CHARMed PyMca,Part II: An evaluation of interlaboratory reproducibility for ED‐XRF analysis of copper alloys

CHARMed PyMca is a protocol for researchers using energy‐dispersive X‐ray fluorescence spectroscopy (ED‐XRF) who wish to maximize interlaboratory reproducibility of quantitative results from copper alloys found in heritage materials. This paper reports the results of a study evaluating the interlaboratory reproducibility expected when following this protocol. Five institutions participated in this study, using nine tube‐based instruments of seven types. A set of 12 reference materials not used for calibration was analysed according to the protocol. The results show an improvement in reproducibility of between 65% and 83%, depending on the element, in comparison with a similar study carried out in 2010. The protocol allowed the consistent reporting of concentrations for 15 elements, in contrast to only eight elements that were reported in the 2010 study. Finally, the protocol was shown to generate accurate quantitative results with a well‐characterized precision. It is hoped the information presented here can help researchers studying heritage copper alloys to collaborate more effectively and with greater confidence in the reproducibility of their results. It should also aid researchers to make realistic and informed decisions about whether or not the degree of reproducibility that they may expect from ED‐XRF makes the protocol suitable for their specific purposes.     

The Copper CHARM Set: A New Set of Certified Reference Materials for the Standardization of Quantitative X‐Ray Fluorescence Analysis of Heritage Copper Alloys*

This paper introduces a new set of certified reference materials designed to aid scientists and conservators working in cultural heritage fields with quantitative X‐ray fluorescence analysis of historical and prehistoric copper alloys. This set has been designated as the Copper CHARM Set (Cultural Heritage Alloy Reference Material Set). The Copper CHARM Set is designed to be used by a wide range of museum‐, art‐ and archaeology‐oriented scientists and conservators to help improve the accuracy and range of their calibrations for quantitative ED–XRF spectrometry of copper alloys, and also increase the number of elements that can routinely be quantified. In addition, the common use of a single core set of the reference materials is designed to significantly improve inter‐laboratory reproducibility, allowing greater data sharing between researchers and thus furthering possibilities for collaborative study.     

For Whom the Bells Fall: Metals from the Cenote Sagrado,Chichén Itzá

Bells of copper and copper alloys and gold–copper alloys were deposited in events at the Cenote Sagrado at Chichén Itzá, Mexico during the site's primary occupation (ad 750–1050) and in later centuries. Housed in three museums in the United States and Mexico, bells (n = 38) were evaluated for traces of fabrication and alteration using Vis–UV–IR optical microscopy. Bulk compositions were determined through p‐ED‐XRF. Phases and compositional variation by depth were characterized through XRD and RBS. The technological styles of bell groups were ascribed to communities of metallurgical practice, from West and Central Mexico to Costa Rica and Panama.     

QUANTITATIVE ANALYSIS OF COPPER-TIN BRONZES BY MEANS OF GLOW DISCHARGE OPTICAL EMISSION SPECTROMETRY

Glow discharge optical emission spectrometry (GDOES) has been used for obtaining calibration curves for copper, tin, lead, silver, antimony, arsenic, zinc and iron from standards including copper-base alloys (Cu 61.33-99.95%) and from bronze Punic coins. The GDOES results were calibrated using atomic absorption spectroscopy with inductively coupled plasma atomization (ICPAAS) to analyse the standards and coins. For all these materials, the results show that via GDOES it is possible to obtain reliable and reproducible calibration curves for copper, tin, lead, zinc and iron with a linear behaviour as a function of the content. The quantitative GDOES and ICPAAS data for the coins have been also compared with those obtained via X-ray fluorescence (XRF) and it is shown that GDOES, ICPAAS and XRF agree well with one another considering that the bronze coins are inhomogeneous to some degree. As an example of GDOES analytical features, a depth concentration profile through a thin bronze patina is shown and briefly discussed.     

ANCIENT EGYPTIAN BLACK‐PATINATED COPPER ALLOYS*

This analytical study aims to investigate ancient Egyptian black‐patinated copper alloys. The study group was selected from the collections of the Egyptian Museum in Cairo and from the Faculty of Archaeology Museum in Cairo University. Examination and analysis were undertaken using optical microscopy (OM), scanning electron microscopy (SEM), energy‐dispersive spectroscopy (EDX), X‐ray diffractometry (XRD), X‐ray fluorescence analysis (XRF) and Fourier transform infrared spectroscopy (FTIR). The analysis results indicated that the black patina contained mainly tenorite (CuO). The study presents evidence of thermal patination and animal glue coating.     

