Search

This study examined the impact of relative humidity (RH) on the corrosion rate of 129 archaeological iron nails from two sites. Oxygen consumption of individual nails in controlled RH was used as a corrosion rate proxy to deliver quantitative data on corrosion rate as a function of RH. This was negligible at 20% RH, slow up to 40% RH for both sites, and increased rapidly at 60% RH for Roman nails from Caerleon (Wales) and at 70% RH for medieval nails from Billingsgate (London). The nails were digested and their chloride content was determined and related to their oxygen consumption at specific RH values. While a generic pattern of corrosion as a function of chloride was identified, for any single concentration of chloride corrosion rate was not predictable. Desiccation is in common use to control post-excavation corrosion of archaeological iron; quantifying how differing levels of desiccation changed corrosion rate provided a scaled tool for identifying corrosion risk, estimating object longevity, and calculating cost benefit for storage options.

Cast iron cannon balls excavated from the wreck of King Henry VIII flagship, the Mary Rose, have been actively corroding on mixed material display at 55% relative humidity (RH). A Cardiff University study has examined corrosion rates of cannon balls treated by hydrogen reduction, alkaline sulfite and Hostacor IT. Oxygen consumption of six cannon balls has been measured at five relative humidities between 20% and 60% RH as proxy corrosion rate. Results show a noticeable increase in corrosion rate at 50% to 60% RH and also returned differences in corrosion rates of cannon balls based on previous treatment. Display RH of 55% for showcases containing organic materials and cast iron is found to be unsuitable for the cannon balls and their display in these conditions is not recommended. This has implications for contextual display at the Mary Rose Trust and elsewhere. Further study with greater sample numbers is providing statistical evidence of the effect of treatment regime on corrosion rate to inform decision making for future treatment, display and storage.

Painted historical wrought iron commonly occurs outdoors exposed to the prevailing climate. Maintaining protective paint layers is an interventive process that often involves removing existing paint layers and repainting. Whilst prior surface preparation greatly affects the longevity of any paint layer, its impact on the metal surface and paint performance has received limited research within heritage contexts, making their selection anecdotal or driven by manufacturers’ guidelines. Historic Scotland-funded research at Cardiff University is quantitatively investigating the effect of surface preparation methods on wrought iron corrosion rates prior to paint application. The feasibility of using historic sample material in this research has been investigated and is reported. Testing historic wrought iron samples in quantitative studies of corrosion offers more direct linkage to heritage scenarios thus facilitating interpretation of results and extrapolation to real time heritage contexts. The use of an oxygen consumption technique to quantitatively determine the corrosion rate of five samples of historic wrought iron in controlled conditions of 90% relative humidity and 20 oC is reported. Results returned corrosion rates indicating a level of reproducibility that, with an error calculation, will allow corroded historic wrought iron to be used for production of test samples to be employed in experiments designed to determine the impact of surface cleaning techniques on the corrosion rate of corroded heritage iron.

Evidence for tinning on corroded archaeological iron is visible in X-radiographs as distinct thin high density white lines. These are used to guide the removal of overlying iron corrosion to reveal the tin layer. Airbrasion with 53µm aluminium oxide, at low pressure and powder flow, was used to remove iron corrosion incrementally from a tinned medieval key. Appearance of the tin layer visible on the radiograph was aided by optical microscopy. At selected points during the cleaning, SEM-BEI imaging with EDA spot analysis was used to determine if the tinning layer had been reached. Comparing optical assessment with SEM-BEI images revealed that the occurrence of tinning could not be detected with the naked eye aided by optical microscopy. No metallic tin was observed and it is presumed to be present as tin dioxide (SnO2). The presence of tinning was confirmed by SEM-EDA spot analysis. It is likely that parts of the tinning layer were lost during airbrasion. This identifies the risk of removing overlying iron corrosion products to expose tinning layers. Conservation strategies for preserving tinning layers on iron need to be formulated according to evidence that tinning cannot be readily detected by optical microscopy and elemental surface analysis is required to confirm its presence.

Objects from King Henry VIII flagship Mary Rose are displayed in a new museum within thematic mixed material displays. Cast iron cannon balls within a showcase representing the gun-deck of the ship were corroding within the 55% relative humidity (RH) that was maintained to prevent damage to wooden gun carriages and other organics in the mixed material display. Determining the corrosion rates of cannon balls, as a function of RH, provided guidance on how fast the cannon balls were corroding while on display and offered data to indicate how lowering the RH would reduce corrosion and prolong survival of the cannon balls. Using oxygen consumption as a comparative proxy for corrosion rate, cannon balls treated byhydrogen reduction, alkaline sulphite desalination and Hostacor™ IT inhibitor were individually incrementally exposed to 20%, 30%, 40%, 50% and 60% RH. Cannon balls treated by hydrogen reduction exhibited negligible corrosion through the full RH range tested, while those treated by alkaline sulphite corroded faster at all humidity values. Hostacor™ IT returned a slow corrosion rate until 60% RH. Initial results reopen the debate about the use of hydrogen reduction as a treatment for cast iron. It was discredited due to the changes it produced in metallographic structure and the cost of its implementation. Research outcomes provided data for producing guidelines for the safer display of cast iron at the Mary Rose Trust.

