Thermochimica Acta 456 (2007) 5663

Thermal and Raman-spectroscopic analysf t

H Klafa, Swd-Will

c Ins ity ofuary 2y 200

eraldiese in

Abstract

Maya Blue, a pigment composed of very low concentrations of natural indigo and the clay mineral palygorskite, is one of the most brilliantblue dyes, intensively used for more than 2000 years in Mesoamerica. It is extremely stable against environmental attacks and was applied by theIndians for inside and outside mural paintings, ceramics, textiles and for colouring their famous codices. In the present paper it was studied as apowder (compared with modern synthetic indigo) and as colour on tissues, a Maya clay head, and fragment IV of the famous Codex Huamantla.Investigations using Raman spectroscopy in the visible and near-infrared range showed a high degree of correspondence among all Maya Blue-carrying sammolecule wfig-trees shoalso true for 2007 Else

Keywords: A

1. Introdu

1.1. Codex

At the evon Humband used tharchaeologa number oby Mexicalections. Hthat they ortorian anddocuments

CorresponE-mail ad

0040-6031/$doi:10.1016/jples and a good agreement with synthetic indigo. Additional spectral lines may be explained by a transformation of the planar indigohen binding to the palygorskite lattice. Thermal investigations of the original amatl paper of the codex and of recent paper fromwed a high similarity and thus proved that this tree was chosen for paper making by Mayas, Aztecs and other Indian tribes. This wasthe codex.vier B.V. All rights reserved.

rt; Codices; Differential scanning calorimetry; Mesoamerica; Raman spectroscopy; Thermogravimetry

ction

Huamantla, fragment IV

nd of his American research expedition Alexanderoldt lived nearly a year (1803/1804) in Mexico Cityis opportunity to visit ruins in the area and to inspectical treasures of the past cultures. He looked at quitef the old pictured manuscripts, which lay unnoticed

n scholars in governmental archives and private col-umboldt bought some of these documents assumingiginated from the Museo Indiano of the Italian his-archaeologist Boturini who collected Mesoamericanin the middle of the 18th Century [1,2]. He ded-

ding author. Tel.: +49 30 833 54367; fax: +49 30 814 99144.dress: biophys@zedat.fu-berlin.de (I. Lamprecht).

icated a collection of 13 fragments of historical hieroglyphicwritings of the Aztecs to the Royal Library to Berlin, nowadaysthe State Library to Berlin (Staatsbibliothek zu Berlin, StiftungPreuischer Kulturbesitz; SBB). Among them are two frag-ments (designated as III and IV) which were part of a muchlarger map of about 8 m2, the Codex Huamantla which waswritten in the end of the 16th or in the 17th Century and foundin the State of Tlaxcala, about 100 km southeast of Mexico City.One has to be careful not to confuse them with fragments IIIand IV kept in the Museo Nacional de Antropologia, InstitutoNacional de Antropologia and Historia, Mexico City (MNA)that preserves six other fragments of this codex.

The two Berlin fragments III and IV measuring102.5 cm 71.5 cm (H B) and 43.5 cm 64.0 cm, respec-tively, were described, explained and if necessary translatedby Seler [3]. Fragment IV shows a peaceful situation withblue water flowing to the right and four persons sitting onblue-painted chairs in front of their thatched-roof houses

see front matter 2007 Elsevier B.V. All rights reserved..tca.2007.02.002artefacts, especially fragment IV oans G. Wiedemann a, Klaus-Werner Brzezinka b,

a Trankebachstrae 17, CH-8712 Stab BAM, Federal Institute for Materials Research and Testing, Richar

titute for Biology, Department of Biology, Chemistry, Pharmacy, Free UniversReceived 18 October 2006; received in revised form 22 Jan

Available online 13 Februar

Paper dedicated to Dr. Antonius Jammers, former GenPreuischer Kulturbesitz zu Berlin who made this of Maya Blue carryinghe Codex Huamantlaus Witke b, Ingolf Lamprecht c,itzerlandstatter-Str. 11, D-12489 Berlin, GermanyBerlin, Grunewaldstrae 34, D-12165 Berlin, Germany007; accepted 4 February 2007

7

rektor der Staatsbibliothek,vestigations possible.

