Arsenical bronze is an alloy in which arsenic is added to copper as opposed to, or in addition to tin or other constituent metals, to make bronze. The use of arsenic with copper, either as the secondary constituent or with another component such as tin, results in a stronger final product and better casting behaviour.
Since copper ore is often naturally contaminated with arsenic, the term "arsenical bronze" when used in archaeology is typically only applied to alloys with an arsenic content higher than 1% by weight, in order to distinguish it from potentially accidental additions of arsenic.
Origins in pre-history
|Ore name||Chemical formula|
Although arsenical bronze occurs in the archaeological record across the globe, the earliest artifacts so far known have been found on the Iranian plateau in the 5th millennium BCE. Arsenic is present in a number of copper containing ores (see table at right, adapted from Lechtman & Klein, 1999) and therefore some contamination of the copper with arsenic would be unavoidable. However it is still not entirely clear to what extent arsenic was deliberately added to copper  and how much its use arose simply from its presence in copper ores that were then treated by smelting to produce the metal.
A possible sequence of events in prehistory involves considering the structure of copper ore deposits, which are mostly sulphides. The surface minerals would contain some native copper and oxidised minerals, but much of the copper and other minerals would have been washed further into the ore body forming a secondary enrichment zone. This includes many minerals such as tennantite, with their arsenic, copper and iron. So the surface deposits would be used first, and with some work deeper sulphidic ores would have been uncovered and worked, and it would have been discovered that the material from this level had better properties.
Using these various ores, there are four possible methods that may have been used to produce arsenical bronze alloys. These are:
- The direct addition of arsenic-bearing metals or ores such as realgar to molten copper.
- This method, although possible, lacks evidence.
- The reduction of antimony-bearing copper arsenates or fahlore to produce an alloy high in arsenic and antimony.
- This is entirely practicable.
- The reduction of roasted copper sulfarsenides such as tennantite and enargite.
- This method would result in the production of toxic fumes of arsenous oxide and the loss of much of the arsenic present in the ores.
- This method has been demonstrated to work well, with little in the way of dangerous fumes given off during it, because of the reactions together among the different minerals.
Furthermore, greater sophistication of metal workers is suggested by Thornton et al. They suggest that iron arsenide was deliberately produced as part of the copper smelting process, to be traded and used to make arsenical bronze elsewhere by addition to molten copper.
Artefacts made of arsenical bronze cover the complete spectrum of metal objects, from axes to ornaments. The method of manufacture involved heating the metal in crucibles and casting it into moulds made of stone or clay. After solidifying it would be polished or in the case of axes and other tools work-hardened by beating the working edge with a hammer, thinning out the metal and increasing its strength. Finished objects could also be engraved or decorated as appropriate.
Advantages of arsenical bronze
Whilst arsenic was most likely originally mixed with copper as a result of the ores already containing it, its use probably continued for a number of reasons. Firstly, it acts as a de-oxidiser, reacting with oxygen in the hot metal to form arsenous oxides which vaporise from the liquid metal. If a great deal of oxygen is dissolved in liquid copper, when the metal cools the copper oxide separates out at grain boundaries and greatly reduces the ductility of the resulting object. It can lead to a greater risk of porous castings due to the solution of hydrogen in the molten metal and its subsequent loss as a bubble (but any bubbles could be forge welded and still leave the mass of the metal ready to be work-hardened).
Secondly, it is capable of greater work-hardening than is the case with pure copper, so that it performs better when used for cutting or chopping. There is an increase in work-hardening capability with increasing percentage of arsenic, and it can be work-hardened over a wide range of temperatures without fear of embrittlement. Its improved properties over pure copper can be seen with as little as 0.5 to 2 wt% As, giving a 10 to 30% improvement in hardness and tensile strength.
Thirdly, in the correct percentages, it can contribute a silvery sheen to the article being manufactured. There is evidence of arsenical bronze daggers from the Caucasus and other artefacts from different locations having an arsenic rich surface layer which may well have been produced deliberately by ancient craftsmen, and Mexican bells were made of copper with sufficient arsenic to colour them silver.
