Next Article in Journal
Modeling Salt Behavior with ECOS/RUNSALT: Terminology, Methodology, Limitations, and Solutions
Next Article in Special Issue
Comparison of the Use of Traditional Solvents and Nanosecond 213 nm Nd:YAG Laser in Thinning Naturally Aged Varnish on a Contemporary Oil Easel Painting
Previous Article in Journal
Contemporary Collecting in a Pandemic: Challenges and Solutions for Documenting the COVID-19 Pandemic in Memory Organizations
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Heritage
Volume 5
Issue 4
10.3390/heritage5040189
heritage-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

What You Clean Is What You Get: A Novel Chemical Cleaning Technique and the Interpretation of Corrosion Products Found in Late Roman Copper Alloy Coins Retrieved from the Sea

by
Maayan Cohen
1,*,
Alexandra Inberg
2,
Dana Ashkenazi
3 and
Deborah Cvikel
1
1
Leon Recanati Institute for Maritime Studies and the Department of Maritime Civilizations, University of Haifa, Haifa 3498838, Israel
2
School of Electrical Engineering, Tel Aviv University, Ramat Aviv 6997801, Israel
3
School of Mechanical Engineering, Tel Aviv University, Ramat Aviv 6997801, Israel
*
Author to whom correspondence should be addressed.
Heritage 2022, 5(4), 3628-3647; https://doi.org/10.3390/heritage5040189
Submission received: 2 November 2022 / Accepted: 18 November 2022 / Published: 21 November 2022
(This article belongs to the Special Issue Cleaning Strategies for Cultural Heritage)
Browse Figures
Review Reports Versions Notes

Abstract

:
Thirteen Late Roman copper alloy coins with a dark concretion layer from the Early Islamic period Ma‘agan Mikhael B shipwreck were chosen to undergo an experimental chemical cleaning and polishing procedure for removing the concretion while limiting the damage to the surviving metal. These coins, and two more without concretion discovered on the beach nearby, were then subjected to a series of non-destructive analyses–visual testing, XRF, multi-focal light microscopy, and Raman spectroscopy–to determine their state of preservation, identify their corrosion products, and acquire information regarding their core material. An additional coin was examined by destructive metallographic light microscopy and SEM-EDS analyses to gain further information concerning the concretion cover. Preservation varied: For some chemically cleaned coins, a shiny orange-coloured metallic surface was exposed, while others were poorly preserved. Moreover, evidence of the stamping process was also observed. The results show that the suggested chemical cleaning treatment could be useful for processing other copper alloy objects retrieved from underwater environments; we propose a 12-step methodology to this effect.

1. Introduction

Corrosion is an inevitable and irreversible process that begins immediately after a metal object’s production [1]. Over time, the corrosion of such objects becomes extensive, and its removal affects their appearance, preservation, and conservation. Improper cleaning may lead to destruction [2,3], and the removal of corrosion products may cause compositional changes on the object’s surface [3,4]. Therefore, the choice of suitable cleaning methods is a matter of considerable significance. In this paper, we use 16 coins from the Ma‘agan Mikhael B (MMB) shipwreck, Israel, to examine a novel chemical cleaning and polishing procedure developed for the removal of the concretion cover typical to archaeological copper objects retrieved from sea environments.

1.1. Corrosion of Ancient Copper Alloy Objects Retrieved from Marine Environments

The corrosion processes of copper alloy objects involve two principal mechanisms: (1) the object’s interaction with the surrounding elements and (2) the dissolution of its copper alloy [5,6,7,8]. However, in practice, numerous conditions and circumstances can intervene, either boosting or hindering corrosion. For instance, ancient manufacturing procedures that entail repeated cycles of thermo-mechanical treatment often result in crystallised and segregated impurities, causing the object’s mechanical properties to deteriorate and accelerating intergranular corrosion [6]. On the other hand, under most circumstances, the addition of tin to copper alloys in antiquity has been shown to curb corrosion, although, under certain conditions, it could have the opposite effect [9]. Either way, these processes often result in an oxidation patina (normally reddish-brown cuprite) covered by a crust of green malachite mineral and various silicates [10].
Although copper alloys have excellent corrosion resistance due to the formation of stable protective cuprite (Cu2O) oxide film, ancient copper alloy objects retrieved from sea environments tend to interact with their sea water surrounding through the years and therefore suffer from long-term corrosion processes [5,6,11]. Such objects are often retrieved covered with a thick encrustation and concretion coating.
According to MacLeod (1989), there are three different types of concretions that are common to ancient copper objects retrieved from shipwrecks. The first type is a mixture of shell fragments and coralline material, which is rather porous and is marked with green copper corrosion products. The second type is an aerobically formed concretion, which is rather dense and is covered with quite uniform thin layers of cemented calcareous materials composed of compounds of calcium and magnesium. Such concretion is typically marked with red-brown corrosion products and may be moderately covered with colonizing materials such as molluscs and oysters. When the second type of concretion is mechanically removed with a flat hammer, a well-preserved copper object is usually revealed. The third type is related to an anaerobic environment or to a heavily polluted atmosphere. For example, the source of sulphide ions may be an activity of naturally occurring bacteria. This type of concretion is dense and is often covered with sand particles and rock fragments. The third type of concretion has a black-grey to deep navy-blue appearance, as is typical of copper sulphide minerals. If a buried copper object has been subjected to an underwater environment with major circulation, for example as a result of strong waves and storms, it is possible to find copper items covered with mixed types of concretions (combination of aerobic and anaerobic corrosion products), resulting in green, red-brown, black, grey, and deep blue layers [12].
The corrosion products of retrieved copper alloy objects may include different compounds, such as tenorite (CuO), cuprite, romarkite (SnO), cassiterite (SnO2), lead (II) oxide (PbO), lead dioxide (PbO2), cerussite (PbCO3), lead (II, IV) oxide (Pb2O3), and lead triplumbic tetroxide (Pb3O4), where each compound has its characteristic colour [5,6]. The corrosion rate and mechanisms of such copper alloy objects buried in underwater environments are complex processes that depends on numerous factors, such as the composition of the alloy, the microstructure of the object, its manufacturing process, water temperature and pH, and other factors [5,11]. Hence, the study of corrosion products of ancient copper alloy coins buried in a marine environment, characterized by the development of a multi-layered surface, has importance in achieving a better cleaning process of the archaeological copper alloy objects retrieved from sea environments [5].

1.2. Removal of Concretion Cover and Corrosion Layers from Archaeological Copper Alloy Objects Retrieved from Marine Environments

Numerous procedures for removing corrosion layers from archaeological copper alloy objects have been devised to date. Thus, for most terrestrial circumstances, our knowledge of procedures and conditions of copper corrosion is sufficiently robust to support a well-informed choice of an appropriate cleaning technique that properly balances the risks and benefits [13]. The most common are mechanical stripping methods (e.g., sandblasting), which are highly effective but often also entail surface damage [10]. Other widespread methods include chemical cleaning, electrolytic reduction, and alkaline dithionite galvanic cleaning [14]. Driven by the demand for minimal destruction, more exacting methods have been sought. Notable among them is laser cleaning, which can be closely controlled and selectively implemented [10,15]. For example, Pini et al. applied laser cleaning to archaeological copper alloy objects [15], and Drakaki et al. used this method to clean a collection of Roman copper alloy coins [10]. However promising, this procedure requires expensive and highly specialised equipment–high-power laser devices with different irradiation conditions and pulsed-emission laser systems–in order to generate a selective, precise, and unintrusive cleaning process [10,15,16,17]. Furthermore, for best results, laser interaction with the copper alloy must be brief to avoid heat conduction into the object’s metal core [10].
However, the above cannot be said for copper alloy objects deriving from marine environments. Very few studies focus on the cleaning process of ancient copper alloy objects retrieved from marine environments, and those that do are usually preoccupied with mechanical procedures [18,19]. For example, Iddan et al. cleaned a bronze powder chamber from an underwater excavation in Akko with abrasive brushing [19]. Consequently, copper objects retrieved from underwater settings do not enjoy the same broad knowledge base and even-headed treatment as those from terrestrial environments. They are more likely to emerge from the cleaning process with damaged surfaces or remaining residue due to over- or underapplication, respectively.
Coins recovered from the MMB shipwreck are covered with black, bluish, or green concretions, about 1 mm thick. These concretions obscure historically significant features and call for cleaning, which is usually conducted by sandblasting [5,18,19]. However, the results are often partial, especially when the concretion is hard, dense, and heterogeneous [5,18]. Accordingly, in this paper we draw on 16 coins from the MMB shipwreck to present and test a chemical cleaning procedure for ancient copper alloy objects from marine environments.
We also use this opportunity to scrutinise the coins’ corrosion. Below, we describe the MMB shipwreck surrounding, as well as the materials and methods used, report the results achieved during the study, and offer new insights into the treatment of copper alloy artefacts found at sea. We propose a 12-step methodology for chemically cleaning and polishing copper alloys.