H.G. Wiedemann et al. / Thermochimica Acta 456 (2007) 5663 57

Fig. 1. (a) Frathe blue chairTotal view ofBerlin, Stiftunarticle.)(Fig. 1a).and blue-pup to the hor symbolsettlementlarger partdescriptionwith this frgment IV of the Codex Huamantla. The bad state of the fragment is clearly seen; alof the second person from left, from the upright standing club of the soldier, andfragment IV of the Codex Huamantla on the freely movable Perspex table in frontg Preuischer Kulturbesitz, Berlin. (For interpretation of the references to colour in

A soldier above the water stream carries a shieldainted club. A line of footprints crosses the sceneouses. It may indicate an often used common trail

ize the migration of the Aztecs that led to theirin the valley of Mexico in the 12th Century [3]. Aof this paper is missing, underlining the often usedof muy maltrado (very maltreated) in connection

agment.

It was atigate this ileave its cltransportedto be subjetions were oof the prectransport taso that it is only a part of a larger piece. Spectra were taken fromthe right end of the blue river (indicated by the red arrows). (b)of the Olympus microscope. By courtesy of Staatsbibliothek zuthis figure legend, the reader is referred to the web version of the

great chance for the authors to be allowed to inves-nvaluable piece of human cultural heritage. It had toimatized treasure for quite a number of hours, to be

straight across Berlin to the research institute andcted to the spectroscopic investigations. All precau-bserved to avoid too intensive irradiation and heatingious codex. Fig. 1b shows the specially developedble to move the paper without touching it.

58 H.G. Wiedemann et al. / Thermochimica Acta 456 (2007) 5663

1.2. Maya clay sculpture

To comoriginal Mahuman arteticity andof the Ethzu Berlin,Museum)ture frombearded mIVCa 2116of the headBlue.

1.3. Blue c

Five bluyears weresame Ethnoband (Ocuc0 and 200from Chiri51292); (ii1889 (IVCfrom Santo45153). In(Wrangler,in the resea

1.4. Pre-C

Pre-Colpaper, mefrom the inFicus padifwith stoneby natural giron-shaped[7].

This papthe Egyptia[7]. The Mby the Otomname ama

of their impadifolia frfrom the codifferences

Thermasamples froat the MNcoupled thetigations,visible effand by enpaper.

1.5. Maya Blue colour

romcolo

romi[8])

merie-Yawithon tblu

e cleneraLazuiantund icatalanticosal protensas ca

l as.

an

as

atta6]. Has a

n 75fina

ternahesen itd int

alyg

ygorinere emte (wrowfelt

givinrnanto t

elpsin gmayttachte tht fai

C ancopind wpare Maya Blue from different origins and periods,ya Blue from the MNA and several Latin-American

facts were included in the investigation. Their authen-age were guaranteed by the donors. By courtesynological Museum in Berlin (Staatliche MuseenStiftung Preuischer Kulturbesitz, Ethnologisches

we got a 10 cm fragment of a Maya clay sculp-the classical period (300900 a.d.): the head of aan, partly decorated with Maya Blue (Catalog No.6) (see Fig. 3). Spots of Maya Blue in several partsand its decoration served as control for a true Maya

oloured threads

e-coloured threads from tissues used in the last 2000investigated. Four of them were friendly gifts of thelogical Museum as with the Maya head: (i) a head-aje) made from camelid wool in the period betweena.d. (No. VA 44827); (ii) a cotton cloth, originatingbote/Peru and the time from 1200 to 1400 a.d. (VAi) a cotton cloth from Barrio de San Paulo/Mexico,a 7980); (iv) a cotton childrens huipil (girls shirt)

Thomas Chicicastenango/Guatemala, 1964 (IVCaaddition, a cotton thread from a modern blue-jeans1999) coloured with synthetical indigo was includedrch.

olumbian paper

umbian Mesoamerican people knew two kinds oftl made from agave (Agave americana) and amatlner bark of different fig-trees, in the majority of casesolia [47]. It could be shown that the bark was beatenimplements to produce pulp, which was held togetherums. The raw paper was set and polished by hot, flatplanches, giving the product a non-blotting surface

er, called huun by the Mayas, was far superior ton papyrus products concerning texture and durabilityaya technique survived the centuries and is still used

i Indians of Mexico who sell such papers under thetes to tourists. Both Mayas and Aztecs used the treesmediate surrounding, the Mayas the fast-growing F.om the warm lowlands (terra caliente), the Aztecsld high plateau (terra fria) and the hill sides. Thus,in Maya and Aztec papers can be readily explained.

l investigations were carried out with milligram-m fragment VI of the Codex Huamantla preservedA, Mexico City. These were supplemented byrmogravimetry (TG)/mass spectrometry (MS) inves-

scanning electron microscopy (SEM) studies forects of bark beating and fungal contaminations,ergy dispersive X-ray analysis for metals in the

A pBlue, amost ppulgiteMesoaby Joscolourocean,

alwaysbecamper miLapisit brillwas foern Yulocal psuffrutnaturadye. Innent was wel[915]

InBlue windigoBlue [1ment wbetweein thethe exnels. TLater oinserte

1.6. P

Palclay mand thgorskilight bfibril,[18],ter. Febetterand hmentsgroupthus aseparaponen[9].

