A multi-analysis characterization of medieval and vernacular coating mortars in rural Valencia (Spain): An experimental study for a Heritage Action Plan

Availableonlineat

ScienceDirect

www.sciencedirect.com

Original article

A multi-analysis characterization of medieval and vernacular coating mortars in rural Valencia (Spain): An experimental study for a

Heritage Action Plan

Juan A. García-Esparza

, Francisco Pardo

, Luis M. Palmero

aDepartmentofMechanicalEngineeringandConstruction,UniversitatJaumeI,Avda.VicentSosBaynats/n,12071CastellóndelaPlana,Spain

bDepartmentofEducationSciences,UniversidadCEUCardenalHerrera,CalleGrecia,31,12006CastellóndelaPlana,Spain

cDepartmentofArchitecturalConstruction,UniversitatPolitècnicadeValència,CamídeVera,s/n,46022València,Spain

a r t i c l e i n f o

Articlehistory:

Received5July2017 Accepted23October2017 Availableonline24November2017

Keywords:

Externalrenderings Limemortars Gypsummortars Vernacularlandscape

a b s t r a c t

Almostallthefac¸adesofruralvernacularconstructionswererenderedinordertoprotectanddecorate themasonrywalls.Therefore,thisstudyhasbeencarriedoutinordertoidentifyandclassifythediffer- entvernaculartechniquesforproducingmortaroverhistoryinagivengeographicalregionthroughthe combinationofpetrophysic,chemicalandorganolepticanalysis,goingontoclassifyanddateconstruc- tionsforwhichtherewerenorecordeddata.Theresultsshowthatmortarsdoindeedcontainawealth ofinformation,whichsituatestheseconstructionsinaspecificperiodofhistoryandallowsrelationships tobeestablishedbetweenconstructionstagesandtechniques.Theresultshavealsodemonstratedthat mortarsaremainlycomposedoflimeratherthangypsumaswaserroneouslyassumedbeforethestudy.

So,thedeterminationofcomponentsandtechniqueshasbeenconsideredacrucialaspecttobetaken intoaccountwhenworkingontheconservation,ofcontemporaryaestheticinterventionsforwhichthe combinationofpetrophysical,chemicalandorganolepticanalysisisnecessarytoguaranteecompatibility betweenexistingmortarsandnewones.

©2017ElsevierMassonSAS.Allrightsreserved.

1. Introduction

Thecharacterizationofoldmortarisofinterestforarchaeol- ogyandarchitecture[1–5],specificallyinthefieldofmonument restorationandconservation[6,7].Inthecaseunderstudy—the analysisof mortars in vernaculararchitecture— ourinterest is focusedonascertainingwhetherornottheinitialhypothesisheld bysomeauthorsthattimeconditioningfactorsdeterminethequal- ityofagivenmortariscorrect,orwhether—asasecondtheory speculates— thequality ofthese mortarsdepends onthepro- ductionprocess and/ortheavailability ofnatural materialsand resources.Inthisregard,recentstudiessuchasthatbyMoropoulou etal.[8]offerapreciseanalysisofthetechnologyandincompat- ibilitiesofcontemporaryinterventionsonhistoricmortars,while Fariaetal.[9]focusesonthesuitabilityanddurabilityofcertain limemortarsoverothers.

∗ Correspondingauthor.

E-mailaddresses:[email protected](J.A.García-Esparza),

[email protected](F.P.Fabregat),[email protected](L.M.P.Iglesias).

Athirdmajorfactorinthisstudyisestablishingtheexactcom- positionofhistoricmortarsintheregionstudiedinordertodate andidentifytherelationshipofthemortarproductiontechnique, andtoprovideruralvernacularconstructionswithrecognitionofa historicidentity,rarelyvaluedorknown.Hencetheneedforfurther knowledgeandthereversalofexternalcoatings’removalprocess whichalterstheconstitutionandexternalappearanceofvernacular buildings(Fig.1).

Given the above, the aim of this study is the petrographic characterizationofthesematerials,identifyingcomposition,tex- ture and porosity, as well as the relationships between them, asfoundinconstructionsin HistoricCentres(HC)or inPeriph- eral Constructions (PC) (Fig. 2). Criteria are selected based on sample characteristics in order to establish typologies that are significantas constructionmortars. Themain aimof this study is to establish a relationship between the typologies obtained and the origin of the mortars: origin of the material, pro- duction process, placement in the building or construction period.

The characterization of mortars is important since the area

—28,799.30 Ha— isintheprocess ofbeingconsideredof spe- cialrelevancebecauseitsvernacularvalues [10].Theplace will https://doi.org/10.1016/j.culher.2017.10.013

1296-2074/©2017ElsevierMassonSAS.Allrightsreserved.

Fig.1.ExampleofplunderingprocessofhistoricalcoatingmortarsinCulla(Spain).

hostajointTerritorialHeritageActionPlanbetweentenmunic- ipalities.AsthemapinFig.2shows,fromeachvillagedepartsa pathwaytothePeakofPenyagolosaMountain.Villagesandpaths hostmorethan200Constructions—houses,cottages,watermills, hermitages,etc. — that havebeen consideredof interest. Thus, therationale oftheworkreliesonunderstandingthehistorical evolutionoftheplaceanditsbuildingsbymeansofthemortars’

morphology.

Indoingso,theauthorsselected36samplesofmortarsfrom thedifferentvillagesandpathways.Sampleswerecodified—HC- USE-001—meaning,HCorCPdependingonthelocationofthe building,USEmeaningtheacronymofthevillagewerethesam- plewastaken, and finally,the numberthat relatesthe sample withthebuilding.In thisregard,fromthe36samples,14were takeninHCand22inPC.Asregardsthelocationandtheimpor- tanceofthebuilding, 8samplesweretakenfromUS-lesUseres village,11inCU-Culla,4inAT-Atzeneta,3inLC-Llucena,2inLD- Ludiente,3inVH-Villahermosa,4inPM-Puertomingalvoand1in TEU-Vistabella.Samplesweretakenfromthosepartsofbuildings where they are better preserved, eaves and jambs, notwith- standingthat surface erosion was intrinsic to the most of the samples.

2. Areaofstudy

TheplaceofstudyisthePenyagolosamountainrange,inthe provinceofCastellón,Spain.Thescatteredmedievalsettlementsin thehighlands,articulatedthroughoutpathways(Fig.1)tookplace duetonecessity,andsubsequentlybyself-sufficiencyexploiting landresources.Thevernaculardwelling,anepitomeofamedieval culturebasedonmobilityandahand-to-mouthwayoflife,came

tobeseenasasymbolofstability[10].Despitemajoreconomicand socialchanges,theautonomoushouseholds,thesurvivingspatial patterns,thebuildings’artsandcrafts,andtheterritorialhistories contributingtothecontextpersisttothisday.So,itisbelievedthat themostofthebuildingsthatstillremaininthelandscapebelong tothe16th–18thcentury[11].

Pre-industrial mortars-making activity had defined specific duties:limekilnworker,mortarmixer,sandcarriers,etc...[12].

