CORR: 00172 - Elsevier...ELSEVIE R PROOF Glossary g0005 Conservation Preservation of cultural...

ELSE

VIER

THIRDPR

OOF

4.43 Preservation of Metallic Cultural HeritageD. WatkinsonConservation Section, School of History and Archaeology, Cardiff University, Cardiff CF10 3EU, UK

� 2010 Elsevier B.V. All rights reserved.

4.43.1 Introduction to Conservation of Metals 3308

4.43.1.1 Cultural Heritage Context 3308

4.43.1.2 Conservation: Definitions and Rationale 3309

4.43.1.2.1 Conservation rationale 3309

4.43.1.2.2 Preservation goals 3309

4.43.1.2.3 Standards in conservation 3309

4.43.1.2.4 Ethics 3310

4.43.2 Corrosion 3310

4.43.2.1 The Influence of Corrosion on Conservation Strategies 3310

4.43.2.2 Corrosion of Archaeological Metals 3310

4.43.2.3 Corrosion of Historical and Modern Metals 3313

4.43.3 Environmental Control of Corrosion: Preventive Conservation 3314

4.43.3.1 Desiccation 3314

4.43.3.1.1 Relative humidity – Threshold corrosion values 3314

4.43.3.1.2 Practical humidity control 3315

4.43.3.2 Deoxygenation 3317

4.43.4 Interventive Treatments 3317

4.43.4.1 Removing Ions that Act as Corrosion Accelerators 3317

4.43.4.2 Washing Methods in Practice 3318

4.43.4.3 Removal of Corrosion Products 3320

4.43.4.4 Electrolytic Techniques 3321

4.43.4.5 Hydrogen Reduction 3323

4.43.5 Coatings 3324

4.43.5.1 Coating Rationale and Research 3324

4.43.5.2 Metal Surfaces and Patinas 3326

4.43.5.3 Preparing Surfaces on Cultural Objects to Receive Coatings 3327

4.43.5.4 Coatings in Conservation Practice 3328

4.43.5.4.1 Acrylic coatings 3328

4.43.5.4.2 Waxes 3330

4.43.5.4.3 Cellulose nitrate 3331

4.43.5.4.4 Silanes 3331

4.43.6 Inhibitors in Conservation 3332

4.43.6.1 Benzotriazole (BTA) 3332

4.43.6.2 Tannins 3334

4.43.6.3 Carboxylates 3334

4.43.7 Painted Metals 3335

4.43.7.1 Removing Paint 3335

4.43.7.2 Refinishing Painted Surfaces 3335

4.43.8 Overview 3337

References 3338

3307

CORR: 00172

ELSE

VIER

THIRDPR

OOF

Glossaryg0005 Conservation Preservation of cultural heritage.

g0010 Dense corrosion product layer (DPL) The

corrosion layer on archaeological iron

objects that approximates to the original

shape of the metal lost.

g0020 Interventive conservation Preservation strategy

that involves physical or chemical change to

an object.

g9000 Marker layer The layer of corrosion that

approximates to the original position of the

metal surface.

g0025 Preventive conservation Preservation strategy

that utilizes techniques that do not involve

physical or chemical change to an object.

AbbreviationsBTA Benzoriazole

CD Current density

CSS Confederate Southern States

DPL Dense corrosion product layer

EDX Energy dispersive X-ray analysis

EIS Electrochemical impedance spectroscopy

HMS Her Majesty’s Ship

RH Relative humidity

SEM Scanning electron microscope

XRD X-ray diffraction

SymbolsEcorr Corrosion potential (volts)

w/w Ratio of weights

s0005 4.43.1 Introduction to Conservationof Metals

s0010 4.43.1.1 Cultural Heritage Context

p0005 Cultural heritage is an industry that significantly con-tributes to the economy of nations via tourism, employ-ment, and taxation. Whole communities rely on it fortheir economic welfare. Cultural objects and structuresdocument developments and events within a society,offering a bridge to the past and providing importantmarkers for individuals, religions, and cultures. To con-tinue their contribution to the welfare, stability, devel-opment, and understanding of societies, cultural objectsmust be actively preserved. Age, origin, and use mean

that historical, archaeological, industrial, and art objectsare often decayed, corroded, and prone to further dete-rioration. The diversity of contexts from which culturalheritage metals originate presents a significant preser-vation challenge. This chapter briefly defines conserva-tion and provides an insight into the goals and ethicalconstructs that underpin its decision-making rationale.This is followed by an overview of how corrosionprocesses relate to treatment design and assessment.Conservation strategies for metals are revealed usingselected examples of treatments to illustrate conserva-tion research and practice.

p0010Metals conservation involves the preservation ofcultural objects and conservators are its practitioners.Their goal is topreserveobjectswhile retainingevidenceof their cultural context and integrity. Ideally thereshould beminimum change to an object while achievingthis and restoration to the former state is rarely carriedout. Refinishing, replacement of parts and repairing ofmetal objects with new corrosion-resistant materials arenot normally options within the conservation process.While preservation appears to be a straightforwardmaterial science problem, involving elucidation of thestructure and corrosion of metals to develop conserva-tion procedures that prevent or control corrosion, it isconstrained by ethics, aesthetics, and cultural contextsthat may complicate, constrain, and ultimately directpreservation strategies.1 The goal of conservation isnormally the preservation of an object using minimumintervention, although in some instances argumentsfor partial or total restoration are valid and profes-sional guidelines aid rationalization here.1

p0015Research into the corrosion and conservation ofhistoric and archaeological metals is developingsignificantly due to the increasing contribution ofdedicated specialists and funding of large collaborativeresearch programs that have a specific focus2,3 and thechallenges of large complex objects. An influx of cor-rosion scientists has improved the methodology ofresearch and the understanding of conservation meth-ods by widespread introduction of electrochemistryinto corrosion and treatment studies.2–4 As conserva-tion is a blend of art and science, it utilizes data andpublications generated by many disciplines to deviseits preservation strategies. Literature sources withinconservation focus mainly on conference proceedingsand research project publications dedicated to specificthemes that provide discussion forums for developing,interpreting, and reporting metallic corrosion andthe conservation process. Metals conferences areorganized and published by national conservationgroups, museums, universities, and the International

3308 Corrosion Management

CORR: 00172

ELSE

VIER

THIRDPR

OOF

Council for Museums Committee for ConservationMetals Working Group (ICOM-CC Metals WG).The Triennial Conference Proceedings of theICOM-CC Metals WG provide the main dissemina-tion route for metals corrosion and treatment research.This is supplemented by a web-based newsletterBulletin of Research on Metals Conservation (BRO-MEC).5 There are few textbooks that either specifi-cally address metals conservation or contain significantsections dedicated to it.6–8 A number of conservationjournals report metals conservation and amongst theseare Studies in Conservation and Reviews in Conservationpublished by the International Institute of Conser-vation; Journal of the American Institute for Conservation;The Conservator (Journal of the Institute of Conserva-tion); and the German Arbeitsblater fur Restauraten.

s0015 4.43.1.2 Conservation: Definitions andRationale

s0020 4.43.1.2.1 Conservation rationalep0020 Conservation of metals can be broadly split into inter-

ventive methods, which involve adding or removingsomething from an object in order to preserve it, andpreventive techniques that aim to prevent corrosion bycontrolling the environment.1,9 Both approaches requirean understanding of the physical and chemical structureof metals, their interactionwith environmental variablesand the properties of their corrosion products. Thisinformation is used to design safe storage environmentsand to develop, implement, and assess interventive treat-ments. Conservation does not have its own universallydefined preservation standards and generally workstowards goals generated to satisfy prevailing situations.

s0025 4.43.1.2.2 Preservation goalsp0025 The ultimate conservation goal of indefinite preserva-

tion has driven conservation rationale for many years.Stabilization is a much used word in conservationliterature and a reassessment of its meaning and con-text are long overdue. Since few metals are inherentlystable in ambient conditions, corrosion prevention formetals and alloys requires some degree of environ-mental control.9 Depending on the degree of controlrequired, significant capital outlay and ongoing energyconsumption may be called for. This will carry acarbon footprint and may be beyond the financialmeans of many museums. Qualitative and quantitativecontrol of environmental parameters, backed by rele-vant research to identify the environmental needs ofthe metals in question, underpins the concept of pre-ventive conservation. Unstable metals are merely pre-vented from corroding using environmental control

and they remain inherently unstable. Since they havenot been stabilized by an interventive treatment, anychanges to the environment could once again favortheir decay. This minimalist approach to preservationis now much favored in conservation practice. Previ-ously, environmental control was normally employedas the end point of an interventive process focused onenhancing the chemical and physical stability of anobject by attempts to remove corrosion acceleratorsand apply coatings and inhibitors. Interventive treat-ments should only be used where there is evidencethat they enhance the stability and lifespan of an objectwithout contravening ethical guidelines.1

p0030Preservation strategies should be designed to sup-port the role and function of the object in a society.This may involve public accessibility, which willrequire provision of oxygen, light, and heat. All ofthese are agencies of decay that can contribute tocorrosion. Access to metal objects may mean it isnecessary to accept the concept of limited lifespan,although this will depend on the metal in questionand its condition. Alternatively, never viewing anunstable metal object while knowing it is preservedin deoxygenated storage may be an acceptable pres-ervation strategy if the mere knowledge of its exis-tence is sufficient for it to fulfill its cultural role. Thismay be the case for iconic objects that form culturalcornerstones within societies and history timelines.

s00304.43.1.2.3 Standards in conservationp0035Few defined preservation and test standards exist

within conservation at the time of writing (2008).The concept of indefinite preservation still pervadesconservation thinking for metals. A move towards arationale that pragmaticallymeasures conservation out-comes as a function of resources, treatment success,cost, and object context is required to facilitate thedefining goals that can be used to create standards. Toproduce attainable standards requires research that pro-vides quantitative evidence of the success of treatment.Within metals conservation there is currently limitedquantitative data that could be translated into hardcurrencies of object longevity, stability, and survival.

p0040Standards grow from attainable and reproduciblequantified outcomes. Gaps in our understanding ofcorrosion processes, treatment mechanisms, and treat-ment success hinder the production of conservationstandards. There is a need for more research thatdefinitively identifies corrosion mechanisms, quantifiestreatments and measures their success relative todefined goals. International agreement on what consti-tutes realistic treatment success is required, as is a

Preservation of Metallic Cultural Heritage 3309

CORR: 00172

ELSE

VIER

THIRDPR

OOF

recognized body that defines or approves standardswith the support of the conservation profession. Thus,if a standard does not prevent decay there must be anindication as to the extent to which it protects fromcorrosion or prolongs lifespan. This requires a defini-tion of lifespan reliant upon identifying the function ofa cultural object and defining when it ceases to fulfillthis function. These are difficult questions but untilthey are answered conservation cannot develop modelsthat quantitatively define goals and offer a frameworkof standards to achieve them. There is a case for theconservation profession to devise its own standards foruse in research and quality assurance, as ISO andASTM standards are often inappropriate for use inconservation research and practice.

s0035 4.43.1.2.4 Ethicsp0045 Ethics both guide and constrain conservation practice.

National guidelines define generic rules for goodpractice that include minimum intervention withobjects, reversibility of conservation materials and oftreatments applied to cultural objects, as well as reten-tion of the cultural integrity of an object and theinformation it contains.1 Situation ethics prevail whendetermining the level of intervention required topreserve objects. Within archaeological conservation,significant intervention is normally essential forunderstanding the technology and function of ametal object or for revealing its shape from a cor-roded metal conglomerate. Ethical dilemmas aboundwhen devising treatment procedures. While a blister-ing paint surface on a working engine comprising castiron and steel might be removed and the surfacerepainted to approved international standards, acces-sioning such an object into a museum changes this andthe paint now constitutes part of the object history. Anethical preservation process for such an object mightinvolve the local stabilization treatment of pits in thepaint layer, followed by application of an unobtrusiveprotective coating and storage in a controlled relativehumidity. This is more expensive to execute and offersa less predictive corrosion control strategy than strip-ping the metal to an SA standard and repainting.

s0040 4.43.2 Corrosion

s0045 4.43.2.1 The Influence of Corrosion onConservation Strategies

p0050 The nature and extent of corrosion strongly influencesconservation strategies. Understanding how metalscorrode within specified environments such as mus-eum interiors, enclosed storage boxes and the

atmosphere is central to developing treatments andidentifying realistic conservation goals. It may be nec-essary to understand more than one corrosion model todevise conservation procedures. For instance, under-standing how archaeological metals corrode duringburial is essential for developing a model of theircorrosion in the atmosphere following excavation.Equally, elucidating the corrosion mechanisms on ahistorical pewter salt cellar facilitates an understandingof how this object corrodes in a museum environment.It may no longer contain salt, but the legacy of itsoriginal function influences its future survival. Corro-sion models also contribute to understanding objectaesthetics and developing conservation ethics. Researchinto corrosion mechanisms informs conservators aboutthe stability of objects, their rate of decay and theirconservation needs.2

s00504.43.2.2 Corrosion of ArchaeologicalMetals

p0055Although reviews of archaeological and historicalmetal corrosion exist,10 aspects of corrosion relevantto the treatments discussed here are briefly outlined.Archaeological metals normally corrode extensivelyduring burial to form corrosion products related totheir burial environment and alloy composition.2,6,7,9,10

Environment, alloy composition and the mechanismsof ionic and electron transfer during corrosion dictatethe nature of the corrosion layer formed and whethercorrosion products are passive or unstable. Smoothdense patinas or voluminous disfiguring corrosionlayers may form. Some corrosion layers can retaingood approximation of the shape of the object andsurface details like wear marks or engraving2,6,7,9,10

(see Figure 1). These layers have been loosely ref-erred to as representing the original surface of theobject2,7 but are now more often called marker layers.2

They are often hard and compact and may be overlaidwith softer more voluminous corrosion.

p0060Conservation is not just about preservation.Archaeological metals are subjected to investigativeconservation that involves analysis designed to answerquestions on technology, provenance, and trade.1,7

Mechanical cleaning employing hand tools like scalpelsand micro air abrasion techniques, which utilize pow-ders such as aluminum oxide, are often employed toremove overlying corrosion to reveal the marker layer.Guidance in this process is helped by X-radiography.By detecting differences in density juxtaposition ofdiffering corrosion products are shown, along withdissimilar metals, technological detail and evidence


CORR: 00172

ELSE

VIER

THIRDPR

OOF

of residual metal7 (see Figure 2). This information-sensitive marker layer may be physically weak andinfused with soluble corrosion accelerators, but it mustremain undamaged by any treatment, whether it has

been revealed or is left coveredwith corrosion products.Chemical treatments that strip off all corrosion pro-ducts to remove the soluble corrosion accelerators theycontain cannot be used to treat archaeological metals, asthe remaining metal rarely represents the original shapeof the object7 (see Figure 3).

p0065Post-excavation corrosion processes usually differfrom those occurring in the burial environment, asthe atmosphere contains more oxygen and is nor-mally less damp than soil.2,10 To develop a preserva-tion plan it is essential to understand how thecorrosion model changes, especially how corrosionproducts form in the burial environment, respond tothe new environment and influence the metal theyare associated with.

p0070The instability of archaeological iron offers a clearexample of how an understanding of corrosion mec-hanisms is essential to treatment design. It is usedhere to illustrate conservation thought processes formetals containing soluble corrosion accelerators. Dur-ing burial, chloride is attracted to anodes as a counterion to balance charge from Fe2+.2,10 In damp burialconditions it exists as a Fe2+/Cl� solution at the metalsurface. Chloride content may be typically 0.1–1.5wt% of object, but can be significantly greater for marinemetals7,10–12 Soluble chloride acts as an electrolytemaking the iron inherently unstable. Overlying thecorroding metal core is a dense corrosion productlayer (DPL) identified as comprising a goethite (a-FeOOH) or sometimes a siderite (FeCO3) matrix inwhich are embedded magnetite (Fe3O4) or maghemite(g-Fe2O3) strips

2 (see Figure 3). This is effectively themarker layer for the surface of the object and is alsopresent in totally mineralized objects, which present noongoing corrosion problem, but remain a conservationproblem due to their fragility and the need to revealtheir original shape.

p0075Moisture and oxygen in the post-excavation envi-ronment facilitates the formation of b-FeOOH

(a)

2mm

(b)

Figure 1f0005 Corrosion products can sometimes retain

significant surface detail. Lead and tin corrosion products

record an assay mark in an archaeological pewter object(a) and a Roman copper alloy stud retains wear mark

scratches in its fine smooth patina (b).