2. The Underwater Site

The MMB shipwreck is located on the Mediterranean coast of Israel, about 35 km south of Haifa and at a depth of 3 m. The seafloor is sandy, and its sediment regime is dynamic (sand is removed and deposited by storms or seasonal movements). To date, five seasons of underwater excavations have been conducted by the Leon Recanati Institute for Maritime Studies. The shipwreck’s maximum measurements are 19.6 m long and 4.9 m wide. The hull remains comprise the keel, endposts, aprons, sternson, framing timbers, hull planks, central longitudinal timbers, stringers, bulkheads, and a mast-step assembly (Figure 1). Walnut (Juglans regia) and fir (Abies sp.) were the dominant wood species used for building the hull. The ship was found to hold numerous and diverse finds, including amphorae and ceramic ware, wooden artefacts, food and faunal remains, bricks and stones, glass fragments, and metal artefacts [20,21]. 14C dates produced for wood and organic samples and typological analyses of the ceramic and glass assemblages place the shipwreck in the seventh–eighth centuries CE, the Early Islamic period in the region [20,22]. This was a period of transition in the Mediterranean. Politically, the region passed from Byzantine to Islamic rule, and ship construction techniques shifted from ‘shell-first’ to ‘frame-based’ techniques.

3. Materials and Methods

3.1. The Coins

A total of 53 coins in various states of preservation were discovered in different parts of the shipwreck. They were systematically documented in the field, retrieved, and registered. According to G. I. Bijovsky of the Israel Antiquities Authority, the coins are Late Roman bronze nummi minted in the mid-fourth century CE [23]. These coins were used as small change and mainly circulated in the eastern region of the Roman Empire during the fourth, fifth, sixth, and, less intensely, seventh centuries CE [18,23,24].
A representative sample of 16 coins was chosen for the current study, as part of an ongoing research concerning the coins retrieved from the MMB shipwreck. The studied 14 coins (nos. 120.3, 143.1, 154.2, 162, 165, 167.1, 168.2, 168.3, 169.1, 176.1, 181, 182.3, 186.2, and 186.3) were entirely or partially covered with concretion, apart from two coins (no. 167.3 and no. 167.9) that were discovered without concretion. The latter two were found on the beach, about 70 m from the shipwreck. Even though they were not found in situ, their typology, composition, and vicinity to the wreck site suggest that they derive from the MMB. As they were well preserved and naturally polished, they did not require cleaning. We decided to use a chemical polish for cleaning only the 13 coins presented in Figure 2, whereas coin 120.3 was cross-sectioned in order to gain further information concerning the concretion layered structure. The goal was to remove the concretion cover, expose the surviving corroded metal underneath, and reveal further information, such as plastic deformations and iconographic motifs.

3.2. The Current Developed Chemical Cleaning Method

As indicated above, sandblasting and other mechanical cleaning procedures are sometimes inadequate for ancient coins. They are insensitive to the concretion layers’ heterogeneity and risk losing valuable information. For example, we applied sandblasting to Coin 120.4 from the MMB shipwreck. The procedure failed to remove the concretion layers in some areas of the coin’s surface and damaged the exposed metal in others [5,18]. Therefore, we devised a chemical procedure to compensate for these shortcomings.
The new chemical cleaning and polishing procedure and recommended methodology we developed here to tackle concretions by interaction with acetic acid organic solvents [25] is specifically for copper alloys retrieved from marine environments after a long burial period. This new procedure comprised several steps, after each of which the coin was washed in deionised (DI) water. First, the coin was immersed in a 12 g/l sodium dodecyl sulphate (SDS; NaC12H25SO4) solution under sonication for 1–2 hours. This treatment caused the concretion to crack and crumble. Next, the coin was moved into a boiling solution of 50% acetic acid for an hour, after which the remaining patina layer was dissolved in a three-times diluted PAN solution at room temperature for 20–30 minutes. By this point, the coin’s core metal was exposed. However, some polishing was still necessary, for which a PAN solution was used (by immersion) for 20–60 seconds. This new method comprises immersing the coin in a boiling solution of acetic acid (CH3COOH) for seven hours, placing it in a diluted peroxyacetyl nitrate (PAN) solution (1H3PO4:5CH3COOH:4HNO3) for two hours at room temperature to dissolve the patina, and carefully rinsing it in water. Lastly, the coin’s surface was polished in a PAN solution for 2 minutes, rinsed, and dried [5].

3.3. Analytical Methods

In order to document the cleaning procedure’s effect and trace the particularities of the corrosive processes that altered the MMB coins, different analytical methods and measurements were applied. Given the coins’ fragility, these methods were required to be sensitive and non-destructive, as is anyway preferable with archaeological objects [26,27]. Our analytical procedure consisted of the following:
(a)
The coins were visually inspected throughout the cleaning process to monitor the procedure’s progress and assess its efficacy (i.e., the extent to which it achieves a clean surface while incurring minimum damage).
(b)
Morphometric data of the coins were collected at the beginning and end of the procedure. This included the coins’ diameter and thickness measured with a digital caliper and their weight attained with a digital balance (precision ± 0.01 g).
(c)
The clean coins’ surfaces, including the exposed metal and remaining concretions, were inspected with a 3D digital Hirox RH-2000 light microscope (LM) equipped with a multi-focus system, high-intensity LED lighting (5700 K colour temperature), and powerful 3D software [28]. It sought to assess the coins’ state of preservation and detect microscopic defects and discontinuities.
(d)
A handheld XRF (HH-XRF) Oxford X-MET8000 was employed to determine the elemental compositions of the coins’ concretion and underlying metal. The XRF was coupled with a Silicon Drift Detector equipped with a 45 kV Rh Target X-ray tube and LE operation mode. The instrument was calibrated with a standard steel calibration sample. Each measurement was collected for 30 seconds, with a measurement spot size of 5 mm in diameter. The measurements were made by comparing independent peaks. For example, the arsenic (As) Kα and Kβ peaks were compared with the lead (Pb) Mα and Lβ peaks, respectively [29]. Light elements, including carbon and oxygen, could not be detected with this HH-XRF instrument.
(e)
Four coins’ concretion covers (nos. 143, 162, 167, and 169) were subjected to Raman spectroscopy, conducted with a Horiba Jobin Yvon LabRam HR spectrometer coupled to an Olympus LM. Raman spectroscopy is a powerful technique for providing information concerning the surface of metals and alloys covered with oxides and corrosion products, and its spectra can identify the different minerals. The analyses were performed using a 532 nm laser-excitation line. A 50 × objective lens was used to identify the concretion products on the coin surface. Two or three spectra were performed on each side of the external concretion. To interpret the Raman analysis results, we used the database of the Center for Nanoscience and Nanotechnology, Tel Aviv University. Additionally, the current results have been compared to the Raman studies in the literature [30].
(f)
The metallographic transverse cross-section (T-CS) of coin no. 120.3 was examined under a scanning electron microscope (SEM) to explore its microstructure morphologies. This analysis was done with an environmental ESEM-FEI Quanta 200FEG instrument combined with energy-dispersive spectroscopy (EDS) in high vacuum mode with an Everhart–Thornley secondary electron (SE) detector. Both secondary electron (SE) and back-scattered electron (BSE) modes were applied. The composition was detected by EDS using a Si (Li) liquid-cooled Oxford X-ray detector (calibrated with standard samples and providing measurements with a first approximation error of 1%) [18,28].