Asregardsthecharacteristicsofmortars,previousauthorsestab- lished theoptimum combinationof limebinders[13,14],while othersestablishedtheconditionsofhydraulicityoflime[15]and otherstheimportanceoftheadditionofsmallfragmentsofcooked clayanditsdosage;mortarsthat arepoorinlimein1/3to1/5 [16,17].

Nonetheless,thenormalproductionof commonlimeinpre- industrialruralsettingswasdoneinaruralkilnwhenthestone wasveryclosetotheconstructionsiteorthewoodwasnearthe quarries.Thiswascheaperthanresortingtokilnsin urbancen- tres.Asa result,mortaroften containedsaltpetre, badlyslaked lime,excessivelycoarseaggregateandvegetableremainsincluding woodcarbonorotherfibres.OthermortarsfoundnearPenyagolosa aremortarswithclayorearthmixes.But,ithasnotyetbeenestab- lishedwhetherthesemortarshaveahighindexofhydraulicityby calcinationasestablishedbyVicatoraresimplybastardmortars madeupofclayearth[18].

The aggregate size comes in two categories, fine or coarse, dependingonwhetheritresemblesdustoraseed[19].Theaggre- gatesaredefinedaccordingtothetypeandqualityoftheclosest stonematerial:siliceousorquartz,sandy,clay,grainedstone,gyp- sum,or calcareouslimestone.Cavanilles(1792–1793)described themountainsofthekingdomofValenciaascalcareousorsandy withsmalloutcropsofgypsum[20].Itisthereforeclearthatthe dosageofthedifferentmaterialsavailableandhowtheywerepro- ducedgreatlyconditionedthemortarquality.

In1785,duetoexcessivecuttingandthinningforcharcoalpro- ductions,anorderbytheNationalCouncilrequestedthatEconomic Societiesgivepreferencetolocationsinwhichcoalstonequarries couldbefound.ThisiswhyCavanillesseesgypsumasawidespread resourceinconstruction[20].Thegypsumusedinruralsettings wasknownasroughorblack,alsomixedwithsand,limeandmetal oxides.Mixedwithsandthisgypsumcoulddoubleitsvolumeand becomeextremelyhard[19].Bastardmortarswereobtainedthat wereonepartgypsumwiththreepartslime-sandmix,withthe increasedvolumeneutralizingthereductionundergonebythelime andusedtorenderexternal wallsavoiding thecracksproduced withtheexclusiveuseoflime.

In general,thebonding agentusedin thesamples collected madeuse ofclay, limeand gypsum, which bonded themortar through different manufacturing, firing or alteration processes.

Among these it is worth noting the binders that only suf- fer physical changes, such as those with clay, or those which experiencechemical transformationssuchasbinderscontaining calciumoxide(CaO)whichinthepresenceofcarbonoxide(CO₂) transforms into calcium carbonate (CaCO₃) and crystallizes as calcite[21].

Thesamplesofthestudyweretakenfromthefoothillsofthe IberianSystem.Theclimateintheareaismarkedbyseasonaland dailythermaloscillations,whichcausemajorstressintheinter- sticesofthemortars,especiallythoseexposedtosolarradiation, whichisintenseinthesummerwithpeaksofmorethan30^◦Cin JulyandAugust.Moreover,thesetensionsworsen inthisregion withthehighnumberofdayswithfreezingtemperaturesatnight withpeaksof−5^◦CmainlyinJanuary;a meanoftemperatures belowzeropractically30daysayear(Table1)[22].Thesecondi- tionsareespeciallyimportantinthecaseoftheseporousmaterials [23].

Fig.2. Mapwiththedispersionofthesamples.

Table1

ClimatedataforVistabelladelMaestrat,PenyagolosaMountain.

Climatedata Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Meantemperature^◦C 2.8 3.5 5.7 8.0 11.7 15.9 19.3 19.5 16.1 10.9 6.3 4.2

Meanmaximumtemperature^◦C 6.6 7.6 10.4 12.8 16.6 20.8 24.4 24.5 20.9 15.3 10.7 8.7

Meanminimumtemperature^◦C −1.0 −0.6 1.0 3.2 6.9 11.0 14.3 14.6 11.3 6.6 2.0 −0.2

NationalAgencyofClimate

3. Methodology

3.1. Characterizationofthemortars

In terms of composition, the mortars can either have pet- rographiccomponentsor a mineralcomposition.In thecase of petrographiccomponents a distinction was made betweenthe aggregatesfrom thebondingphase, determiningtheirmineral- ogyand identifyingceramic,coal,and vegetablefragments.The chemicalbindercanbehomogeneousinappearanceorshowaten- dencytoformlumps,pellets,andnodulesordiffusegrains.Thisis connectedwiththepurityofthelimeandthedegreeofhomoge- nizationofthemixduringthemortarproductionprocess.

Thedosageorbondingagentratioallowsthetexturetypetobe identifiedasgranular,floatingormassive.Inthisregardthepet- rographiccharacterizationexecutedincorporatestheobservations uponsightandthosemadeusingthebinocularloupe[24],asemi- quantitativechemical analysisandapetrophysicalanalysis.The petrophysicaloneexaminedporosityandsizeandshapecharac- teristics.Followingthepetrographiccharacterization,fivemortar typeswereestablished,subsequentlysortedintothreetypologies andtwosub-typologies.Thecharacteristicsandfrequencyofeach typeallowtheirrepresentationwithinthetotalanalyzedsample tobeassessed.

3.1.1. Petrophysicalanalyses

Thesetestshelpusunderstandthebehaviourofstonematerials whenincontactwithwater,andtheircapacitytoabsorbitorcause ittoevaporate.Asinmostcases,alterationprocessesdependon watercirculationinsideporoussolids,thesetestsareimportant forascertainingtheageand durabilityofmortarsdependingon theirexposure.Thesetestswereperformedonallmortarsamples.

Somesamplesdidnotprovideresultsgiventhelackofcoherence, especiallyinthecaseofsamplesrichinclayandlowinchemical binderastheylosttheirconsistencywhenintroducedinwater.

Thepetrophysicalanalyseswereperformedonirregularsam- ples,alwaysusingportionsascompactaspossibleinordertoavoid thelossofhighpercentagesofsamplewhen submergingthem.

Duetotheirirregularity,thesampleswerepreparedinparalleliped shapesfollowinginternationalguidelines[25–29].Thetesttubes bestsuitedtothetestwerechosentakingintoaccountthespecific limitationsofthesesamples.Thetestsfordeterminingthephysi- calpropertiesofthesampleswereperformedinaccordancewith EuropeannormsUNEEN1097-3fordeterminingapparentden- sityandporosity;UNE-EN1097-6fordeterminingparticledensity andwaterabsorptionandUNE-EN1097-7fordeterminingthereal densityofthefillerusingthepycnometermethod(Table2).

AccordingtoTable2,theabsorptioncoefficient,C_abs(%),iscalcu- latedbydeducingtheweightofthedrysamplefromtheweightof thesaturatedsample.Theresultisdividedbytheweightofthedry sampleandthevalueismultipliedby100.Itisusedtocharacterise thebehaviourofmortarsincontactwithwater.Thefreezingcoef- ficient,K_h(%),iscalculatedbydeducingopenPorosityfromtotal Porosityandthevalueismultipliedby100.Iftheresultisequal orhigherthan75%itcanbesaidthatthesampleofmortarcanbe potentiallyfrozen.