Archaeological copper alloy objects form unique corrosion structures according to their manufacture and burial environment. Categorisation has identified protective Type I structures and chloride-containing Type II structures which may lead to bronze disease. Using visual examination to determine which of these exist on objects dictates conservation decisions; unfortunately, there is limited evidence linking surface features directly to corrosion risk. In this study, the chloride content of 39 archaeological copper alloy coins was determined non-destructively using prompt gamma-ray activation analysis. Their surfaces were examined through Raman spectroscopy, reflective transformation imaging and visually using descriptors reflecting Type I, Type II and ‘bronze disease’ surfaces. When subjected to high relative humidity, none of the coins exhibited bronze disease despite containing chlorides and showing characteristics of Type II structures. The results highlight the difficulty of determining risk and taking informed decisions for the management of archaeological copper alloy objects.

Flame cleaning followed by wire brushing is a common treatment for wrought iron; the flame combusts existing coatings and spalls oxides while wire brushing removes any debris, producing a sound surface for recoating. Although frequently applied, little is known about the effects of the treatment on the substrate material and its post-treatment corrosion rate. This study reports scanning electron microscope backscattered electron imaging (SEM-BEI) and oxygen consumption corrosion rate testing of wrought iron flame cleaned by three practitioners. Wrought iron samples treated by two practitioners corrode up to 4 times faster than uncleaned control samples. Samples cleaned by one practitioner exhibit no increase in corrosion rate. Torch fuel type and temperature attained by the iron are identified as parameters potentially contributing to differences in corrosion rates. Comparing oxide morphology of treated and untreated samples reveals extensive cracking and fragmentation following flame cleaning. This offers multiple pathways for ingress of oxygen and water to the metal core to support corrosion. This data simultaneously calls into question the ethics of flame cleaning and offers evidence that a ‘safe’ method exists, prompting further research into this popular treatment.

Brunel’s ss Great Britain ocean going liner is an iconic historic ship located in Bristol (UK). At its launch in 1843 it embodied multiple steps forward in engineering and shipbuilding. It was the biggest ship in the world, built with an iron hull using an innovative box girder hull design that was the forerunner for all future ship design and was screw driven. After an eventful life as a liner, emigrant vessel, sailing ship and finally a floating warehouse in the Falkland Islands, it was returned to Bristol in 1970 to rest in the dry dock in which it was constructed, where it become a visitor attraction. Traditional painting techniques failed to offer sufficient protection from corrosion for the corroded hull open to the elements. The ss Great Britain Trust explored ways to preserve the ship, finally settling on a procedure that involved glazing the dock at water level to provide an enclosed space for controlling corrosion of the lower hull by desiccation, leaving the less corroded upper hull exposed to the atmosphere and maintained by traditional painting methods. This addressed the problem of the instability of the salt infested lower hull area. Research at Cardiff University determined the impact of relative humidity (RH) on the corrosion rate of iron to inform the design of the desiccation plant. Mass gain experiments examined the impact of ferrous chloride and the corrosion bearing ferric oxyhydroxide akaganéite, which were both identified on the hull, on corrosion rate of iron identifying corrosion would be negligible at a target RH of 20%. Plant was designed to maintain this value by desiccating and channelling air around the dock. This paper reports the performance of the plant by looking at the spatial distribution of RH on the surface of the hull, inside it and within the roofed in dry dock area. Temperature control within the dock is a challenge during summer months due to solar gain through the glass roof and the welfare, safety and comfort of visitors entering this area has to be considered, as well as operation of the plant. Relationships between corrosion rate, plant performance, RH and temperature are discussed and set against the cost of maintaining the desiccation levels and the ethical constructs of preserving heritage.