4. Results

4.1. Visual Observations

Aside for the two well-preserved specimens found on the beach, all remaining 14 coins were covered with thick (about 1 mm) concretion. It consisted of sand particles and marine organisms embedded in a fine, usually black matrix (Figure 3a), sometimes accompanied by dark blue spots (Figure 3b). In one instance (no. 168.3), the concretion had small black, dark green, and dark grey sectors (Figure 3c,d), typical of internal copper and lead oxides-rich concretion layers [5]. Another coin (no. 167.1, Figure 2h) was coated with a unique green-turquoise concretion. Viewed at higher magnifications, a more variegated pattern was observed, comprising areas of black, dark green (Figure 4a,b), and bright green compounds (Figure 4c). In some areas of this concretion, a blue-turquoise mineral was also observed (Figure 4b, arrow).
The two well-preserved coins (nos. 167.3 and 167.9, Figure 5) found on the beach had no concretion cover, possibly due to natural erosion in the marine environment. Thus, they did not need to be cleaned and were readily identifiable. However, their external surfaces were covered with dark green, bright green, turquoise, brown, yellow, white, grey, and black oxides (Figure 5).
LM and SEM observations of the T-CS of coin no. 120.3 recorded a layered structure comprised of three components. The inner layer comprised the copper alloy core consisting of dark copper-rich and bright lead-rich areas (Figure 6b, centre of image, BSE mode). The second layer enveloped the core with oxides and corrosion products, while the third outer layer was a dark concretion cover with some pores and cracks (Figure 6b).
Heritage 05 00189 g004 550
Figure 4. Multi-focal LM images of coin no. 167.1 (side A): (a) general view of the coin’s surface; (b) dark green-black concretion and blue-turquoise minerals (white arrow); (c) an area of a pale green compound.
Figure 4. Multi-focal LM images of coin no. 167.1 (side A): (a) general view of the coin’s surface; (b) dark green-black concretion and blue-turquoise minerals (white arrow); (c) an area of a pale green compound.
Heritage 05 00189 g004
Heritage 05 00189 g005 550
Figure 5. Multi-focal LM of coins no. 167.3 and no. 167.9, which were found without a concretion coating: (a) a stamped image of two soldiers (coin no. 167.3, side A); (b) a portrait of Constantine II (coin no. 167.3, side B); (c) a stamped image of two soldiers (coin no. 167.9, side A); (d) a portrait of the House of Constantine I (coin no. 167.9, side B).
Figure 5. Multi-focal LM of coins no. 167.3 and no. 167.9, which were found without a concretion coating: (a) a stamped image of two soldiers (coin no. 167.3, side A); (b) a portrait of Constantine II (coin no. 167.3, side B); (c) a stamped image of two soldiers (coin no. 167.9, side A); (d) a portrait of the House of Constantine I (coin no. 167.9, side B).
Heritage 05 00189 g005
Heritage 05 00189 g006 550
Figure 6. Metallographic images of coin no. 120.3′s transverse cross-section (T-CS, no etching, BSE mode): (a) LM view of the coin’s copper alloy (top) oxide and corrosion products (middle), dark concretion (bottom); (b) SEM view of the coin’s copper alloy consisting of dark (copper) and bright (areas) covered, in turn, by a layer of oxides and corrosion products, which are enveloped by a dark concretion layer. The dark areas in the copper alloy regions according to LM (Figure 6a) and the bright areas in the copper alloy regions according to SEM BSE mode (Figure 6b) are both lead precipitations.
Figure 6. Metallographic images of coin no. 120.3′s transverse cross-section (T-CS, no etching, BSE mode): (a) LM view of the coin’s copper alloy (top) oxide and corrosion products (middle), dark concretion (bottom); (b) SEM view of the coin’s copper alloy consisting of dark (copper) and bright (areas) covered, in turn, by a layer of oxides and corrosion products, which are enveloped by a dark concretion layer. The dark areas in the copper alloy regions according to LM (Figure 6a) and the bright areas in the copper alloy regions according to SEM BSE mode (Figure 6b) are both lead precipitations.
Heritage 05 00189 g006
After removing the concretion and corrosion layers according to the procedure described above, a shiny orange-yellow metallic surface was exposed on some coins, two of which–154.2 (Figure 2e, side B) and no. 168.3 (Figure 2l, side A)–retained traces of their stamped designs. On others, however, oxide and dark concretion layers remained (Figure 7a–c, coins nos. 181, 186.1, and 162.8), and for yet other coins, none of the original metal was found, probably because of severe corrosion or inadvertent damage caused by the chemical procedure. Thus, coins nos. 167.1 (Figure 2i) and 181 (Figs. 2o, 7a) broke during cleaning due to their poor state of preservation.
The chemical cleaning and polishing of coin no. 165′s surface uncovered parts of shiny orange metal, although many parts remained below a thin black concretion layer (Figure 2g). Similarly, chemical cleaning and polishing revealed portions of shiny orange-yellow metal of coin no. 176.1. However, adherent brown, green, white, and grey minerals, as well as some black concretion, were also left (Figure 2n). As of coin no. 186.1, cleaning and polishing removed most of its black concretion and adherent minerals, although some remained (Figure 7b). In addition, a notch was observed on coin no. 186.1 (Figure 7b) and other coins (e.g., nos. 154.2, 181). The cleaning and polishing of coin no. 168.2 resulted in an alternating morphology that shifted between areas of shiny orange-yellow metal and areas of bright green, dark green, and black mineral cover. It may have resulted from the coin’s microstructural heterogeneity (Figure 7c). Coin no. 154.2 also showed a thin green mineral layer (Figure 7d).

4.2. Morphometric Features

The measurements of coins’ diameter, thickness, and weight, before and after the concretion removal, are presented in Table 1. The considerable reduction in the coins’ thickness and weight indicates that corrosion was severe, explaining why most coins failed to manifest stamped designs.