DShygrosof bouinent role in decorating the codex is played by Mayarant composed of indigo and some clay components,nent among these being palygorskite (synonym: atta-. It was a widely used blue colour in Pre-Columbianca for about 2000 years and introduced in a papercaman et al. [9] with the sentence Magnificent blue

a turquoise tone similar to that of the Caribbeanhe shores of the Maya homeland. Maya Blue is note but may change to turquois or greenish hues. Itar in the 19th Century that Maya Blue was not a cop-l and had no relations to natural ultramarine, groundli or Lazurite. Its unprecedented stability which keptover 12 centuries is astonishing. While palygorskiten the soils of the Mayan kingdom, specially in North-n (Sakluum) [10,11], indigo originated from theXinquilit (Indigofera suffructicosa, sometimes also) [12]. Due to a lack of heavy metals and the use ofducts Maya Blue was an environmentally friendlyive research on Maya Blue and its indigo compo-rried out in the last decades, applying spectroscopicmicroscopic and other physico-chemical methods

early paper, when the composition of Mayastill unknown, van Olphen produced a syntheticpulgite complex with a colour very similar to Mayae could show that the essential point for a stable pig-thermal treatment for several days at temperaturesand 150 C. Because of the very low dye content

l product he assumed that indigo was absorbed atl palygorskite surfaces only and not inside its chan-observations were confirmed for natural Maya Blue.

was shown in an X-ray investigation that indigo alsoo the palygorskite channels [10].

orskite

skite, the inorganic component of Maya Blue, is aal with the chemical formula (Mg, Al)2Si4O104H2Opirical formula Mg1.5Al0.5Si4O104H2O [17]. Paly-ith priority to the name attapulgite) is yellowish,

n to pure white and has a micro crystalline, fine-like structure (with a thickness of 3060 nm)

g its surface a fabric or leather looking charac-dez et al. showed that an orthorhombic model fitshe palygorskite structure than a monoclinic oneto explain the properties of archaeological pig-eneral and also of Maya Blue [19,20]. Its siloxylproduce covalent bonds with organic reagents andthem firmly to the clay [19]. All experiments to

e indigo of Maya Blue from its palygorskite com-led because of an intimate structural relationship

d DTG analyses of palygorskite indicated a loss ofc and zeolithic water up to 200 C followed by a lossater between 250 and 450 C. At even higher temper-

H.G. Wiedemann et al. / Thermochimica Acta 456 (2007) 5663 59

atures hydroxylic units were given of, in total all endothermicprocesses [18].

In a recwere compinside theclature functionalrials, readyby these reson the basals and popigments.

2. Materia

2.1. Therm

DifferenMettler insland). Thedeterminat1 K min1,having a 2were perfoples in alumTG 50 cou(MSC 200;gravimetriccodex papethe edge ofmogravime

2.2. Rama

The Rawith (i) adevice FRAtion by a Nat 1064 nmter (DILORcooled CCpled to a Bin micro-mlong workiion laser (Ipower of 5a laser spot10 and 300of about 1 c

2.3. Colou

SynthetIndigo bluespheres (on a microsonly availa

Head3009liche Mseumthe re

A ave taexic

ca (s

rtefa

ma

illionam

ell inmicrery oryl wcope stage. It could freely move on several ball-bearingsecond polyacryl table connected with the spectrometer.the fragment could be manoeuvred by the microscopecontrol in both directions with an accuracy of a few

eters and without touching it. Spectra were taken frome chair of the second person from left (upper right cornerCodex in Fig. 1b), from the upright standing club of the(lower right corner), and the right end of the blue riverleft corner).reparation of the present paper a Mexican artefact fromssical period, 300900 a.d. was lent by the Berlin Eth-cal Museum (Catalogue No. IV Ca 21166, Acten No.7). It represented the head of a bearded man with several