Figure 2f0010 X-radiography records the presence of a dissimilar metal inlay (silver) hidden under corrosion of uneven density on

this Roman Scabbard. The density distribution on the radiograph indicates that no metallic iron remains.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

(akaganeite) in the acidic chloride rich solutionsited at anodes on the metal surface. Chloride isincorporated into tunnels within the hollandite-typeb-FeOOH crystal structure and adsorbed onto itssurface.2,10 Growth of these voluminous tower-likeb-FeOOH crystals pushes off the overlying DPL andis catastrophic for the object as an information resource,as it destroys evidence of the original surface and shapeleaving only an unevenly corroded metal core (seeFigure 4). While conservation strategies aim to revealthe shape of the object they must also prevent thecorrosion of metallic iron, as the consequent formationof b-FeOOH would remove the information retainingDPL. Totally mineralized iron is stable, as there is nomore metal to oxidize and this has allowed solubleanodically held chloride to diffuse out of the object.

p0080 Corrosion of nonferrous archaeological metalspresents different conservation problems. Most arealloyed with other metals as; bronze (copper/tinsometimes with added lead); brass (copper/zinc);

pewter (lead/tin); and precious metals (silver/gold,silver/copper, gold/copper). This facilitates prefer-ential corrosion.7,10 Thus, gold may lose the copper itis alloyed with and be covered in copper corrosion

(a)2 mm 1 mm

(b)

1 mm

(c) (d)

Figure 3f0015 A cross section of a medieval nail shows it retains a substantial metal core and differing corrosion layers (a).

Images (b) and (c) show how the dense corrosion product layer (DPL) retains the shape of the nail, which is overlaid with acorrosion matrix containing soil. Stripping the nail of all corrosion would lose the shape of the object and its surface detail.

Totally mineralized iron retains the shape of the object as a corrosion layer, visible in this Roman nail (d).

Figure 4 f0020Corrosion at the metal/DPL interface results in

loss of overlying corrosion products and the information

they contain.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

products. Chemically removing these copper corro-sion products leaves a soft porous copper depletedlayer of gold on the surface of the alloy. Mechanicalremoval of corrosion leaves copper corrosion pro-ducts embedded in the metal surface alongside thegold, but produces a stronger surface.

p0085 Corrosion processes, products, and rates can varyaccording to the proportions of metal within alloys andthe burial conditions. This can prove advantageous,although bronzes often decuprify the corrosion ofalloyed tin, producing an insoluble and poorly mobiletin oxide to act as a marker layer for the original surfaceof the metal.2 Lead forms very stable lead carbonates inneutral to alkaline burial conditions and insoluble chlo-ride compounds that are stable following excavation.9

Silver is also stable after excavation due to the highlyinsoluble silver chloride and silver sulfide formed on itssurface.10,13 Lead and tin have similar corrosion poten-tials and both corrode in pewter to produce a lamellarwart-infested surface with powdery corrosion productsthat are unlikely to form marker layers. Copper thatforms sparingly soluble cuprous chloride (CuCl) nextto the metal surface can be unstable following excava-tion if there is sufficient water to hydrolyze it to volu-minous Cu2OH3Cl known as ‘bronze disease.’2,6,7,8,10

Conservation procedures will be dictated by the stabil-ity of the corroded metals, corrosion mechanisms, andthe presence or absence of marker layers.

s0055 4.43.2.3 Corrosion of Historical andModern Metals

p0090 Historicalmetals are found in outdoor and indoor envir-onments and may be decorative objects d’art, functionalartifacts, or large industrial items. Metals kept indoorssince their fabricationmay have very little corrosion andretain their original surface finish. Conservation aimsto retain such surfaces. This can be achieved by moni-toring environments, simple good housekeeping andenvironmental control in relation to corrosion thresh-old values for the metals in question.1,9,14

p0095 The context of the object is important and thenature of a corrosion patina (see Section 4.43.5.2)may be seen as an important aesthetic value of theobject or part of its life history. This will dictatewhether it is removed or retained during conserva-tion and heavily influences the choice and success oftreatments. Analysis aids decision making in conser-vation. Sometimes it may be necessary to replacemetals if they are performing a structural function.Analysis detected dezincification of brass (70:30) innails used to join the planks in the hull of the yacht

Asgard (see Figure 5). Built in 1905 by Colin Archerthe renowned Norwegian shipwright for Robert Ers-kine Chalders the yacht is now an important part ofIrish history. The analysis provided information thataided conservation decisions for nails that otherwiseappeared to have only surface corrosion.

p0100Modern metals increasingly occur in museum col-lections and are the fabrication materials of manycultural icons and monuments. These include zinc,aluminum, magnesium, and their alloys.15 Conserva-tion of protective and technologically informative sur-face finishes on metals like aluminum may be equallyas important as preserving the metal itself.16 Air qual-ity, relative humidity, and temperature are importantfor metal stability within museums. The Oddy test17

was devised to determine if materials used to constructboxes and showcases give off gas pollutants that willcause metals within them to corrode over long-timeperiods. It can also be applied to show if metals and

(a)

500 µm

(b)

2 mm

Figure 5 f0025SEM/EDX mapping of zinc brass hull nails from

the yacht Asgard revealed surface dezincification (a) which

is not visible on the nail (b). This analysis contributed todesigning the yacht’s conservation plan carried out at the

National Museum of Ireland. (SEM scale bar 500 mm).


CORR: 00172

ELSE

VIER

THIRDPR

OOF

alloys corrode in polluted environments, thus alumi-num and Duralmin (Al/Cu) corrode in the presence offormic acid and magnesium and AZ31 (Mg/Al/Mn/Zn alloy) corrode in the presence of a wide range ofmaterials that include oak, plywood, formic acid andwool, which may all be found in museum environ-ments.15 Lead and copper are also susceptible to cor-rosion from carbonyl pollutants.2,9,10,18 Equally studiesmust examine what occurs when objects are handled,as chloride corrosion products on zinc coins and IndianBidri ware (zinc alloy) in museum collections wereattributed to handling.15 The influence of handlingon corrosion is evident for tarnishing of silver by sulfurcompounds, where fingerprints can act as a focus forcorrosion (Figure 6). Conservation has moved someway towards developing standards for showcasedesign and the use of corrosion-safe constructionmaterials, but much more remains to be done.

p0105 Outdoor corrosion patterns on metals are predom-inantly influenced by alloy composition, environment,climate, and the object morphology.19 Out of doorsiron produces a corrosion layer mostly comprisedof loose a FeOOH rather than form the shape retain-ing DPL that occurs on archaeological objects2 (seeFigure 3). Sulfurous gases can produce a FeSO4 cor-rosion cycle that is humidity sensitive. While conser-vation of iron that has corroded in the atmosphere mayinvolve removal of corrosion layers, rather than theirretention, outdoor copper alloys may develop patinasconsidered to be aesthetically desirable and conserva-tion planning takes account of this2,6 (see Section4.43.5.2). Humid air containing SO2 will attack copperalloys dissolving passive layers and supporting electro-chemical corrosion of the alloy,19 which is problematic

for statuary and architectural fittings. Establishing theprecise role of patinas and corrosion product layers inthe corrosion process is currently underway for manymetals,2 which will facilitate better long-term planningand improve treatment design.

s00604.43.3 Environmental Control ofCorrosion: Preventive Conservation

s00654.43.3.1 Desiccation

p0110A common preservation strategy for cultural metals isto prevent their electrolytic corrosion by controllingone of the agencies essential for this process to occur.Normally, removal of moisture is used to prevent elec-trolyte formation, although in some instances theoxygen necessary to support cathodic processes andform corrosion products is removed. Removal of elec-trolyte ions from corroded metals involves interventivetreatment. It is easier, cheaper, and more user friendlyto control corrosion by desiccation than to removeoxygen. The degree of desiccation necessary to preventcorrosion differs according to the corrosion mechanismof the particular metal or alloy and dictates the cost ofcorrosion control as capital outlay on plant, fuel, andpersonnel. Despite the importance of understandingcorrosion mechanisms for optimizing preventive con-servation strategy, only chloride infested iron has beenextensively researched to identify relative humidity(RH) no-corrosion thresholds for reactions within itscorrosion mechanism.

s00704.43.3.1.1 Relative humidity – Threshold

corrosion valuesp0115Ambient humidity can supply moisture to solvate ions

within corroded metals and create electrolytes. Con-servation of metals examines corrosion as a function ofRH rather than absolute humidity.6,8,10,14 It is wellknown that hygroscopic or deliquescent salts such assodium chloride can lower the RH threshold for cor-rosion to occur.10 The RH at which the formation ofelectrolytes begins is the threshold corrosion value.20

On corroded iron the surface adsorbed chloride onhygroscopic b-FeOOH becomes mobile within atmo-spheric moisture and corrodes iron in contact with it at15% RH20 (see Figure 7). Additionally, ferrous chlo-ride may form in the low pH and high Fe2+/Cl�

environment created as damp chloride infested iron,dries10 (see Figure 8). Any FeCl2�4H2O that formscontains enough water to corrode iron in contactwith it, but once FeCl2�2H2O forms at 20%, RHcorrosion ceases20 (see Figure 7). Storage standards

Figure 6f0030 The influence of a fingerprint on the pattern of

sulfur tarnishing on this silver surface is clearly visible.

A protective coating on the surface would haveprevented this.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

for chloride infested iron at 20 �C have been definedas: at 12% RH no corrosion; 15–20% RH b-FeOOHcauses slow corrosion of iron; above 21% RH both b-FeOOH and FeCl2�4H2O contribute to corrosion20

(see Figure 7). Corrosion increases significantlyabove 25%19–21 and several times faster above 30%RH than at 25%.20,21

p0120 In contrast controlled storage of corroded ironfrom historical contexts has received scant attentionin conservation research, as it will tolerate higherhumidity values before incurring corrosion if it ischloride free. No conservation studies have examinedhow iron contaminated with sulfates from SO2 pollu-tion would respond to RH.

p0125 Corrosion of archaeological copper alloys canoccur when cuprous chloride (CuCl) close to themetal surface hydrolyses to form basic cupric chloride(Cu2OH3Cl) polymorphs, usually clinoatacamite oratacamite with an accompanying molar volume in-crease.6 This pressures or undermines the overlyinghard and dense corrosion layers that predominantlycomprise malachite (CuCO3�Cu(OH)2), breaking theircontact with the CuCl/Cu2O mixture next to themetal and ultimately causing their exfoliation. Scott6

observed no reaction over the relative humidity range42–46% but rapid corrosion at 70%, leading him torecommend storage below 45% relative humidity in

contrast to the anecdotal 39% commonly cited inconservation. The CuCl/Cu corrosion model meritscloser scrutiny to define both the critical no-corrosionhumidity and corrosion rates above this value.

p0130Unless it has been already corroded by aggressiveions such as chloride and sulfate, aluminum offersminimal conservation challenges provided it is storedat around 55% relative humidity.16 Reaction of chlo-ride infested aluminum alloys to relative humidity hasnot been examined in relation to safe storage thresh-olds. Similarly, the role of soluble tin chlorides in thecorrosion of tin and tin alloys merits detailed study,since tin chloride solutions are prone to hydrolysis.

s00754.43.3.1.2 Practical humidity controlp0135Humidity control either employs active dehumidifica-

tion using mechanical plant or passive desiccationusing desiccants like silica gel to control environmentin fixed storage or display areas.9,14,21–23 A significantadvantage of controlling corrosion of cultural metals bydesiccation is that they remain easily accessible, espe-cially if an entire store room is desiccated as opposed toindividual storage boxes. Although maintaining lowhumidity using mechanical desiccation plant demandsfinance and energy, it is capable of controlling largespaces that will require only one or two telemetricsensors to monitor relative humidity. These can imme-diately and remotely detect any failure to maintaintarget relative humidity on a macroscale, which allowsfor rapid remedial response.Thusmechanical corrosioncontrol is a low risk corrosion control strategy forcultural metals, provided resources exist to underpinits continuing operation and management.

p0140In contrast the small scale control of climate by silicagel for storage within enclosed polyethylene boxes andfor display in sealed showcases appears to offer acheap conservation strategy for small metal objects.Gradual hydration of the desiccated silica gel willoccur as a function of air exchange rate, external relativehumidity and initial moisture content of the metalsbeing desiccated.21–24 This eventually raises internalhumidity within the controlled space to values thatsupport corrosion, which will speed up as internalrelative humidity continues to rise until it equilibrateswith the external environment. Themetals now rest in acontinually aggressive environment. Determining whento replace the silica gel relies upon knowing the thresh-old corrosion value of the metals in question and detect-ing when the interior of the box exceeds this value.

p0145Research to evaluate the complex equation ofstorage, hardware performance, air exchange, moni-toring, maintenance time, and human error is needed

4.064.044.02

43.983.963.943.92

0 2000 4000 6000Time (min)

Bal

ance

rea

din

g (g

)35%

27.5%22%19%

(a)

25%

21%16% 15%

12%

4.05

4

3.95

3.9

3.850 4000 8000 12 000 16 000

Time (min)

Bal

ance

rea

din

g (g

)

(b)

Figure 7f0035 Relative humidity influences the corrosion of iron

powder mixed with FeCl2�2H2O (image 1) and bFeOOH

(image 2); with no-corrosion points at low humidity andsignificant corrosion as 30% relative humidity is

approached, as shown by weight gain due to oxidation.

Initial weight loss (b) is due to dehydration of bFeOOH as it

equilibrates with its environment.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

to offer a realistic and practical assessment of thiscorrosion control strategy. Enclosure design largelycontrols air exchange rate which is the governingfactor in passive desiccation systems.24 Assessmentof enclosure performance can produce predictivelife spans for specified passive desiccation storagesystems.21,24,26 Showcases designed for desiccationmaintained relative humidity below 20% for 1 yearwithout a change of their silica gel desiccant, whenexternal relative humidity fluctuated between 60%and 95%.22,24 As telemetric monitoring of showcase

performance is easy to initiate and financially viable,this system offers potentially reliable low energyconsumption corrosion control for unstable metalson display. The major cost will be an initial outlayto meet the showcase specification. Further workcould establish performance of the showcase withage and use. This requires real-time testing thatmay take several years. Attempting to define long-term effectiveness by extrapolating a one year studyor short-term laboratory results cannot provide adefinitive guide to performance.

(a) (b)

(c)

Figure 8f0040 Brunel’s iconic SS Great Britain was the first ocean going screw driven iron hulled passenger liner and is now

preserved by a mechanical desiccation system that maintains relative humidity to 20% and allows visitors to walk aroundthe hull. Reproduced from Watkinson, D.; Tanner, M.; Turner, R.; Lewis, M. The Conservator 2005, 29, 73–86.25


CORR: 00172

ELSE

VIER

THIRDPR

OOF

p0150 The performance of polyethylene storage boxes asdesiccated depositories for archaeological iron hasbeen studied as functions of manufacturer, age, use,silica gel performance, and ambient relative humidity.21

The study confirmed earlier studies on threshold cor-rosion values20 and determined most air leakage wasvia lid seals, making storage protocols important inrelation to the stacking of boxes compromising theirseals. Storage in polythene boxes suffers from a relianceon the limited accuracy of moisture sensitive indicatorsto detect the internal relative humidity of boxes for gelregeneration cycles. Individual telemetric monitoringof the interior of the box is financially prohibitivefor the several hundred polyethylene boxes normallyhoused in a storeroom. Good management is also nec-essary if the gel is to be changed as required. It isessentially a low cost, but high risk corrosion controlstrategy when the frailties of human nature are takeninto account.

p0155 Relating corrosion rate to increasing relativehumidity provides opportunity for pragmatic man-agement decisions.25 The degree of corrosion controlimplemented can be matched to resources availableto produce a package that will reduce, but not pre-vent corrosion. This reasoning was used to controlcorrosion of Brunel’s famous 1843 wrought ironsteamship SS Great Britain, whose chloride-riddenlower hull is preserved in a mechanically desiccateddry dock maintained at 20% relative humidity with agoal of retaining its structural integrity for at least100 years25 (see Figure 8). This allowed public accessto the hull and avoided employing costly andenvironmentally unfriendly alkaline washing meth-ods (see Section 4.43.4.2) in attempts to removechloride corrosion accelerators from the 324 ft longhull. The conservation challenge required a scientificinvestigation to identify the influence of RH on cor-rosion, followed by an engineering project to designan envelope to maintain the operating RH.