4.3. Chemical and Mineralogical Compositions

The results of the XRF chemical analysis are summarised in Table 2. They demonstrate that the principal constituents of the black concretion cover are copper (≤65.7 wt.% Cu), lead (≤9.9 wt.% Pb), and silicon (≤59.3 wt.% Si), often occurring with sulphur (≤16.0 wt.% S). Additional elements detected include tin (Sn), zinc (Zn), arsenic (As), magnesium (Mg), silver (Ag), and iron (Fe). The delicate grey concretion on coin no. 168.3 consists of copper (64.4–65.0 wt.% Cu), silicon (3.5–11.8 wt.% Si), and sulphur (17.0–20.1 wt.% S), alongside tin, zinc, silver, and iron, while the green concretion on coin no. 167.1 comprises 72.3 wt.% copper, 3.8 wt.% lead, 4.7 wt.% tin, and 18.2 wt.% sulphur. Other elements detected include zinc, silver, and iron. Therefore, the green concretion colour results from a higher concentration of copper oxide than the black and grey concretions.
The SEM-EDS analysis (scan area of 300 µm × 300 µm) of the metallographic sample of coin no. 120.3 revealed composition of (78.3 wt.% Cu, 17.6 wt.% Pb, and 4.1 wt.% Sn). The SEM-EDS analysis of coin 120.3 (line scan of the T cross-section) found that the core metal composition was 32–94 wt.% copper, 0–40 wt.% lead, up to 14 wt.% oxygen, up to 8 wt.% chlorine, and up to 2 wt.% sulphur, the oxides and corrosion products layer was rich in copper (10–88 wt.% Cu), lead (1–68 wt.% Pb), oxygen (4–30 wt.% O), chlorine (up to 14 wt.% Cl), and sulphur (0–30 wt.% S), while the dark concretion cover consisted primarily of copper (about 70 wt.% Cu) and sulphur (about 30 wt.% S).
XRF analysis of the core metal of the two well-preserved coins (nos. 167.3 and 167.9; Figure 5) indicated that their surface was principally copper (88.9–94.6 wt.% Cu) with low ratios of other elements (1.6–6.9 wt.% Pb, 1.8–2.8 wt.% Sn, 0.1 wt.% Zn, 1.0–1.9 wt.% Ag, and up to 0.3 wt.% of S and Fe, Table 2). Conversely, the chemically cleaned and polished coins were found to be copper–lead alloys, consisting of 54.6–75.2 wt.% copper, 6.7–25.2 wt.% lead, 2.0–9.9 wt.% tin, 0.1 wt % zinc, up to 1.5 wt.% arsenic, 0.6–1.2 wt.% silver, 2.2–18.1 wt.% silicon, up to 13.7 wt.% sulphur, and 0.2–1.2 wt.% iron, typical of fourth–seventh century CE Byzantine bronze nummi coins. Similarly, the SEM-EDS chemical and composition mapping analyses of coin no. 120.3 demonstrated that the base metal is mostly copper and lead, alongside low ratios of tin, arsenic, bismuth (Bi), oxygen, chlorine, sulfur, and silicon (Figure 8).
Raman analysis of four coins’ concretion layers (nos. 143.1, 162, 167, and 169; Figure 9) demonstrates that their mineralogical composition consists primarily of quartz (Figure 9, peaks at 128, 206, 264, 390, 465, 697, 808, and 1162 cm−1) and K-feldspathic sands: sanidine (colourless to white)/microcline (white, grey, salmon-pink, and bluish green) [KAlSi3O8; peaks at 512, 652, 997, and 1125 cm−1] [31]. Nevertheless, coloured corrosion products of bronze, such as azurite (dark to pale blue) [2CuCO3∙Cu(OH)2; peaks at 287, 402, 542, 748 and 1461 cm−1], Egyptian blue [CaCuSi4O10; peaks at 478, 990 and 1040 cm−1] [32], malachite (green) [CuCO3∙Cu(OH)2; peaks at 178, 354, 510 and 1498 cm−1], atacamite [CuCl2∙3Cu(OH)2; peaks at 120, 150 and 512 cm−1], basic copper (II) ethanoate [Cu(CH3COO)2∙Cu(OH)2; peak at 1524 cm−1] [31,32,33], and cerussite [PbCO3; peak at 860 cm−1] [33] were also found. These results are in close agreement with the elements detected by the XRF chemical analysis.

5. Discussion

The current study tested a novel chemical cleaning method on archaeological thickly encrusted copper-alloy coins from a marine environment. It consisted of several steps and different solutions to dissolve the concretion. Our results suggest that this method is particularly effective when the concretion is hard, dense, and heterogeneous. However, when corrosion is severe, the cleaning process may result in undesirable damage, including broken and illegible specimens. Moreover, for copper alloy objects with a thin and sparse concretion cover, this method might be too aggressive, calling for more diluted solutions, shorter immersion periods, or both. This issue should be further examined in future studies. Further investigation may also be required to determine if corrosion processes accelerate after cleaning.
The corrosion products of ancient copper-lead bronze objects from shipwrecks are characterised by a multi-layered cover with varying compositions and structures. Their core consists of metal (mostly copper and lead), whereas the first internal corrosion layers typically comprise oxides, such as cuprite, tenorite, and lead oxides (PbO, PbO2, Pb2O3, Pb3O4). The outer corrosion layers, conversely, usually consist of copper and lead carbonates and silicates [6,34,35,36]. The XRF chemical analysis of the MMB shipwreck coins’ outer corrosion surfaces revealed that they were mainly composed of copper, silicon, and sulphur compounds (Table 2), indicating the presence of copper silicates and copper sulphates due to the continued interaction of the coins’ metal with the sea environment. However, carbon could not be detected due to the XRF’s instrumental limitations.
The high ratios of copper, lead, and tin and the low ratio of silicon recorded after the procedure was completed indicate that it successfully uncovered the core metal (Table 2, XRF analysis results). Interestingly, most specimens are characterised by large amounts of lead (13.2–27.9 wt.% Pb), probably due to economic constraints and inflation that started in the late Roman Empire [18,23]. Exceptional in this regard are the two well-preserved coins found on the beach that are notable for their comparatively high copper-to-lead ratio, suggesting that high lead frequencies might be responsible for poorer preservation (Table 2). Perhaps and notwithstanding the thorough use of DI water, the chemical procedure affected the composition of the inner concretion layers. However, the current XRF results agree with the chemical SEM-EDS analysis results of coin no. 120.3 T-CS.
In the current research, we used Raman spectroscopy to study the corrosion products of the copper–lead coins retrieved from the MMB shipwreck, and we gradually removed the concretion and oxide layers in search of a deeper understanding of the corrosive processes. It is known that such concretion covers protect the coins and impede corrosion [5,18,27]. However, when the alloy on the surface dissolves in submerged settings, metal ions are likely to migrate through cracks to the aquatic environment. Moreover, turbulent water flows may accelerate the concretion cover’s dissolution and promote further corrosion until a new concretion coating is formed. Thus, the coins in the MMB shipwreck may have gone through several cycles of concretion formation, dissolution, and corrosion. On the other hand, the shipwreck’s excellently preserved organic materials (such as ropes and baskets) indicate that the wrecking event was rapid and that the ship was quickly covered in sand. Thus, it is likely that the corrosion process was not accelerated.
The external concretion layers of all examined coins were predominantly comprised of quartz and sanidine/microcline that probably derive from the sand and the seawater. However, Raman analysis (Figure 9) demonstrated that corrosion products of copper alloy are also present. These products include azurite, Egyptian blue, malachite, atacamite, basic copper (II) ethanoate, and cerussite [31,32,33].
The occurrence of these minerals and the composition of the concretion layer, more generally, are a function of a wide range of environmental conditions such as temperature, salinity, pH, and the presence of oxygen and bacteria. Thus, for instance, sulphur (up to 18.2 wt.% S in the external black concretion and up to 20.1 wt.% S in the fine grey concretion) suggests the dissolution of the mineral langite [Cu4(SO4)(OH)6·2H2O], which in turn suggests the existence of cracks and pores in the concretion layer. Copper ions would have diffused outward through these cracks and pores and reacted with the dynamic marine environment, producing minerals such as Egyptian blue [5]. Basic copper (II) ethanoate probably formed as a result of coins’ contact with organic contaminants in the seawater or with wood.
Unlike all other coins from the shipwreck whose concretions were black (Figure 3), coin no. 167.1 had a green-turquoise concretion layer (Figure 4). This peculiarity in colour is attributable to higher copper and tin ratios and lower silicon and sulphur ratios, which probably derive from this coin’s core metal (Table 2).
The two well-preserved coins (nos. 167.3 and 167.9, Figure 5) found on the beach with no concretion indicate a natural polishing process [37,38]. However, these coins’ external surfaces manifested different corrosive products: for example, black-grey to brown tenorite, brownish-orange cuprite, greenish atacamite, and light greyish to green-blue cerussite [11,38].
Most of the black concretion and minerals were successfully removed from the coins. However, most coins still carried remains of various oxides and dark concretions (Figure 7). On some coins, a shiny orange-yellow metallic surface of the copper alloy was exposed, and, occasionally, evidence of the stamping process by plastic deformation was observed (Figure 2e, 2l) [18,39]. Variations in the coins’ elemental compositions may explain the diversity in their states of preservation (Table 2). As suggested above, poor preservation seems to correlate with high lead ratios (Table 2). Smelted primary lead typically contains trace elements of zinc, arsenic, and silver. The presence of more than 1000 ppm of silver in the copper–lead alloy frequently indicates that the lead was smelted but not refined [40,41]. The presence of silicon, sulfur, and iron most probably relates to soil contamination and corrosion products [42].
Concerning copper alloy composition, it seems that mechanical and chemical cleaning procedures achieve similar results. The coins’ compositions found in the present study (Table 2) are similar to those of a previous published XRF analysis of other MMB-shipwreck copper coins that were mechanically cleaned with a Pulsar Zero II particle blast machine (66.1–91.7 wt.% Cu, 0.2–21.1 wt.% Pb, 0.1–3.9 wt.% Sn, 0.3–16.8 wt.% Si, along with small amounts of Zn, As, Sb, Ag, S, and Fe) [17].
Notwithstanding their comparatively high copper ratio, the two coins from the beach have a composition similar to that of the other 13 coins. The fact that they underwent natural polishing and were well-preserved while the others were found covered with concretion and were poorly preserved can be explained by their pre-burial conditions, different alloys composition, and the non-equilibrium environmental conditions.
LM observations demonstrate that the coins’ microstructure consists of two phases that formed during the solidification process: copper-rich bright areas and lead-rich dark areas [5,43,44,45]. Elsewhere, Cohen et al. (2018) observed α-copper grains with few twins and dark preferred-oriented lead precipitations (according to LM) in the T-CS of coin no. 120.3 [18]. The presence of twins indicates that the coin was heated before being stamped. Lead’s solubility in copper is very low, as indicated by the equilibrium copper–lead phase diagram [36]. Therefore, lead tends to form ‘islands’ in copper alloys [36,46]. Thus, the high lead ratios in the coins studied here and the presence of small amounts of zinc, arsenic, and silver suggest that they were produced from recycled metal, probably due to economic constraints [24].
Given these observations, we propose a 12-step methodology for chemically cleaning and polishing copper alloy coins (and other objects retrieved from marine environments) covered with thick concretion: (1) Measure the coins’ diameter, thickness, and weight. (2) Detect the composition and morphology of the concretion cover’s surface with a multi-focus LM and SEM. (3) Select at least one coin for destructive metallographic (T-CS) analysis to examine the morphologies and microstructures of the coins’ concretion, corrosion products, and base metal; this should be followed by LM and SEM-EDS analysis of the coin’s inner corrosion layers and the core metal. (4) Use XRF analysis to determine the elemental compositions of the coins’ concretion cover. (5) Apply Raman analysis to discern the minerals that compose the coins’ external concretion cover. (6) Select several coins for mechanical (sandblasting) cleaning [5,18]; they will serve as a reference group. (7) Apply the chemical cleaning procedure described above to the rest of the set. (8) Measure the coins’ weight and dimensions. (9) Inspect the coins’ surfaces with the naked eye, a multi-focus LM, and SEM. (10) Conduct XRF analysis of the coins to determine their elemental compositions and to evaluate the procedure’s effectiveness. (11) Apply a protective coat to the coins’ surfaces–e.g., restore the patina or artificially coat the coins [47,48,49]. Lastly, (12) compare the mechanically and chemically cleaned coins and their surface compositions to the metallographic T-CS sample alloy.