ecorations on the forehead, the left eyebrow, the right earthe head ornaments (Fig. 2). These spots served as furtherl for authentic Maya Blue. To keep and adjust the head, al support was constructed for a sensitive and reproduciblevement under the ulwd objective of the microscope.r information about true May Blue was collected fromhreads from 4 different periods of the 2000 years of thisent paper [15] the narrow channels of palygorskiteared with the modern nanotubes. The chemical bondsfibre-like structures produce in modern nomen-a hybrid organicinorganic nanocomposite. Suchhybrids are considered to be highly innovative mate-to enter a multitude of different applications. Guidedults a synthetic dye named Mayacrom was produced

is of natural plants and clays, avoiding heavy met-isonous additives as they are found in many usual

l and methods

al analysis

t thermal analytical methods were carried out withtruments (Mettler-Toledo AG, Greifensee, Switzer-thermosystem TA 400 was used for TG and DTG

ions of ancient Aztec paper with a scanning rate ofthe samples being in closed aluminium crucibles

mm hole in the cover and at still air. The DSC scansrmed with a Mettler DSC 25 on 4 mm diameter sam-

inium crucibles at still air and 1 K min1. A Mettlerpled with a Balzers mass spectrometer MS-CubeTM

Balzers AG, Liechtenstein) served for the thermo-/mass spectroscopic investigations of the originalr. A specimen of a few milligram was obtained fromfragment VI kept at the MNA. Only DSC and ther-tric curves are shown in this publication.

n spectroscopy

man spectroscopic investigations were carried outFT-IR-spectrometer IFS 66v along with a Raman

106 (BRUKER, Ettlingen, Germany) with excita-d:YAG-laser (DPY 421, ADLAS, Lubeck, Germany), and (ii) with a DILOR XY Raman spectrome-, Bensheim, Germany) equipped with a nitrogen

D-camera (charged coupled device) as detector cou-H-2 microscope (OLYMPUS, Hamburg, Germany)ode technique (objective 50 and 50 ulwd (ultrang distance)). The 514.5 nm excitation line of an Ar-LA 120-1, Carl Zeiss, Jena, Germany) with a radiantmW provided an irradiance of about 0.5 mWm2 atdiameter of 2m. Integration periods varied betweens and 2250 accumulations at a spectral resolutionm1.

rs

ic indigo was obtained from FLUKA (No. 56980;; indigotin; Buchs, Switzerland) in the form of small4060m). They were used as such or smearedcopic slide in a thin layer. Authentic Maya Blue wasble in traces of a few micrograms originating from

Fig. 2.period (of Staatches Mulegend,

the MNadhesion a MAmericodex.

2.4. A

Theeral mIts sizeside won theperiphpolyacmicroson a s

Thus,stagemicromthe bluof thesoldier(lower

In pthe clanologi1141/9blue dand incontrospecia3D-moFurthecloth tof a bearded man, fragment of a Maya clay figure, classical00 a.d.) (spots of measuring indicated by letters). By courtesyuseen zu Berlin, Stiftung Preuischer Kulturbesitz, Ethnologis-

, Berlin. (For interpretation of the references to colour in this figureader is referred to the web version of the article.)

nd was preserved between glass slides or fixed on anpe. Further specimens of true Maya Blue were foundan head sculpture and on cloth threads from Latinee Section 1.3) and of course on fragment IV of the

cts

in and very precious artefact (insurance value sev-Euros) was fragment IV of the Codex Huamantla.

ounts to 43.5 cm 64 cm with a large hole from oneto the centre (Fig. 1). Because of the limited spaceoscope stage (less than 10 cm) only blue spots at thef the fragment could be used. A transport table fromas specially constructed and firmly attached to the

60 H.G. Wiedemann et al. / Thermochimica Acta 456 (2007) 5663

colour in Latin America (see above). All were dyed with bluecolour.

3. Results

3.1. Invest

3.1.1. IndiSynthet

with pale g466 and 16experimentcompared woccurred annot due to itraces of cofor these dewith the res

Two miMNA couland excitatby a high, usive tape topublished bon authentbecause of

3.1.2. MayDue to th

the four chwith FT Ra(1064m)ated two brto bands frbands cann3000 cm1of the samInvestigatioditions broof indigo,intensities.indicate thasettings. Prsurface isundergroun

Excitatirich of bancould be shmaterial, ba good sigand that thThe relativtissues (seeand 1593 cin the signseen in IRthe flat ind

amanAll foolour

rskitt themantural

Bluefou

of Med inn Bligo

m1threachannt tothe d199089) wtheticboth

d agreement with data published in the literature [2224].