The heritage ironwork community is debating the practical and moral disadvantages of coating wrought iron with two-pack epoxy coating systems versus their potential for corrosion prevention. Their promotion by manufacturers as highly engineered, long-lasting ‘wonder-coatings’ is being called into question and weighed against the issues of cost, requirement for blasting substrates to Sa2.5 (near white metal), a relatively short pot life, reversibility challenges, and the lack of compatibility between the inflexible cured polymer and the dimensionally thermo-responsive ironwork which is reported to cause cracking of the epoxy resin at joints with resultant ingress of water. Currently absent from this discussion is empirical evidence of the impact of epoxy resin coatings on the corrosion rate of wrought iron as compared with other coating types. A project investigating the surface treatment and protective coating of historic wrought iron is under way at Cardiff University with sponsorship from Historic Scotland. The effect of surface preparation methodology prior to coating on the corrosion rate of the uncoated substrate was reported at EUROCORR 2012. Current work aims to provide a cost benefit comparison between the use of two-pack epoxy resin coating systems and oil-based systems. This paper presents the results of corrosion rate investigations and electrochemical impedance spectroscopy (EIS) analyses of samples coated with a commonly used epoxy resin system and two household name oil-based systems. Mid-19th century rolled plate wrought iron samples were prepared by blasting to Sa2.5 (near white metal) with aluminium oxide powder or by wire brushing, as per the coating manufacturers’ recommendations. Similarly prepared samples were contaminated with synthetic seawater (BS3900-F10:1985) to simulate wrought iron exposed to a coastal environment. All three coatings were applied as per the manufacturers’ instructions. The samples were sealed within individual glass reaction vessels containing silica gel conditioned to 90% RH. Depletion of oxygen within the vessels over 12 months was remotely recorded using a World Precision Instruments OxyMini oxygen meter to measure the quenching of fluorescence of an oxygen sensor spot attached to the wall of each vessel. Oxygen consumption of the coatings was determined using controls and subtracted from that of the wrought iron samples to give the corrosion rates of the metal substrates. Comparative performance of the coatings was investigated using EIS to support the real time long-term tests employing oxygen consumption. Interpretation of the outcomes will build towards evidence based advice for contractors choosing coatings for corrosion prevention of heritage wrought iron monuments in coastal and inland contexts.

A real time 5-year outdoor study compares the performance of three paint systems (traditional oil based; acrylic; epoxy resin) applied to naturally corroded wrought iron, with their surfaces prepared to Swedish Standard Sa2.5 or Steel Surface Preparation Standard ST3. Interim data at 2 years reports on gloss, colour change, pull-off testing and EIS measurements. Traditional oil-based paint exhibited significant visual change; a modified alkyl paint underwent minor but visually undetectable changes and an epoxy/acrylic coating remained unchanged. The oxide layer in ST3 controlled adherence of the coatings.

Mill scale, the oxide layer formed on the surface of iron and steel during hot rolling and forging, is well understood for modern steels but less so for ancient and historic ferrous material. Recent studies on atmospheric corrosion of historic wrought iron reinforcements in French gothic cathedrals identified surviving mill scale in corrosion layers and studied its effect on corrosion processes. Anecdotal evidence from conservators working on heritage iron also attests to the presence of mill scale and they report that it has a protective effect on the wrought iron substrate. Better understanding of the properties of mill scale will lead to improved conservation rationales and the design of evidence based procedures. Determining whether mill scale represents the original surface of corroded ironwork, if it survives only as fragmented layers and whether its presence influences corrosion rate will develop this understanding, as will determining its impact on the adhesion of protective coatings applied over it. Surface preparation advice from coating and steel manufacturers and the use of industrial standards conflict with the experience of heritage ironworkers and conservation ethics. Removing all oxide layers to ‘optimise’ performance of protective coating systems inevitably leads to complete loss of the information contained therein, yet the evidence that removal of oxide layers reduces post coating corrosion rates is limited. Work at Cardiff University aims to characterise mill scale on wrought iron and mild steel samples and assess how its removal impacts on corrosion rates and the adhesion of coatings. The morphology and composition of mill scale on 2nd century AD archaeological forged nails, mid/late 19th century rolled wrought iron bar and plate and modern mild steel samples have been investigated. Location and thickness of ‘mill scale’ layers and their degree of continuity have been determined using optical microscopy and SEM-EDS. X-ray diffraction has been used to characterise mill scale in the context of the corrosion profiles on un-cleaned, cleaned to mill scale, and cleaned to bright metal samples. The results give insight into the nature of mill scale on archaeological and historic wrought iron produced using differing technologies. The corrosion behaviour of these samples is being investigated using Electrochemical impedance spectroscopy and corrosion rates derived by measuring oxygen consumption of samples sealed in controlled high relative humidity environments. Effect of mill scale retention on adhesion of coating systems to samples cleaned to mill scale and bright metal is tested using a standard hydraulic pull-off rig. The decision whether to remove or retain mill scale during a conservation process can now be made with an enhanced understanding of its impact on corrosion rate and coating performance.