6. Conclusions

The chemical procedure developed and examined here was designed to remove the concretion layers that cover archaeological copper alloy coins from marine environments. Although some concretion and oxides remained, the procedure was usually successful in exposing the shiny orange-coloured metallic surface of the core metal. This chemical cleaning procedure efficiently removed the concretion and oxide layers without visible damage to the coin’s core and therefore might prove helpful for future numismatic conservation efforts. Moreover, with the 12-step methodology outlined above, this procedure can be applied to other archaeological copper alloy objects from marine environments, especially when a thick and hard concretion layer needs to be removed in order to observe and study the core metal. Furthermore, our analysis indicated that the external concretion layer of the examined coins mainly comprised of quartz and K-feldspathic sands (sanidine and microcline minerals) and that the plastic deformations observed on the coins’ surfaces are due to the stamping process applied to them.

Author Contributions

M.C. defined the research questions. D.A. wrote the introduction. A.I. was in charge of the experiment and formal analysis. M.C. and D.C. were in charge of the archaeological aspects of this research. All authors discussed the results and contributed equally to the manuscript. Conceptualisation: M.C.; Data curation: D.A.; Investigation: All authors; Methodology: A.I.; Writing—original draft: All authors; Writing—review & editing: All authors. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Israel Science Foundation (grant no. 1891/16) and the Honor Frost Foundation.

Data Availability Statement

Not applicable.

Acknowledgments

The underwater excavations (IAA permits G-41/2016, G-40/2017 and G-26/2018) and research of the Maʻagan Mikhael B shipwreck are supported by the Israel Science Foundation (grant no. 1891/16), the Honor Frost Foundation, a Dov Shafir Fellowship, a Natan Rotenstreich Scholarship, a Sir Maurice Hatter Fellowship, the Research Authority of the University of Haifa, Kibbutz Ma‘agan Mikhael, and anonymous donors to whom the authors are grateful. The authors are thankful to G. I. Bijovsky and N. Iddan for their valuable assistance.

Conflicts of Interest

The authors declare that they have no conflict of interest.