AscanbeobservedinTable2,theporosityofmortars tends tobehigh,althoughveryvariable,rangingbetween10%and40%, withsomesamplesevenreaching50%inthisstudy.Theporesare oftenintergranular,duetothelowcontentofbondingagentor therelativelylowqualityofthemortarproduction.Theporosym- metricaldistributionofmortarsiscomplexandthereareusually severalmaximumradiiofaccesstothepores.Byanalyzingdata fromTable2andapreviousworkfromdelaTorre[30]withsim- ilarsamples,mortarswithhighlimedosagesandclaymineralsin itsdosage,type 2in thisstudy,havea maximumofverysmall sizes,lessthan0.1␮m.andtheirporousvolumepercentageislow, between7and28%.Othermortarswithlowerdosages,type1with gypsumasthemainbondingelementtogetherwithlimeandtype 3withlimeasbondingagentbutwithoutclaymineralsintheir composition,assuggestedbydelaTorre[30],havetheirmaximum inlargersizes(1to0.1␮m).

3.1.2. Chemicalanalysis

3.1.2.1. X-rayfluorescence. Thechemicalcompositionofthetotal sampleinamortarcanbepoorlyrepresentativegiventhehetero- geneoussamples.Asalreadystated,theaggregateismadeupof differenttypesofrock.Inadditionthebondingagentratioisnot constantinallsamples,astheproductionforpre-industrialmate- rialsandprocessescouldhavevariedgreatly.Despitethenumber ofmortarsamplesstudiedand thevariety ofconstructionsand periodstowhichthesebelong,thecharacteristicsfoundinthem couldbeconsideredrelativelyuniform.

Traceelementcontentisalsoimportantwhenexaminingmor- tars, their groups and classification. However, this is greatly influencedbytheheterogeneityofthesematerialsandpossible contaminationprocesseswhichmayhaveoccurredbetweenthe timeofexecutionandthepresent[30].Twentychemicalelements wereanalyzed,althoughthetablesonlyfeaturethoseabovedetec- tionlevels,sothatdifferencesbetweenthevarioussamplescanbe identified.Therefore,thevaluesinFig.3areasemi-quantitative approximationtomortarcompositionwhich,althoughsomewhat homogenous,stilldisplaystheintrinsicheterogeneityofmortar,as canbeobservedintheorganolepticanalysis.

Thesemi-quantitativeassessmentofthechemicalanalysiswas executedusingX-RayFluorescence(XRF)–S4Pioneerspectrom- eter,BrukerfromUniversitatJaumeI.Amargin oferrorof+5%

wasassumedfortheanalysis[31].Thesemi-quantitativeanalysis methodwaschosenasitprovidedaveryapproximatequantifica- tionoftheconcentrationofthedifferentelementsinthesamples usingalgorithms basedonfundamental parameters of exciting, emittinganddetectingX-rays.

3.1.2.2. X-raydiffraction. X-raydiffraction wasusedto obtaina structuralcharacterizationusingtheBrukerD4-Endeavordiffrac- tometertocharacterizethecrystallinecomponents.Thecrystalline phases of sample components were identified and quantified accordingtotemperature,dehydrationandphasetransitions.Dia- gramshavemadeitpossibletocomparetheresultsobtainedfrom thesamplesof this studywithothers previouslyanalyzed, and alsotodate and classifythem accurately.The interpretationof themineralogical analysis wasperformed by semi-quantitative analysis and following a percentage of approximation. The

Table2

Petrophysicpropertiesofthesamples.

Designation VR(cm³) Da(g/cm³) DR(g/cm³) C(%) P(%) Cabs(%) Kh(%)

2015CH-US-001 2.00 − 2.63 − − − −

2015CH-US-002 2.89 1.36 2.38 57.23 42.77 25.91 85.04

2015CH-US-003 1.53 1.47 2.60 56.46 43.54 23.06 82.15

2015CH-US-004 2.45 2.35 2.72 86.27 13.73 2.40 63.11

2015CH-US-005 0.66 1.11 2.63 42.31 57.69 30.64 70.78

2015CH-US-006 1.28 1.72 2.72 63.37 36.63 12.97 71.64

2015CH-CU-001 2.79 1.37 2.58 53.24 46.76 24.34 77.78

2015CH-CU-002 7.62 1.62 2.63 61.35 38.65 14.26 71.23

2015CH-CU-003 2.84 2.06 2.59 79.55 20.45 13.72 161.14

2015CH-CU-004 0.69 1,14 2.54 44.95 55.05 4.54 52.49

2015CH-CU-005 11.04 1.76 2.51 70.36 29.64 9.11 68.59

2015CH-CU-006 6.68 1,52 2.59 58.74 41.26 20.62 80.66

2015CH-CU-007 7.74 2.52 2.68 93.82 6.18 0.48 55.46

2015CH-CU-008 3.59 1.65 2.53 65.15 34.85 13.55 73.59

2015CH-CU-009 11.30 1.80 2.53 71.25 28.75 5.01 59.30

2015CP-AT-001 3.45 1.45 2.65 54.85 45.15 21.20 75.96

2015CP-AT-002 3.31 1.66 2.63 63.17 36.83 18.58 86.28

2015CP-AT-003 1.98 1.52 2.54 60.00 40.00 23.26 89.68

2015CP-AT-004 2.30 1.41 2.44 57.79 42.21 26.20 88.70

2015CP-CU-001 4.46 1.49 2.52 59.31 40.69 23.95 89.23

2015CP-CU-002 2.59 1.43 2.56 55.94 44.06 22.62 78.97

2015CP-LC-001 3.50 1.35 3.07 44.03 55.97 28.03 75.56

2015CP-LC-002 2.85 1.72 2.59 66.59 33.41 14.91 81.20

2015CP-LD-001 3.35 1.61 2.61 61.58 38.42 15.89 74.97

2015CP-LD-002 2.55 1.28 2.91 44.04 55.96 28.07 73.63

2015CP-LD-003 2.10 1.19 2.60 45.85 54.15 37.50 84.83

2015CP-TEU-001 5.89 1.51 2.61 57.75 42.25 24.84 89.86

2015CP-US-001 2.32 1.68 2.52 66.48 33.52 17.78 89.94

2015CP-US-002 2.69 1.40 2.55 54.79 45.21 28.61 89.36

2015CP-US-003 0.41 1.42 2.56 55.41 44.59 22.86 78.50

2015CP-VH-001 3.66 1.35 2.60 51.77 48.23 27.70 81.57

2015CP-VH-002 5.28 1.63 2.58 63.31 36.69 16.30 78.53

2015CP-VH-003 4.92 1.70 2.56 66.22 33.78 15.23 80.93

2015CP-PM-001 2.69 1.46 2.59 56.28 43.72 24.71 84.97

2015CP-PM-002 2.08 1.52 2.72 56.06 43.94 24.07 85.65

2015CP-PM-003 6.95 − 2.77 − − − −

2015CP-PM-004 4.63 − 2.70 − − − −

2015CP-PM-005 4.57 − 2.69 − − − −

VR:realvolume;Da:apparentdensity;DR:realdensity;C:compactness;P:porosity;Cabs:absorptioncoefficient;Kh:freezingcoefficient.SamplesCP-PM-003,004and005 didnotproviderelevantresultsduetotheirlackofcoherence.

diffractogramshavebeeninterpretedwiththehelpofthecom- puterapplicationEVAofthefirmSocabimand theJCPDScards, untilidentifyingthemineralspresentineachoneofthesamples.