Arsenical bronze, sites and civilisations
Arsenical bronze was used by many societies and cultures across the globe. Firstly, the Iranian plateau, followed by the adjacent Mesopotamian area, together covering modern Iran, Iraq and Syria, has the earliest arsenical bronze metallurgy in the world, as previously mentioned. It was in use from the 4th millennium BC through to mid 2nd millennium, a period of nearly 2,000 years. There was a great deal of variation in arsenic content of artefacts throughout this period, making it impossible to say exactly how much was added deliberately and how much came about by accident. Societies using arsenical bronze include the Akkadians, those of Ur, and the Amorites, all based around the Tigris and Euphrates rivers and centres of the trade networks which spread arsenical bronze across the Middle East during the Bronze Age.
The Chalcolithic-period hoard from Nahal Mishmar in the Judean desert west of the Dead Sea contains a number of arsenical bronze (4–12% arsenic) and perhaps arsenical copper artifacts made using the lost-wax process, the earliest known use of this complex technique. "Carbon-14 dating of the reed mat in which the objects were wrapped suggests that it dates to at least 3500 B.C. It was in this period that the use of copper became widespread throughout the Levant, attesting to considerable technological developments that parallel major social advances in the region."
Sulfide deposits frequently are a mix of different metal sulfides, such as copper, zinc, silver, lead, arsenic and other metals. (Sphalerite (ZnS2), for example, is not uncommon in copper sulfide deposits, and the metal smelted would be brass, which is both harder and more durable than copper.) The metals could theoretically be separated out, but the alloys resulting were typically much stronger than the metals individually.
The use of arsenical bronze spread along trade routes into North western China, to the region Gansu – Qinghai, with the Siba, Qijia and Tianshanbeilu cultures. However it is still unclear as to whether arsenical bronze artefacts were imported or made locally, although the latter is suspected as being more likely due to possible local exploitation of mineral resources. On the other hand, the artefacts show typological connections to the Eurasian steppe.
The Eneolithic period in Northern Italy, with the Remedello and Rinaldone cultures in 2800 to 2200 BCE, saw the use of arsenical bronze. Indeed, it seems that arsenical bronze was the most common alloy in use in the Mediterranean basin at this time.
In South America, arsenical bronze was the predominant alloy in Ecuador and north and central Peru, because of the rich arsenic bearing ores present there. By contrast, the south and central Andes, southern Peru, Bolivia and parts of Argentina, were rich in the tin ore Cassiterite and thus did not use arsenical bronze.
The Sican Culture of north western coastal Peru is famous for its use of arsenical bronze during the period 900 to 1350 AD. Arsenical bronze co-existed with tin bronze for in the Andes, probably due to its greater ductility which meant it could be easily hammered into thin sheets which were valued in local society.
Arsenical bronze after the Bronze Age
The archaeological record in Egypt, Peru and the Caucasus suggests that arsenical bronze was produced for a time alongside tin bronze. At Tepe Yahya its use continued into the Iron Age for the manufacture of trinkets and decorative objects, thus demonstrating that there was not a simple succession of alloys over time, with superior new alloys replacing older ones. There are few real advantages metallurgically for the superiority of tin bronze, and early authors suggested that arsenical bronze was phased out due to its health effects. It is more likely that it was phased out in general use because alloying with tin gave castings which had similar strength to arsenical bronze but did not require further work-hardening to achieve useful strength. It is also probable that more certain results could be achieved with the use of tin, because it could be added directly to the copper in specific amounts, whereas the precise amount of arsenic being added was much harder to gauge due to the manufacturing process.
Health effects of arsenical bronze use
Arsenic is an element with a vaporization point of 615 °C, such that arsenical oxide will be lost from the melt before or during casting, and fumes from fire setting for mining and ore processing have long been known to attack the eyes, lungs and skin.
Chronic arsenic poisoning leads to peripheral neuropathy, which can cause weakness in the legs and feet. It has been speculated that this lay behind the legend of lame smiths, such as the Greek god Hephaestus.
A well-preserved mummy of a man who lived around 3,200 BCE found in the Ötztal Alps, popularly known as Ötzi, showed high levels of both copper particles and arsenic in its hair. This, along with Ötzi's copper axe blade, which is 99.7% pure copper, has led scientists to speculate that he was involved in copper smelting.