Wavelength‐dispersive X‐ray fluorescence analysis of ancient glasses*

The application of X‐ray fluorescence to the study of the composition of ancient glasses has been limited by the large amount of material required for samples. In this paper the setting of a reduced sampling XRF method for the analysis of ancient glasses is described. The method involves the preparation of glass beads by melting the sample with lithium tetraborate flux. By means of reference and synthetic samples, the method allows regression curves to be set, covering broad ranges from ppm to high concentrations. By means of a single program, up to 31 elements of interest in the study of ancient glass are analysed. The reproducibility, sensitivity and reliability of the method are discussed. The results demonstrate that 0.2 g of glass is sufficient to obtain accurate and sensitive analyses for most of the elements of interest.     

Compositional study of bronze vessels from the Derveni tombs of central Macedonia of the fourth century bce using energy‐dispersive micro‐X‐ray fluorescence (EDμXRF) spectrometry

The chemical composition of 61 copper alloy vessels was studied non‐invasively by energy‐dispersive micro‐X‐ray fluorescence (EDμXRF) spectrometry. The vessels were found in five tombs (A, B, D, Z and H) at Derveni in central Macedonia, Greece, which belong to the cemetery of the ancient city of Lete. The tombs are dated to the last quarter of the fourth century bce . The opulence of the numerous grave offerings denotes the high social status of the deceased within Macedonian society. The material under study was divided into two main groups according to their appearance and use. The first group included 44 vessels with a golden appearance for use in banquets; the second group consisted of 17 dark vessels for domestic use. The aim of this research was to study the vessels' technology according to their chemical composition. Both golden and dark vessels consisted of a binary Cu‐Sn alloy and a slight amount of impurities (Fe, Ni, Zn, Pb, As, Co, Sb, Ag, Ti). It is probable that during construction the ancient metalsmiths were taking into account criteria such as the desired appearance, cost of metals, future use and physical properties such as resistance to corrosion. Banquet vessels had a consistent chemical composition in all their different parts compared with those for domestic use. The only chemical element used systemically for alloying purposes, except Sn, was Pb in order to improve the castability of the bases and handles of the vessels. The study provides additional information about the metalworking of bronzes in central Macedonia in the fourth century bce, as well as about the access Macedonian metalworkers had at the time to the raw materials for such use.     

A Comparison Study of Middle Bronze Age II Daggers and Their Rivets as a Tool for a Better Understanding of Their Production

The Middle Bronze Age II is a period during which there exists a contemporaneous usage of arsenic copper and tin bronze for metal weaponry production. In order to learn more about the alloys used in this period, the blades and rivets from 65 daggers of two significantly different types, which were discovered at the Rishon LeZion (RL) cemetery, Israel, were tested by the non‐destructive method of X‐ray fluorescence (XRF). The results reveal new knowledge of the alloys selected for dagger and rivet production, both of which represent fine examples of the Middle Bronze Age II Southern Levant in metal industry.     

Never Judge A Gold Object by Its Surface Analysis: A Study of Surface Phenomena in a Selection of Gold Objects From the Staffordshire Hoard

With the introduction of commercially available handheld XRF analysers, there is a growing problem that the results, simply taken at face value, are regarded as representative of the bulk gold alloys of the objects analysed in museums and used for treasure analysis. It is well known that surface analysis is problematic, and the aim of this study was to test whether non‐destructive surface analysis could be useful in grouping the very large numbers of fragmentary pieces in the Anglo‐Saxon Staffordshire Hoard. Surface and subsurface analyses of 16 objects from the Hoard were carried out by X‐ray fluorescence (XRF) and scanning electron microscopy with energy‐dispersive X‐ray analysis (SEM–EDX). The results indicate that there are significant but inconsistent levels of enrichment of the gold at the surface of many of these objects, due to the loss of both copper and silver. The significant loss of silver (up to 50% relative in some cases) suggests that some form of deliberately induced depletion gilding was carried out during manufacture to remove both silver and copper from the surface. In three cases, an increase in silver at the surface compared to the core metal was detected: this might tentatively be explained by redeposition from contact with silver objects during burial. These results have clear implications for the reliability and interpretation of surface analysis data of gold alloys from this period.     

A Preliminary Study on the Loss of Iron and Arsenic in the Re‐Melting of Iron‐Bearing Arsenical Copper

Re‐melting experiments were conducted with specimens made of high‐arsenic copper alloys containing lead, iron and sulphur. The melting treatment at 1000°C in ambient conditions for 5 min was found to cause an almost complete removal of iron and sulphur along with a notable reduction of arsenic levels. Evidence was found that the preferential oxidation of iron and arsenic was responsible for this change in composition. The treatment also brought about perfect dissolution of speiss particles scattered throughout the original specimen, suggesting the possibility of the addition of speiss to molten copper for the production of arsenical copper.     