3.1.3. Petrographic-mineralogicalanalysis

XRDwasusedtoobtainthefirst approximationtoamortar composition identifyingthebonding agent. However,it is nec- essary to use organoleptic methods, which allow the physical characteristicsoflimeandgypsummortarstobeunderstoodupon sightandtouch–hardnessandtint.Visualanalysis,supplemented withaLeicaMZAPOmicroscopeandsubsequentlyXRF,allowed numerousintermediatecasestoberecognised.Theseincludedlime mortarsattackedbyalteredgypsum,bastardmortarsoflimeand gypsumorgypsummortarswithsmallamountsofcalcite.

Visualexaminationwasusedtoidentifythevariationsingran- ulometryoftheaggregateusedindifferentsamples.Thetypesof aggregateobservedwerediverse,fromnearbylimestoneandsand- stone.Duringtheextractionofsamplesconsiderablepetrographic similaritiescouldbeobservedbetweenthemasonryandtheaggre- gateofthemortars.Someexceptionswerefoundincertainsamples, whichwerethoughttobefromrelativelyrecentmortarsgiventhe appearanceoftheconstructions.

Visualexaminationofthesamplesmadeitpossibletoidentify components,suchasremainsofterracottaorfiredclay,thepres- enceofclay,remainsofcarbonised matterand evenremainsof organicstructures—hyphaeorfungalmycelium—whosepresence mayhavebeenduetoanalterationresultingfrombiodeterioration.

Therewasawiderangeoftonesofthesamplesastheywereaffected

bythetypeofchemicalbinder,aggregate,andthesmallimpurities ofclaysorironoxides(Fig.4).

4. Resultsanddiscussion

Theresultsobtainedshowitispossibletoestablishaninitial classificationdependingonthebinder.Thesecanbedistinguished intolimemortarandgypsummortardependingonthechemical classificationobtainedusingX-rayfluorescence.Allsamplesana- lyzedarelimemortarexceptCP-LC-001,CP-LD-002,CH-US-001, whicharegypsummortars.Itshouldbenotedthatamongthelime mortarstherearetwosamples,CP-PM-004andCH-CU-003,witha highersiliceouspercentagethancalciumoxide(Table3).

Themineralogicalcharacterizationofthesamplesrevealeda prevalence of certain mineralogical phases. These can be clas- sified into three types: hydrated calcium sulphate, quartz and calcite(Table4).Thepresenceofothermineralphasesalongside thechemicalcharacterizationshowsthepresenceoffeldsparand claymineralsorassociates.Followinganinitialorganolepticand petrophysicalexaminationandtakingintoconsiderationthemain mineralphasesandchemicalcomposition ofthesamples,three majorsampletypologieswiththeirsubtypologieswereestablished.

TheSEM/EDXanalysisofthesampleCP-LC-002inFig.5shows spectrawithanintensesignalofcalcium(Ca),sulfur(S)andsilicon (Si).IfwelookatthemappingsofFig.5candditcanbeseenthe distributionofthesetwoelements(Ca)inthethreelayersand(S) justinthelayer1.Therefore,itwouldbeagypsummortarwith siliceousarid.Theresultisconfirmedinanotherareaofthesame

90 80 70 60 50 40 30 20 10 0 Typology 1 2015CH-US-001 2015CP-LC-001 2015CP-LD-002 Typology 2 (2a) 2015CP-AT-004 2015CH-CU-001 2015CP-PM-002 2015CP-PM-004 2015CP-US-002 Typology 2 (2b) 2015CH-CU-002 2015CH-CU-005 2015CH-CU-007 2015CH-CU-008 2015CH-CU-009 2015CH-US-005 2015CH-US-004 2015CH-US-006 2015CP-US-001 2015CP-VH-003 Typology 2 (2c) 2015CH-US-002 2015CH-US-003 2015CP-AT-002 2015CP-AT-003 Typology 3 2015CP-AT-001 2015CP-CU-001 2015CP-CU-002 2015CP-LD-001 2015CP-LD-003 2015CP-US-003 2015CP-VH-001 2015CH-CU-003 2015CP-PM-001 2015CP-LC-002 2015CP-PM-003 2015CP-PM-005 2015CP-TEU-001 2015CP-VH-002

CaO (%) Fe2O3 (%) SiO2 (%) SO3 (%) Al2O3 (%)

Compactness 55-60% -Presence of clay -14-15th Century –13.8% of total sample Compactness 55-60% -No presence of clay -19-20th Century –11.2% of total sample

Compactness 65-80% -Presence of clay -16-17th Century –27.8% of total sample Compactness 50-60% -No presence of clay -16-17th Century –38.8% of total sample

No presence of clay -

Compactness 45% - 18th Century-8.4% of total sample

Fig.3.Maincharacteristicsofthetypologies.

layer,beinginthiscasethelargestsiliconpeakforhavingselected anareawithahigherconcentrationofaggregates.

4.1. Type1

Thefirst type corresponds tosamples where thegypsum is themainbondingagent,asinterpretedfromtheX-raydiffraction wheretheprevalentcrystalphaseis hydratedcalciumsulphate comparedtootherphasessuchasorthoclaseandquartz.SO3 in

samplesCP-LC-001andCP-LD-002ishigherthan40%,asconfirmed by X-ray fluorescence. SiO2 and AlO3 maintain the coherence betweenbothmaterialcharacterizationtechniques,showingthe presenceofquartzsand.Finally,itshouldbenotedthatthepres- enceofCaOandMgOformspartoftheaggregatesmakingupthe mortars.

In organoleptic terms, within the first typology the role of gypsumis thatofbondingagentalongsidelime,CP-LC-001and CP-LD-002,orasaggregateoralterationmineralCH-US-001.This

Fig.4. Organolepticcharacterizationofthesamples.

sampleisdifferenttothepreviousones,asthefirstsarebastard mortarsofgypsumandlime,whilethelatterhasabondingagent thatismadeupalmostexclusivelyoflime,andthelimefillsthe pores.Forthisreasonitisassumedthatthelastsamplemusthave beenoriginallyalime-basedmortar,althoughitisnowsodete- rioratedthatithasmostlybeenreplacedwithalterationgypsum [32].Thisisnotedespeciallyduringopticalmicroscopyandvisual analysis(Fig.4).Asabondingagentitappearsintheformofamass

offineacicularcrystals.Asanaggregateorimpurity,intheformof non-calcinatedfragments,itappearswithtexturescharacteristicof naturalgypsumandwhenitisanalterationmineralitfillsinpores andcrack.Mortarswithagypsum-bondingagentarenoticeably hardiftheyareingoodcondition,andunderthemicroscopethey showagreatnumberofimpurities:fragmentsofnon-calcinated naturalgypsum,andovercookedgypsum,aswellasclayandmarl nodules[33,34].Ingeneral,type1samplespresentsomesmalland

Table3

IdentificationofdescriptivevaluesofsamplesbyX-RayFluorescence(XRF).