Fragment IV of the Codex Huamantlagment IV of the Codex Huamantla is an extremely pre-bject which was only allowed to leave the climatized safesState Library for a few hours, considerably insured anded by two specialists of the library. A short pre-experimentT-Raman spectroscopy showed no results, neither forlue nor for the supporting fig-tree paper. Therefore, it was

d to concentrate on a region between 1000 and 1800 cm1aser excitation at 514.5 nm. The experimental set-up issed above and seen in Fig. 1b.

signal-to-noise ratios (SNR) and a high fluorescenceound with weak Raman bands were typical for theed spectra. Prolonging the duration of exposure by theg laser light to reduce the fluorescence prior to the Ramanrements and shortening the counting time at high numbersand discussion

igations by Raman and IR spectroscopy

go and authentic Maya Blue from the MNAic indigo in form of small globules ( 4060m)reenish and slightly red hues showed bands between90/1700 cm1, corresponding to those determinedally and calculated by Tatsch and Schrader [21]. But,ith their data, some additional bands or band shifts

d some bands were missing. These differences werendirubin, an isomeric form of indigo. It might be thatntamination or varying syntheses were responsibleviations. This indigo spectrum is compared in Fig. 5ults from the threads, the clay head and fragment IV.

crogram specimens of original Maya Blue from thed only be investigated with microsample techniqueion in the visible. The obtained spectra were markednstructured background with some lines of the adhe-which the pigment was fixed to. Similar effects werey Coupry et al. [22]. Unfortunately, the investigation

ic Maya Blue powder could not be followed furtherinsufficient quantities of material.

a clay heade negative results mentioned above, investigations of

osen blue spots of the sculpture head were performedman spectroscopy and excitation in the near-infraredwith intensities between 50 and 500 mW. They gener-oad bands (2002100 and 25003500 cm1) similarom the red sculpture surface without colour. Theseot be connected with indigo. The wide band aroundis a so-called warming band due to overheating

ple. These NIR excited spectra are not shown here.ns of a piece of Blue Jeans tissue under identical con-

ught a spectrum rich with lines equivalent to thosecorresponding well in wave numbers and relativeAdditional lines originate from cotton. These resultst indigo can be detected at the chosen experimentalesumably, the indigo concentration on the sculpturevery low and its signals are obscured by the broadd of palygorskite.

on in the visible range (514.5 nm) presented spectrads and identical for all 4 spots A to D (Fig. 3). Itown that the spectra did not originate from the clay

ut from Maya Blue itself. All lines were very small,n that there was no heating-up by laser excitatione historical artefacts were treated with much care.e line intensities were equal to those from historicalbelow). Moreover, some additional lines (e.g. 1380

m1) were present in all four spectra, but missingals of synthetic indigo. They were similar to linesexcited indigo spectra. This points to the fact thatigo molecule was transformed when bound to the

Fig. 3. R(Fig. 2).the red c

palygopredicthe Rais a na

3.1.3.The

yearsincludmoderfor ind1700 con allSomepigmeter ofJeans/ico/18or synity forin goo

3.1.4.Fra

cious oof theattendwith FMaya Bdecidewith ldiscus

Badbackgrobtainexcitinmeasuspectra from the 4 indicated blue spots of the Maya clay fragmentur spectra are identical, originating from Maya Blue and not fromof the head.

e lattice and that the mutual exclusion rule, whichoccurrence of additional lines either in the IR or inspectra, were no longer valid [14]. But as Maya Blueindigo dye, impurities might also be responsible.

coloured tissue and threadsr blue-coloured camelid-wool threads spanning 2000

esoamerican and South American history werethese Maya Blue investigations, together with a

ue Jeans tissue. The most intensive Raman bandson wool or cotton (597, 1249, 1364, 1574 and) determined by Coupry et al. [22] could be detecteddsas well as on the clay head and the fragment IV.ges might have occurred during the fixation of thethe tissue due to a change in the symmetry charac-ye. The spectra of the 3 most recent threads (Blue; childrens huipil/1964; cotton cloth from Mex-ere very similar, although it is not clear if naturalindigo was used for the 1889-sample. The possibil-dyes exists. The results were compared and found

H.G. Wiedemann et al. / Thermochimica Acta 456 (2007) 5663 61

Fig. 4. RamanHuamantla. Sand shifted ve

of accumulbecame potive spectraa constantthemselvesated Fig. 4are more or

contributiothose of syabove forwere signifispectrum oin the litera

The spehead, excefragment liwell-knowncontrast tocan be curthe indigobinding tostructure apossible withe comparindigo, Fig

3.2. Therm

3.2.1. PapUnfortu

from the Befragment VThe resultsFig. 7 presfrom the MTwo mainposition of

aman spectra with the 20 most prominent lines for synthetic indigo, 5reads, the Mayan clay head and fragment IV of the Codex Huamantla.

