQuattrocento.InthiswayIhopetothrowlightonsomeoftheconventionalassumptionswhichcolour(orrather’decolorize’)ourviewofthepast.

• JørgenHein(Arthistorian,Curator,DeDanskeKongersKronologiskeSamlinger,RosenborgSlot)“Polychromewaxsculptureofthe17th and 18th century”

AbstractThepointofdepartureforthispaperistheseriesofportraitbustsofmembersoftheDanishroyalhouse,keptintheRoyalCollectionsatRosenborgCastleandtheMuseumofNationalHistory at theCastle of Frederiksborg.These busts date from theperiod 1669–1772.Themajorityaretakenfromdeathorlifemasks,whileafewaremodelled.Mostaredressedinclothes belonging to the person portrayed. The busts were set up either in the private royal apartmentsorexhibitedintheCollectionofTreasuresatRosenborg.From1690sonwardstheywereallonshowintheso-called’HallofHeroes’oftheRoyalCabinetofCuriosities. The busts point both forward and back in time. Looking back in time, they lead to the cult ofthedeadinclassicalantiquity,acultinheritedbytheMedievalPeriod,asinthefiguresineffigieknownfromroyaltombsinFrance,EnglandandSpain,aswellas inthevotivegifts, the portraits of patrons and the representations of saints in the catholic Church. The way forward in time leads from the Cabinets of Curiosities of royals and aristocrats, to the travellingWaxCabinetswhichfromtheendofthe17th century showed the populace in and aroundLondonandParisdepictionsofboththepowerfulandthemisshapen;theprecur-sorsoftheMadameTussaud’sofourowntime. Waxisacheapandeasilyavailablematerialwhichmaybedyedand/orpaintedon.Itisextremelywell-suitedtothedepictionofhumanskin,musculatureandveins.Theeffectmay be heightened with by the addition in natural or synthetic materials of such features asbeardstubble,eyebrows,hairandglasseyes.Artistically,theendresultisapolychromesculptureofgreatrealism,apttofoolthesensesandinvitethoughtsontheMagic–espe-cially in a time before the advent of photography.

Page 96: Tracking Colour

96Other CPN activities

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

Waxishoweveralsoabrittleandeasilyperishablematerial;asaconsequenceonlyfewworkssurvive.Itisthereforedifficulttojudgetheextenttowhichthepolychromyofthepreserved pieces is representative and whether such works influenced polychrome sculp-tureinothermaterials.Toshedsomelightonsuchissues,Ipresentsomeselected17th and 18th century polychrome sculptures in small formats, of stone, alabaster and ivory.

• AnnaSchramVejlby(Arthistorian,Copenhagen)“RedCheeks.Colourandsculpture1750–1860”

AbstractTheperiod1750–1860isinmanyrespectsatransitionalone,bothasregardspolychromesculpture and its reception. InmypaperIdescribetheprocessofdevelopmentinsculpture,fromtheglaringwhite-ness of neoclassicism to discreetly red cheeked godesses and on to a complete polychromy – which by the mid 19th century led some critics to describe as timid those sculptors who did not employ a full range of colours. Sensual,lifeimbuingcolour,changingwiththelightandthetimeofday,wasdefinitelynot appreciated by neoclassical sculptors, architects and philosophers of art. They were bentondepictingEternityandTruth,subjectsincommensuratewitheverchangingcolour,especially in such a reprehensible combination of art forms as that seen in polychrome sculpture.Forartistswhoadmiredtheperfectformsoftheancientmasterpieces,cladinnoble white marble, the dawning knowledge of ancient polychromy was not welcome. ByandlargeWinckelmannignoredtheinstancesofpolychromywhichhecameacross,and seems only to have found polychrome works of a kind not Greek of the best period. Herderpraisedsculptureasbeingpureform,unfetteredbythescintillatingillusionismofcolour.AndevenQuatreméredeQuincy,authorofthefirstworkonancientsculpturalpoly-chromy, seems to have preferred the chryselephantine statues, untainted by applied col-our. Iwillbepresentingworksrepresentativeofthecultofwhitemarble–butmoreespeciallythoseexceptions(suchasworksbyCanovaandGibson)whichatoneandthesametimeconfirmtheruleandpointforwardtowardsanartdevotedtootherendsthantherepre-sentationoftheEternallyTrue.MypointofdeparturewillbethearttheoreticaltraditionwhichfromtheRenaissancepassedjudgementoncolourasbeingsensualandtemporal,asopposedtothetimelessnessoflineandform.Atraditionwhichthrivedintherarefiedairsofneoclassicalartandarttheory.Butwhichexperiencedachangeofcircumstancesinthemid-19th century, a time when polychrome sculpture was hailed by some for the very reason that it placed the works in question close to life as lived, here and now.

• FlemmingFriborg(Arthistorian,Director,NyCarlsbergGlyptotek)“Colourful,repressed–andtempting?ColourinEuropeansculptureofthe19th century”

AbstractMycontributiondealswithEuropeansculptureintheperiodfromc.1800–1910.Throughdiscussion and interpretation ofworks from that period, I attempt to focus on, and de-scribe

• Colourasproblematicinsculpturefromc.1800on • Romanticistattitudestocolourandsensualperception,insculptureandingeneral • Theimpactofarthistoryandthehistoryofideasonthedevelopmentofmodernsculpture

‘incolour’

Page 97: Tracking Colour

97Other CPN activities

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

• Thebeginningsofamodernisticupendingofthemimeticrelationofsculptureto’reality’ • Colourasanexponentof‘TheNew’inthevisualarts,especiallyintheartofGauguinand

theFauvistes/theExpressionist, includingthereceptionofthisapproachbyPicassoandBraque

Bywayofintroduction,Iwilloutlinethemainelementsoftheoryandpractisewithwhichthe art of sculpture had to deal at the opening of the 19th century in regard to the use of colour.Romanticistfascination(combinedwithmoraldread)ofsensualperceptionissetupagainstclassicisticidealsofwhitenessandspiritualaswellasformalpurity;followingonthis,Iofferananalysisoftwoparticularlyrelevantworks(Gibson’s’TintedVenus’(1851)andCordier’s‘JewessofAlgiers’(1862).