Designation MgO AI2O3 SiO2 P2O5 SO3 K2O Cao MnO Fe2O3 Na2O Cl PPC U Sum Observation

CP-LC-001Typology1 3.71 3.61 8.98 − 38 1.34 34.4 0.015 1.48 − − 7.52 % 99.7 Highratesof

gypsumand magnesium

CP-LD-002Typology1 1.49 3.54 9.73 − 40 1.64 33.9 0.028 1.53 − − 6.98 % 99.5 Highratesof

gypsum CH-CU-001Typology2a 0.425 6.91 23.2 − 0.364 4.54 44.9 − 3.02 0.503 1.34 14.3 % 100 Highpresenceof

potassiumoxide CH-US-006Typology2b 0.334 2.12 7.19 0.088 1.59 0.914 49.4 0.029 2.02 0.096 0.11 34.26 % 98.4 Theonlyonewith

phosphates CP-AT-002Typology2c 0.403 1.83 6.18 − 0.22 0.985 56.6 0.022 2.33 0.038 − 31.30 % 100.2 Highratesof

calcium CH-CU-003Typology3 0.379 11.6 40 − 0.206 4.64 27.7 − 4.57 − − 8.52 % 99.0 Highrateofsilica CP-TEU-001Typology3 0.586 7.11 23 − 0.0728 2.49 39.8 0.28 13.1 − − 14.8 % 102.3 Highratesof

Aluminaandiron CP-PM-001Typology3 0.491 2.94 16.4 − 0.411 2.32 58.8 − 3.06 0.077 − 15.5 % 100 Highratesof

calcium

Table4

MineralogicalcompositionofsamplesbyX-RayDiffraction(XRD).

CP-LD-002 CH-CU-001 CH-CU-005 CP-TEU-001 CP-PM-001

Typology1 Typology2a Typology2b Typology3 Typology3

Quartz + ++++ +++ ++++ ++++

Calcite − ++ ++++ +++ +

Illite/muscovite − (+) − (+) (.)

Orthoclase + (+) + + +

Albite − − − − (+)

Gypsum +++++ − − − −

Kaolinite − + (+) − −

+++++:>75%;++++:>50%;+++:>25%;++:>15%;+:>10%;(+):presence(5–2%);(.):(<2%);−:nopresence.

Fig.5.CP-LC-002.Type1.OpticalMicroscopyandScanningElectronMicroscopy(SEM/EDX).

mediumnon-calcinatedlimenodules,remainsofcalcinatedcoals andoccasionallyfiredclay.Intermsofappearance,thesesamples arepredominantlywhitish(seeFig.6).

Thelimeusedinmostofthemortarswasnon-hydraulic.Inaddi- tion,itwasalsorathergreasy,thatistosay,ithadalowcontentin

Mg.Thiswasconfirmedviaasemi-quantitativeXRFanalysis.Car- bonatedlimetendstotaketheformofcalcite,althoughinsome instancesofmixeswithhighlimecontent,apartofitprecipitates intheformofaragonite[35].Mineralswhichmaysuggestsome degreeofhydraulicnature(tobermoriteorettringite)werefound

Fig.6.XRDpatternofsampleCP-LD-002.

onlyinsamplesCP-LC-002,CP-US-001,CH-CU-007,CP-TEU-001 duetothehardness,thegreyishappearanceofthechemicalbinder andtheadditionofvitreousreactiveaggregateorgroundbrick.

Thesesampleswereinitiallyclassifiedastype4giventheocca- sionalmedium-sizedlimenodulestheypresented,andtheabsence of vegetableremains exceptin CH-CU-007,withmedium-sized aggregateandclaynodules.Mineralogicalcriteriawerefinallyused toclassifythefirstthreesamplesastype2(2b)andCP-TEU-001as type3(seeFig.4).

Withintype1 wefindsampleswithoccasional variationsin colour,especiallyduetothetoneoftheaggregate.Intype2and3, insampleCP-VH-003thelimenodulesarelarge,whileinsamples CP-PM-003andCP-PM-005thelimenodulesaresmallandtheir hardnessandconsistencyarelowerthantherest.Finally,sample CP-VH-002veryoccasionallypresentsvegetableremains,CH-US- 004showsremainsoffiredclayandinCP-VH-003theaggregate isfinerthanintherest.Despitethedifferencesintheorganoleptic characteristicsofthesamples,duetothedifferentmanufacturing processesandavailabilityofmaterials,chemicalcompositionand petrophysicaltests,compactnessandporosymmetryofthetotal sample,acertainsimilaritycouldbesuggestedbetweenthesam- plesoftype2(2a)andsomeoftype3.

4.2. Type2

Thesecondtypologyisthatofsamples,whichhavelimeasa bondingagent,butthemainmineralogicalphaseiscalcitefollowed byquartzwithapresenceoffeldsparandkaolinite.Thesecharac- teristicallyincludelargeamountsofsilicaandlimeoxide,aswell aspotassiumoxideandaluminiumoxide.Thissecondlargegroup ischaracterisedbyincludingclaymineralsinitscomposition,as describedinthefollowingsamples.

CH-CU-001servesasa sampleexamplefor thesecondlarge group;limemortarwithamainmineralogicalphaseofquartzwith apresenceofclay;itischaracterizedbyitslimebondingagentbut themainmineralogicalphaseisquartzfollowedbycalcitewith a presence oforthoclase, illite and kaolinite. Thesecharacteris- ticallyincludelargeamountsofsilicaandlimeoxideaswellas potassiumoxideandaluminiumoxide.Themineralogicalcharac- terizationshowsthepresenceofclaymineralsaswellasassociated ones,asisconfirmedbyXRF(seeFig.7).

CH-CU-005isanexampleofsamplesfromthethirdlargegroup;

limemortarwithamainmineralogicalphaseofcalcite.Thesam- plestudiedshowsthatthelimemortarisduetothewidespread presenceofcalcite(56%)withahighpercentageofquartz(35%) showingalargepresenceofquartzsandinthesample.TheXRFfor thesampleconfirmsandisinagreementwiththeseanalysesasit showslargeamountsofCaOandSiO₂.Itshouldbenotedthatthere isahighpresenceofKO2(4.54%)inthesample,whichexplainthe presenceoforthoclaseintheXRD.Thereforethepresenceofclay orotherassociatedmineralssuchasthekaoliniteprovidedbyAlO₃ andSiO2isalsoobserved(seeFig.8).