in between 350 and 500 C. The characteristics of thepes are very similar. The two prominent peaks occur atd 422 C for the codex and 290 and 410 C for the newrespectively. Additional shoulders are seen at 372 C inper graph and 420 C in the lower one, which shows a

spectra of the three chosen spots in the fragment IV of the Codexpectra were normalized to the same intensity in the strongest bandrtically for better legibility.

ations, the SNR could be increased and evaluationsssible. Moreover, a combination of several consecu-with a small spectral shift kept the Raman bands at

place, while disturbances were shifted and annealed. A combination of these different techniques gener-for the 3 blue spots indicated in Fig. 1a. These spectraless identical and show Maya Blue bands only and nons from palygorskite. Again, some lines additional tonthetic indigo appeared as discussed and explainedblue tissues. Moreover, the relative line intensitiescantly increased at several wave numbers. A mean

f fragment IV is compared in Fig. 5 with those foundture and with synthetic indigo, threads and clay head.ctra of fragment IV correspond to those of the Mayapt for a weak additional line at 1695 cm1. Somenes are broader than those of the head, a phenomenon

from spectra of amorphous substances or solute inthose of well-crystallized samples. Such disturbancesed by thermal treatment. It might therefore be thatin Maya Blue on the clay head obtained a stronger

Fig. 5. Rtissue th

of ligntwo slo289 anpaper,the uppalygorskite by mild firing and thus a more regularnd narrower lines. Such treatment is of course notth paper as a substratum for Maya Blue. To facilitateison of the two main artefacts with pure synthetic. 6 depicts the three spectra separately.

al investigations

er of fragment IVnately, it was not possible to obtain a paper samplerlin fragments III or IV. Therefore, investigations onI paper from the MNA were included in this research.of the thermal analysis are shown in Figs. 7 and 8.

ents DSC curves for the original Huamantla paperNA (top) and for recent fig-tree material (below).

exothermic peaks are prominent: oxidative decom-cellulose at still air ranges from 240 to 350 C, that

further one

Fig. 6. CompMayan clay hat 389 C hidden in the steep ascent to the lignin

arison of Raman spectra of pure synthetic indigo (bottom), theead (centre) and fragment IV of the Codex Huamantla (top).

62 H.G. Wiedemann et al. / Thermochimica Acta 456 (2007) 5663

Fig. 7. DSC curves of a 2-mg piece of paper from the Codex Huamantla (MuseoNacional de Mexico) (top) and of a recent paper made from the inner barkof Ficus padifolia (bottom). Please notice the small endothermic peak around150 C mentioned in the text (see also insert). For both samples: Y-axis inarbitrary units, heating rate 1 K min1, still air, exothermic effects upwards,endothermic downwards. Adapted from [6].

peak. Therknown endlower trackIt is usuallmaterial duof the barkacid in theproduction

The theheated at sindicated inA first losswater, the mand can bethe last oneof lignin. T

Fig. 8. Thermifolia (3.760 mweight loss oair. The derivfrom [6].

4. Conclusion

In recenpresent topworld heritscientific sin ancient tthermal anwere applieclassical peolis in Yucared and yelRaman specrystal strustructure ofby the samRaman spe

Not onlyCity are liscopies fromtioned in th

ion,nisvisi

hrouginaa co

of tch crlinamanof s

threathe

sampatiohintpoine are as yet no explanations for these peaks. A well-othermic effect is clearly indicated at 153 C in theand weakly detectable at 147 C in the upper one.

y seen in papers produced by breaking the raw barkring the mechanical treatment. When the cell wallsare destroyed calcium oxalate forms from oxalic

cell liquid and calcium in the water used for paper.rmogravimetric curves of recent Ficus-tree materialtill air show three distinct steps of mass loss, betterthe first time-derivative of its original curve (Fig. 8).

of about 7% up to 180 C is due to dissipation of freeost prominent step with about 67% occurs at 265 C

ascribed to the oxidative mass loss of cellulose, whileof 26% at 398 C corresponds to an oxidative loss

he residual ash content is very small.

collectBoturiboldtswent tare orishowsa copyout muthe BeThe Rdegree(head,minatepapercertificIV, allresultsogravimetric curve (TG) of paper made from bast of Ficus pad-g). The three steps around 100, 270 and 390 C correspond to a

f 6.6, 67.3 and 26.0%, respectively. Heating rate 1 K min1, stillative slope (DTG) makes the three steps better visible. Adapted

von HumboCodex Hua

Acknowled

This papseveral insgrateful toopportunityof historicaby Mrs. LeNicklisch.