A Second Investigation into the Chemical Composition of the Roman Provincial (Procuratorial) Coinage of Judaea,AD 6–66*

An investigation of the chemical composition of the first Roman provincial coinage of Judaea, minted in ad 6–66, was conducted. A total of 103 copper‐alloy coins were analysed by ICP‐AES. It was determined that different copper alloys were used for the coinage, a leaded tin‐bronze and a pure tin‐bronze alloy. The investigation also showed that the copper alloy was made in four different formulae with regard to the alloying elements added to copper. Trace element profiles point to the existence of a shared pool of metal for Roman coins and metalwork.     

TECHNOLOGY AND PROVENANCE OF A COLLECTION OF ISLAMIC COPPER‐BASED OBJECTS AS FOUND BY CHEMICAL AND LEAD ISOTOPE ANALYSIS*

Chemical and lead isotope analyses were utilized to determine the composition, technology and origin of a collection of Islamic copper‐based objects found in Jordan. The atomic absorption spectrometry results show that the objects were made of different types of copper‐base alloys that contain various amounts of zinc, tin and lead. The use of brass, highly leaded brass and quaternary alloys of Cu‐Zn‐Sn‐Pb in the manufacture of everyday, household objects strongly points to Islamic traditions. The lead isotope compositions of the objects match very well that of the copper ore mined from the Dolomite‐Limestone‐Shale unit of the Arabah copper mines.     

Middle Bronze Age II Battleaxes from Rishon LeZion,Israel: Archaeology and Metallurgy

Thirteen bronze battleaxes from Middle Bronze Age II graves at Rishon LeZion, Israel were analysed by ED XRF at multiple surface locations in order to determine their metallurgical composition. Six of these were further subjected to neutron diffraction using an ENGIN‐X diffractometer in order to determine bulk phase composition. The results indicate that the previously established geographical and chronological variability in Sn–Cu (with occasional Pb) and As–Cu alloys found using the former method may be an artefact of preservation and conservation. In addition, the varying homogeneity determined by the latter reflects special treatment for improving on the metal cast.     

Evaluation of Quantitative XRF Analysis Applied to Determine Cobalt Sources in Chinese Blue‐and‐White Porcelain

A method using the ratios between MnO, Fe₂O₃ and CoO to differentiate the cobalt sources for Chinese blue‐and‐white porcelain was developed in Oxford in the 1950s using X‐ray fluorescence (XRF) analysis directly on the glaze. In this paper, six blue‐and‐white porcelain sherds from the Luomaqiao kiln were analysed by XRF on the glaze and by scanning electron microscopy with energy‐dispersive spectrometry (SEM‐EDS) in cross‐section. The ratios between MnO, Fe₂O₃ and CoO calculated by quantitative XRF and EDS analyses are different. The analysis depths for MnO, Fe₂O₃ and CoO are < 60 μm by XRF analysis. However, the average glaze thickness of samples is > 400 μm, and the MnO, Fe₂O₃ and CoO mainly remain in the lower layer of the glaze, which is beyond the analysis depths of XRF analysis. The limitations of major and minor quantitative analyses for differentiating cobalt sources are discussed.     

NON‐DESTRUCTIVE ANALYTIC METHOD USING XRF FOR DETERMINATION OF PROVENANCE OF ARCHAEOLOGICAL OBSIDIANS FROM THE MEDITERRANEAN AREA: A COMPARISON WITH TRADITIONAL XRF METHODS*

A non‐destructive analytical method using wavelength‐dispersive X‐ray fluorescence (WDXRF) that allows the establishment of the provenance of archaeological obsidians was developed and a comparison with the classical XRF method on powders is discussed. Representative obsidian samples of all the geological outcrops of archaeological interest of the Mediterranean area (Lipari, Pantelleria, Sardinia, Palmarola and the Greek islands of Melos and Gyali), were analysed with the normal procedures used in rock analysis by XRF (crushing, powdering and pelletizing). The non‐destructive XRF analysis was instead conducted on splinters taken from the original geological pieces, with the shape deliberately worked to be similar to the refuse usually found at archaeological sites. Since the analysis was conducted on the raw geological fragment, intensity ratios of the suitably selected chemical elements were used, instead of their absolute concentrations, to avoid surface effects due to the irregular shape. The comparison between concentration ratios (obtained by traditional XRF methods) and the intensity ratios of the selected trace elements (obtained from the non‐destructive methodology) show that the different domains of the chemical composition, corresponding to the geological obsidians of the source areas, are perfectly equivalent. In the same way, together with the geological splinters, complete archaeological obsidians, from Neolithic sites, may be analysed and their provenance may be determined. The proposed non‐destructive method uses the XRF method. Due to its sensitivity, low cost and high speed, it is surely an extremely valid instrument for the attribution of the provenance of the archaeological obsidian from Neolithic sites.     