Inorganoleptictermsthesamplesoftype2(2a)and3aresim- ilarinappearanceastheyaremadeupofachemicalbinderthat isexclusivelylime.Itcanbestatedthatthepetrographiccharac- teristicsoflimemortarscanbemonotonous.Thetypeofaggregate isalmostalwaysthesame,subangularorsubroundedfragments oflimestoneandsandstone.However,followingavisualinspec- tionseveralcharacteristicfeaturesidentifiedthedifferentsamples, providingfurtherinformationontheirdateandformofproduction.

Thesamplesclassifiedastype2(2a)andwhosemineralogicalcom- positionappearsinFig.3couldbeconsideredbetterqualitylime mortars,asbytouchthesampleissurprisinglyhardandcoherent, duetothefineaggregateofthegranulometriccomposition,which insomesampleshasbeenmixedwithfiredclayandinotherswith

Figs.7–8.XRDpatternofsamplesCH-CU-001andCH-CU-005.

claynodules–exceptinsamplesCP-PM-001andCP-PM-004.The claydetectedinseveralsamples,includingCH-CU-005,couldbe consideredafineaggregatealthoughitdoeshavebondingpower.

Finally, also for the small and medium-sized lime nodules, exceptinsamplesCH-CU-002,CH-CU-005andCH-CU-009where calichesareslightlylarger.Itslimitedporosityisobserved—inthis casefissured—underthemicroscope,asistheexcellentunionin theaggregate-bondingagentcontactsofsamplesCH-US-004,CH- CU-005,CH-CU-007andCH-CU-009.Cross-referencingthisdata withthepetrophysicalanalysesthesesamplesareamongthemost compactofthetotalsample.Finally,itshouldbenotedthatthis typedoesnotcontainremainsofcalcinationoroccasionalvegetable remainspossiblyresultingfrompollution,exceptforsamplesCH- CU-002, CH-CU-005, CH-CU-008, CH-US-005 and CH-US006. In termsofthechemicalcomposition,itshouldbenotedthattype 2sampleshaveahighpercentageofCaO,above50%,andbelow 20%forAl₂O₃andSiO₂.

4.3. Type3

Thethird typology correspondstothesampleswithlimeas bondingagentandamineralogicalphasewithquartzfollowedby calcitewithapresenceoffeldsparandillite.Thesecharacteristically containhighamountsofsilicaandlimeoxide,aswellaspotassium oxide,sodiumoxideandaluminiumoxide.Thisthirdmajorgroup doesnotincludeclaymineralsinitscomposition,asdescribedin thefollowingsamples.

CP-PM-001:thecharacterizationofthesampleshowsthatthis isalimemortarinwhichquartzsandspredominateovercalcite, followedbyorthoclaseandalbite,withthepresenceofillitealso detected.ThesemineralogicalphasesobtainedusingXRDarein keepingwiththeXRFwhichshowaconsiderablequantityofCaO whoseoriginisattributedtocalciteaswellasofSiO₂comingfrom thequartzsand,orthoclaseandalbite.JustasNaO2andAl2O3come fromalbite,illiteprovidesKO₂andFe₂O₃.Illitearisesfromthermal alterationofmuscovite,andisalsooneofthemineralsassociated withclay(possiblyduetoonsitecontaminationorduetothemor- tar).Orthoclaseandalbitearefeldspars,whichcontributesilicato thecomposition.Asregardschemicalanalysis,itisworthhighlight- ingthemajorPPCduemostlytotherudimentaryprocedurewith nothermalcontrolresultingfromtheproductionoflimeinakiln.In organolepticterms,remainsofcoalorinertmaterialscanbefound (seeFig.9).

CP-TEU-001:thesampleshowstheprevalenceofquartzover calciteintheXRD,thesearefollowedbyorthoclaseandillite.The presenceofthesemineralogicalphasesisinkeepingwiththeXRF analysisshowinglargeamountsofSiO2andCaO(23%and29.8%).

Thedatashowtheuseofquartzsandsinthemortars,aswellas carbonates.Thepresenceoforthoclaseandillitephasessupport theconsiderablepresenceofK2O,Al2O3andFe2O3.Chemicaland mineralogicalcharacterizationshowsthatthemortarislime.As regardschemicalanalysis,itisworthhighlightingthemajorPPC duemostlytotherudimentaryprocedurewithnothermalcontrol resultingfromtheproductionoflimeinakiln(seeFig.10).

Inorganoleptictermstwoofthemaindifferencesidentifiedin thesamplesclassifiedwithintype 3,comparedtothose oftype 2(2a),arethesizeofthenodulesofnon-calcinatedlime,asthey aremedium-sized,largeorverylarge,andabundant.Thisisalso thecaseofsamplesCP-CU-001andCH-US-003,whichdonothave aggregatesintheformoffiredclayorclayearthinthesamples, exceptincasessuchasthoseofsamplesCP-AT-001orCP-CU-001.

Therefore,themortars withinthis type correspondtoasimilar typology,withmediumlimedosages,contrastingwiththesizeof thelimenodulesandnotentirelycoherentwiththesesamples.This, combinedwiththeabsenceofclayorterracotta,suggeststhatthese mortarsbelongtoadifferentconstructionperiodtothosereflected

intype2(2a).Theshadesofthesetype3mortarsareochre,light brownandoccasionallywhitish.Incontrast,mostofthoseanalysed visuallywithintype2(2a)areapinkishochrecolour,perhapsdue toarelativecontentofreddishclayorironoxides.Thechemical compositionofthesemortarsdisplaysahigherpercentageofAl2O3

andSiO₂,above20%,andalowerdosageofCaO,below50%.

Toconcludethevisualexaminationofthemortarsitshouldbe notedthattwosamplesofmortarthatdifferedgreatlyfromthe restwerefound.Thesewereaprioriclassifiedastype5.Asregards themineralogicalcompositionsnomajordifferenceswerefound, exceptthattheyweretwoofthemortarswiththehighestcontent inCaO,between54and57%.However,duringorganolepticanalysis twoveryhomogenoussamplesareobserved,assuggestedbythe appearanceoftheconstructions,whichwereprobablymadeinrel- ativelyrecentmortars,CP-AT-002andCP-AT-003.Whentouched, bothappearratherhard,althoughnotashardastype2samples CH-US-004,CH-CU-005,CH-CU-007andCH-CU-009.Theypresent averyregularandvarieddosageofaggregatewithanalmostcom- pleteabsenceoflimenodules,withoutremainsoralterationsand inahomogenouswhitishandgreyishcolour.Inanycase,aswith type4(seethedescriptionintypology1),mineralogicalcriteria werefinallyusedtoidentifymortarsoftype5asavariantoftype 2(2c).

4.4. Synthesisofdata

Theresearchhasemployedcomplementaryanalysis,qualitative andsemi-quantitative,thatinformsoneanother.Thatiswhythe authorsbelievedthattheexplanationandanalysisofthesamplesis betterunderstoodbyjoiningtheoutputsofthedifferenttechniques andexplainingthemalltogetherthroughsection4.

Theanalysisoftypology1demonstratesthatsomenodulesof limepresentinthesampleswasbadlycookedandthatiswhytheir compactnessistheloweroftheseries–45%.Thepresenceofgyp- summayhavepreventedthemfrombeingeroded;gypsumwas extensivelyusedatthattimeasrenderinginthemajorcities[34].