s0080 4.43.3.2 Deoxygenation

p0160 Ageless™ is a commercial oxygen absorber that hasbeen used in conservation to deoxygenate enclosedspaces.26 It is expensive to use and the exothermicreaction that expends oxygen produces water, whichshould offer no corrosion risk provided the storageenvironment remains oxygen free. Despite technicalchallenges, high cost, and reduced accessibility, archae-ological iron has been stored in heat sealed low lineardensity polyethylene tube, with barrier properties 2000times better than polypropylene, deoxygenated with

oxygen scavenging RP-A™ (mordenite/CaO/PE/activated carbon).27 This study showed that ironobjects, which had been pretreated in aqueous NaOHto wash out chloride (see Section 4.43.4.2), remainedvisibly unchanged over a 5-month period but untreatediron had a 12% failure rate attributed to poor seals andbag punctures. Although the researchers predict a stor-age lifespan of 4–6 years, the high initial cost of materi-als and labor and a one-shot storage system make itunlikely that it will displace desiccation in popularity.A long-term quantitative testing program comparingits effectiveness, ease of use and global cost with desic-cation techniques would define the relative merits ofthese two treatments. Using a flow of nitrogen gasthrough an enclosed container offers a no-corrosionenvironment for damp high status archaeologicalobjects prior to their conservation.

s00854.43.4 Interventive Treatments

s00904.43.4.1 Removing Ions that Act asCorrosion Accelerators

p0165Interventive treatments aimed at reducing orpreventing corrosion of unstable metals mostly focuson removing chloride and other soluble ions that arecapable of acting as electrolytes in high humidityenvironments. Entirely removing soluble ions shouldprevent corrosion, while their partial removal enhancesthe stability of the object by reducing electrolyte avail-ability. Additionally, some washing treatments may aimto remove or modify sparingly soluble or insolublecorrosion products that promote corrosion.11,28

p0170From an ethical standpoint, merely enhancing sta-bility without controlling environment to prevent cor-rosion may not be considered an appropriate corrosioncontrol strategy. Even a small amount of corrosionoccurring at anode sites on the metal can interferewith adhesion of important overlying corrosion layers,whose loss destroys the object as an informationresource despite the survival of the metal core (seeFigures 2–4). Delaying this for a short time by reduc-ing corrosion rate could be viewed as a very limitedgoal, when measured against the ideal of indefinitepreservation. Equally, treatments that partially removecorrosion accelerators could be seen as unnecessary,since post treatment storage must be similar to thatprovided for untreated metals. Why bother trying toremove electrolytes if the objects still need to be sub-jected to the same stringent storage conditions? Onepossible advantage is that failure to maintain the con-trolled environment will lead to a slower corrosion


CORR: 00172

ELSE

VIER

THIRDPR

OOF

rate, as compared to untreated iron objects. Quantita-tive understanding of treatment efficiency and effec-tiveness is essential for facilitating informed decisionsabout their use.

p0175 To optimize treatments designed to remove corro-sion accelerators that are normally predominantlycomprised of chlorides, it is necessary to: identify thecorrosion mechanism; devise appropriate chlorideremoval procedures; quantify the removal of chlorideas an absolute value of chloride contained in the metal;use this value to calculate extraction efficiency; deter-mine the extent to which the metal is stabilised.11

While the concept of removing soluble ions to preventelectrochemical corrosion is logical, no studies providequantitative assessment to show how stable metals likeiron become following attempts to remove their solu-ble chloride. Equally, there is limited information as towhat percentage of chloride present within objects isextracted by a particular treatment and how consis-tently this occurs.11,12,29

s0095 4.43.4.2 Washing Methods in Practice

p0180 Washing methods are applied to metals to solvatesoluble corrosion products and soluble ions adsorbedonto corrosion products. Iron is the most unstablearchaeological and historical metal, due to its abilityto attract chloride and retain it in a highly solubleform.10 Washing methods will solvate chloride ionsand this allows them to diffuse out of the object.7,11

This process is hampered by the DPL limiting solu-tion access to the iron surface, where wash solutionsmust enter pits and micro cracks to exchange withchlorides at anode sites. Chloride removal is diffusioncontrolled, slow, and is unlikely to be 100% efficient,although efficiency can be improved using chemicaladditives.11,12,29 Evidence from a qualitative study ofthe long-term stability of treated archaeological ironrevealed simple washing with hot deionized water tobe a poor stabilization system, with 68% of treatedobjects recorroding.30 Recent study quantitativelyconfirms washing in deionized water is an unpredict-able treatment that never extracts large amounts ofchloride11 (see Figure 9).

p0185 Their ease of use and low aggression to iron hasmeant that inhibited wash solutions have been chosento treat large objects. HMS Holland was the first sub-marine used by the British Navy. It was fortunate thatshe lost her tow en route to the breakers yard and sank,as when she was discovered and salvaged many yearslater she had become an important historical object.Her 19m length offered challenges to removing the

chloride that infused her structure. Health and safetyand cost augmented against the use of strongly alkalinewash treatments (see below) and electrolytic desalina-tion (see Section 4.43.4.3). A purpose built treatmenttank containing 820 000 dm3 of Na2CO3 was used toprovide an inhibited wash relying on diffusion toextract chloride that was continually monitored.31

Treatment was followed by storage in a building con-structed for this purpose maintained at 35% RH toslow down any corrosion from residual chloride. Whilenew study indicates storage at 25% RH would slowiron corrosion significantly more than 35% RH,11,20,21

any environmental control program must be related tothe finance available to sustain it and 35% RH willrestrict corrosion better than higher values. This is atypical finance/corrosion control cost benefit assess-ment that is a factor in the conservation equation. It isoften used to influence selection of treatment optionsand should be considered when developing standardsof treatment. Equally object size may dictate whichtreatments are feasible, as in the case of the Holland.

p0190Alkaline wash systems act as inhibitors and opti-mize chloride removal.8,11,12,29 Inhibition providedby alkalis such as NaOH reduces corrosion duringtreatment. This releases chlorides from their chargebalancing role at anodes and allows them to diffuseout into the wash solution, while at the same time thealkali supplies OH� ions to replace them and associ-ate with Fe2+. Solvation of solid ferrous chloride andsurface adsorbed chloride on b-FeOOH occurs,

1 2 3 4 5

Treatment

0

20

40

60

80

100

Chl

orid

e ex

trac

ted

Maximum

AverageMinimum

Figure 9 f0045Influence of treatment solution on percentage

chloride removal from Roman and Medieval archaeological

iron nails. For each treatment method columns show: best

extraction; average extracted; worst extraction. Totalchloride in nails determined by post treatment digestion.

Treatment methods: (1) NaOH/Na2SO3 0.5M aq. (17 nails);

(2) NaOH 0.5M/deoxygenated with nitrogen (10 nails);

(3) NaOH0.5Maq. (16 nails); (4)Water/deaeratedwith nitrogen(10 nails); (5) Water (10 nails). Reproduced fromWatkinson, D.;

Al Zahrani, A. The Conservator 2008, 31, 75–86.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

reducing b-FeOOH hygroscopicity and preventingpost-treatment corrosion of iron in contact with it.32

In strong alkali b-FeOOH can transform and this willadvantageously release chloride into the wash solu-tion.28 Since the ionic size of chloride occluded intunnels within b-FeOOH crystals exceeds the diame-ter of tunnel entrances it remains trapped and presentsno post-treatment corrosion threat, unless conversionof b-FeOOH to a-FeOOH releases this chloride.Post-treatment hydrolysis of residual b-FeOOH toa-FeOOH can occur but is little studied. A 23-year-old dry sample of b-FeOOH was assayed as being 1%a-FeOOH and 99% b-FeOOH.32 A major study cur-rently underway aims to identify physicochemicaltransformations of corrosion layers as a function oftreatment regimes, using microanalysis techniques tocharacterize the corrosion products.33 This will signif-icantly contribute to an understanding of treatmentand post-treatment corrosion.

p0195 Further improvements in chloride extraction canbe produced by deoxygenated NaOH solutions11 (seeFigure 9). Passivation of iron in oxygenated NaOHsolutions is likely to be incomplete, as NaOH needs toreach all anodes sited at the metal surface to be effec-tive. Consequently, chloride continues to be held as acounter ion and is difficult to wash out. DeoxygenatingNaOH solutions with an SO3

2� oxygen scavenger ornitrogen gas stops corrosion and improves chlorideextraction.11 Anecdotal reporting indicates undesir-able softening of corrosion products in NaOH solu-tion,30,34 but no studies offer quantitative evidence ofthis or address the influence of any chemical treatmentresidues on iron corrosion products. This merits furtherattention. Objects should be washed post-treatment indeoxygenated water to flush out chemicals.

p0200 A promising variation on alkaline washing currentlyunder investigation is the use of a subcritical high pres-sure alkali treatment for treating the American CivilWar submarine Hunley, which was recovered from amarine environment where she had lain since she sankimmediately after becoming the first submarine to sinka warship.12,29 Treatment transforms corrosion promot-ing b-FeOOH to the a-FeOOH, Fe3O4, and Fe2O3

chloride-free phases that are nonaggressive to iron.p0205 Other inhibitors have been tried, notably aqueous

washing of archaeological iron with ethylene diamine.Inhibition and a high solution pH (11.5 at 5% (v/v) aq.)aim to prevent corrosion during treatment and aidchloride removal. It is an inferior chloride extractoras compared to 2%NaOH aq.,34 but has been cited asoffering good long-term stability to iron in a qualita-tive study of treated objects.30,34

p0210There is limited information available regardingthe amount of chloride removed by treatments as afunction of total chloride within the iron, since deter-mining residual chloride requires post-treatmentdigestion of objects. The small amount of quantifieddata available indicates that some wash methods aresignificantly better than others and consistently extractwithin fixed percentages of chloride.11,12,29 The limitof detection for measuring chloride extracted intowash baths cannot guarantee iron is chloride free.11,12

Although it is important to determine whether smallresidues of chloride are capable of causing significantcorrosion of iron, testing the susceptibility of washediron to corrosion has received little attention. Forguaranteed no-corrosion all washed iron should bestored in a controlled environment to a standardequaling that used for unwashed iron, but this makesstorage costs the same as for untreated iron.

p0215Studies that qualitatively examine the stability ofwashed and unwashed iron within museum collectionshave offered comparative assessment of treatmenteffectiveness.30,35,36 While these types of survey offer agenuine insight into treatment outcomes, they are lim-ited by their retrospective nature, wide range of un-controlled variables and a lack of quantified data.Computer monitoring of RH now offers an opportunityfor precise recording of storage environment and newstudies of long-term performance of treated objects areoverdue. It would also be worth interpreting past stud-ies using new data available on corrosion mechanisms.2

Until quantitative long-term post-treatment stabilitytesting is carried out and outcomes are linked to chlo-ride residues within objects, washing treatments remainempirical applications best described as having anunpredictable and unknown capacity to remove chlo-ride and enhance object stability. Such studies willprove hard to fund because of their time commitment.

p0220Aluminum alloys are becoming increasingly com-mon in museum exhibits and include aircraft and boatsrescued from extreme environments. Once the highlyprotective nonconducting Al2O3 layer on aluminumfails the metal will corrode. This may cause pittingor intergranular corrosion according to alloying andmetallography. Chlorides exacerbate corrosion byattacking the Al2O3 film and concentrating in pitswhere they cause low pH, hydrolysis of Al3+ andvoluminous disfiguring corrosion product at themouth of the pit.10 The shape of the object is obscuredand ultimately lost with eventual perforation of themetal. Washing to remove soluble chloride can becarried out on objects retaining significant amountsof metal, but those without large quantities of metal


CORR: 00172

ELSE

VIER

THIRDPR

OOF

will be beyond recovery as corrosion products are non-adherent and friable.

p0225 Chemically aided washing was innovatively usedto treat a Duralumin (3–5% copper) seaplane floatbelonging to a Junkers W33 infused with chloride,exhibiting extreme pitting and redeposited copper onits surface that could act as a cathode for continuingcorrosion of the aluminum.37 Washing conditionswere carefully manipulated using (NH4)2SO4/NH3

buffer solution to complex Cu2+ ions to Cu(NH3)42+

and E0 of the system was lowered to allow corrosionthat removed redeposited copper, provided oxygenwas present.37 Thermodynamic data revealed Al2O3

was stable in the treatment solution. Its presence onthe metal surface protected aluminum at the treat-ment pH of 9.6 and chloride diffused out of the objectfollowing hydrolysis reactions within pits. As withmost washing methods employed in conservation,assessment of the success of the treatment was quali-tative and reported that the object appeared to bestable 4months after treatment.

p0230 Washing has been employed for copper alloys todeal with the problem of CuCl hydrolyzing to formCu2(OH)3Cl polymorphs that damage and disfigurecopper alloy objects.6,10 Fortunately Cu2(OH)3Cl isnot hygroscopic and does not cause corrosion of metal-lic copper in contact with it, although its growth canundermine and lever off overlying patina. Not all cop-per alloys containing chloride are unstable. MacLeod38

argues that there is a minimum chloride ion concentra-tion required to support corrosion. The low solubility ofCuCl and its positioning within or under Cu2O meanschloride solvation by simple aqueous washing can lastyears. Washing for up to 2 years in sodium sesquicar-bonate (NaHCO3�Na2CO3) at 5% (w/v) pH 10 hasbeen employed to slowly solvate CuCl and supplyOH� to form cuprous oxide (Cu2O).

6,9 Basic coppercarbonate (CuCO3Cu(OH)2) and chalconatroniteNa2Cu2CO3�2H2O may also form and darkening ofpatina may make the treatment undesirable whenaesthetics are an important consideration. It is nowmore likely to be used as an inhibited wash solutionfollowing other treatments like electrolysis (see Section4.43.4.4).

s0100 4.43.4.3 Removal of Corrosion Products

p0235 Stripping techniques may be employed when ethicalarguments allow removal of corrosion, such as prepar-ing lightly corroded industrial and historical steelsto receive protective coatings. They may also beused to free archaeological objects from corrosion

conglomerates provided substantial amounts of metalremain, as heavily mineralized objects will lose theiridentity when corrosion is removed.8,39 While thereare several studies looking into the action of strippingagents there is no single definitive comparative assess-ment that identifies a preferred method for specifiedalloys. Conservators must evaluate studies or them-selves test stripping agents relative to the object theyare planning to treat. That stripping is ethicallyacceptable at all, illustrates how situation ethics dictateconservation procedures.

p0240Traditional acid stripping solutions are aggressive tometal and corrosion products alike and are now largelyignored, except where it is ethical to remove all corro-sion products. Copper alloy coins have been separatedusing complexing agents exhibiting low aggression tometallic copper to remove copper corrosion products.Citric acid is reported as being best in this respect,when compared to alkaline glycerol, alkaline Rochellesalt, sodium hexametaphosphate, and sulfuric acid.40

Elsewhere alkaline Rochelle salt is reported as beingless aggressive than reagents like methanoic acid, alka-line glycerol and alkaline dithionite.39 Electrochemicalassessment of these reagents would help indicate whichreagents offer the best removal of corrosion for theleast aggression to the parent alloy. Dealloying is aproblem. Predictably, metallic lead is lost from copperalloys by alkaline stripping systems and organic acids.Stripping corrosion from copper alloys38 and iron41 hasbeen achieved using thiourea (SC(NH2)2) inhibitedcitric acid washes. Thiourea controls the aggressiveaction of citric acid by complexing unstable Cu(I)species, to prevent them forming metallic copper, andby forming stable complexes with Cu(II).8,38 Phospho-ric acid (H3PO4) has been used for stripping ferrousmetals to leave a ferric phosphate protective film, as hashydrochloric acid with hexamine corrosion inhibitor.41

All stripping methods require post-treatment washes toremove residual chemicals that would continue to cor-rode metals. This is often followed by application of aprotective coating.

p0245Following citric acid/thiourea treatment of copperalloys, washing in sodium sesquicarbonate is used toneutralize acid and remove CuCl left in crevices andcracks, but this is likely to be incomplete.38 Washingmarine copper alloys revealed the benign nature ofthis treatment method, as tin, zinc, and lead com-pounds were not removed from either the corrosionproducts or copper alloy, whereas washing in deio-nized water removes metallic lead as the purity ofthe water prevents formation of protective PbCO3.