References

  1. Di Turo, F. Limits and perspectives of archaeometric analysis of archaeological metals: A focus on the electrochemistry for studying ancient bronze coins. J. Cult. Herit. 2000, 43, 271–281. [Google Scholar] [CrossRef]
  2. Serghini-Idrissi, M.; Bernard, M.C.; Harrif, F.Z.; Joiret, S.; Rahmouni, K.; Srhiri, A.; Takenouti, H.; Vivier, V.; Ziani, M. Electrochemical and spectroscopic characterisations of patinas formed on an archaeological bronze coin. Electrochim. Acta 2005, 50, 4699–4709. [Google Scholar] [CrossRef]
  3. Denker, A.; Bohne, W.; Opitz-Coutureau, J.; Rauschenberg, J.; Röhrich, J.; Strub, E. Influence of corrosion layers on quantitative analysis. Nucl. Instrum. Methods. Phys. Res. B. 2005, 239, 65–70. [Google Scholar] [CrossRef]
  4. Di Francia, E.; Lahoz, R.; Neff, D.; de Caro, T.; Angelini, E.; Grassini, S. Laser-cleaning effects induced on different types of bronze archaeological corrosion products: Chemical-physical surface characterisation. Appl. Surf. Sci. 2022, 573, 150884. [Google Scholar] [CrossRef]
  5. Inberg, A.; Ashkenazi, D.; Cohen, M.; Iddan, N.; Cvikel, D. Corrosion products and microstructure of copper alloy coins from the Byzantine-period Ma‘agan Mikhael B shipwreck, Israel. Microchem. J. 2018, 143, 400–409. [Google Scholar] [CrossRef]
  6. Ingo, G.M.; De Caro, T.; Riccucci, C.; Angelini, E.; Grassini, S.; Balbi, S.; Bernardini, P.; Salvi, D.; Bousselmi, L.; Gener, M.; et al. Large scale investigation of chemical composition, structure and corrosion mechanism of bronze archeological artefacts from Mediterranean basin. Appl. Phys. A. 2006, 83, 513–520. [Google Scholar] [CrossRef]
  7. Manti, P.; Watkinson, D. Corrosion phenomena and patina on archaeological low-tin wrought bronzes: New data. J. Cul. Herit. 2022, 55, 158–170. [Google Scholar] [CrossRef]
  8. Robotti, S.; Rizzi, P.; Soffritti, C.; Garagnani, G.L.; Greco, C.; Facchetti, F.; Borla, M.; Operti, L.; Agostino, A. Reliability of portable X-ray Fluorescence for the chemical characterisation of ancient corroded copper-tin alloys. Spectrochim. Acta Part B At. Spectros. 2018, 146, 41–49. [Google Scholar] [CrossRef]
  9. Papadopoulou, O.; Vassiliou, P. The influence of archaeometallurgical copper alloy castings microstructure towards corrosion evolution in various corrosive media. Corros. Mater. Degrad. 2021, 2, 227–247. [Google Scholar] [CrossRef]
  10. Drakaki, E.; Karydas, A.G.; Klinkenberg, B.; Kokkoris, M.; Serafetinides, A.A.; Stavrou, E.; Vlastou, R.; Zarkadas, C. Laser cleaning on Roman coins. Appl. Phys. A. 2004, 79, 1111–1115. [Google Scholar] [CrossRef]
  11. Ashkenazi, D.; Inberg, A.; Langgut, D.; Hendler, N.; Cvikel, D. Brass–iron couple and brass–iron–wood ternary system of metal objects from the Akko 1 shipwreck (Israel). Corros. Sci. 2016, 110, 228–241. [Google Scholar] [CrossRef]
  12. MacLeod, I.D. Formation of marine concretions on copper and its alloys. Int. J. Naut. Archaeol. 1982, 11, 267–275. [Google Scholar] [CrossRef]
  13. Korenberg, C.; Baldwin, A. Laser cleaning tests on archaeological copper alloys using an ND: YAG Laser. Laser Chem. 2006, 2006, 75831. [Google Scholar] [CrossRef] [Green Version]
  14. Al-Sadoun, A.; Abdel-Kareem, O. Authentication and conservation of selected metal objects excavated from al-serein, near makkah, saudi arabia. Mediterr. Archaeol. Archaeom. 2019, 19, 23–38. [Google Scholar]
  15. Pini, R.; Siano, S.; Salimbeni, R.; Pasquinucci, M.; Miccio, M. Tests of laser cleaning on archeological metal artefacts. J. Cult. Herit. 2000, 1, S129–S137. [Google Scholar] [CrossRef]
  16. Hrnjic, M.; Angurel, L.A.; Lahoz, R.; Grassini, S.; Angelini, E.; Schiavon, N.; de la Fuente, G.F. Near-IR laser cleaning of Cu-based artefacts: A comprehensive study of the methodology standardization. In Proceedings of the 1st International Conference on Metrology for Archaeology, Benevento, Italy, 22–23 October 2015; pp. 389–394. [Google Scholar]
  17. Garbacz, H.; Fortuna-Zalesna, E.; Marczak, J.; Koss, A.; Zatorska, A.; Zukowska, G.Z.; Onyszczuk, T.; Kurzydlowski, K.J. Effect of laser treatment on the surface of copper alloys. Appl. Surf. Sci. 2011, 257, 7369–7374. [Google Scholar] [CrossRef]
  18. Cohen, M.; Ashkenazi, D.; Bijovsky, G.I.; Inberg, A.; Klein, S.; Cvikel, D. Copper alloy coins from the Byzantine-Period Ma‘agan Mikhael B Shipwreck, Israel: Metallurgical characterization. Metallogr. Microstruct. Anal. 2018, 7, 542–560. [Google Scholar] [CrossRef]
  19. Iddan, N.; Ashkenazi, D.; Klein, S.; Cvikel, D. Metallurgical analysis of a bronze powder chamber retrieved from an underwater excavation in Akko (Israel): An application of novel minimally destructive field multi-focal metallography. Archaeol. Anthropol. Sci. 2022, 14, 131. [Google Scholar] [CrossRef]
  20. Cohen, M.; Cvikel, D. Ma-agan Mikhael B, Israel: A preliminary report of a Late Byzantine–early Islamic period shipwreck. Int. J. Naut. Archaeol. 2019, 48, 189–207. [Google Scholar]
  21. Creisher, M.; Goren, Y.; Artzy, M.; Cvikel, D. The Amphorae of the Ma’agan Mikhael B Shipwreck: Preliminary Report. Levant 2019, 51, 105–120. [Google Scholar] [CrossRef]
  22. Natan, E.; Gorin-Rosen, Y.; Benzonelli, A.; Cvikel, D. Maritime trade in Early Islamic-period glass: New evidence from the Maʻagan Mikhael B shipwreck. J. Archaeol. Sci. Rep. 2021, 37, 102903. [Google Scholar] [CrossRef]
  23. Bijovsky, G.I. Gold Coin and Small Change: Monetary Circulation in Fifth–Seventh Century Byzantine Palestine; Università di Trieste: Trieste, Italy, 2012. [Google Scholar]
  24. Kahanov, Y.; Ashkenazi, D.; Cvikel, D.; Klein, S.; Navri, R.; Stern, A. Archaeometallurgical analysis of metal remains from the Dor 2006 shipwreck: A clue to the understanding of the transition in ship construction. J. Archaeol. Sci. Rep. 2015, 2, 321–332. [Google Scholar] [CrossRef]
  25. Turner-Walker, G. The nature of cleaning: Physical and chemical aspects of removing dirt, stains and corrosion. In Proceedings of the International Symposium on Cultural Heritage Conservation, Tainan, Taiwan, 6–8 November 2012. [Google Scholar]
  26. Crosera, M.; Baracchini, E.; Prenesti, E.; Giacomello, A.; Callegher, B.; Oliveri, P.; Adami, G. Elemental characterisation of surface and bulk of copper-based coins from the Byzantine-period by means of spectroscopic techniques. Microchem. J. 2019, 147, 422–428. [Google Scholar] [CrossRef]
  27. Di Turo, F.; Parra, R.; Piquero-Cilla, J.; Favero, G.; Doménech-Carbó, A. Crossing VIMP and EIS for studying heterogeneous sets of copper/bronze coins. J. Solid. State. Electrochem. 2019, 23, 771–781. [Google Scholar] [CrossRef]
  28. Ashkenazi, D.; Cvikel, D. A journey into the microstructure: Using a multifocal 3D digital light microscope to study archaeological artefacts retrieved from shipwrecks. Digit. Appl. Archaeol. Cult. Herit. 2020, 16, e00129. [Google Scholar] [CrossRef]
  29. Ashkenazi, D.; Bunimovitz, S.; Stern, A. Archaeometallurgical investigation of thirteenth–twelfth centuries BCE bronze objects from Tel Beth-Shemesh, Israel. J. Archaeol. Sci. Rep. 2016, 6, 170–181. [Google Scholar] [CrossRef]