A COMBINED MULTI‐ANALYTICAL APPROACH FOR THE STUDY OF ROMAN GLASS FROM SOUTH‐WEST IBERIA: SYNCHROTRON μ‐XRF,EXTERNAL‐PIXE/PIGE AND BSEM–EDS

An integrated, multi‐analytical approach combining the high sensitivity of SR‐μXRF, the light element capability of PIXE/PIGE under a helium flux and the spatial resolution of BSEM + EDS was used to characterize chemical composition and corrosion of glass samples (first to fourth centuries ad ) from an important, but scarcely investigated, Roman region of south‐west Iberia (southern Portugal). The geochemical trends and associations of major, minor and trace elements were investigated to shed light on production techniques, the provenance of raw materials and decay mechanisms. The results, while confirming a production technique common to Roman glasses throughout the Empire—that is, a silica‐soda‐lime low‐Mg, low‐K composition, with glass additives as colouring and/or decolouring agents (Fe, Cu, Mn, Sb)—show at one site high Zr–Ti contents, suggesting a more precise dating for these glasses to the second half of the fourth century. The Ti–Fe–Zr–Nb geochemical correlations in the pristine glass indicate the presence of minerals such as ilmenite, zircon, Ti‐rich Fe oxides and columbite in the sands used as raw materials for the glass former: these minerals are typical of granitic‐type source rocks. The unusually high K content in the corrosion layers is consistent with burial conditions in K‐rich soils derived from the alteration of 2:1 clays in K‐bearing rock sequences.     

ART‐P‐MAP Neural Networks Modeling of Land‐Use Change: Accounting for Spatial Heterogeneity and Uncertainty

Spatial land‐use models over large geographic areas and at fine spatial resolutions face the challenges of spatial heterogeneity, model predictability, data quality, and of the ensuing uncertainty. We propose an improved neural network model, ART‐Probability‐Map (ART‐P‐MAP), tailored to address these issues in the context of spatial modeling of land‐use change. First, it adaptively forms its own network structure to account for spatial heterogeneity. Second, it explicitly infers posterior probabilities of land conversion that facilitates the quantification of prediction uncertainty. Extensive calibration under various test settings is conducted on the proposed model to optimize its utility in seeking useful information within a spatially heterogeneous environment. The calibration strategy involves building a bagging ensemble for training and stratified sampling with varying category proportions for experimentation. Through a temporal validation approach, we examine models’ performance within a systematic assessment framework consisting of global metrics and cell‐level uncertainty measurement. Compared with two baselines, ART‐P‐MAP achieves consistently good and stable performance across experiments and exhibits superior capability to handle the spatial heterogeneity and uncertainty involved in the land‐use change problem. Finally, we conclude that, as a general probabilistic regression model, ART‐P‐MAP is applicable to a broad range of land‐use change modeling approaches, which deserves future research.     

Handheld Portable X‐Ray Fluorescence of Aegean Obsidians

We tested two portable XRF instruments (with different technologies) using two correction schemes (‘soils’ and ‘mining’) with both factory‐set calibrations and linear regression calibrations derived from published data. All four Aegean obsidian sources, including Sta Nychia and Dhemenegaki on Melos, were distinguished in each case. The newer instrument provided better ‘off‐the‐shelf’ accuracy than the older instrument, but calibrations negated these differences. The newer model also offered superior precision for most elements, despite measuring specimens for shorter times. Both correction schemes for the newer instrument resulted in almost equal precision, meaning that either may be used for successful Aegean obsidian sourcing.     

The Composition and Microstructural Variation of the Bronze Age Metal Ornaments from Lower Silesia (South‐West Poland): Chemical Analytical and Archaeological Aspects

Metal ornaments from Únětice (Mierczyce, Tomice, Jordanów and Opatowice) and Urnfield culture (Szprotawa, ?arek and Wroc?aw ?erniki) sites in south‐west Poland are made of bronze of varying composition, with the exception of the early Bronze Age Przec?awice site, where the copper items are found in graves from stages III/IV. The combined XRF and electron microprobe study of 37 ornaments (mostly pins) shows that those excavated from Únětice culture graves usually consist of cored dendrites, plus a Sb‐rich phase in some. The copper objects from Przec?awice correspond in composition to the East Alpine Copper or to the Ösenring copper and were produced from fahlore. Those from the Urnfield culture sites consist of homogeneous bronze. The Únětice culture ornaments were produced with little control of the effect of varying bronze composition on its microstructure. The results of XRF analyses of heterogeneous metal artefacts vary depending on the proportion of phases in the analysed site, and should be combined with micrometre‐scale analytical data and microstructural information yielded by electron microprobe methods. The same refers to deeply weathered or corroded objects made of homogeneous metal, which contain irregularly dispersed decomposition products.