38

Washing of iron in inhibited chromate or nitrite


CORR: 00172

ELSE

VIER

THIRDPR

OOF

solutions following citric acid/thiourea stripping is notenvironment-friendly, so ecologically superior alkalimetal salts of organic carboxylic acids (secacic acid –NaOOC[CH2]8COONa) have been used to providewashes of similar efficiency.41 They can also providelong-term inhibitive properties (see Section 4.43.6.3)and offer promise for widespread use in conservationfollowing more in-depth testing.

p0250 Chelating and sequestering agents, such as thesodium salts of ethylenediaminetetraacetic acid(EDTA), have been used extensively in conservationin attempts to dissolve corrosion products withoutattacking metal. However, they can be slow to actand may be too selective in their action on corrosionproducts. Careful selection of commercial products isessential to obtain suitable pH values for optimizingcomplexation. A study examining reaction of disodiumedeterate (2NaEDTA) on copper corrosion productsand copper metal revealed that chelation rate variedaccording to corrosion product.42 This would result indifferential removal of corrosion products leavingexposed copper open to the treatment solution. Fortu-nately, metallic copper was barely attacked by2NaEDTA over a 30-h period.42 Dissolved sodiumsalts of EDTA have also been frequently usedto remove corrosion from iron and lead and itsalloys.6,8,31,43 Using compresses allows stripping agentsto be used in a controlled manner. Degrigny44 treateda Citroen caterpillar tracked vehicle used on the trans-Sahara crossing of 1922/23 by this method. It stillretained much of its original paint layer. Corrodedareas that had sustained paint loss were treated withcompresses of tetrasodium edeterate (4NaEDTA) toremove iron corrosion products. This was followed byapplication of inhibitor (see Section 4.43.6.2) and aprotective wax coating (see Section 4.43.5.4.2) toprovide a typical stripping-inhibition-coating treat-ment regime. All treatment methods were assessedusing electrochemical measurements prior to theiruse. Treatment success was gauged in the usual quali-tative manner of visually checking for corrosion as afunction of time and no corrosion was visible after6months of indoor display. Developing standards forassessing time weighted visual success of treatmentsset against conservation criteria would produce a use-ful tool for conservation practice.

p0255 Removal of corrosion from copper alloys may alsobe achieved by alkaline dithionite (Na2S2O4) solu-tions, which provide electrons to reduce corrosionproducts to finely divided metallic copper.8,38 Weakheavily corroded objects may be consolidated by thisconversion, but there is also a chance they will

disintegrate. The skill and knowledge of the conser-vator are essential as treatment is fast, lasting only afew minutes followed by a wash of several days toremove residual chemicals.8

s01054.43.4.4 Electrolytic Techniques

p0260Electrolysis is employed in conservation either toentirely remove corrosion products from a metal orto aid diffusion of soluble corrosion accelerating ionsinto an electrolyte, while retaining corrosion layers onthe metal. These differing outcomes are achieved bycareful control of current density (CD) and in someinstances current delivery. The metal is immersed in asuitable electrolyte and made cathodic relative to aninert anode, which protects it from corrosion duringelectrolysis. Different metals are electrolyzed in dif-fering ways according to the nature of their corrosionproducts and the treatment goal. It is an extremelyuseful technique provided its use matches ethical goalsand research identifies optimum treatment methodol-ogy. Its context within conservation has only beenquantitatively established over the past 15 yearssince the application of potentiometric electrochemi-cal analysis within conservation research.

p0265Procedures vary from metal to metal. Iron isimmersed in an electrolyte in which it will remainpassive, such as NaOH. Using an imposed current tocathodically protect the iron frees chloride from itscounter ion role, allowing it to solvate and diffuse outof the object.8,12,29 Chloride is drawn to the inertanode and delocalized from the object by diffusion.Maintaining a low CD retains the information richDPL by avoiding polarization that would producehydrogen gas which physically dislodged corrosionlayers. Reduction of FeOOH corrosion products toFe3O4 can occur and the accompanying increasein density increases porosity which is said to contrib-ute to facilitating diffusion of chloride out of theobject.7,8

p0270Most quantitative assessment of chloride removalby electrolysis has been carried out on cast iron,whose even corrosion matrix is different to the lamel-lar structure of corroded wrought iron.2,8,29 Tests onsmall cast iron chloride infested samples reveal sig-nificant migration of chloride to the anode duringelectrolysis.29 Quantitative measurement has shownthat electrolysis (CD 10mAdm�3) in NaOH solutionis no better at extracting chloride than aqueous wash-ing in NaOH solution.12 While washing is easier tocarry out than electrolysis, full inhibition of ironcorrosion in a 1% NaOH wash solution cannot be


CORR: 00172

ELSE

VIER

THIRDPR

OOF

guaranteed in the same way as an imposed currentprotects during electrolysis.

p0275 Electrolytic treatments are often favored for largeiron objects, as they will remain cathodically protectedduring the long treatment times required to removedeep-seated chloride. Treatment may take from 1 to4 years for large cast iron cannon and wrought ironanchors from marine contexts45 and even very smallcast iron samples required over 200 days to desali-nate.12,29 As with most metal treatments, there is noreport of any quantitative long-term assessment ofpost-treatment stability of electrolyzed iron as a func-tion of relative humidity. Intentionally using a highCD to remove all corrosion layers may be ethical forcertain industrial and historical objects, provided theircorrosion layers contain no information and theyretain substantial metal cores. Such treatment wouldusually be a precursor for the application of protectivecoatings (see Section 4.43.5).

p0280 An advantage of electrolytic treatments is theirdegree of control, as the imposed potential can bematched to specific reduction processes. They arealso useful for large objects, as electrolytic treatmentscan be run for long periods of time with little inputfrom the conservator thereby saving labor. Circum-stance must dictate whether electrolysis is possible.HMS Minerva was a big gun monitor launched in1915 and sold as a visitor attraction in 1984. Removalof the high levels of chloride inside her hull wasachieved by flooding it with Na2CO3 electrolyte andmaking it cathodic (35mAm�2) relative to stainlesssteel mesh anodes placed directly above it.31

p0285 Electrolysis can be used to remove all corrosionfrom lightly corroded lead to reveal fine details on themetal surface. This is normally carried out in sulfuricacid electrolyte. A brief reversal of the electrode polesupon completing treatment makes the lead anodic andforms an invisible PbSO4 coating that has protectiveproperties in the atmosphere.43,46 The final ‘freshmetal’appearance of lead treated in this way would influenceany decision to use this method. Electrolysis of copperalloys to remove potentially reactive CuCl and all othercorrosion products is now considered to be aggressiveand difficult to control, with a high risk of redepositingreduced copper. Simple electrolysis of copper alloys inwater is advocated as deserving of more research, as itcan retain the copper corrosion product patina and offercathodic protection of the metal during treatment.6

p0290 The inherent insolubility of the silver corrosionproducts Ag2S and AgCl prevents the formation ofelectrolytes on corroded silver exposed to high relativehumidity. Consequently their removal is largely an

aesthetic act or an information revealing strategy,according to whether the layer is a thin Ag2S tarnishor a thicker layer of AgCl/Ag2S that might be found onarchaeological silver objects. Once silver is returned toa shiny metal finish the challenge is to apply coatingsor control environments to prevent sulfur-based pol-lutants tarnishing it (see Section 4.43.5). Electrolysisremoves silver corrosion products within a wide rangeof operating parameters using electrolytes such as for-mic acid (5–30%, w/w) or NaOH (5–15%, w/w) withCD of 0.3–20mAcm�2 at 3–12V.6,13

p0295Very low current densities have been used toslowly reduce corrosion on mineralized brittle silverobjects to metallic silver, which is retained in situ toproduce a consolidative effect.13 Conflicting reportsof success make this a high-risk strategy, as weakmineralized silver objects may fall apart during treat-ment due to reduced coherence of their corrosionlayers. Potentiometers have been employed to applypotentiostatic cleaning of silver.13 This method uses athird reference electrode to control the potential ofthe cathode and a polarization plot is used to identifyreduction potentials of the corrosion products, whichare then reduced by setting the object potential tomatch the reduction potential of the corrosion prod-uct. It is a finely controlled system that avoids anyunwanted reactions such as hydrogen reduction andis useful for tarnish (Ag2S) removal.

p0300Once corrosion begins on aluminum and its alloysthey corrode rapidly in the atmosphere. Electrolysiscan be used to remove chloride corrosion acceleratorsand unstable corrosion products. Degrigny47 devel-oped a treatment that cathodically polarized alumi-num to constant potential in a buffered deaeratedslightly acid citrate solution, with treatment para-meters optimized to avoid pitting corrosion from thechlorides extracted into the treatment solution andcathodic corrosion of the metal. Composite aluminumobjects associated with other metals like iron are prob-lematic, as significant galvanic couples can be estab-lished during treatment. For this type of compositemetal object electrolysis is modified. Active polariza-tion of the composite object in an inhibited solutionprotecting the aluminum alloy from corrosion pro-duces a potential that facilitates formation of a protec-tive magnetite coating on the iron fittings.48 This isfollowed by polarization of the object in citric acidsolution (pH 7) to remove aluminum corrosion pro-ducts and chloride. Finally, polarization in deionizedwater washes out chemicals introduced during treat-ment. The object is dried and coated with a suitableprotective coating (see Section 4.43.5.4.2).


CORR: 00172

ELSE

VIER

THIRDPR

OOF

p0305 Treatments are often fine tuned to the needs of anobject. Adaptation of electrolyzing aluminum omittedthe magnetite forming step to treat a WWII JapaneseOscar Ki43II fighter retrieved from a saline swamp.49

The design utilized a swimming pool, 55 000 l of waterand 800 kg of chemicals with expanded stainless steelmesh anodes conformed to the shape of the aircraft andsited 30 cm above its surface (see Figure 10). Theinevitable post-treatment flash rusting of ferrous com-ponents was dealt with by a tannic acid inhibitor.Original paint and penciled notes on the cockpit fasciawere masked prior to treatment. Success was assessedby ongoing observation, as opposed to quantifiablemeasurements of change or residual chloride in objects.

p0310 Imposed currents are used industrially to protectlarge homogenous steel structures, but heterogeneityand corrosion in historical metal objects producediscontinuities that limit applications of this methodfor corrosion control. The American Civil War sub-marine CSS Hunley was recovered in 2000 andstored in a holding tank containing chilled (10 �C)filtered water (see Figure 11).50 Storage involves thehull being protected by an impressed current moni-tored by five electrodes, with two large anode seg-ments running the length of the tank. Comparingcorrosion rates of an unprotected mild steel testprobe with a similar probe with an imposed currentindicated that the post-excavation corrosion rate hadbeen reduced by a factor of 8 and now matched pre-disturbance corrosion rates recorded on the seabed.50

Protection of the hull by sacrificial anodes was rejectedas a preservation option, as it would have required 50cumbersome 20 kg anodes to be attached to the hull.50

Impressed current was used to reduce corrosion of the

USS Monitor’s steel structure from 254mmyear�1 toas little as 25mmyear�1.51 Measuring the in situ corro-sion rates of ship and aircraft wrecks to identify therole of galvanic couplings, concretions, coatings, andenvironment on their longevity has been extensivelystudied by MacLeod52 to develop predictive corrosioncontrol strategies in marine contexts.

s01104.43.4.5 Hydrogen Reduction

p0315Removal of chloride from iron using hydrogen reduc-tion at temperatures of over 800 �Cwas once seen as away forward for large chloride contaminated marineobjects but is now rarely used. The treatment uses ahydrogen/nitrogen gas mix to reduce chloride bear-ing compounds at around 350 �C, iron oxides above570 �C and fully volatilizes iron chlorides and anyNaCl present on marine iron at 800 �C.8,30 Highcapital outlay, stringent safety measures designed toprevent oxyhydrogen explosions and ethical concernsthat changes to metallographic structure compromisetechnological data and object history limit its use.It may be argued that it is still a viable treatmentfor cast iron cannon from marine contexts, wheretechnology is well reported and where objects maybe considered to be effectively untreatable in anyother way due to the amount of deep-seated chloridewithin them. The technique is reported to providegood post-treatment stability, but suggestion that themethod can offer stabilization by reducing corrosionproduct to metallic iron is challenged by the instabil-ity of the reactive pyrophoric iron it produces.53

Another heat assisted expensive reduction processused in conservation is gas plasma. It has been used

Figure 10f0050 Electrolysis of Japanese ‘Oscar’ aircraft (a) and completed fuselage (b).49 Image courtesy of Australian WarMuseum.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

ELSE

VIER

THIRDPR

OOF

to aid chloride removal from iron by reducingFeOOH to denser Fe3O4 to produce a more porouscorrosion product layer that aids removal of chloridein subsequent aqueous alkaline treatments.54

s01154.43.5 Coatings

s01204.43.5.1 Coating Rationale and Research

p0320Conservation coatings are normally applied as barriersto combat ingress of gaseous, aqueous, or particulateagencies of decay. They may be treatments in theirown right or form part of a treatment regime, such aswhen they are applied as moisture barriers followingattempts to remove chloride from iron using washingtreatments. Coatings applied to silver must have goodgas barrier properties to reduce ingress of H2S andthereby prevent tarnish.13 Resistance to vapors such asvolatile carbonyl pollutants is important for lead, cop-per alloys, zinc, and aluminum.15,18,43

p0325The role of a coating, environment in which it mustbe effective and the nature of the metal surface dictatecoating choice. An absence of standardized proce-dures for testing coating effectiveness on culturalmetals hinders the use of published material for com-paratively identifying the most suitable coating for usein a specific context. While some popular coatings stillmostly have qualitative data to support their use, thereexists a considerable amount of useful informationwithin conservation literature and there is a movetoward providing quantitative data and standardizedtesting.3 Both qualitative and quantitative assessmentof coating performance within conservation nowemploys electrochemistry and modern methods ofinstrumentation in conjunction with increased collab-oration with corrosion scientists.2–4,55

p0330The concept of manufacturing conservation spe-cific coatings is largely a redundant approach, sinceconservation viability is cultural rather than com-mercial as there is no significant financial gain forpotential manufacturers in such a small market. In-creasing collaboration with industrial partners wouldbenefit conservation and reveal its links to industrialapplications. Both industry and conservation seek lowtoxicity, easy application, predictive performance,and low cost for their coatings. Additionally conser-vation has stringent ethical guidelines that requirecoatings to have enhanced life span, long-termreversibility and minimal visible impact to the objectupon application or ageing. Object context may mod-ify these goals to some extent. Coatings that offergood protection from corrosion, but are aestheticallydispleasing, may be acceptable for objects in long-term storage provided they are easy to remove fordisplay purposes. Since the conservation ideal is forindefinite preservation coatings should be long lastingand low maintenance during their lifespan. Demandfor long-term durability and guaranteed reversibility

(a)

(b)

(c)

Figure 11f0055 The H.L. Hunley submarine was lifted from the

seabed in a metal frame (a) that was then stored in a large

holding tank (b) protected by an impressed current (c).50

Images courtesy Friends of the Hunley.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

remain at odds with coating design. While cross-linking polyurethanes have renowned barrier proper-ties their irreversibility, without significant mechanicalaction coupled with aggressive solvents, rules themout for most conservation applications. Since minimalintervention with an object is contravened by applyinga coating and by its future removal, its application isonly ethical provided there is evidence that it signifi-cantly improves corrosion protection.

p0335 Conservation coatings must often perform well onporous, uneven oxidized metal surfaces that may befragile and chemically unstable. No coating is perfectand this type of surface will result in a large number ofcoating imperfections. This complicates the testing ofcoatings for conservation applications, as producingtest samples with suitably corroded surfaces to matchcorrosion patinas developed over years or centuries inthe ground or air may be impossible. The alternativeof using original materials suffers from reproducibilityproblems and may be subject to ethical constraints. Insome instances actual objects are used to test products.The baton from the bronze equestrian statue of thefamous fifteenth century mercenary Bartolomeo Col-leoni (see Figure 12), which is sited opposite theOspedale San Marco in Venice, was used to comparethe performance of organo-silane coatings (see Section4.43.5.4.4), synthetic waxes (see Section 4.43.5.4.2) andacrylic (see Section 4.43.5.4.1) coatings in tandemwithtests on naturally aged metal samples.56 This approachlinks the intrinsic properties of coatings to real-lifeapplications and contexts. Samples are not reproduciblebut performance can be linked to tests on prepared andreproducible samples. Two-phase test systems, wherematerials are laboratory tested on reproducible samplesby accelerated and natural ageing, then tested on objectsin real life environments were employed in thePROMET project.3 This system should be applied togenerate evidence-based advice for practical conserva-tors regarding contextual use of materials.

p0340 Coatings on metals are used extensively withinconservation, but the absence of industry standardsmakes it difficult to identify the best coating for definedcontexts. Much published work produces useful stand-alone data. This usually compares groups of coatingsusing highly specific goals and personalized testregimes to deal with the multiple numbers of variableswithin the test procedures. The demands and com-plexity of coating use in metals conservation meansthat many studies naturally attempt to review manyaspects of a chosen group of coatings. A stepped studythat examines and compares the performance of coat-ings in relation to a single variable then moves onto the

next would gradually build into a detailed review ofmaterials. Unfortunately, this approach demands sig-nificant time and resource commitment and wouldnecessarily be a team project requiring long-termmajor coordination. Conservation research is not lav-ishly funded. Collaborative research projects are cur-rently underway or recently completed.2,3 Their workcould offer platforms for movement towards industryrecognized standards for testing. Significant coordina-tion and standardization were evident in the recent EU6th Framework PROMETproject.3 It sought to identifysuitable coatings for use on metals within museumssurrounding the Mediterranean basin and aimed toidentify a methodology for developing and testing inhi-bitors and coatings for use on iron and copper alloys inmuseum environments. Additionally it was lookingtowards environment friendly inhibitor systems.