Thistypologyonlyrepresentsthe8.4%ofthetotalsample,represen- tativeoftwospecificplacesinthecountryside.Giventhefindings ofCavanilles,asregardsthelackofcoalforcalcinatingthelime- stone,thistypecouldbelongtoaveryspecificperiod,conditioned bytheavailabilityofthematerialandthewidespreadoccupation oftheterritory,bothcharacteristicfeaturesofthe18thcentury.

Typology2wasthemostcomplexandthatiswhyitwasfinally subdividedinthreesub-typos.Type2(2a)presenteda55-60%of compactness,nonethelesssomeofthesamples,assampleCH-CU- 001showsinthephasesoftheXRD,sufferedfromdecaydueto theirexposition;thissamplesaremainlylocatedinthecountry sideandintheupperzonesoftheareaofanalysis.Inthistype2, fundamentaldifferencesinthesemi-quantitativeanalysisoftwo samplesshowsthatCU-001XRDpatternhasmuchmorepercent- ageofSiO2thanCU-005XRDpatternwhichshowsmoreCaCO3. Nonethelessthetwosampleshavethetwocomponentsintheir constitution.Then,thequalitativeorganolepticvisualizationrefers tothesamples’aggregatesaslimestone(CU-005)andsandstone (CU-001).Samplesofthistypedidnotusegypsum,nonetheless clayandfibresarepresentinthemostofthem.Thistyperepresents the13.8%ofthetotalsampleand,thestratigraphicunitfromwhich thesampleswereextracted,possiblydatestheminthe14th–15th centuries.

Type2(2b),themostcompactsamples—65%–80%—,usedclay butnotgypsuminitscomposition,astype2 (2a).Thissamples didnotpresentseveredecay,themostofthemarelocatedinHC andrepresentthe27.8%ofthetotalsample.Byanalyzingthetype ofbuildingsandtheircorrelationwithtypology3,itcanbestated thatthistype2bbelongstothe16thand17thcenturies.Type2(2c) samplesmaywellbeofpoorerqualityandexecution—55–60%of

Figs.9–10.XRDpatternofsamplesCP-PM-001andCP-TEU-001.

compactness—.Samplesdonotpresentgypsumnorclayintheir composition,besidestheywerefoundinbothHCandPCrepresent- ingjustthe11.2%ofthetotalsample.Theiroriginprobablydates backtothe19thand20thcenturies.

Intypology3,thenomenclatureofthesamplesestablishalink betweenthe morphology of mortars and its location, which is mainlydisperseincottagesinPCandwarehousesinHClocatedin theupperpartsofthevillageofCulla.Thistyperepresentsalmost the40%ofthetotalsampleanditischaracterizedbyaquitegood stateofdecayinspiteofthebadqualityofthemortar.Typology 3ischaracterizedbytheabsenceofgypsumandclay.Thelimited coherenceofthemortarsuggestsamoreartisanalandlessrigor- oustechnique,asitcannotbelinkedsolelytotheruralkilnsorcity kilns,giventhatthesesamplesarefoundinbothsettings.These conclusions,andthearchivedata,promptustoconcludethatthey areconnectedwithalessperfectedformofconstructionswhich wasprobablylaterthantype2(2a)samples,betweenthe16thand 17thcenturies.

5. Conclusions

Tosumupitcanbestatedthatthemortarusedintheregion since theMiddle Ages it hasbeenmade oflime mortar.Given theabsenceofsignificantdifferencesbetweencomponentsit is assumedthatthisdidnotdependonwhethertheconstructions wereinHCorPC.Exceptionsandparticularitiesareobservedin thecomponentsofsomemortars and theirmanufacturingpro- cess,confirmedthroughtestsandanalyses.Althoughwedonot haveanexactdateforthemortars,itcanbestatedthatthereare fivedifferentmortartypologiesandfivedifferentiatedconstruction periodsthatcanbedefinedbytheclassesobtainedwiththeorigin ofmortars;originofmaterial,productionprocess,locationwithin thebuildingorconstructionperiod.

Accordingtotheclassificationanddatingfromthisstudy,the initialcriterionestablishedregarding16thand17thcenturymor- tarscanbeassumedasvalid.However,thestatementthatmedieval mortarsarepoorerqualitycannotbeconsideredasvalid.Inour caseweconsiderthesamplestobemoreheterogeneousaspre- viousauthorsindicated,but thedosageandarehighlycompact accordingtothechemicalanalysisundertakeninthestudy.Infact, bycarefullyexaminingthecompositionofthedifferentsamples,it canbestatedthatthemaindifferencesarefoundbetweenHistoric Centres(CH)andPeripheralConstructions(CP).Thedosageestima- tionofCHsamplesisricherinaggregatethanCPones.Proportions mayvaryfromvaluesnearto1/1CHand1/2CPforthe14thto15th centuriesand16thto17thcenturies.Inthislastperiodthepropor- tioncandropeventuallyto1/3insomeCPsamples.Fromthe18th centuryon,mortarsdonotshowsuchdifferences.Giventheabove, itcanbeconcludedthattheperiodofconstructiondeterminesthe qualityofthemortar,whileorganolepticanalysisshowsthatmor- tarqualityalsodependsontheprocessofproductionand/orthe availabilityofmaterialsandnaturalresources;thelocation.

Thecharacterizationofmortarshasallowedtheresearchersto mapandestablishtypologiesrelatedtoalocation,toaspecifictime inthehistory,andbetweenthemselves.Thecharacterizationhas allowedtheresearcherstounderstandtheorigin,themateriality andtheproductionprocessofmortars.Thisisofutmostimpor- tancesincea TerritorialHeritageAction Planwillassume some retrofitactions.Theseactionswillimplytheuseofnewmortars, dosagesandcompositions.So,whatthestudyhasdemonstrated istheimportanceofheterogeneityinagreatruralarealookingfor futurematerialcompatibility.Anynewsampleofmortarinthearea cannowbeclassifiedandunderstoodinsideamoreglobalprocess ofconservation.Theimportanceofthestudythenreliesonthelink- agesbetweeneventsandepochsratherthanonthematerialityof

buildings.So,theanalysisprovidesmorerelevancetotheterritory, andtotherelationshipsbetweenbuildings,waysoflifeandspace.

Funding

ThisworkwassupportedbytheRegionalGovernmentofCastel- lón anddeveloped bytheteamoftheChair ofHistoricCentres andCulturalRoutesatUniversitatJaumeI.Itwasalsosupported withfundingfromtheproject“WorldHeritageCulturalLandscapes List. Keysfor theiridentificationand criteria for theirmanage- ment”fromtheMinistryofEconomyandCompetitivenessinSpain, MINECO.CSO2015.65787-C6-6-P.Finally,theauthorswouldalso liketothankthereviewersfortheirinvaluablecomments.

References

[1]F.Gutiérrez-Solana,M.Jáuregui,R.Bohigas,P.Sarabia,Análisisdemorteros medievalesdedoscastillosdeCantabria(Espa ˜na),Mater.Construcc.39(213) (1989)37–45.