Our spetime Manathe Manuscto let the extreasure aning the speand Dr. Jut years some interesting papers connected with theic appeared showing the common interest in thisage. Berke [25] published an excellent historical andurvey of the Invention of blue and purple pigmentsimes including Maya Blue. Various techniques likealysis, Raman spectroscopy and model calculationsd. Two years earlier Maya wall paintings of the lateriod (900 a.d.) were described, found in an acrop-tan/Mexico. The whole set of Mayan pigments from

low over green and blue to black was investigated byctroscopy [26]. Powder synchrotron diffraction andcture refinements were applied to correct the assumedpalygorskite and Maya Blue [17], recently extendede authors with computation, thermal analysis andctroscopy [27].the original codices preserved in the MNA in Mexico

ted in J.B. Glass catalogue [28], but also numerousdifferent centuries. The Codex Huamantla is men-

e catalogue with 8 fragments left over from Boturinis6 fragments in the MNA, 2 in the SBB in Berlin. Asactivities lay more than half a century before Hum-t to Mexico and as his collection was dispersed andgh several hands the question arose which fragmentsl and which are copies. Figure 39 in Glass cataloguemposition of the (Mexican) fragments III and IV andhe Berlin fragment III. But this copy is done with-are in many details. Already Seler remarked [3] thatparts are written on original amatl fig-tree paper.

spectra of the 3 points on fragment IV show a highimilarity with those of the certified ancient artefactsds) proving that original Maya Blue was used to illu-codex. Although it was not possible to investigate ale of the Berlin fragments and thus to get a further

n about the age and the material of fragments III ands from the literature together with our spectroscopict to the fact that the Berlin fragments bought by A.ldt in Mexico are part of the original former Boturinimantla and no copies.

gments

er would not have been possible without the help oftitutions and a number of friendly persons. We arethe Ethnological Museum in Berlin-Dahlem for theto investigate the rather old Maya clay head, the gift

l threads with authentic Maya Blue and the supportna Bjerregaard, Dr. Maria Gaida and M.A. Andrea

cial thanks go to the Berlin State Library, its at thatging Director Dr. Antonius Jammers and the Head ofript Division Dr. Eef Overgaauw for their readinesstremely precious codex fragment leave its climatizedd be transported to the research institute for perform-ctroscopic investigations. Mrs. Dorothea Barfknechtlia Bispinck helped to prepare the action and Mrs.

H.G. Wiedemann et al. / Thermochimica Acta 456 (2007) 5663 63

Margit Hundertmark and Dr. Renate Schipke accompanied thejourney and handled the codex. To all of them our gratitude isdue.

References

[1] Indianische Handschriften und Berliner Forscher. Handbuch zur Ausstel-lung. Universitatsbibliothek, Freie Universitat Berlin, Berlin (Indianmanuscripts and Berlin scholars. Handbook to an exhibition. UniversityLibrary, Free University of Berlin, Berlin), 1988, p. 70.

[2] A. von Humboldt, Vue des Cordillie`res et monumens (sic) des peuplesindige`nes de lAmerique, Paris, 1810.

[3] E. Seler, Die Mexikanischen Bilderhandschriften Alexander von Hum-boldts in der Koniglichen Bibliothek zu Berlin (The Mexican manuscriptsof Alexander von Humboldt in the Royal Library at Berlin), Berlin, 1895.

[4] R. Schwede, Uber das Papier der Maya Codices (About the paper of theMaya codices), Richard Bertling, Dresden, 1912.

[5] W. von Hagen, The Aztec and Maya Papermakers, Hacker Art Books, NewYork, 1977.

[6] H.G. Wiedemann, A. Boller, Thermal analysis of Codex Huamantla andother Mexican papers, J. Therm. Anal. 46 (1996) 10331045.

[7] H.G. Wiedemann, Paper technology from Egyptian, Chinese, and Mayancultures, in: Proc. of the Workshop on State-of-the-Art of Thermal Analysisheld at NBS, Gaithersburg, MD, May 2122, Nat. Bur. Stand. Spec. Publ.580 (1979) 201217 (Issued May 1980).

[8] Rompp Online (www.roempp.com) version 2.10, Thieme Chemistry, 2006.[9] M. Jose-Yacaman, L. Rendon, J. Arenas, M.C.S. Puche, Maya Blue paint:

an ancient nanostructured material, Science 273 (1996) 223225.[10] L.A. Polette, G. Meitzner, M.J. Yacaman, R.R. Chianelli, Maya Blue: appli-

cation of XAS and HRTEM to materials science in art and archaeology,Microchem. J. 71 (2002) 167174.