’Naturalismvs.Idealism’isdiscussedinthelightofanascentshiftofaestheticalvalues.Ashiftcloselyconnectedwiththeadventofmajor,modernnationalstatesandcolonialpow-ers, involving changed conditions for the world of art and its representational framework. Onthisbackground,Iexaminetherelationsofsculpturetothehumanbeing,inbodyandinmind,asrevealedbypredominantlyFrenchworksofartbetween1870and1900suchasDegas’‘BalletDancer,14years’(1879–81).Novelchoicesofmaterialsandtechniquesaswellastheapplicationofindustrialmeansofshaping(alsointhedecorativearts)servetocomofshapingservetocombinecolourmorestronglywithformintheperiodathand.Thisex-pansioninthehandlingformsandmaterialsby‘classic’artformsisthenfollowedinworksbyBraque,PicassoandDuchamp.

• Dr.MikaelWivel(Arthistorian,Copenhagen)“Modernismandthereturnofpolychromy”

AbstractAttheendofthe19thcentury,DanishsculptorswerestillpinnedundertheweightoftheBertelThorvaldsentradition.Itwasjustasmuchacaseofin-breedingasofapathy.Onlywhenagroupofamateursinthefieldgottogetherinthemid1880sdidaweatherchangeoccur.Their forumwas theworkshopof JohanWallmannatUtterslevnearCopenhagen.Here,experiments inceramicswereunderway.TheinitiativewastakenbythearchitectThorvaldBindesbøll,butthepeopleto lettheirtalents looseandreintroducecolour intoDanishsculpturewerepainterssuchasTheodorPhilipsen,EliseKonstantin-Hansenand,notleast,Niels,JoakimandSuzetteSkovgaard.J.F.Willumsenbelongedaswell,thoughhenevercametoUtterslev.. Polychromesculpture isthusanintegratedpartofthebreakthroughofModernisminDenmark.Itissignificantthatthatpainters,notsculptors,weretheonestorecognizeitsrejuvenatingpotential.Neoclassicismhadrelegatedpolychromesculptureto,atbest,therearranksofthevisualarts,forcommonpeople.Itsrediscoveryandacceptanceisoneofthe earliest instances in the 20thcenturyoftheinspirationunorthodoxartistsmightfidinthe so-called trivial arts. Asaresult,polychromyhasbeenpresentinDanishsculptureeversince,andthatwithsuchforcethatithasneardrivenmarbleandbronzeoffthefield.ModernistssuchasJaisNielsenandAdamFischerwereamong thefirst to takeup thechallenge fromUtterslev,butsincethenindependentsculptorslikeMogensBøggild,HenrikStarcke,JørgenHaugen-Sørensen,BjørnNørgaard,KurtTrampedachandChristianLemmerzhaveeachintheirownwayexploredtheartisticpotentialofpolychromywithastoundingoriginality.

Page 98: Tracking Colour

98Other CPN activities

tracking colour – The polychromy of Greek and Roman sculpture in the Ny Carlsberg GlyptotekPreliminary Report 1, 2009 – the copenhagen polychromy network

• BjørnNørgaard(Visualartist)“Iscolouranoptimizerofform–ordoesitreduceform?”

• MikkelScharff(HeadofDepartment)TheRoyalDanishAcademyofFineArts,TheSchoolofConservation“PolychromesculptureinAntiquityandinEuropeanMiddleAges: Aviewfromconservationscience”

AbstractKnowledge of the structure, the materials and their deterioration in polychrome sculpture isachievedbysystematicanddetailedexaminationanddocumentation.,accompaniedbyrelevantchemicalandphysicalanalyses.Inconnectionwiththisitis–asalwaysinheritagepreservation–ofthefirstimportancetocollaboratecloselywithallrelevantstakeholdersandspecialists,inordertoensuretheintegrationofresultsfromthefieldsofconservationscience,archaeology,chemistryandphysics.Inthesphereofconservationofvisualart,theterm‘technicalarthistory’isusedtodescribethecoordinationofdatafromhumanisticsand the technical sciences. TheSchoolofConservationinCopenhagenis interestedinthefurtherdevelopmentofaspects of conservation scientific research. Ithas therefore joined theCopenhagenPoly-chromy Network with enthusiasm. Our aim is to obtain increased knowledge of pigment remains on ancient sculpture, to understand the structure, the materials and their deterio-ration.Wealsoaimtotestanumberofmethodsofexaminationandanalysisonancientsculpture. Knowledge of the polychromy of Greek and Roman sculpture may increase our understanding of the technological background for medieval use of colours on buildings, interiorsandavarietyofobjects.Itisourhopethatwemayatthesametimecontributenewdatatoafieldofresearchnowinitsformativephase.Finally,itishopedthatincreasedknowledge will help in planning preventive measures designed to preserve the remains of polychromy still found on ancient sculptures.