[2]M.Vendrell-Saz,S.Alarcón,J.Molera,M.García-Vallés,Datingancientlime mortarsbygeochemicalandmineralogicalanálisis,Archaeometry38(1)(1996) 143–149.

[3]A.B.Leslie,J.J.Hughes,Bindermicrostructureinlimemortars:implicationsfor theinterpretationofanalysisresults,Q.J.Eng.Geol.Hydrogeol.35(3)(2002) 257–263.

[4]F.Casadio,G.Chiari,S.Simon,Evaluationofbinder/aggregateratiosinarchae- ologicallimemortarswithcarbonateaggregate:acomparativeassessmentof chemical,mechanicalandmicroscopicapproaches,Archaeometry47(4)(2005) 67–689.

[5]S.Pavía,S.Caro,Aninvestigationofromanmortartechnologythroughthe petrographicanalysisofarchaeologicalmaterial,Constr.Build.Mater.22(2008) 1808–1811.

[6]J.I.Álvarez,A.Martín,P.J.García-Casado,I.Navarro,A.Zornoza,Methodol- ogyandvalidationofahothydrochloricacidattackforthecharacterization ofancientmortars,CementConcreteRes.29(7)(1999)1061–1065.

[7]O.Cazalla,M.J.delaTorre,Morterosderestauraciónymorterosantiguos.Téc- nicasdeestudio,in:R.Villegas,E.Sebastián(Eds.),Metodologíadediagnóstico yevaluacióndetratamientosparalaconservacióndelosedificioshistóricos, InstitutoAndaluzdelPatrimonioHistórico,Comares,Granada,2003,pp.36–47.

[8]A.Moropoulou,A.Bakolas,K.Bisbikou,Investigationofthetechnologyofhis- toricmortars,J.Cult.Herit.1(1)(2000)45–48.

[9]P.Faria,F.Henriques,V.Rato,Comparativeevaluationoflimemortarsfor architecturalconservation,J.Cult.Herit.9(3)(2008)338–346.

[10]J.A.García-Esparza,Areworldheritageconceptsofintegrityandauthenticity lackingindynamism?AcriticalapproachtoMediterraneanautotopicland- scapes,LandscapeRes.(2017)(Editingprocess).

[11]J.A.García-Esparza,P.Altaba,Re-readinglandscape.Anapproachtothehistor- icalandarchitecturalconfigurationofPenyagolosaarea,in:J.A.García-Esparza (Ed.),Landscape,ethnographyandritualsatPenyagolosa,UniversitatJaumeI, CastellóndelaPlana,2017,pp.185–206.

[12]G.Navarro,LaindustriadelaconstrucciónenlospaísesdelaCoronadeAragón (ss.XIII–XVI),in:S.Cavaciocchi(Ed.),L’ediliziaprimadellarivoluzioneindus- triale.,AttidellaXXXVISettimanadiStudidell’IstitutoInternazionalidiStoria Economica,LeMonnier,Florence,2004,pp.167–207.

[13]C.Sambricio,Laarquitecturaespa ˜noladelailustración,ConsejoSuperiorde ColegiosdeArquitectos,Madrid,1986.

[14]B.Bails,TratadodeArquitecturaCivil,2,ColegiodeAparejadoresyArquitectos detécnicosdeMurcia,Murcia,1796,pp.369–401.

[15]L.J.Vicat,Recherchessurlemorteries”,CasaGoujon,in:LibrerodesusAltezas Reales,LaDuquesadeBerry,Paris,1818(“Investigacionesexperimentales sobrelascalesdeconstrucción,loshormigonesylosmorterosordinarios”.

INTEMAC.InfoprintS.A.,1999).

[16]I.Gárate-Rojas,Artesdelacal,Munilla-Lería,Madrid,2002,pp.415.

[17]M.J.Cassinello,Effectofmortarjointthicknessondeformabilityinmedieval stonewalls,Mater.Constr.56(284)(2006)69–80.

[18]M.Fornés,Gurrea,Observacionessobrelaprácticadelartedeedificar,Car- rerizo,Valencia,1841.

[19]F.GeryLobez,Tratadodeconstruccióncivil,Laminervaextreme ˜na,Badajoz, 1898.

[20]N.LaRoca,Canteras,minasycortesobservadosenlosviajesdeCavanilles, CuadernosdeGeografía(1997)425–454.

[21]F.J.Alonso,R.Bustamante,C.Díaz,J.Monjo,I.Salto,Glosariodemorteros, ReCoPar6(2009)33–40.

[22]AEMET.,AEMETClimaticdatainSpain2015.(accessedon:16thOctober2015;

availableat:http://www.aemet.es).

[23]J.M.García-deMiguel,Procesosdedegradacióndelapiedra,in:L.Mas-Guindal (Ed.),ProcedimientosytécnicasconstructivasdelPatrimonio,Munilla-Lería, Madrid,1999,pp.175–183.

[24]B.Agulleiro,J.J.Alonso,Estudiodelresiduoinsolublederocascarbonatadas espa ˜nolas,Bol.RealSoc.Esp.70(1972)29–54[HistoriaNatural(Sec.Geológ- ica)].

[25]RILEM,Recommendedteststomeasurethedeteriorationofstoneandtoassess theeffectivenessoftreatementmethods,in:Commission25-PEM:protection eterosiondesmonuments,1980,pp.175–253.

[26]ISMR,Suggestedmethodsforlaboratorytestingofargillaceousswellingrocks, Int.J.RockMech.MiningSci.26(5)(1981)415–427.

[27]ASTM,Construction,in:AnnualBookofASTMStandards,ASTM,Pennsylvania, 1988,pp.8.

[28]NORMAL, Assorbimen to d’aqua per immersion totale, in: Capacità di imbibizione,1981,pp.7–81.

[29]UNE,CatálogodenormasUNE1993,InstitutoEspa ˜noldeNormalización,Centro SuperiordeInvestigacionesCientíficas,Madrid,1985.

[30]M.J.delaTorre,EstudiodelosmaterialesdeconstruccióndelaAlhambra, UniversidaddeGranada,Granada,1995.

[31]R.M.Mellinger,QuantitativeX-raydiffractionanalysisofclayminerals,in:

Anevaluation,SaskatchewanResearchCouncilG-79,Saskatchewan,1979,pp.

1–46.

[32]M.J.delaTorre,J.Rodríguez,E.Sebastian,Presenciadeyesoenmorterosy hormigonesnazaríesenlaAlhambra,Bull.Soc.Esp.Mineral.14(1)(1991)3–4.

[33]T.Kawiak,Gypsummortarsfromatwelfth-centurychurchinWíslica,Poland, Stud.Conserv.36(3)(1991)142–180.

[34]V.LaSpina,C.Mileto,F.Vegas,ThehistoricalrenderingsofValencia(Spain):an experimentalstudy,J.Cult.Herit.145(2013)544–551.

[35]J.M.García-Ruíz,J.L.Amorós,Sobrelaprecipitaciónpolimórficadelcarbonato cálcico,Stud.Geol.36(1980)193–200.