[11] L.Polettein Mayahttp://ww

[12] D. ReineBluetgorskite

[13] E. Stoks

[14] K. Witke, K.-W. Brzezinka, I. Lamprecht, Is the indigo molecule perturbedin planarity by matrices? J. Mol. Struct. 661/662 (2003) 235238.

[15] N. Massetti, Mayan BlueNanotech pigment from the 8th Century. MayanBlue pigment: Vegetable or mineral? Its nanostructure provides the longsought answer. http://www.nanovip.com/forums/showthread.php?t=742.

[16] H. van Olphen, Maya Blue: a clay-organic pigment? Science 154 (1966)645646.

[17] G. Chiari, R. Giustetto, G. Ricchiardi, Crystal structure refinements of paly-gorskite and Maya Blue from molecular modelling and powder synchrotrondiffraction, Eur. J. Miner. 15 (2003) 2133.

[18] R.L. Frost, Z. Ding, Controlled rate thermal analysis and differential scan-ning calorimetry of sepiolites and palygorskites, Thermochim. Acta 397(2003) 119128.

[19] M.E. Fernandez, J.A. Ascencio, D. Mendoza-Anaya, V. Rodriguez Lugo,M. Jose-Yacaman, Experimental and theoretical studies of palygorskiteclays, J. Mater. Sci. 34 (1999) 52435255.

[20] G. Strubel, H. Zimmer, Lexikon der Mineralogie, Thieme Stuttgart, 1982.[21] E. Tatsch, B. Schrader, Near-Infrared Fourier transform Raman spec-

troscopy of indigoids, J. Raman Spectr. 26 (1995) 467473.[22] C. Coupry, G. Sagon, P. Gorguet-Ballesteros, Raman spectroscopic inves-

tigation of blue contemporary textiles, J. Raman Spectr. 28 (1997)8589.

[23] H.G.M. Edwards, D.W. Farwell, A.C. Williams, FT-Raman spectrum ofcotton: a polymeric biomolecular analysis, Spectrochim. Acta A 50 (1994)807811.

[24] G.N. Andreev, B. Schrader, H. Schulz, R. Fuchs, S. Popov, N. Handjieva,Non-destructive NIR-FT-Raman analyses in practice. Part 1. Analysesof plants and historic textiles, Fresenius J. Anal. Chem. 371 (2001)10091017.

[25] H. Berke, The invention of blue and purple pigments in ancient times,Chem. Soc. Rev. 36 (2007) 1530.

[26] P. Vandenabeele, S. Bode, A. Alonso, L. Moens, Raman spectroscopiclysista A 6GiustGobetdy, J.. Glatropo, N. Ugarte, R. Chianelli, In-situ identification of palygorskiteBlue samples using synchrotron X-ray powder diffraction.

w.srs.dl.ac.uk/arch/ssrl/maya-blue.html.n, P. Kohl, C. Muller, The nature of the colour centres in Mayahe incorporation of organic pigment molecules into the paly-lattice, Z. Anorg. Allg. Chem. 630 (2004) 97103.tad, Quantum secrets of Maya Blue, New Sci. 151 (1996) 17.

ana

Ac[27] R.

R.stu

[28] J.BAnof the Maya wall paintings in EkBalam, Mexico Spectrochim.1 (2005) 23492356.

etto, F.X. Llabres i Xamena, G. Ricchiardi, S. Bordiga, A. Damin,to, M.R. Chierotti, Maya Blue: a computational and spectroscopicPhys. Chem. B 109 (2005) 1936019368.ss, Catalogo de la Coleccion de Codices. Museo Nacional delogia, Instituto Nacional de Antropologia e Historia, Mexico, 1964.

Thermal and Raman-spectroscopic analysis of Maya Blue carrying artefacts, especially fragment IV of the Codex HuamantlaIntroductionCodex Huamantla, fragment IVMaya clay sculptureBlue coloured threadsPre-Columbian paperMaya Blue colourPalygorskite

Material and methodsThermal analysisRaman spectroscopyColoursArtefacts

Results and discussionInvestigations by Raman and IR spectroscopyIndigo and authentic Maya Blue from the MNAMaya clay headBlue coloured tissue and threadsFragment IV of the Codex Huamantla

Thermal investigationsPaper of fragment IV

ConclusionAcknowledgmentsReferences