  30. Adar, F. Introduction to interpretation of Raman spectra using database searching and functional group detection and identification. Spectroscopy 2016, 31, 16–23. [Google Scholar]
  31. Oancea, A.V.; Bodi, G.; Nica, V.; Ursu, L.E.; Drobota, M.; Cotofana, C.; Vasiliu, A.L.; Simionescu, B.C.; Olaru, M. Multi-analytical characterization of Cucuteni pottery. J. Eur. Ceram. Soc. 2017, 37, 5079–5098. [Google Scholar] [CrossRef]
  32. Bell, I.M.; Clark, R.J.; Gibbs, P.J. Raman spectroscopic library of natural and synthetic pigments (pre- ≈1850 AD). Spectrochim. Acta. A. Biomol. Spectrosc. 1997, 53, 2159–2179. [Google Scholar] [CrossRef]
  33. McCann, L.I.; Trentelman, K.; Possley, T.; Golding, B. Corrosion of ancient Chinese bronze money trees studied by Raman microscopy. J. Raman. Spectrosc. 1999, 30, 121–132. [Google Scholar] [CrossRef]
  34. Kosec, T.; Ćurković, H.O.; Legat, A. Investigation of the corrosion protection of chemically and electrochemically formed patinas on recent bronze. Electrochim. Acta 2010, 56, 722–731. [Google Scholar] [CrossRef]
  35. He, L.; Liang, J.; Zhao, X.; Jiang, B. Corrosion behavior and morphological features of archeological bronze coins from ancient China. Microchem. J. 2011, 99, 203–212. [Google Scholar] [CrossRef]
  36. Ingo, G.M.; Riccucci, C.; Guida, G.; Pascucci, M.; Giuliani, C.; Messina, E.; Fierro, G.; Di Carlo, G. Micro-chemical investigation of corrosion products naturally grown on archaeological Cu-based artefacts retrieved from the Mediterranean Sea. Appl. Surf. Sci. 2019, 470, 695–706. [Google Scholar] [CrossRef]
  37. Cvikel, D.; Cohen, M.; Inberg, A.; Klein, S.; Iddan, N.; Ashkenazi, D. Metallurgical characterisation of brass sheet from the 19th-century Akko Tower Wreck (Israel). Mater. Charact. 2017, 131, 175–187. [Google Scholar] [CrossRef]
  38. Ashkenazi, D.; Inberg, A.; Cvikel, D. Analysis of naturally etched surface of brass sheathing from a nineteenth-century shipwreck. J. Min. Metall. B. 2018, 54, 101–110. [Google Scholar] [CrossRef]
  39. Kossolapov, A.J.; Chugunova, K.S. Residual stress in struck and cast coins. Insight NDT Cond. Monitor. 2020, 62, 139–144. [Google Scholar] [CrossRef]
  40. Craddock, P.T. Early Metal Mining and Production; Edinburgh University Press: Edinburgh, Scotland, 1995. [Google Scholar]
  41. Ehya, F.; Lotfi, L.M.; Rasa, I. Emarat carbonate-hosted Zn–Pb deposit, Markazi Province, Iran: A geological, mineralogical and isotopic (S, Pb) study. J. Asian Earth Sci. 2010, 37, 186–194. [Google Scholar] [CrossRef]
  42. Ingo, G.M.; Angelini, E.; Bultrini, G.; Calliari, I.; Dabala, M.; De Caro, T. Study of long-term corrosion layers grown on high-tin leaded bronzes by means of the combined use of GDOES and SEM+ EDS. Surf. Interface Anal. 2002, 34, 337–342. [Google Scholar] [CrossRef]
  43. Davidde, B. Methods and strategies for the conservation and museums display in situ of underwater cultural heritage. Archaeol. Marit. Mediterr. 2004, 1, 136–150. [Google Scholar]
  44. Chakrabarti, D.J.; Laughlin, D.E. The Cu−Pb (copper-lead) system. Bull. Alloy Phase Diagrams. 1984, 5, 503–510. [Google Scholar] [CrossRef]
  45. Griesser, M.; Kockelmann, W.; Hradil, K.; Traum, R. New insights into the manufacturing technique and corrosion of high leaded antique bronze coins. Microchem. J. 2016, 126, 181–193. [Google Scholar] [CrossRef]
  46. Torrisi, L.; Caridi, F.; Giuffrida, L.; Torrisi, A.; Mondio, G.; Serafino, T.; Caltabiano, M.; Castrizio, E.D.; Paniz, E.; Salici, A. LAMQS analysis applied to ancient Egyptian bronze coins. Nucl. Instrum. Methods. Phys. Res. B 2010, 268, 1657–1664. [Google Scholar] [CrossRef]
  47. Argyropoulos, V.; Giannoulaki, M.; Michalakakos, G.P.; Siatou, A. A survey of the types of corrosion inhibitors and protective coatings used for the conservation of metal objects from museum collections in the Mediterranean basin. In Proceedings of the the International Conference on Strategies for Saving Indoor Metallic Collections with a Satellite Meeting on Legal Issues in the Conservation of Cultural Heritage, Cairo, Egypt, 25 February–1 March 2007; pp. 1–5. [Google Scholar]
  48. Adriaens, A.; Dowsett, M.; Leyssens, K.; van Gasse, B. Insights into electrolytic stabilization with weak polarization as treatment for archaeological copper objects. Anal. Bioanal. Chem. 2007, 387, 861–868. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  49. Ashkenazi, D.; Nusbaum, I.; Shacham-Diamand, Y.; Cvikel, D.; Kahanov, Y.; Inberg, A. A method of conserving ancient iron artefacts retrieved from shipwrecks using a combination of silane self-assembled monolayers and wax coating. Corr. Sci. 2017, 123, 88–102. [Google Scholar] [CrossRef]
Heritage 05 00189 g001 550
Figure 1. The Ma‘agan Mikhael B shipwreck, looking towards the bow of the ship (Photo: A. Yurman).
Figure 1. The Ma‘agan Mikhael B shipwreck, looking towards the bow of the ship (Photo: A. Yurman).
Heritage 05 00189 g001
Heritage 05 00189 g002 550
Figure 2. Selected coins from the Ma‘agan Mikhael B shipwreck (A and B refer to a coin’s obverse and reverse faces, respectively): (a,b) coin no. 143.1 before and after partial cleaning; (c,d) coin no. 154.2 before and after cleaning; (e) coin no. 162 before cleaning; (f) coin no. 165 after cleaning; (g,h) coin no. 167.1 before and after cleaning; (i,j) coin no. 168.2 before and after cleaning; (k) coin no. 168.3 after cleaning; (l) coin no. 169.1 before cleaning; (m) coin no. 176.1 after cleaning; (n) coin no. 181 after cleaning; (o,p) coin no. 182.3 before and after cleaning; (q) coin on. 186.2 before cleaning; and (r,s) coin 186.3 before and after cleaning (Photo: A. Efremov).
Figure 2. Selected coins from the Ma‘agan Mikhael B shipwreck (A and B refer to a coin’s obverse and reverse faces, respectively): (a,b) coin no. 143.1 before and after partial cleaning; (c,d) coin no. 154.2 before and after cleaning; (e) coin no. 162 before cleaning; (f) coin no. 165 after cleaning; (g,h) coin no. 167.1 before and after cleaning; (i,j) coin no. 168.2 before and after cleaning; (k) coin no. 168.3 after cleaning; (l) coin no. 169.1 before cleaning; (m) coin no. 176.1 after cleaning; (n) coin no. 181 after cleaning; (o,p) coin no. 182.3 before and after cleaning; (q) coin on. 186.2 before cleaning; and (r,s) coin 186.3 before and after cleaning (Photo: A. Efremov).
Heritage 05 00189 g002
Heritage 05 00189 g003 550
Figure 3. Multi-focal LM images of concretions on coins nos. 169.1 and 168.3: (a) the external concretion coating of coin no. 169.1, including embedded dark blue particles; (b) the external concretion coating of coin no. 169.1 at high magnification, showing sand particles and marine organisms; (c) view of internal concretion coating on coin no. 168.3; (d) higher magnification of internal concretion coating on coin no. 168.3, note black, dark green, dark grey and light grey oxide colours.
Figure 3. Multi-focal LM images of concretions on coins nos. 169.1 and 168.3: (a) the external concretion coating of coin no. 169.1, including embedded dark blue particles; (b) the external concretion coating of coin no. 169.1 at high magnification, showing sand particles and marine organisms; (c) view of internal concretion coating on coin no. 168.3; (d) higher magnification of internal concretion coating on coin no. 168.3, note black, dark green, dark grey and light grey oxide colours.