Akaganéite (β-FeOOH) corrodes iron in contact with it due to its hygroscopicity and the mobility of its surface adsorbed chloride.1,2 The high hygroscopicity of β-FeOOH means corrosion can occur down to 15% relative humidity.3 Aqueous washing removes either some or all surface adsorbed chloride, according to the washing process and its temperature, reducing the ability of β-FeOOH to corrode iron.1 Chloride occluded in the crystal structure of β-FeOOH, does not corrode iron since it is trapped in the Hollandite tunnel structure of the crystal.1,4 Research into its properties of β-FeOOH mostly employs synthesised samples, produced either by precipitation or solid state corrosion, as both processes mirror how it might be formed on objects. Since sufficiently pure naturally formed β-FeOOH is difficult to obtain in quantities suitable for use in experiments, it is often asked whether synthetically produced β-FeOOH reflects how the naturally formed product impacts on corrosion of archaeological objects. This study compares natural β-FeOOH from a 1st century AD Roman shipwreck at Saintes-Maries-de-la-Mer, with β-FeOOH, synthesised by acid hydrolysis of FeCl3 solution5 and by a solid state method of FeCl2.4H2O/iron powder mixture to high humidity6 (Turgoose 1983). Their crystal structure is identified, the impact of each on the corrosion rate of iron was determined quantitatively and their chloride content was measured. Corrosion rates are compared to recently published data on the corrosion rates of over 100 archaeological iron objects.7

Controlling post-excavation corrosion of chloride (Cl-) infested archaeological iron presents a major storage problem for museum professionals. This corrosion is stimulated by Cl- containing corrosion products, such as akaganeite (βFeOOH), whose hygroscopicity and mobile surface adsorbed Cl- corrode iron at low relative humidity (RH) 1. Renewed interest in inhibitors to control corrosion of archaeological iron has centred on low toxicity eco-friendly compounds. Recent study suggests that the extract of Emblica officinalis (Indian gooseberry) has inhibitive properties for iron2. This is attributed to the high level of the strongly hydrolysable gallic (C6H2(OH)3COOH) acid in E. officinalis extract dissolving βFeOOH to produce Fe2+, which it complexes to form magnetite (Fe3O4) covered in iron gallate. This transformation releases Cl- ions from βFeOOH into solution, thereby removing electrolyte ions that drive iron corrosion and interfere with the effectiveness of many inhibitors. Simply cold washing βFeOOH removes most of its Cl- and significantly slows its ability to corrode iron at high RH3. The study reported here examines quantitatively the impact of washing βFeOOH with various aqueous concentrations of E. officinalis, over differing time periods, to assess the ability of βFeOOH to corrode iron. βFeOOH transformation was investigated using FTIR and Micro-Raman spectroscopy. βFeOOH was cold washed with selected aqueous concentrations of E. officinalis juice for various time periods and dried. Controls involved washing βFeOOH solely in water. Washed βFeOOH samples were individually mixed with Fe powder and sealed into their own dedicated reaction vessel containing silica gel conditioned to 80% RH and an oxygen sensor, which recorded oxygen concentration via light projected through the vessel wall. Oxygen depletion was used to represent Fe corrosion rate. Results revealed the impact of E. officinalis on the corrosion of iron by βFeOOH, offering insight for optimising concentration and wash time.

Museums worldwide house collections of archaeological copper alloy (CuA) objects that are potentially unstable due to incorporation of cuprous chloride (CuCl) formed in the burial environment. With sufficient atmospheric moisture, post-excavation hydrolysis of CuCl, a process often referred to as ‘bronze disease’, forms voluminous basic copper chlorides (Cu2(OH)3Cl). These disrupt overlying patinas, degrading the artistic, aesthetic and technological value of collections. To prevent the formation of Cu2(OH)3Cl and corrosion of unstable archaeological CuA objects, relative humidity (RH) levels around artefacts are often controlled to below published thresholds. Maintaining RH can be time consuming and resource-intensive, therefore a thorough understanding of the risk to unstable objects in a given environment is critical to produce sustainable management procedures tailored to preservation aims and budgets. Current guidelines for safe storage provide a range of RH targets without quantified evidence of corrosion risk relative to RH. This lack of definitive guidance has led to highly variable practices internationally in the management of desiccation for collections. This paper presents the results of AHRC SWWDTP funded doctoral research which is building an understanding of the risk associated to unstable objects at different RH levels. To simulate post-excavation corrosion processes occurring in unstable objects, Cu and CuCl powder mixtures have been subjected to the RH range 20-90% and their corrosion rates measured quantitatively using the Cardiff University oxygen consumption technique1,2,3. X-ray Diffraction and Raman spectroscopy have been used to examine the composition of the powders and the corrosion products formed. The results provide insight into the corrosion process of unstable objects and illustrate the efficacy of using RH to control post-excavation corrosion. A solid baseline for undertaking further measurements has been formed which will inform future management guidelines.