p0345Many researchers modify international standardsor use them unchanged for testing materials anddesigning experiments. Overall, there remains a

Figure 12 f0060The baton from the right hand of the fifteenth

century statue of the codotteri Bartolomeo Colleoni in

Venice was used for trialing protective coatings.Reproduced from Joseph, E.; Letardi, P.; Mazzeo, R.;

Prati, S.; Vandini, M. InMetal 07, Book 5, Protection of Metal

Artefacts, Interim Meeting of the ICOM-CC Metal WGAmsterdam, 17–21 September 2007; Degrigny, C., Van

Langh, R., Joosten, I., Ankersmit, B., Eds.; Rijksmuseum:

Amsterdam, 2007.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

need for developing test procedures that support con-servation goals and facilitate comparative performancetesting of coating materials for use in conservation.A conservation-specific accelerated corrosion test stan-dard could be used to model the demands of conserva-tion practice. Coating performance in its operationalenvironment is of crucial importance, as it determinesits maintenance requirements and life-span. Althoughlaboratory studies, instrumental analysis and electro-chemistry can establish intrinsic performance charac-teristics of coatings according to a set of fixed variables,they cannot offer quantitative assessment of real-timein situperformancewithinmuseum stores, display casesand outdoor environments. Laboratory testing is usefulfor establishing which coatings are unsuitable for useand for comparative performance testing in specifiedconditions. Coatings identified as good performerswithin the laboratory may fail in the field becausecertain parameters could not be effectively replicatedin the laboratory. However, appropriately designedaccelerated ageing normally offers an indication ofthe relative performance of materials.

p0350 Within testing processes examining coating perfor-mance a recurring theme is that application methodsoften dictate whether a coating performs well or badly.The procedure here may be critical. Inappropriatecoating methods may negate differences between coat-ings. A coating with intrinsically better barrier proper-ties may fail at the same rate as one that offers poorerbarrier properties because of application methods.Standardizing application methods to ensure variablecontrol is an appropriate experimental procedure, butit may not reflect how the coating will perform inpractice where the condition of the object, surfacemorphology and positioning can dictate whether appli-cation is by spray, brush, or cloth. How coatings per-form as a function of their application method is asimportant as their intrinsic properties and should forma major part of testing. A coating with only mediocrebarrier properties may perform well if its physicalproperties suit a particular application procedure.Spraying is often a favored coating method, but coat-ings of the cellulose nitrate lacquer Frigiline™ onsilver were found to fail unevenly due to thicknessvariations and discontinuities that resulted from spray-ing.57 Brushing is known to offer preferential failure introughs created by the brush stroke and solid waxcoatings may have to be applied as polishes. The natureof the surface of the object may also favor differentmethods of application and mobile self-healing filmsmay be essential in certain environments. The coatingitself may dictate which application methods are

possible to use. This overall complex equation mustbe effectively worked into research studies if results areto be of maximum practical use.

s01254.43.5.2 Metal Surfaces and Patinas

p0355A major hurdle to the successful application of coat-ings within conservation is surface preparation. Onlyin specific circumstances is it ethical to remove patinasand corrosion from surfaces. Within museumsequipped with dedicated micro- or macroenviron-mental control and monitoring, bare metal surfaceson historical metals may be left uncoated. However,without environmental controls silver tarnishes rap-idly in the presence of very low concentrations ofsulfur pollutants like H2S and forms Ag2S

8,10,13,57 andlead and copper alloys are attacked by low concentra-tions of organic acids.2,6,8,10,15,18,19,47 Preventive con-servation is the preferred option to control pollutantavailability and attack, but coatings offer an alternativestrategy where sources of pollution cannot be elimi-nated. Potential for disaster exists in mixed displays orwith composite objects, such as when an unstablecellulose acetate object emits acetic acid into anenclosed showcase it shares with a lead object.

p0360Patinas on many metals offer an informationresource and provide aesthetic attraction. They maybe deliberately applied or naturally formed corrosionlayers. Deliberately patinated or polished copperalloy statuary is often modified by corrosion frompollution, rain, and particulates that form corrosionproducts that may undermine, obscure, or destroy theoriginal patina and outermost metal surfaces6,19 seeFigure 12). The extent of corrosion will relate tometal composition, climate, location, preexistingpatina and the solubility, morphology, uniformity,and adherence of new corrosion products.6,10,19 Runmarks, differential corrosion from pitting, differingcolors and textures produced by climate, pollution,composition, and object morphology can entirelyruin object aesthetics and offer a porous unevensurface that will not accept a continuous coating.Wear marks on patinas can be important to retain aspart of the history or present life of an object (seeFigure 13). Whether total or partial removal ofpatina is appropriate, in return for improved objectstability and surface preparation for coating, presentsethical and aesthetic dilemmas for the conservator.

p0365Apart from ethical considerations, establishing therole of a patina in the corrosion process can help decidewhether to remove it as part of a corrosion controlstrategy. Equally an understanding of patina stability


CORR: 00172

ELSE

VIER

THIRDPR

OOF

and protectiveness contributes to developing corrosioncontrol strategy by offering a measuring point formonitoring, assessing, and predicting patina deteriora-tion with time.58 Nondestructive in situ analysis hasbeen used to identify variance of electrochemical sta-bility within patinas to locate areas that offered highstable electrode potential relative to the base metal, asthey provide good corrosion protection.59 Resultsrevealed that unsightly black crusts offered more pro-tection than visually acceptable blue/green corrosionproduct layers. This creates an aesthetic dilemma forconservation strategies based on patina retention.Understanding the contribution of corrosion layers tothe corrosion process makes it easier to determine ifinterventive treatments are a better corrosion controloption than noninterventive monitoring strategies.

p0370 Disfiguring, unstable and aesthetically displeasingpatinas formed by corrosion may be removed, either

to reveal metal or a selected corrosion product layerin preparation for repatination and/or adding a pro-tective coating.6 Repatination may even form part of acorrosion control strategy, as some patinas may offer adegree of protection by acting as partial barriers tomoisture, electrical conduction, and gases.6 Instrumen-tal analysis and historical research to determine theoriginal appearance of a metal object is a prerequisiteto selecting a repatination process. Working objectsand industrial equipment that never developed apatina or were frequently repainted may in someinstances be stripped to the metal surface, providedtheir history is not compromised by loss of data such aspaint layer sequences. Analysis of paint layers woulddetermine if this was likely to occur. Stripping pro-cesses like electrolysis, chemical dissolution and com-plexation, which remove corrosion prior to applyingcoatings should provide for lower maintenance ofcoatings, all other factors being equal. Yet there remainethical dilemmas concerning the stripping of historicalmetals.60 During its working life a cog may haveuncorroded teeth that developed a surface finishfrom work, with a corrosion patina on its ‘noncontact’body that can also be associated with its working life;should conservation retain both these layers?

s01304.43.5.3 Preparing Surfaces on CulturalObjects to Receive Coatings

p0375Surface preparation of objects may take the form ofmechanically removing disfiguring corrosion toreveal the ‘original’ surface of the object or the markerlayer for this. In archaeology it is normal to usemechanical systems to remove corrosion to reveal theobject shape and surface detail.7 For historical metalspreparation may involve mechanical cleaning systems,chemicals, or electrolysis. Precious metals may bepolished to a high shine as the goal of coating is toensure that future repolishing occurs as infrequentlyas possible. In all cases it is best to remove solublecorrosion accelerators before applying coatings.

p0380Outdoor statuary is cleaned according to the goalsof the conservation process, taking account of ethicsand aesthetics. Preparing surfaces by stripping patinasis normally unacceptable due to ethical constraintsand aesthetic outcomes (see Figure 14). All processesmust involve minimum loss of material. Choosing acleaning method normally relates to how it physicallymodifies surfaces and alters appearance, rather than itsimpact on metal stability by removing corrosion accel-erators or making a surface more coating friendly,unless it is being prepared for paint. Cleaning methods

(a)

(b)

5 mm

Figure 13f0065 Constant touching of this copper alloy hand

has created a wear mark patina that now forms part of the

object history and presents a dilemma for conservation

planning (a). Achieving a continuous coating onarchaeological copper alloys is problematic due local

corrosion and differential patina (b).


CORR: 00172

ELSE

VIER

THIRDPR

OOF

will influence the effectiveness of coatings, as leavingcorrosion products and patinas in situ influences theiradhesion and distribution.

p0385 Most studies qualitatively compare the effects ofcleaning methods on specific metal surfaces. Dryparticulate abrasion, pressurized water and chemicalshave variously been compared for cleaning patinatedcopper.61 One study serves to illustrate some of thedecision making associated with cleaning surfaces. Itcompared cleaning of a patinated copper alloy tokenby abrasion techniques employing glass beads, walnutshell, corn cobs, and sodium bicarbonate deliveredat pressures between 1.38� 105 and 5.52� 105 Pawith simple abrasion using Al2O3, pressurized waterand chemical cleaning by complexation (solutioncomprising KNaTartarate, 2NaEDTA, NaOH, andfumed silica applied in a cellulose ether gel pack).61

Postcleaning samples were exposed to synthetic rainand their corrosion was monitored.

p0390 Results of these tests reveal the general dilemmaproduced when trying to balance effectiveness withaesthetics and ethics. Although chemical cleaningoffered the most stable substrate, as observed after1 year, it produced the most visually altered surfaceby removing all overlying green corrosion productsto leave only orange/brown Cu2O in situ. Predictablythe softest corrosion products disappeared from themechanically abraded objects, which were less stablethan the chemically treated sections. Pressurizedwater produced the least visual change, but removedthe most soluble corrosion products leaving the metalactively corroding after a year. None of the methodsentirely stabilized the metal and various corrosion

products were left in situ, neither did they removeall corrosion accelerators. Other workers have ex-pressed concern at indiscriminate surface prepara-tions that fail to recognize the importance oforiginal coatings on metals like aluminum.16 Glassbeads have been cited as being too damaging onfinishes on aluminum alloys, but airbrasion can beretained by use of plastic beads at 1.0� 105 Pa.16

p0395Preparing historical metals and objects d’art thattraditionally carry a shine for the application of coat-ings involves either chemical or physical removal oftarnish. Removing H2S tarnish from silver can beachieved by chemicals or abrasion.13 Chemicals areaggressive and may lead to differential corrosion andsurface enrichment of alloys such that loss of copperalloyed with silver leaves a soft and porous silverenriched surface. They may also leave residues thatare too aggressive and these have been tested to deter-mine their impact on the object and develop a protocolfor treatment. Polishes remove metal and produce sur-face scratching. Studies of polishing methods concen-trate on the loss of metal incurred and scratching of thesurface, with CaCO3 being a favored inert polish forsilver.62 Chemicals are normally only preferred forheavy tarnish due to the risks of preferential leachingof alloying components and the need for acidifiedcomplexing agents to remove Ag2S.

12 Potentiometricmethods mentioned earlier can also be used to removesilver tarnish (see Section 4.43.4.4). Focus should beon identifying the least interventive method that offersthe best surface for receiving a coating.

s01354.43.5.4 Coatings in Conservation Practice

p0400There are a small range of generic coatings in generaluse within conservation and there is no definitivecomparative study which clearly identifies the mostsuccessful coating for any given context. Preferencesand trends exist for various chemical groupings andproducts.3 Coatings are reviewed below by chemicalaffiliation, rather than their application to specificmetals. Trade names are used in general discussion,having initially defined the chemistry of the product.

s01404.43.5.4.1 Acrylic coatingsp0405The Paraloid™ range of acrylics, known as Acryloid™

in USA, is commonly used within conservationpractice. Other acrylics are used according to availabil-ity, which means geography can dictate choice ofmaterials. Survey reveals that only a few coatingsare routinely used, with Paraloid B72™ (70 methylmethlacrylate/30 ethyl acrylate copolymer) being oneof the most popular general purpose surface coatings

5 mm

Figure 14f0070 This contrast between a surface retaining its

uncleaned patina and a section cleaned by airbrasion with

aluminum oxide starkly reveals the tension between best

practice for preparing surfaces to receive coatings and theaesthetic and ethical constructs of conservation.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

ELSE

VIER

THIRDPR

OOF

for cultural metals. Its long-term reversibility andapplication as a coating, consolidant, and adhesivemake it a workhorse within conservation and acomparative performance standard in many test pro-cedures.3 Performance of Paraloid B72™ and an eth-ylene based commercial coating Poligen ES 91009™were compared, both alone andwith added inhibitor ascoatings on fresh and precorroded coupons of steelexposed to cycling g (90% at 35 �C to 55% at 25 �C)to simulate extreme museum environmental fluctua-tions.3 Performance was assessed qualitatively by visualinterpretation and quantitatively using EIS and polari-zation techniques.3,63 Poligen ES 91009™ was moreeffective than Paraloid™, and added inhibitors failedto significantly improve the performance of either coat-ing. Long-term exposure of similar test coupons inmuseum environments confirmed that although Para-loid B72™ gave ‘quite satisfactory results,’ it failedat edges producing filiform corrosion.3 This was recog-nized as being problematic where much cultural mate-rial like edged weapons and armor had multiple thinedge sections. The importance of environment wasrevealed in these tests as matched sample sets per-formed differently over similar exposure times withinmuseums in differing Mediterranean countries. Appli-cationmethodwas again revealed as being an importantvariable. In one test, a 250mm layer of polyurethane(Rylard™ boat varnish) performed much better thaneither Paraloid B72™ (40mm) or an innovative physicalvapor deposition system (<1 mm).3 Standardizing testmethodology by employing aged coupons and usinglaboratory and onsite real-time testing in monitoredenvironments offers a solid test platform for culturalmetals and for theevolution of assessment standards. Yetoutcomes discounted certain products due to outrightfailure, it proveddifficult to quantify the performance ofsome of themore successful coatings. Overall, the studyrevealed weaknesses in the performance of traditionalconservation materials and identified the newly testedPoligen ES 91009™ as offering promise for the future.

p0410 The inability to separate lead organ pipes from oaksupporting frames emitting organic acids in EuropeanBaroque organs led to a research program that testedthe ability of coatings and inhibitors to prevent thecorrosion of lead by organic acids.64,65 Using an aque-ous washing strategy to remove soluble lead methano-ates and ethanoates acting as electrolytes failed toreduce the corrosion rate upon reexposure of the leadto the methanoic (170 ppb Pb(CHOO)2) and ethanoicacid (195 ppb Pb(CH3COO)2).