Heritage 05 00189 g003
Heritage 05 00189 g007 550
Figure 7. Multi-focal LM of several coins after chemical cleaning and polishing: (a) coin no. 181, note cracked surface; (b) coin no. 186.1, note notch (arrow); (c) coin no. 168.2, note variegated surface, including black concretion remains, brown, green, and yellow oxides, and exposed metal; and (d) coin 154.2, note exposed copper metal and patches of light green oxide.
Figure 7. Multi-focal LM of several coins after chemical cleaning and polishing: (a) coin no. 181, note cracked surface; (b) coin no. 186.1, note notch (arrow); (c) coin no. 168.2, note variegated surface, including black concretion remains, brown, green, and yellow oxides, and exposed metal; and (d) coin 154.2, note exposed copper metal and patches of light green oxide.
Heritage 05 00189 g007
Heritage 05 00189 g008 550
Figure 8. Elemental SEM-EDS analysis (bright areas) of coin no. 120.3′s copper alloy: (a) copper; (b) lead; (c) tin; (d) arsenic; (e) bismuth; (f) oxygen; (g) chlorine; (h) sulfur; and (i) silicon.
Figure 8. Elemental SEM-EDS analysis (bright areas) of coin no. 120.3′s copper alloy: (a) copper; (b) lead; (c) tin; (d) arsenic; (e) bismuth; (f) oxygen; (g) chlorine; (h) sulfur; and (i) silicon.
Heritage 05 00189 g008
Heritage 05 00189 g009 550
Figure 9. Raman spectra of the external concretion coating of coins nos. 143.1, 162, 167.1, and 169.1.
Figure 9. Raman spectra of the external concretion coating of coins nos. 143.1, 162, 167.1, and 169.1.
Heritage 05 00189 g009
Table
Table 1. Dimensions and weights of the copper alloy coins, with and without concretion layers.
Table 1. Dimensions and weights of the copper alloy coins, with and without concretion layers.
Coin No.Coin DescriptionDiameters (mm)Thickness (mm)Weight (g)
120.3Including the black concretion layer [17]17.10 × 17.103.002.66
143.1Including the black concretion layer15.76 × 16.30––
Totally corroded, no metal survived–––
154.2Including black concretion layer16.13 × 16.352.751.74
After removal of concretion layer14.30 × 15.160.990.84
162Including black concretion layer15.79 × 16.233.781.42
165Including black concretion layer13.65 × 16.131.440.86
After removal of concretion layer12.51 × 14.180.960.58
167.1Green-turquoise concretion15.01 × 16.612.101.47
167.3Clean, well preserved17.16 × 17.511.372.04
167.9Clean, well preserved14.73 × 15.101.181.31
168.2Including black concretion layer15.98 × 16.321.741.07
After removal of concretion layer12.10 × 13.160.550.33
168.3Including black concretion layer (possibly internal)14.37 × 15.441.921.25
169.1Including black concretion layer20.10 × 20.394.333.21
176Including black concretion layer15.11 × 16.181.441.41
After removal of concretion layer14.88 × 14.911.161.04
181Including black concretion layer14.14 × 15.512.250.87
After removal of concretion layer11.17 × 12.850.730.40
182.3Remains of black concretion layer15.23 × 16.693.421.49
186.2Remains of black concretion layer14.79 × 15.381.941.36
After removal of concretion layer13.97 × 14.780.750.89
186.3Remains of black concretion layer14.59 × 15.142.861.33
After removal of concretion layer13.80 × 14.341.210.79
Table
Table 2. XRF chemical analysis of the coins before and after surface chemical cleaning (A and B refer to a coin’s obverse and reverse faces, respectively).
Table 2. XRF chemical analysis of the coins before and after surface chemical cleaning (A and B refer to a coin’s obverse and reverse faces, respectively).
Coin No. and Its Examined AreaComposition (wt.%)
CuPbSnZnAsMgAgSiSFe
120.3/A, a shiny metallic surface [17]71.310.01.70.5––0.19.96.30.2
154.2/A, with black concretion51.64.70.7–0.5–0.623.210.38.4
154.2/B, with black concretion58.63.20.80.10.5–0.719.08.38.8
154.2/B, metal after partial surface chemical cleaning58.327.92.9–2.6–1.73.7–2.9
162/A, with black concretion31.60.1––––0.359.38.20.5
162/B, with black concretion35.80.2––––0.351.011.90.8
165/A, with black concretion, after partial surface chemical cleaning67.613.22.10.1––0.72.213.70.4
165/B, with black concretion, after partial surface chemical cleaning59.48.92.00.1––0.618.110.60.3
167.1/A, with green concretion72.33.84.70.1––0.418.2–0.5
167.1/A, metal after chemical cleaning and polishing treatment75.26.79.90.1––1.26.7–0.2
167.3/A, naturally polished coin (Figure 5a, centre of the coin, average of two measurements)90.15.62.80.1––1.2–0.20.2
167.3/B, naturally polished coin (Figure 5b, centre of the coin, average of two measurements)94.22.02.40.1––1.1–0.10.1
167.9/A, naturally polished coin (Figure 5c, centre of the coin, average of two measurements)94.21.91.80.1––1.9––0.1
167.9/B, Naturally polished coin (Figure 5d, centre of the coin, average of two measurements)94.21.81.90.1––1.9––0.1
168.2/A, with black concretion49.26.10.30.1––0.330.312.51.2
168.2/B, with black concretion39.99.90.4––1.60.539.46.91.4
168.3/A, with fine grey concretion64.40.81.40.1––0.311.817.04.2
168.3/B, with fine grey concretion65.01.51.80.1––0.33.520.17.7
169.1/A, with black concretion45.20.20.1–––0.543.310.40.3
169.1/B, with black concretion43.40.2–0.10.10.10.844.110.50.7
176/A, with black concretion remains, after partial surface chemical cleaning54.625.23.2–––0.53.411.91.2
176/B, with black concretion remains, after partial surface chemical cleaning56.019.92.60.1––0.57.313.10.5
181/A, with black concretion59.61.60.90.1–17.80.412.96.50.2
181/B, with black concretion50.47.51.50.10.56.20.729.43.10.6
181/B, metal covered with oxide after chemical cleaning and polishing treatment63.120.53.00.11.5–1.19.8–0.9
182.3/A, with black concretion48.61.00.50.1––0.435.113.80.5
182.3/B, with black concretion44.11.20.60.1––0.543.19.60.8
186.2/A, with black concretion60.54.11.30.1––0.720.712.20.4
186.2/B, with black concretion65.79.32.20.1––0.75.816.00.2
186.3/A, with black concretion57.58.81.30.1––1.523.56.90.4
186.3/B, with black concretion38.61.20.60.1––0.852.65.80.3
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Cohen, M.; Inberg, A.; Ashkenazi, D.; Cvikel, D. What You Clean Is What You Get: A Novel Chemical Cleaning Technique and the Interpretation of Corrosion Products Found in Late Roman Copper Alloy Coins Retrieved from the Sea. Heritage 2022, 5, 3628-3647. https://doi.org/10.3390/heritage5040189

AMA Style

Cohen M, Inberg A, Ashkenazi D, Cvikel D. What You Clean Is What You Get: A Novel Chemical Cleaning Technique and the Interpretation of Corrosion Products Found in Late Roman Copper Alloy Coins Retrieved from the Sea. Heritage. 2022; 5(4):3628-3647. https://doi.org/10.3390/heritage5040189

Chicago/Turabian Style

Cohen, Maayan, Alexandra Inberg, Dana Ashkenazi, and Deborah Cvikel. 2022. "What You Clean Is What You Get: A Novel Chemical Cleaning Technique and the Interpretation of Corrosion Products Found in Late Roman Copper Alloy Coins Retrieved from the Sea" Heritage 5, no. 4: 3628-3647. https://doi.org/10.3390/heritage5040189

Article Metrics

Back to TopTop