Effective management of heritage assets relies on decision-making which is underpinned by empirical evidence of impact of treatments on long term survival prospects of materials. Historic wrought iron presents a particular problem for conservation. It occupies a niche position between heritage and engineering, is frequently exposed to outdoor atmospheric corrosion and, in the case of bridges, gates and similar structures, may be required to perform a distinct function. Sector guidance to direct practices is based on anecdotal evidence and established methods. British Standards relate to modern steels hence application to historic ferrous metals is complicated by differences in metallurgy and lack of concession to conservation ethics.\ud This study generates empirical evidence of the effects of five surface preparation methods and three protective coating systems on the corrosion rate of historic wrought iron samples. Immersion in sodium hydroxide solution and blasting with crushed walnut shells are found to reduce corrosion rates of uncoated wrought iron. Aluminium oxide and glass beads blasting increase corrosion rate but offer removal of contaminants and a keyed surface for coating adhesion. Flame cleaning increases corrosion rate by almost four times the uncleaned wrought iron corrosion rate.\ud A two-pack epoxy resin coating system with polyurethane topcoat applied over substrate surfaces blasted to Sa2.5 (near white metal) and a surface tolerant single-pack alkyd coating applied over coherent oxide layers successfully prevented corrosion for almost two years in high static relative humidity environments. An alkyd system applied over Sa2.5 blasted surface does not significantly reduce corrosion rate of the uncoated substrate. A cost benefit approach to interpreting the empirical results in relation to practicalities of applying the treatments is advocated.\ud The methods developed for standardising historic sample material and measuring oxygen consumption of coated samples as proxy corrosion rate offer scope for further work in this area. A standardised approach to testing permits correlation of test data between workers in this area to generate a database of empirical data to inform decision-making.

Heritage preservation often seeks to fulfil an unrealistic goal of stopping decay, whereas controlling decay rate is a more realistic option. To develop pragmatic management based on limiting decay of heritage material requires predictive strategies. These can only be developed by quantifying the decay rates of heritage materials as functions of their intrinsic and extrinsic factors. This data facilitates prediction of object longevity and incremental loss of heritage value, which provides management with an evidence base for decision making that maximises use of resources and provides cost benefit. This evidence based management approach and the challenge it faces are discussed by relating the corrosion rate of archaeological iron objects to management strategies for its preservation. A combination of quantitative analytical methods developed a strong evidence base for managing the preservation of archaeological iron. The corrosion rates of over 100 archaeological iron objects were determined as a function of relative humidity by recording the oxygen consumption of individual objects. This data was then qualitatively linked to the physical damage that corrosion produced in objects using criteria anchored assessment and was quantitatively related to object chloride content using digestion. Predictions for loss of heritage value as a function of corrosion could then be made. Prompt Gamma Activation Analysis (PGAA) at the Budapest Neutron Centre was later employed to link chloride distribution within objects to the physical damage produced by corrosion to facilitate refinement of the predictive corrosion model. English Heritage has published sector guidelines for storing archaeological iron using this research.

Controlling the post-excavation corrosion of archaeological iron nails infused with chloride ions from their burial environment is a longstanding problem. While desiccation is proven to prevent corrosion, it requires significant outlay and ongoing management for its continued success1. Desalination enhances stability but treatment effectiveness is unpredictable2. Inhibitors potentially offer a labour saving cheap solution for long-term storage of archaeological iron objects, with the ideal treatment being immersion in the inhibitor to deliver an object that is stable in aggressive relative humidity (RH). Unfortunately, inhibitors may change object aesthetics, pose health risks and there is limited quantitative evidence-based assessment of their effectiveness on archaeological iron. Recently, attention has turned to low toxicity natural inhibitors. Linear sodium carboxylates have shown effectiveness on modern steel analogues and archaeological surfaces when tested by electrochemical immersion techniques3. This study determined quantitatively the impact of a standardised dodecanoic acid treatment on the corrosion rate of 20 archaeological nails from the Roman site at Caerleon (South Wales). Prior to treatment, individual corrosion rates were determined by sealing each nail in its own reaction vessel with the interior controlled to 80% RH, then remotely recording oxygen depletion inside the vessel via light striking an oxygen sensor within it. Treatment of each nail involved an aqueous 0.1M dodecanoic acid solution, with pH adjustment to manipulate solubility and prevent gelling. Each nail was immersed in its own 100ml dodecanoic acid solution for 24 hours followed by controlled 12 hour desiccation prior to re- measuring oxygen consumption rates at 80% RH. No object ceased to consume oxygen. Seventeen nails (85%) returned lower oxygen consumption rates but with no discernible pattern, as both major and minor rate reductions occurred. Three nails (15%) had increased corrosion rate. The nature, form and location of chloride in archaeological iron and corrosion layer composition and morphology introduce a wide range of sample variables that are considered when discussing outcomes. Although reproducibility limitations exist, this methodology offers an evidence based route for predicting the success of an inhibitive system in its working context.