65 Coatings alone pro-duced limited protection, with Paraloid B72™ offeringno protection, microcrystalline wax providing uneven

protection and various inhibitors had little effect.64,65

This is of concern as Paraloid B72™ has long beenused as a coating on lead. Treatment must be effectivefor the long term, due to the cost and logistics ofcoating the organ pipes. Nanotechnology is nowbeing tested in a preventive treatment using Ca(OH)2nanoparticles to control acid emission from thewood.64

p0415Incralac™ lacquer comprises Paraloid B44™(ethyl methacrylate/butyl acrylate copolymer) withadded epoxidised soya bean oil as a leveling agentand benzoriazole as a UV absorber6 and is industrydesigned for protecting copper alloys. It has beenextensively used both on archaeological and outdoorcopper alloys since the 1960s. Its application aestheti-cally changes bronzes by darkening their surface.56

However, outdoor exposure is likely to result in opticalchanges to most coatings as they weather and collectdirt. Good electrochemical impedance measurementsrecorded for fresh Incralac™ coatings on polished andunpolished bronze, implied good coverage and lowporosity, but after natural ageing on a rooftop in Can-berra for 4 years impedance was the same as foruncoated metal.57 Although two coats improved per-formance the time-related failure remained the same.Poor long-term performance was also found in 10-year-old Incralac™ coating on a gilt bronze statue inNew York.6 It was entirely cracked and its insolubilitywas likely due to cross linking exacerbated by the lossof the BTAUV stabilizer which was absent in the agedlacquer, which required strong solvents and physicalintervention to reverse it. Inhibitors such as benzotri-azole (see Section 4.43.6.1) have been used to primebronze surfaces prior to applying Incralac™.6 Themanufacturer reports a 5-year lifespan with removalinstructions recognizing its reduced solubility withage, but a 2-year lifespan on outdoor copper alloymonuments is more common. The outdoor perfor-mance of Incralac™ is limited when measured againstconservation goals of low maintenance, good protec-tion, and reversibility but it continues to be used in theabsence of any quantitative proof of significantly bet-ter performing alternatives.6 Survey suggests Incra-lac™ is now less commonly used on historical andarchaeological metals inside museums, apparently infavor of Paraloid B72™.3

p0420Solvent-based acrylic resins and their aqueous dis-persions have been tested as H2S barriers in a programcomparing them to vinyl acetates, cellulose nitrate andmicrocrystalline wax.66 Acrylic dispersion systemsperformed better than the acrylic resin solutions andwere equal performers with vinyl acetate and cellulosenitrate, but the defining factor in protectiveness of all


CORR: 00172

ELSE

VIER

THIRDPR

OOF

the coatings tested was their evenness and thickness.Uneven coatings produced differential tarnishing.Microcrystalline wax (Renaissance Wax™) producedthe worst result because its solid state made it difficultto apply. The conflict between good experimentalprocedure and practical use of coatings is reflectedby drying coatings for 8months prior to testing themwith H2S, as this is unlikely to reflect the procedurewithin museum contexts.

p0425 Fragile metal objects are held together by impreg-nating them with polymers using immersion systems.Acrylic polymers like Paraloid B72™, microcrystallinewaxes and occasionally, epoxy resins are used for frag-ile iron.7,36 Unsurprisingly, long-term survey revealsthat iron objects washed to remove chloride and thenconsolidated with epoxy resins survive better thanwashed iron that remained unconsolidated.36 Epoxyresin coating offers a degree of protection againstmoisture ingress and its strength as an adhesive retainsthe physical integrity of the iron even if it corrodes.36

While the insolubility of epoxy resins appears to con-travene the central conservation concept of reversibil-ity, its use is measured against increased longevity ofthe iron as a cultural resource. The influence of ethicson treatment choice takes account of both the prevail-ing situation and the ultimate goal of prolonging thefunctional life of an object.

p0430 In reality even the use of a reversible consolidant iseffectively an irreversible process. Since its use implic-itly acknowledges it is essential for retaining the physi-cal integrity of an object, overcoming consolidant/metal interactions to remove it would likely result inobject fragmentation. While there have been no conser-vation studies on the influence of epoxy resins on thecorrosion of metals, less viscous short chain aliphaticepoxy resins suited to consolidation processes producemore OH groups upon cross-linking than aromaticepoxy oligomers, which means they attract morewater from the atmosphere and this may facilitategreater corrosion. Trapping moisture within ironobjects by using consolidants and coatings runs therisk of creating microclimates that may accelerate andlocalize corrosion.

s0145 4.43.5.4.2 Waxesp0435 Microcrystalline waxes are commonly used as coatings

in conservation, especially on smooth even steel sur-faces such as armor3 andpatinatedoutdoormonuments.6

Recent Ecorr studies revealed one such wax offered littleor no protection when compared to the acrylic ParaloidB72™ and polyethylene wax Poligen ES™ on smoothsteel in contact with electrolytes.3 Interpreting this

failure illustrates the importance of taking account ofenvironment and metal substrate, as when microcrystal-line waxes were tested on lead exposed to atmospherescontaining organic acid vapor they performed signifi-cantly better than Paraloid B72™.65

p0440Renaissance™ and Cosmoloid 80H™ waxes arethe most commonly cited microcrystalline waxes inconservation literature. Typically in conservation,particular commercial products tend to be preferredfor use over long time periods.3 In what is often anonevidence-based manner the perception growsthat they are ‘proven’ products. Replacing popularcoatings that are no longer being manufactured pre-sents a problem. An extensive study was set up tofind the best commercially available alternative for amicrocrystalline wax being phased out of production.A range of microcrystalline waxes and a low meltingpoint (80–100 �C) polyethylene wax were comparedusing electrochemical impedance spectroscopy andimmersion in 0.1M NaCl, supported by real-timeatmospheric exposure.55,67 BeSq 2095™ microcrys-talline wax performed similarly to TWA 2095™which was being phased out. Real-time testing wasan essential part of this study and identified completefailure of the waxes over a 4-year exposure period. Asreported for many other test procedures it was theapplication method that once again provided a signif-icant influence on performance. BeSq 2095™ micro-crystalline wax performed better when applied in amolten state, which produced crystalline lamellae,whereas polyethylene wax offered best protectionwhen buffed onto bronze surfaces.

p0445Protection of bronze ethnographic and artisticobjects with waxes provides a transparent coatingwith a degree of color saturation that meets aestheticgoals. Sticky aesthetically displeasing mobile wax coat-ings that alter color saturation of surfaces are not usedin conservation, unless the object is to be stored.A number of commercial products comprising waxand volatile corrosion inhibitor additive were rankedas better performing coatings for outdoor bronzesculpture, as compared to the conventional microcrys-talline wax Besq 195™ using electrochemical imped-ance spectroscopy.67 Dinitrol 4010™ performed best.It is used by the aerospace industry to protect enginesin storage.68 and it is a preferred choice for electro-lyzed aluminum alloys,16,47,68 although a note of cau-tion was expressed on the use of inhibited commercialwaxes for aluminum alloys without first examiningtheir relationship with the corrosion processes takingplace.16 A Focke-Wulf 190 aero engine treated byelectrolysis was coated with a 40mm layer of Dinitrol


CORR: 00172

ELSE

VIER

THIRDPR

OOF

4010™ post treatment,68 which has good protectiveproperties and is favored by the aircraft industrybecause it has no adverse affect on rubber and plastics.

p0450 Wax coatings are considered to be high mainte-nance options for outdoor bronzes, as compared toacrylics. However, they are used as sacrificial topcoatson outdoor bronze statuary where they protect theunderlying acrylic layer from degradation. A three-phase system of Benzotriazole inhibitor primer,Incralac™ main coat and microcrystalline wax top-coat has been employed by a number of workers andoffers a more robust protective system.6

s0150 4.43.5.4.3 Cellulose nitratep0455 Before the introduction of the Paraloid™ range of

acrylics, cellulose nitrate was commonly used in con-servation as an adhesive and a coating. It has provenlong-term reversibility and its properties and decayhave been extensively reported. Long-term perfor-mance studies discounted its effectiveness as a coat-ing on archaeological iron.3,34 Although it has been inlong-term use as an antitarnish coating on silver thecellulose nitrate lacquer Agateen™ only equaled theperformance of acrylic and vinyl acetate emulsions inlaboratory tests.66 As with many other coating studieson silver, tarnishing from sulfur was strongly influ-enced by coating methods and the quality of theirapplication.57,66 Anecdotal reporting suggests recoat-ing silver to protect against tarnishing is necessaryevery 10 years. This is an expensive and time consum-ing exercise for a large silver collection, so coatinglifetime has been explored using accelerated ageingtests.69 These revealed that light and relative humidityhad a pronounced effect on the lifetime of the com-monly used cellulose nitrate lacquer Frigilene™.Extrapolating the results of the accelerated tarnishingtests indicated that Frigiline™ should still protectfrom tarnishing after 10 years in mid-range relativehumidity values. Calculations did not account forthermal effects inherent in accelerated ageing tests,so lacquer lifetime was expected to be greater than10 years’ exposure at room temperature.

p0460 Appropriately these laboratory tests are being eval-uated by English Heritage using real-time monitoringof the Waterloo silver centerpiece in Apsley House,London. In several years’ time the final outcome willoffer insight into the effectiveness and value of the testprocedures, as much as the performance of Frigiline™lacquer as a gas barrier. Focused pragmatic studiessuch as this offer tangible evidence for choosing mate-rials fit for task and for devising long-term conserva-tion planning and budgeting. The reported testing

supplied sufficient information for the production ofa predictive conservation strategy. Benign impact onmetal, proven reversibility, and long operational life-span are the main essentials for conservation planning,which, in this instance, revealed that cellulose nitratewas a good choice for antitarnish coatings on silver.Although investigations into the surface chemistryrelationships between cellulose nitrate and silverwould offer insight into mechanisms of protectionand failure, it is not necessary in order to devise apreservation strategy.

s01554.43.5.4.4 Silanesp0465While organosilicon compounds are extensively used

in industry,70 they have received limited attention inconservation practice. It appears that there is muchscope for employing them on cultural metals ashydrophobic barriers, as they are largely invisibleon surfaces, are good water repellents and are capableof bonding to corrosion products.70 Barrier propertiesof silanes can be improved with the inclusion ofinorganic additives like silica plates, which canreduce water permeability and aid reversibility.Eventually silanes will be broken down by waterreaching the metal–silane interface, where it reversesthe Si–O–Me bond that was formed by hydrolysis toprotect the metal. Applying thicker films cannot beused to counter this as they tend to be too brittle andapplication difficulties are experienced.70

p0470Organic–inorganic polymer systems produced byhydrolysis and condensation reactions of alkoxysilanes and organo-functional alkoxy silanes havebeen tested on bronze, fresh, and precorroded steelin the form of the Ormocer™ (ORganically MOd-ified CERamic) family of lacquers produced by theFrauhofer Institute in Munich.71 Good adhesion tometal surfaces and their hydrated corrosion productsshould occur due to the presence of Si–OH andSi–O–R groups in the polymer and an organic poly-meric network results from the cross linkable func-tional groups of alkoxy compounds (R–Si(OR)3).Ormocer™ lacquers can be modified to producediffering elasticity, by reacting their main cross link-able component glycidoxypropyl trimethoxy silanewith alkoxy and hydroxy silanes.71

p0475These lacquers significantly outperformed waxesin laboratory-based accelerated (SO2) corrosion testsand outdoor exposure using various internationalstandards.72 Application method and concentrationproved important to performance. Monolayers(4–8 mm) and bilayers (10–12 mm), applied by spray,darkened metals and the monolayer performed better


CORR: 00172

ELSE

VIER

THIRDPR

OOF

than the bilayer on patinated metal. This data indi-cates that it may be a promising coating for corrodedmetals, but technical and ethical boundaries must becrossed when using it. Ormocer™ requires methy-lene chloride paint stripper to reverse it. Testing thislacquer in SO2 contaminated environments reflectsits initial conception for use within industrial con-texts, where polluted environments offer significantthreat. Despite the siting of many cultural metalobjects in urban contexts surprisingly few studieswithin conservation examine coating performancein SO2 contaminated conditions.

p0480 As in other areas of conservation comparativestudies dominate conservation research into silanes.An EU research project (ARCHITECH) seeking toidentify better coatings for outdoor art works isexamining silane coatings, copper oxalate patinasand increasing the thickness of Cu2O layers toimprove the natural protectiveness of patina on cop-per alloys.56 Comparing selected silanes to the com-monly used Incralac™ acrylic coating revealedthem to perform only equally well, which is notencouraging given the overall weak outdoor perfor-mance of Incralac™ (see Section 4.43.5.4.1). Thefact that Incralac™ altered the chroma of the metalsurface more than the silanes56 does not seem tooffer strong support for using silanes. The fact thatOrmocer™ showed better adhesion than Incralac™in high humidity72 may be of use, but such condi-tions are likely to more readily induce its hydrolysis.The PROMET project included Silane A (5%g-mercatopropyltrimethoxysilane; 2%bis-(trimethox-ysilylpropyl)amine; 1% hydrated tetraethoxysilane;92% ethanol) in its comparative testing regime.3 Thisoffered best protection for bronze in long-term real-time testing, as compared to Paraloid B72™ andselected corrosion inhibitors. However, all systemsfailed over the test period. There are many difficultiesto overcome if silanes are to be successfully used oncultural metals.

s0160 4.43.6 Inhibitors in Conservation

p0485 Inhibitors are used selectively and, in many cases,empirically in conservation practice, either alone orin combination with coatings. While their use isgoverned by the usual ethical constraints of appear-ance and reversibility, fashion also has an input. Forinstance the blackening of iron by tannate inhibitorsmay be deemed to be ethically acceptable on archae-ological iron, as it could be argued that visual changes

do not stray too far from the appearance of the gray/black DPL layer retained on objects.2 This encapsu-lates the flexible nature of ethics and aestheticswithin conservation, as the slight darkening of patinason copper alloys caused by the use of benzotriazole iscited as being of concern by some authors. Whetherinhibitor use in conservation can be reconciled withthe goal of stabilizing metals is debatable, as inhibi-tors slow rather than prevent corrosion.

p0490No inhibitors have been specifically developed foruse in conservation practice. They are borrowed fromindustrial contexts, where they have been assessed foruse in specific operational environments on particularalloys. As with protective coatings inhibitors arerequired to be effective in the presence of corrosionproducts. This presents difficulties, as soluble corrosionaccelerators like chloride ions normally interfere withtheir action. Low inhibitor toxicity is also a require-ment, as objects remain in the public domain wherethey must be accessible and easy to handle. This tendsto rule out many vapor phase inhibitors, as these areoften based on volatile and toxic amine base com-pounds. Attempts have been made to find less toxicalternatives and cheap natural inhibitors. Recentlyextract of seed oil from the cactus Oputunia ficus indicawas used to formulate an inhibitor that contained longchain fatty acids, triethanolamine, and potassiumhydroxide3 but it failed to offer protection when com-pared to a range of coatings. This is likely due to poorfilm formation, as a continuous film reportedly con-stitutes its inhibitive properties. In the same tests,adding corrosion inhibitors to films failed to improvetheir protective properties significantly.3

s01654.43.6.1 Benzotriazole (BTA)

p04951,2,3-Benzotriazole is the most successful commercialinhibitor used in conservation practice. It is applied toprevent corrosion of chloride contaminated copperalloys.6 There is some discussion whether it is theinhibitive properties of BTA or the barrier propertiesof Cu–BTA films that infer protection to patinatedchloride containing copper alloys. The Cu(I)–BTAinhibitor complex films formed on copper, copperalloys, and Cu2O surfaces have excellent adhesionfrom primary and secondary bonding and likely playthe dominant role in the protection of patinated cop-per alloys by limiting water, ion, and oxygen ingress toreactive metal and mineral surfaces.2,73,74 The filmtolerates chloride ions and low-pH environmentsand reacts with CuCl to produce a BTA–chloridecompound that is stable at high humidity72 (see


CORR: 00172

ELSE

VIER

THIRDPR

OOF

Figure 15). Compared to a range of other nitrogen orsulfur-containing organic compounds it is more effec-tive (99%) at inhibiting hydrolysis of CuCl toCu2(OH)3Cl and although both 2-mercapto-benz-imidazole (98% effective) and 2-mercapto-benzothi-azole (97% effective) approach its inhibitive properties,they form aesthetically unacceptable white and yellowcomplexes with CuCl.75 The black-green coating BTAproduces when it reacts with CuCl is probably whyBTA is noted for darkening patinas on copper alloys.This is a small trade-off for its proven effectiveness atpreventing corrosion.

p0500 There is no definitive agreed and tested protocolfor its use. It is normally used as a 3% solution inalcohol or 1% in water, with no quantitative evidenceto support suggested treatment times that range frompaintbrush application through a few hours in vac-uum to several days soaking. Archaeological anddeeply pitted copper alloys often prove difficult tostabilize using BTA solutions, as hydrolysis withinpits creates a low pH that interferes with the forma-tion of an effective Cu–BTA coating. This is over-come by pretreatment of pits with Na2CO3 but at anaesthetic cost of producing brown-colored spots.6

p0505 Factors influencing copper–BTA reactions and thenature of the resulting film include the condition andoxidation state of the reacting surface, potential, tem-perature, pH, and chloride and oxygen concentrations.Thus Cu(I) complexes formed in acid corrosive envir-onments tend to be thicker, less polymerized, andmore permeable to oxygen than equivalent filmsformed in neutral and deaerated conditions.73 Thisdiversity in film formation indicates that there would

be considerable benefit in tests designed to examinethe nature of the films formed to determine optimumapplication conditions. For instance, it appears thatdeoxygenated solutions may offer an advantage overcurrent oxygenated treatment environments and shorttreatment times could explain why BTA sometimesfails to inhibit corrosion. Tests on artificial patinashave shown that it takes several days for reactionwith brochantite (Cu4SO4(OH)6), which is a commoncorrosion product on outdoor statuary in urban areas,to go to completion.38,74 Although the acid by productof this reaction lowers pH as concentrations of BTAincrease, a dilemma exists as treatment must ensurethere is excess BTA present to repair any damage tothe protective BTA polymeric film formed.74

p0510Use of BTA on outdoor statuary requires a regularmaintenance program, as BTA is likely to be lost fromeffects of rain-wash due to its solubility and volatiliza-tion because of its low vapor pressure. The use of BTAas a primer coated with protective lacquer(s) shouldprolong its effectiveness. Additionally, BTA also formsPb–BTA and PbO–BTA compounds as crystallinepolymeric films which have been shown to protectlead within leaded bronzes from corrosion by organicacids76 and Zn(II)–inhibitor complexes have beendetected on copper alloy.2 In contrast electrochemicalmeasurements showed silicon bronzes had lowerorganic inhibitor efficiency as silica is poorly reactivetowards BTA.2 It is clearly important to know whatcopper alloy is being treated to assess likely inhibitoreffectiveness.

p0515Whether copper alloys routinely require treat-ment with BTA is questionable. What percentage ofobjects would have remained stable, even if theyhad not been treated with BTA, is unknown. Manypatinated copper alloys treated by BTA appear sta-ble in museum environments, although this obser-vation remains unsupported by quantitative data.A study of surface chemistry reactions with thetypical corrosion profiles found on copper alloyobjects and quantitative long-term studies of cop-per alloy object stability following treatment withBTAwould determine its inhibitive powers relativeto the conservation goals of longevity and pred-ictive success.

p0520Despite its proven inhibitive success, workers stillseek alternatives to BTA mainly due to toxicityworries and its high cost. While no better inhibitorfor cultural corroded copper alloys has been recorded,synergistic inhibitive effects have been identifiedwhenusing BTA with other inhibitors that included5-amino-2-mercapto-1,3,4-thiadazole.77 Caution is

Figure 15f0075 Benzotriazole was specifically applied inconservation practice to combat ‘bronze disease’; the

growth of voluminous Cu2OH3Cl polymorphs at the Cu2O–

CuCl interface beneath shape-retaining overlying corrosion

layers. Their loss is visible on this Roman pin.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

required when mixing inhibitors as the formation ofdiffering BTA complexes can lower performance ofBTA.75 Recently BTA with a hydrophobic alkyl sidechain (C6-BTA) has been tested on the basis that theside chain will better repel aqueous electrolyte.2

A more novel use for BTA is as an inhibitive washingprocedure for marine iron, where it forms Cu(I) com-plexes and Cu(II) species and prevents corrosion whilethe water wash extracts chloride from CuCl.38 A bigadvantage of using a BTA preservation strategy oncopper alloys is that it can be applied without anysurface preparation, which is ideal where patina ispart of the intrinsic value of the object. Ensuring accessto the Cu2O layer that is normally situated next to themetal or integrated with CuCl sited there will providethe best opportunity for treatment success, due to theability of BTA to react with this oxide. Unfortunately,removing the overlying green brochantite/antlertite/malachite patinas to reveal orange/red Cu2O wouldbe visually unacceptable in many instances.

s0170 4.43.6.2 Tannins

p0525 As conservation seeks to balance its goals with globalconcerns regarding toxicity and carbon footprints,testing low toxicity inhibitors derived from naturalsources like plant extracts is of interest. Tannin plantextracts have been used intermittently in conser-vation since the 1960s and are reported to act asrust converters, leaving a black inhibitive film onthe metal. Treatment involves either immersion intannin solutions or, more likely, painting onto objects.Currently there is no quantitative in-depth study oftheir action within conservation contexts or any datathat offers quantified guidance for optimum treat-ment concentrations or application procedures.

p0530 It is well known that condensed tannins appliedeither in water or solvent to rust-covered iron offerinhibitive properties by forming ferric–tannate com-plexes that act as insoluble barriers and phosphoricacid is said to improve inhibitive properties. Tannate-coated iron performed well in long-term storageaccording to a qualitative survey of archaeologicaliron.30 While this provides useful information, post-treatment stability studies for determining treatmentsuccess should be controlled and semiquantitativewith clear links to environmental variables, ratherthan retrospective examinations of treated objects.

p0535 Iron stripped by 4NaEDTA was coated with amixture of phosphoric and tannic acids to inhibit fur-ther corrosion, following electrochemical testing todetermine optimum concentrations and combinations

of inhibitor.44 A recent study identifying optimumconcentrations of tannin and phosphoric acid forinhibiting corrosion of rusted iron in a 3.5% NaClsolution indicates the importance of research andtesting to optimize tannin applications to meet theprevailing circumstances. In this study, Ecorr valuesshowed that phosphoric acid reduced the inhibitiveproperties of mangrove tannins at low pH (0.5 and2.0), yet when used alone at higher pH (5.5) it offeredimproved inhibition as compared to tannin/phosphatemixtures.78 Corroding archaeological iron has varyingpH across its surface due to hydrolysis of anodicallyproduced Fe2+ and its thick corrosion layers. Anytesting of tannins for use in conservation practiceshould be tailored to the corrosion model extant.

p0540As with other protective surface coatings testingthe stability of treated objects at typical storagehumidities is necessary and, until this is achieved,application of tannins in conservation will retain adegree of empiricism. To date, no inhibitor has beenshown to have effective long-term inhibitive actionon chloride-contaminated archaeological iron retain-ing its corrosion layers and exposed to mid to highrelative humidities. Commercial tannate-based inhi-bitors have been tested and shown to be effective forinhibiting internal corrosion of boilers in the workingnineteenth-century paddle steamer PS Enterprisebetween its weekly use on lake Burley Griffin, Aus-tralia79 (see Figure 16). In these closed systems theboiler retains its water and inhibition results from acombination of reducing the availability of oxygen,formation of iron tannate film and precipitation ofdissolved salts as sludge.

s01754.43.6.3 Carboxylates

p0545Sodium carboxylates (CH3(CH2)n� 2COONa – usu-ally n¼C10 or C12) derived from vegetable oils havebeen used on copper alloys and iron to form inhibitivecopper and iron carboxylates.2,3,41,80–82 Like tannates,they are environmentally friendly. They have beenlinked to conservation in tests on corroded iron andcopper coupons and blank standards.41,80 Potentiody-namic curves revealed they offered slightly betterinhibition than mimosa tannin solutions and consider-ably better protection than phosphates tested on bareclean steel coupons.80 A further advantage is that thenanomeric hydrophobic iron carboxylate soap layerformed by reaction with iron cations79 is not aestheti-cally disfiguring, as is a black iron tannate layer. Car-boxylates readily reverse in ethanol, which may meanthey are best suited to indoor applications. Long-term


CORR: 00172

ELSE

VIER

THIRDPR

OOF

real-time testing using precorroded metal couponsindicated that sodium decanoate offers temporary pro-tection on partially oxidized historic steels and may beconsidered as an alternative to Paraloid B72™.3 It alsooffered good protection for copper alloys.

p0550 Carboxylates were originally tested in conservationas inhibitors for preventing corrosion of lead by vola-tile organic acids.81 Polarization plots produced bymodeling corrosion of inhibited lead by these acidsindicated that sodium decanoate and undecanoateprotected best against corrosion, whereas phosphateinhibitors actually increased corrosion rates.65 Real-time X-ray diffraction studied the resistance of leadcarboxylate (CH3(CH2)8COO)2Pb) films to aceticacid vapor and revealed considerable protective prop-erties, but ultimate failure.82 Carboxylates appear todelay corrosion by organic acids, but ultimately theycannot prevent it. As in other areas of conservation,contextual needs dictate preservation strategies, whilethese tests showed carboxylates as being unsuitable forpreserving church organs, they may be consideredsuitable where lead is more accessible and easy tomonitor to determine if retreatment is necessary.

s0180 4.43.7 Painted Metals

s0185 4.43.7.1 Removing Paint

p0555 Conservation ethics normally dictate that original orlater paintwork that offers either a record of the history

of the object or milestones in its life should be pre-served. The preservation of fragmentary original paintsurfaces is challenging, as pitting and corrosion under-mining paint are a threat to paint integrity and aredifficult to stabilize. In such instances environmentalcontrol may be the only option guaranteed to preventongoing corrosion beneath the paint layer. There areinstances where ethical arguments support refinishingthe surfaces of cultural objects either to their originalspecification or to an improved standard in order tooffer the best opportunity for longevity. This shouldnot compromise the future interpretation of the object.

p0560Refinishing may be acceptable when there is nooriginal paint layer remaining and replicating an origi-nal finish offers significantly increased object longevity.A combination of refinishing and preservation of orig-inal surfaces may be adopted in some instances.83 If thecondition of the metal allows, refinishing offers oppor-tunity for highly specified surface preparation. It mayalso be possible to apply modern paint systems if visualappearance rather than replication of the original rec-ipe is deemed to be the most important interpretivefactor. Stripping to the bare metal for repaintingshould involve minimal loss of the original metal.

s01904.43.7.2 Refinishing Painted Surfaces

p0565Preparation of historic metal surfaces to receive paintmust always minimize loss of metal. Commercial con-servation programs often adhere to international

Figure 16f0080 Tannates are used to protect the boilers in this 19th-century paddle steamer belonging to the National Museum

of Australia. Image courtesy of David Hallam.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

standards for preparing surfaces, such as SSPC,NACE, and Swedish Standards. This offers the clienta defined measuring point within the contract of work.Repainting the upper hull of the SS Great Britain wasjustified on the basis of continual repaints during theworking life of the ship and the presence of a failingmodern paint layer. The hull above the waterline wasto be exposed to the British climate and merited thebest possible protection from the elements.25 It wasstripped to near white metal (SA2.5) using high pres-sure water lances (2500 bar) and for weaker areascrushed Australian Garnet (8 bar) in preparation forpainting (see Figure 17). Although this removed sev-eral microns of original wrought iron, it was reasonedto be ethical as this thickness of metal would inevitablybe lost through future corrosion if preparation was notso stringent. The paint system chosen was typical forcreating a durable, long-lived outdoor protective coat-ing on a well prepared surface. A three-phase treat-ment comprising a 2-pack zinc-rich epoxy primerfollowed by 2-pack high-build main coat paint systemwith a urethane UV-resistant topcoat, produced a totalpaint thickness of 225–250mm25 (see Figure 17). Thepreferred method of application involved an airlessspray at 207� 105 Pa with brushing when the spraycould not be used. Life expectancy of this layer withappropriate maintenance is predicted at 15 years.

p0570 Often it is not possible to prepare corroded metalsurfaces to a high standard. In this instance originallead-based paints are preferred, both to replicateoriginal paint layers and because of their perceivedlongevity. Since health and safety considerationsmake their manufacture costly and their use hasenvironmental implications, aluminum-based primersare normally used instead. Ferro zinc (HMG Paints)is a rust conversion paint that was used to paint themonitor HMS Minerva after electrolysis.31 The rangeof historic alloys and the inability to effectively pre-pare many surfaces beyond SA 1 or 2 means that eachsurface offers a unique coating problem. Conse-quently conservation designs often test paint adher-ence using a pull-off test (ASTM D 4541-02 or ISO4624) before applying it to an entire surface. Reportson the longevity of painted surfaces relative to theprepainting condition of the surface and the paintsystem applied are not reported and this offersanother area for research that could act as a platformfor designing industry standards. Silanes are used forpreparing metals for paint within industry,70 but thishas not been reported in conservation practice.

p0575 Where sections of metal have entirely corrodedaway, decisions regarding a course of action have to

account for ethical, technical, and corrosion consid-erations. Structural problems require structural solu-tions but care must be taken not to create galvaniccells by repairing with alloys of differing potentialfrom the original metal. Equally, where structural

(a)

(b)

(c)

Figure 17 f0085Surface preparation to SA2.5 standard on the

SS Great Britain and the completed paint schedule.25

Images courtesy of Eura Conservation Ltd.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

integrity is not a concern it is possible to use glassreinforced plastics and epoxy systems.25

s0195 4.43.8 Overview

p0580 The overall equation for preserving cultural metals isa complex mixture of material science, context, and

ethics. During the past 40 years, conservation hasdeveloped links with corrosion science that have ledto greater understanding of corrosion processes,which, in turn, supports investigation into treatmentmechanisms and methods. This has been extensivelyboosted over the past 15 years by increased applica-tion of modern instrumental analysis in conservation

(a)

(b)

Figure 18f0090 Challenges in metal conservation extend beyond museum interiors to whole sites such as housing at Valparaiso

(UNESCO World Heritage site) and abandoned factory settlements like Stromness (South Georgia) whaling station. Images

courtesy of Eura Conservation Ltd.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

research. Much greater opportunity exists for collab-oration between conservators and corrosion scientiststo develop a synergistic relationship for solving theproblems of preserving ‘metallic heritage.’

p0585 Within conservation there are normally multiplepreservation options, either due to insufficient evi-dence to identify one treatment as being significantlybetter than another or because of differing, butequally valid ethical arguments allow for a range ofconservation routes and outcomes. Thus, within thisreview, large iron chloride-infested naval vesselswere treated in four different ways in attempts tostabilize them. HMS Holland was washed in an inhib-itive solution to extract chloride31; CSS Hunley willbe treated by a newly researched washing methodemploying alkali and pressure12,28,29; HMS Minervawas electrolyzed to remove chloride31; SS GreatBritain used environmental control to prevent corro-sion.11,25,32 All methods are either supported by newresearch or utilize universally applied treatment tech-niques. They also have a strong treatment rationalewith practical arguments to support the course ofaction taken. While there will always be divergingviews and arguments on appropriate ethical strategies,there remains the opportunity to ensure treatmentsare researched and optimized. Conservation is a stagedprocess with systematic treatments. There is initially arationalizing process to decide a course of action,which is then split into treatment stages. As an exam-ple, chloride removal from iron might be followed byan inhibitive treatment, over which a coating is appliedand the result is subjected to long-termmonitoring thatassesses outcomes and provides data for future action.Large-scale complex contextual combinations willchallenge this approach. At Valparaiso ad hoc steelhouses form a world heritage site (see Figure 18).They are lived in and offer the challenge of corrodingsheet steel, galvanized structures and various paintregimes. Besides finding safe, proven treatment proce-dures, complex decisions exist on retention of paint, setagainst a backdrop of what the occupants want andexpect, relative to their standard of living. Should thestructures eventually become a museum or continueto be lived in? Deserted, but equally daunting, arethe remains of a whaling station at Stromness, SouthGeorgia. Remoteness, climate influences, and costoffer significant challenges here. An overarching con-sideration is climate change and how it will influencecorrosion.

p0590 Conservation should define standards of preserva-tion, starting with costly indefinite survival of metalsdown to the guarantee of only a few months

preservation. Developing such a scale goes beyondthe science of preservation to its ethics. It may beargued that national icons providing cornerstones insocieties merit the cost of indefinite preservation, butcommonplace archaeological iron that is to be ana-lyzed, recorded, and published is preserved only forthe duration of the analysis and publication process.Counter arguments would focus on the importance offuture research into material such as archaeologicaliron. Suggesting a scale of importance will necessitateextensive ethical debate about how to define impor-tance. In the modern world of diminishing resources,carbon footprints and concern over environmentalimpact, there are difficult decisions to make andresponsibilities to accept for the action taken.Focused coordinated research in collaboration withcorrosion scientists can contribute towards informingthese decisions for metals preservation.

References

1. Caple, C. Conservation Skills; Routledge: London, 2000.2. Dillmann, P.; Beranger, G.; Piccardo, P.; Mathiesen, H.,

Eds. Corrosion of Metallic Heritage Artefacts:Investigation, Conservation and Prediction of Long-termBehaviour; Woodhead: Cambridge, UK, 2007.

3. Argyropoulous, V. Ed. Metals and Museums in theMediterranean; Protecting, Preserving and Interpreting;TEI of Athens: Greece, 2008.

4. Costa, V. In Conservation Science 2002; Papers from theConference held in Edingburgh, Scotland 22–24 May2002. Townsend, J. H., Eremin, K., Adriaens, A., Eds.;Archetype Publications: London, 2002.

5. icom-cc.icom.museum/Documents/WorkingGroup/Metals/Bromec.

6. Scott, D. A. Copper and Bronze in Art; Corrosion,Colourants and Conservation; Getty Trust Publications,Getty Conservation Institute: USA, 2002.

7. Cronyn, J. The Elements of Archaeological Conservation;Routledge: London, 1990.

8. Pearson, C. Conservation of Marine ArchaeologicalObjects; Butterworths: London, 1987.

9. Ashley-Smith, J. Risk Assessment for ObjectConservation; Butterworth-Heinmann: Oxford, 1999.

10. Selwyn, L. Metals and Corrosion: A Handbook for theConservation Profesional; Canadian ConservationInstitute: Ottowa, 2004.

11. Watkinson, D.; Al Zahrani, A. The Conservator 2008, 31,75–86.

12. Drews, M. J.; de Vivies, P.; Gonzalez, N. G.; Mardikian, P.In Metal 04; Proceedings of the International Conferenceon Metals Conservation, Canberra, 4–8 October 2004;Ashton, J., Hallam, D., Eds.; National Museum of Australia:Canberra, Australia, 2004.

13. Costa, V. Rev. Conservat. 2001, 2, 18–34.14. Cassar, M. Environmental Management; Guidelines for

Museums and Galleries; Routledge: London, 1994.15. Green, L. R.; Thickett, D. In Saving the Twentieth Century;

The Conservation of Modern Materials, Proceedings of aConference Symposium 91 – Saving the Twentieth


CORR: 00172

ELSE

VIER

THIRDPR

OOF

Century, 15–20 September 1991; Grattan, D. W., Ed.;Canadian Conservation Institute: Ottowa, Canada, 1993.

16. Adams, C.; Hallam, D. In Saving the Twentieth Century;The Conservation of Modern Materials, Proceedings of aConference Symposium 91 – Saving the TwentiethCentury, 15–20 September 1991; Grattan, D. W., Ed.;Canadian Conservation Institute: Ottowa, Canada, 1993.