Akaganeite (βFeOOH) occludes chloride within its crystal structure and adsorbs it onto it surface during its formation. βFeOOH was synthesised by solid-state and aqueous precipitation routes. The amount of adsorbed chloride removed by a single aqueous wash was measured. The impact of the synthesised βFeOOH on the oxidation rate of iron was determined by using a remote oxygen sensing technique to record oxygen concentrations within a sealed vessel, internally controlled to 80% relative humidity and containing mixture of βFeOOH and iron powder. The oxidation rate of iron was directly related to the mass% of adsorbed chloride removed by the aqueous wash. This contributes to understanding of both iron corrosion and the development of conservation treatments.

Upon excavation archaeological iron normally contains soluble chloride, making it highly unstable in its post-excavation environment. Long-term treatments aimed at controlling archaeological iron corrosion include aqueous washing methods designed to remove soluble chloride. Most often these involve using either NaOH or NaOH/Na2SO3 to create alkaline conditions that aid the removal of chloride ions. Chloride occurs within objects in a number of forms: as counter ions in solution to balance Fe2+ produced at anodes on the metal surface when sufficient moisture is present in the object; as highly soluble FeCl2.4H2O when conditions are sufficiently dry; adsorbed onto insoluble βFeOOH and occluded in its crystal structure; adsorbed onto αFeOOH in small amounts (0.2% maximum). While soluble chloride bearing compounds can cause corrosion in high relative humidity (RH), hygroscopic compounds such as FeCl2.4H2O and βFeOOH are problematic at low humidity. Mobile surface adsorbed chloride on hygroscopic βFeOOH allows it to corrode iron at humidities as low as 15% RH and FeCl2.4H2O corrodes iron to 20% RH. While FeCl2.4H2O readily solvates in water or NaOH (aq), washing βFeOOH in deionised water for a few hours has been shown to remove most of its soluble surface adsorbed chlorideand greatly reduces its ability to corrode iron.1 The work here explores how short term washing βFeOOH in NaOH solutions of differing molarities impacts upon its ability to corrode iron at high RH. A range of NaOH molarities were used to provide differing hydroxyl ion concentrations in the wash solutions, which were standardised to run for 24 hours per sample. The corrosion rate, at 80% RH, of iron powder mixed with washed βFeOOH samples was recorded by measuring oxygen consumption of samples in reaction vessels, using a remote sensing technique. Unwashed βFeOOH samples and βFeOOH washed in deionised water were used as controls and comparators. The amount of chloride removed in each 24 hour wash was measured and related to the corrosion rates of iron mixed with the washed βFeOOH samples. Thedataset was used to identify relationships between removal of chloride from βFeOOH as a function of NaOH molarity and subsequent changes to its ability to corrode iron. A complex set of variables, which include morphology and composition of the corrosion product layers, chloride location and form and concentration, mean optimum removal of chloride from archaeological iron by alkaline wash treatments can take months. This study adds to understanding of such treatments by offering clear guidance on how short-term washes of archaeological objects in NaOH impact on the ability of insoluble βFeOOH to corrode iron.

Corrosion profiles extant on archaeological copper alloys from aerated marine contexts often comprise CuCl beneath or intermixed with a Cu2O layer, which is normally overlaid by a range of green blue Cu2OH3Cl polymorphs. Depending on prevailing conditions, other chloride bearing compounds and basic copper carbonates, may also be present. Aesthetically pleasing patinas can be disrupted by post excavation hydrolysis of CuCl producing voluminous Cu2OH3Cl polymorphs. This can be controlled by using the inhibitor benzotriazole, except in the presence of low pH caused by hydrolysis. Alternatively, the removal of soluble chlorides (Cl-), using various aqueous alkaline wash solutions, offers an approach to fulfilling the aim of controlling post-excavation corrosion. Enhancing washing effectiveness by using subcritical conditions, that are claimed to aid desalination, has been experimentally studied for marine archaeological iron but has received limited attention for copper alloy desalination. This research investigates the potential of subcritical fluid technology to release Cl- and transform compounds that are commonly present on marine archaeological copper alloys. The impact of the treatment variables pH, temperature and time on analogue pressed pellets of the commonly occurring corrosion products, atacamite (CuCl2OH3Cl), clinoatacamite (CuCl2OH3Cl), malachite Cu3(CO3)2(OH)2, nantokite (CuCl) and cuprite (Cu2O) was studied. Sets of analogues comprising 3 pellets were treated in subcritical conditions at pH8 (NaHCO3) and pH10 (NaCO3) for up to 480 minutes, with analysis of the concentration of Cl- and pH values in the eluate at time intervals of 15, 30, 45, 60, 120 and 480 minutes. The surface and cross section of samples was analysed with SEM-EDS and ground pellet samples with XRD to detect whether transformation had occurred. To translate this into contexts for treating naturally forming corrosion product profiles, analogues were produced by corroding copper alloy coupons in a brackish environment for 12 months. Finally, real artefact samples were treated. These tests used the same subcritical methodology employed for the pressed corrosion product pellets, recording pH and Cl-concentration at the same time values. Patina and corrosion layer characteristics were recorded before and after treatment using digital photography. The outcome offered insight into the combination of variables that provided the most effective Cl- extraction from Cl- bearing pellets and identified transformations that would impact on colour change. Aesthetic impact was determined by the appearance of the coupons and real artefact samples pre and post treatment. This indicated that the change caused by the most effective parameters for Cl- extraction would likely be unacceptable in any practical context for archaeological objects.