17. Green, L. R.; Thickett, D. Stud. Conservat. 1995, 40, 145.18. Thickett, D.; Bradley, S.; Lee, L. In Metal 98; Proceedings

of the international conference on metals conservation.Draguignan-Figanieres, France, 27–29 May 1998;Mourney, W., Robbiola, L., Eds.; James and James:London, 1998.

19. Leygraf, C.; Graedel, T. Atmospheric Corrosion; WileyInterscience, 2000.

20. Watkinson, D.; Lewis, M. R. T Stud. Conservat. 2005, 50,241–252.

21. Thickett, D.; Odlyha, M. In Conservation of ArchaeologicalMaterials: Current Trends and Future Directions.Williamsburg, Virginia, 13–17 November 2005. AmericanInstitute for Conservation Postprints 9. Washington, DC,in press.

22. Thickett, D.; Luxford, N. In Metal 07, Book 5, Protection ofMetal Artefacts, Interim Meeting of the ICOM-CC MetalWG Amsterdam, 17–21 September 2007; Degrigny, C.,Van Langh, R., Joosten, I., Ankersmit, B., Eds.;Rijksmuseum: Amsterdam, 2007.

23. Weintraub, S. In Object Speciality Group Postprints 9,2002; American Institute for Conservation: Washington DC,2002.

24. Thickett, D.; Luxford, N. The Conservator 2006, 29, 19–34.25. Watkinson, D.; Tanner, M.; Turner, R.; Lewis, M. The

Conservator 2005, 29, 73–86.26. Green, L.; Blackshaw, S. In Metal 95; Proceedings of the

International Conference on Metals Conservation Semuren Auxois, 25–28 September 1995; Macleod, I.,Pennec, S. L., Robbiola, L., Eds.; James and James:London, 1997.

27. Mathias, C.; Ramsdale, K.; Nixon, D. In Metal 04;Proceedings of the International Conference on MetalsConservation, Canberra, 4–8 October 2004; Ashton, J.,Hallam, D., Eds.; National Museum of Australia: Canberra,Australia, 2004.

28. de Vivies, P.; Cook, D.; Drews, M. J.; Gonzalez, N.;Mardikian, P.; Memet, J. B. InMetal 07, Book 5, Protectionof Metal Artefacts, Interim Meeting of the ICOM-CC MetalWG Amsterdam, 17–21 September 2007; Degrigny, C.,Van Langh, R., Joosten, I., Ankersmit, B., Eds.;Rijksmuseum: Amsterdam, 2007.

29. Gonzalez, N.; Cook, D.; de Vivies, P.; Drews, M. J.;Mardikian, P. In Metal 07, Book 3, Use of ElecrtochemicalTechniques in Metal Conservation; Interim Meeting of theICOM-CC Metal WG Amsterdam, 17–21 September 2007;Degrigny, C., Van Langh, R., Joosten, I., Ankersmit, B.,Eds.; Rijksmuseum: Amsterdam, 2007.

30. Selwyn, L. S.; Logan, J. A. In International Council forMuseums Committee for Conservation 10th TriennialMeeting, 22–27 August 1993; Washington, D. C.,Bridgland, J., Eds.; International Council for Museums,James and Jame: Washington, DC, 1993.

31. Barker, D. In Conservation Science Heritage Materials;May, E., Jones,M., Eds.; RSC Publishing: Cambridge, 2006.

32. Watkinson, D.; Lewis, M. R. T. In Materials Issues in Artand Archaeology VII, Symposium Proceedings of theMaterials Research Society of America; Vol. 852;Vandiver, P. B., Mass, J. L., Murray, A., Eds.; MaterialsResearch Society: Warrendale, PA, 2005.

33. Reguer, S.; Neff, D.; Remazeilles, C.; Guilmeinot, E.;Nicot, F.; Pele, C.; Meguelati, M.; Mirambert, F.;

Dillmann, P.; , Refait, P.; et al. In Metal 07, Book 2,Innovative investigation of metal Artefacts; Interim Meetingof the ICOM-CC Metal WG Amsterdam, 17–21 September2007; Degrigny, C., Van Langh, R., Joosten, I.,Ankersmit, B., Eds.; Rijksmuseum: Amsterdam, 2007.

34. Selwyn, L. S.; Argyropoulos, V. Stud. Conservat. 2005, 50,81–99.

35. Keene, S. V In Ancient and Historic Metals: Conservationand Scientific Research; Scott, D. A., Podanay, J.,Consadine, B., Eds.; Getty Conservation Institute: LosAngles, 1991; pp 249–264.

36. Keene, S. V.; Orton, C. Stud. Conservat. 1985, 30, 1–7.37. MacLeod, I. D. Stud. Conservat. 1983, 28, 1–7.38. MacLeod, I. D. Stud. Conservat. 1987, 32, 1–13.39. Thickett, D.; Enderly, C. In The Interface Between Science

and the Conservator; British Museum Occasional Paper No.116, Bradley, S., Ed.; The British Museum: London, 1997.

40. Merk, L. E. Stud. Conservat. 1978, 33, 1522.41. Otiento-Alego, V.; Thurrowgood, D.; Creagh, D. C.;

Bailey, G. In Conservation Science 2002; Papers from theConference held in Edingburgh, Scotland, 22–24 May2002; Townsend, J. H., Eremin, K., Adriaens, A., Eds.;Archetype Publications: London, 2002.

42. Huda, K. Stud. Conservat. 2002, 47, 211–214.43. Costa, V.; Urban, F. Rev. Conservat. 2005, 6, 48–62.44. Degrigny, C. Stud. Conservat. 1995, 40, 227–236.45. Guilminot, E.; Baron, G.; Memet, J. B.; Huet, N.; le Noc, E.;

Roze, J. P. In Metal 07, Book 3, Use of ElecrtochemicalTechniques in Metal Conservation; Interim Meeting of theICOM-CCMetal WG Amsterdam, 17–21 September 2007;Degrigny, C., Van Langh, R., Joosten, I., Ankersmit, B.,Eds.; Rijksmuseum: Amsterdam, 2007.

46. Degrigny, C.; Le Gall, R. Stud. Conservat. 1999, 3,157–169.

47. Degrigny, C. In Saving the Twentieth Century; TheConservation of Modern Materials, Proceedings of aConference Symposium 91 – Saving the TwentiethCentury, 15–20 September 1991; Grattan, D. W., Ed.;Canadian Conservation Institute: Ottowa, Canada, 1993.

48. Hallam, D. L.; Adams, C. D.; Bailey, G.; Heath, G. A InMetal 95, Proceedings of the International Conference onMetals Conservation Semur en Auxois, 25–28 September1995; Macleod, I., Pennec, S. L., Robbiola, L., Eds.; Jamesand James: London, 1997.

49. Bailey, G. T. In Metal 04, Proceedings of the InternationalConference on Metals Conservation, Canberra, October4–8, 2004; Ashton, J., Hallam, D., Eds.; National Museumof Australia: Canberra, Australia, 2004.

50. Mardikian, P. Int. J. Nautical Archaeol. 2004, 33, 137–148.51. Nordgren, E.; Goncalves, P.; Shindelholtz, E.; Brossia, C. S.;

Yunovich, M. In Metal 07, Book 3, Use of ElecrtochemicalTechniques in Metal Conservation; Interim Meeting of theICOM-CC Metal WG Amsterdam, 17–21 September 2007;Degrigny, C., Van Langh, R., Joosten, I., Ankersmit, B., Eds.;Rijksmuseum: Amsterdam, 2007.

52. MacLeod I. D. Conservat. Mgmt Archaeol. Sites 2006, 7,203–223

53. Birchenall, E. C.; Meussner, R. A. In ; Brown, F. B., Ed.;Corrosion and Metal Artifacts: A Dialogue BetweenConservators, Archaeologists and Corrosion Scientists;National Bureau of Standards Special Publication,Vol. 479, 1977.

54. Schmidt-Ott, K.; Boissonnas, V. Stud. Conservat. 2002,47, 81–87.

55. Price, C.; Hallam, D.; Aston, J.; Heath, G.; Creagh, D. InMetal 95, Proceedings of the International Conference onMetals Conservation Semur en Auxois, 25–28 September1995; Macleod, I., Pennec, S. L., Robbiola, L., Eds.; Jamesand James: London, 1997.


CORR: 00172

ELSE

VIER

THIRDPR

OOF

56. Joseph, E.; Letardi, P.; Mazzeo, R.; Prati, S.; Vandini, M.InMetal 07, Book 5, Protection of Metal Artefacts, InterimMeeting of the ICOM-CC Metal WG Amsterdam, 17–21September 2007; Degrigny, C., Van Langh, R.,Joosten, I., Ankersmit, B., Eds.; Rijksmuseum:Amsterdam, 2007.

57. Thickett, D.; Hockey, M. In Conservation Science 2002;Papers from the Conference held in Edingburgh, Scotland,22–24 May 2002; Townsend, J. H., Eremin, K.,Adriaens, A., Eds.; Archetype Publications: London, 2002.

58. Rosales, B.; Vera, R.; Moriena, G. Corros. Sci. 1999, 41,625–651.

59. Cicileo, G. P.; Crespo, M. A.; Rosales, B. M. Corros. Sci.2004, 46, 929–953.

60. Prytulak, G. In Industrial Collections: Care andConservation; Proceedings of a Conference hosted byThe Council of Museums in Wales and The UnitedKingdom institute for Conservation, Cardiff, 9–11 April1997; Dollery, D., Henderson, J., Eds.; CMW and UKIC:Cardiff, 1999.

61. Lins, A. In Dialogue 89 – The Conservation of BronzeSculpture in the Outdoor Environment: A Dialogue AmongConservators, Curators, Environmental Scientists, andCorrosion Engineers; Drayman-Weisser, T., Ed.; NationalAssociation of Corrosion Engineers: Houston, 1992.

62. Wharton, G.; Maish, S. L.; Ginell, W. S J. Am. Inst.Conservat. 1990, 29, 13–31.

63. Cano, E.; Bastidas, D. M.; Argyropoulos, V.; Siatou, A.In Strategies for Saving our Cultural Heritage; InternationalConference on Conservation Strategies for Saving IndoorMetallic Collections, Cairo, 25 February–1 March 2007;Argyropoulous, V., Hein, A., Abdel Harith, M., Eds.;Technological Educational Institute of Athens: Athens,Greece, 2007.

64. Aslund, A.; Bergsten, C. J.; Chiavari, C.; Martini, C.;Niklasson, A.; Prandstraller, D.; Svensson, J. E. InMetal 07, Book 5, Protection of Metal Artefacts, InterimMeeting of the ICOM-CC Metal WG Amsterdam, 17–21September 2007; Degrigny, C., Van Langh, R., Joosten, I.,Ankersmit, B., Eds.; Rijksmuseum: Amsterdam, 2007.

65. Chiavari, C.; Martini, G.; Poli, D.; Prandstraller, D. InMetal 04, Proceedings of the International Conference onMetals Conservation, Canberra, 4–8 October 2004;Ashton, J., Hallam, D., Eds.; National Museum of Australia:Canberra, Australia, 2004.

66. Grabow, N.; Smith, C.; Grissom, C.; Brostoff, L. In Metal07, Book 5, Protection of Metal Artefacts, Interim Meetingof the ICOM-CCMetal WG Amsterdam, 17–21 September2007; Degrigny, C., Van Langh, R., Joosten, I.,Ankersmit, B., Eds.; Rijksmuseum: Amsterdam, 2007.

67. Otieno–Alego, V.; Heath, G.; Hallam, D.; Creagh, D. InMetal 98, Proceedings of the International Conference onMetals Conservation, Draguignan-Figanieres, France,27–29 May 1998; Mourney, W., Robbiola, L., Eds.; Jamesand James: London, 1998.

68. Degrigny, C. Stud. Conservat. 1995, 40, 10–18.69. Luxford, N.; Thickett, D. In Metal 07, Book 5, Protection of

Metal Artefacts, Interim Meeting of the ICOM-CC MetalWG Amsterdam, 17–21 September 2007; Degrigny, C.,

Van Langh, R., Joosten, I., Ankersmit, B., Eds.;Rijksmuseum: Amsterdam, 2007.

70. van Ooij, W. J.; Zhu, D.; Stacy, M.; Seth, T.; Mugada, J.;Gandhi, J.; Puomi, P. Tsinghua Sci. Technol. 2005, 10(6),639–664.

71. Piltz, M.; Romich, H. In Metal 95, Proceedings of theInternational Conference onMetals Conservation Semur enAuxois, 25–28 September 1995; Macleod, I., Pennec, S. L.,Robbiola, L., Eds.; James and James: London, 1997.

72. Seipert, B.; Pilz, M.; Kiesenberg, J. In Metal 98,Proceedings of the International Conference on MetalsConservation, Draguignan-Figanieres, France, 27–29 May1998; Mourney, W., Robbiola, L., Eds.; James and James:London, 1998.

73. Brostoff, R. In Metal 95, Proceedings of the InternationalConference on Metals Conservation Semur en Auxois,25–28 September 1995; Macleod, I., Pennec, S. L.,Robbiola, L., Eds.; James and James: London, 1997.

74. Turgoose, S.; Duncan, S. In Dialogue 89 – TheConservation of Bronze Sculpture in the OutdoorEnvironment: A Dialogue Among Conservators, Curators,Environmental Scientists, and Corrosion Engineers;Drayman-Weisser, T., Ed.; National Association ofCorrosion Engineers: Houston, 1992.

75. Faltermeir, R. T. Stud. Conservat. 1999, 44, 121–128.76. Sharma, V. C.; Shankar, U.; Singh, T. Stud. Conservat.

2003, 48, 203–209.77. Golfimistou, S.; Merkel, J. F In Metal 07, Book 5,

Protection of Metal Artefacts, Interim Meeting of theICOM-CCMetal WG Amsterdam, 17–21 September 2007;Degrigny, C., Van Langh, R., Joosten, I., Ankersmit, B.,Eds.; Rijksmuseu: Amsterdam, 2007.

78. Rahin, A. A.; Rocca, E.; Steinmetz, J.; Kassim, M. J.Corros. Sci. 2008, 50, 1546–1550.

79. Greiner, A.; Hallam, D.; Thurrowgood, D.; Creagh, D.In Metal 07, Book 5, Protection of Metal Artefacts,Interim Meeting of the ICOM-CC Metal WG Amsterdam,17–21 September 2007; Degrigny, C., Van Langh, R.,Joosten, I., Ankersmit, B., Eds.; Rijksmuseum:Amsterdam, 2007.

80. Hollner, S.; Mirambet, F.; Rocca, E.; Steinmetz, J. InMetal 07, Book 5, Protection of Metal Artefacts, InterimMeeting of the ICOM-CC Metal WG Amsterdam,17–21 September 2007; Degrigny, C., Van Langh, R.,Joosten, I., Ankersmit, B., Eds.; Rijksmuseum:Amsterdam, 2007.

81. Rocca, E.; Steinmetz, J. Corros. Sci. 2001, 43, 891–902.82. Dowsett, M.; Adiaens, A.; Scotte, B.; Jones, G.;

Bouchenoire, L. In Metal 07, Book 5, Protection of MetalArtefacts, Interim Meeting of the ICOM-CC Metal WGAmsterdam, 17–21 September 2007; Degrigny, C., VanLangh, R., Joosten, I., Ankersmit, B., Eds.; Rijksmuseum:Amsterdam, 2007.

83. Thurowgood, D.; Sattler, U.; Wehner, K. InMetal 07, Book 1,When Archaeometry and Conservation Meet; InterimMeeting of the ICOM-CC Metal WG Amsterdam, 17–21September 2007; Degrigny, C., Van Langh, R., Joosten, I.,Ankersmit, B., Eds.; Rijksmuseum: Amsterdam, 2007.


CORR: 00172

Non-Print Items

Abstract:The ethics and rationale that underpin the preservation of cultural metals are outlined and theirinfluence on the conservation process is examined via a review of treatment strategies illustrated byexample. Citing current and formative research into treatments illustrates the role of corrosionscience in conservation. Topics that merit further study and research are highlighted whereverappropriate.

Keywords: Conservation; Corrosion; Culture; Heritage; Metals; Preservation; Research; Treatment

Author and Co-author Contact Information:

David WatkinsonConservation SectionSchool of History and ArchaeologyCardiff UniversityCardiff CF10 3EUUK

CORR: 00172

CORR: 00172 - Elsevier...ELSEVIE R PROOF Glossary g0005 Conservation Preservation of cultural...

Documents

Transcript of CORR: 00172 - Elsevier...ELSEVIE R PROOF Glossary g0005 Conservation Preservation of cultural...