In high humidity βFeOOH corrodes iron in contact with it, likely due to the mobile chloride on its surface providing an electrolyte. Studies have shown that washing βFeOOH does not entirely remove chloride, which remains occluded in its crystal structure. Since occluded chloride is not mobile in water it follows that washed βFeOOH should not corrode iron in contact with it. This study reports washing naturally formed βFeOOH and synthetic βFeOOH formed by the Fe/ferrous chloride protocol using a Soxhlet wash system and by stirring in cold water until no further chloride extraction occurs. The corrosion of iron powder mixed with βFeOOH is recorded as oxygen consumption in sealed reaction vessels at 60%, 70% and 80% RH and this is compared to corrosion rates of unwashed βFeOOH at the same RH values. Desalination methods are employed within conservation to remove chloride from archaeological and historical iron. The outcomes of this study offer insight into the impact of this washing process on the post-treatment corrosion of iron by washed βFeOOH and provides guidance on whether temperature differences during treatment will impact on the removal of surface adsorbed βFeOOH. It also offers data on the amount of chloride adsorbed onto the surface of naturally formed and synthetic βFeOOH and whether this influences the rate of corrosion of iron.

Corrosion of archaeological metals, particularly iron and copper alloy artefacts, is an ongoing problem for conservation and collections care. If not managed, corrosion can lead to reduction in value or complete loss of artefacts and collections. This paper presents the results of a long-term research programme at Cardiff University which investigated corrosion rates linked to humidity levels and best practice in creation of desiccated microclimates for corrosion prevention. Surveying sector practices in the post-excavation storage of archaeological metals has revealed the complexity of the decision-making process and a distinct lack of evidence-based guidance to direct protocols. Immediately post-excavation, free water in corrosion product layers can create high humidities and drive destructive electrochemical corrosion. Advice on drying techniques is limited and conflicting, leading to ad hoc practices and consequent danger to objects. Once dry, chloride-bearing compounds mean archaeological and marine iron artefacts can remain unstable down to 15% relative humidity (RH). Therefore, for most museums and archaeological units, long-term corrosion control is by desiccated storage reliant on creating and maintaining low RH microclimates in plastic boxes. Success of these microclimates is driven by air exchange rates of boxes which are in turn dictated by box design and size. Along with the mass of silica gel included, these variables determine the lowest RH achievable and its longevity. Without evidence of the influence of these variables, effective management of storage procedures is impossible. This paper delivers new data on the influence of post-excavation drying, storage box variables, mass of silica gel and gel regeneration cycles in successful creation of desiccated microclimates for medium and high RH external store environments. Combining this with corrosion rate data for iron and copper alloy objects between 20-80% RH allows predictions to be made about the risk to artefacts of following a range of common protocols. Guidance on best-practice drying and storage procedures to minimise corrosion and enhance object longevity are now offered to the heritage sector. The research updates previous, generic guidance on storage box selection and silica gel use. Results of surveying practice indicate that the go-to guidance remains First Aid for Finds, the most recent edition of which was published in 1998. Advice on silica gel per volume of box in that publication was based on contemporary practice rather than evidence-based data and no guidance on box selection was offered beyond the ubiquitous Stewart Sealfresh. The synergy of conservation science and practice reported here combines laboratory experimentation using climatic chambers, oxygen consumption corrosion rate testing and air exchange measurements with an extensive survey of sector practice and close liaison with end users to produce pragmatic guidelines for practitioners and managers. Supporting cost benefit decision-making in storage box selection and silica gel regeneration cycles, these guidelines will allow managers of archaeological metalwork collections to design bespoke storage protocols which have the potential to extend lifetimes of collections. Assessment of risk to objects can be weighed against hardware and human resource costs and variables manipulated to design workable, case-specific solutions to a widespread problem.

Corrosion of archaeological copper alloys during terrestrial burial normally produces a corrosion product profile comprised of a mixed layer of either CuCl/Cu2O or SnO2/CuCl/Cu2O for bronzes, which is overlaid by a corrosion product layer that is normally predominantly comprised of CuCO3CuOH2. If CuCl has been hydrolysed post excavation, Cu2OH3Cl polymorphs may also be present. Aesthetically it is important to expose the shape of an object and to either retain or expose the colour and texture of patinas. Mechanical removal (cleaning) of the outer layers to reveal shape and enhance aesthetics can be labour intensive, while chemical methods are difficult or impossible to control. Laser generated energy potentially offers a controlled method for achieving ethical and aesthetic goals but it’s ability to remove or transform commonly encountered copper corrosion products is unknown. The energy a laser produces and its delivery to surfaces is governed by a wide range of variables. Their individual and collective impact on copper corrosion products must be known for laser treatment of archaeological copper alloys to employ evidence-based operating parameters, which deliver predictive outcomes. This project examined the impact of a Q-switched Nd:YAG 1064nm laser on selected individual corrosion products and their mixtures, which typically occur in corrosion profiles found on archaeological copper alloys. The analogues comprised: CuCO3CuOH2; CuCl; Cu2O; 50/50 CuCl/CuCO3CuOH2mix; and metallic copper. Results made it possible to predict the likely impact of laser ablation on patinas found on archaeological copper alloys. During the experiment only the amount of fluence (energy dispersed over cm2) was controlled, as the same fluence can be replicated when using other lasers, by adjusting energy and spot size. Analysis of the experimental outcomes employed Raman and FTIR spectroscopy to detect transformations of copper compounds and colourimetry to assess colour changes. This data offers insight into both physical and aesthetic changes that could occur from the use of lasers on archaeological objects.

Within conservation practice a few specific coatings are utilised for a very wide range of applications, based mostly on sketchy evidence of their long term performance, ageing properties, ethical factors such as reversibility and aesthetic considerations that include transparency. Hence the acrylic copolymer Paraloid B72 and the microcrystalline wax Cosmolloid 80H are extensively used as indoor and outdoor coatings across a broad spectrum of metals from archaeological to historical contexts. Evidence of their performance is largely qualitative, based on observations made in environments with uncontrolled and often unmonitored variables. This approach is suited to recording performance in real life storage and display contexts but it lacks reproducibility and does not deliver insight into how specific variables such as relative humidity, application methodology, surface preparation and contaminants impact on coating performance. This study examines the performance of Paraloid B72, Cosmolloid 80H and a silane polymer Siliglide 10, as protective coatings on modern steel continuously exposed to fixed high humidity. Test samples were cut from the rear door of a Saracen armoured personnel vehicle circa 1950’s. Coating procedure was carefully controlled and assessed to offer both standardisation and alignment with commonly used application procedures in the heritage sector. Similarly, surface preparation was uniform and standardised. Both uncontaminated and chloride contaminated surfaces are used as test platforms for the coatings. Long exposure periods of months to years were employed. Quantitative data is collected by measuring the oxygen consumption of individual samples in controlled relative humidity. The rationale was to deliver evidence of how these coatings may perform in conditions where relative humidity can reach high levels but there is no direct exposure to rain, thus including open shelters and uncontrolled stores used to store historic and industrial objects. The research programme aimed to deliver practical outputs for project partners and end users at the Tank Museum Bovington UK and Historic Scotland. The Arts and Humanities Research Councils sponsored the study. Results to date indicate that all three coatings perform similarly on surfaces that are cleaned. On surfaces that have undergone chloride contamination post-cleaning the silane was best at preventing corrosion, measured by oxygen consumption followed by Paraloid B72 then Cosmolloid 80H.

Archaeological iron excavated from damp soil usually contains free water within corrosion product layers. This water can create high humidity environments around objects post-excavation and form an electrolyte which enables electrochemical corrosion processes to occur. Archaeologists and conservators frequently store freshly excavated objects in plastic boxes and make decisions as to whether to dry objects before storage and whether to use silica gel to desiccate the environment. Guidelines on best-practice protocols for drying and immediate post-excavation storage are limited and conflicting in the advice they offer, leading to ad-hoc practices, no standardisation of procedures and consequent danger to objects. Once dry, chloride-bearing compounds within corrosion product layers can cause iron objects to remain unstable down to 15% relative humidity (RH). For most museums and archaeological units, effective, long-term corrosion control is by desiccated storage which relies on being able to create and maintain low RH microclimates in plastic boxes. Variables driving the success of microclimate creation and retention are the air exchange rate of the box and the mass of dry silica gel contained. These are affected in turn by factors inherent in box design and size. Data generated at Cardiff University shows that generic recommendations for storage box selection do not reflect differences in their performance. This paper delivers new data on the influence of post-excavation drying, storage box variables and mass of silica gel on the possibility of creating desiccated microclimates for immediate and long-term safe storage of archaeological ironwork. Combining this with corrosion rate data for iron objects at humidities from 20-80%RH allows predictions to be made about the risk to objects from following a range of common protocols. Maintenance regimes for regeneration of silica gel can be projected and cost-benefit assessed based on balancing humidity and risk against staff resources. Guidance on best-practice drying and storage procedures to minimise corrosion and enhance object longevity can now be offered to archaeologists, conservators and the wider museum sector.