Modern uses of arsenical bronze
Arsenical bronze has seen little use in the modern period. It appears that the closest equivalent goes by the name of arsenical copper, being defined as copper with under 0.5 wt% As, below the accepted percentage in archaeological artefacts. The presence of 0.5 wt% arsenic in copper lowers the electrical conductivity to 34% of that of pure copper, and even as little as 0.05 wt% decreases it by 15%. Therefore, there is no demand for copper containing arsenic in electric wires etc., one of the major modern uses of copper and steam engine boilers are no longer made from it, leading to no modern use.
- Charles, J. A. (January 1967). "Early Arsenical Bronzes – A Metallurgical view". American Journal of Archaeology. 71 (1): 21–26. JSTOR 501586.
- P Budd and B S Ottoway. 1995. Eneolithic Arsenical copper – chance or choice? In: Borislav Jovanovic (Ed) Ancient mining and metallurgy in southeast europe, International symposium, Archaeological institute, Belgrade and the Museum of mining and metallurgy, Bor, page 95.
- Thornton, C.P.; Lamberg-Karlovsky, C.C.; Liezers, M.; Young, S.M.M. (2002). "On pins and needles: tracing the evolution of copper-based alloying at Tepe Yahya, Iran, via ICP-MS analysis of Common-place items.". Journal of Archaeological Science. 29 If a great deal of oxygen is dissolved (29): 1451–1460. doi:10.1006/jasc.2002.0809.
- Lechtman, H.; Klein, S. (1999). "The Production of Copper–Arsenic Alloys (Arsenic Bronze) by cosmelting: Modern Experiment, Ancient Practice". Journal of Archaeological Science. 26 (26): 497–526. doi:10.1006/jasc.1998.0324.
- De Ryck, I.; Adriens, A.; Adams, F. (2005). "An overview of Mesopotamian bronze metallurgy during the 3rd millennium BC" (PDF). Journal of Cultural Heritage. 6 (6): 261–268. doi:10.1016/j.culher.2005.04.002.
- Tylecote, R.F. (1992). A History of Metallurgy (2nd ed.). London: Maney publishing. ISBN 0-901462-88-8.
- Lechtman, Heather (Winter 1996). "Arsenic Bronze: Dirty Copper or Chosen Alloy? A View from the Americas". Journal of Field Archaeology. 23 (4): 477–514. doi:10.2307/530550. JSTOR 530550.
- Thornton, C.P.; Rehren, T.; Piggot, V.C. (2009). "The production of speiss (iron arsenide) during the Early Bronze Age in Iran.". Journal of Archaeological Science. 36 (36): 308–316. doi:10.1016/j.jas.2008.09.017.
- Ryndina, N. 2009. The potential of metallography in investigations of early objects made of copper and copper-based alloys. Journal of the historical metallurgy society. 43, 1-18.
- The Nahal Mishmar Treasure at Metropolitan Museum
- Jianjun Mei, page 9 in Metallurgy and Civilisation, Eurasia and beyond, ed: Jianjun Mei and Thilo Rehren. Proceedings of the 6th international conference on the beginnings of the use of meals and alloys (BUMA VI), 2009,Archetype publications, London.
- Eaton, E. R. 1980. Early metallurgy in Italy. In: ed. W. A. Oddy, Aspects of early metallurgy, occasional paper 17, British Museum Publications, London.
- Hörz, G.; Kallfass, M. (December 1998). "Metalworking in Peru, ornamental objects from the Royal Tombs of Sipan". Journal of Materials. 50 (12): 8. doi:10.1007/s11837-998-0298-2.
- Harper, M. (1987). "Possible toxic metal exposure of prehistoric bronze workers". British Journal of Industrial Medicine. 44 (44): 652–656. doi:10.1136/oem.44.10.652.
- Age determination of tissue, bone and grass samples from Ötztal Ice Man (PDF; 476 kB)
- <Please add first missing authors to populate metadata.> (16 September 2002), Iceman's final meal, BBC News
- "Arsenical copper carpenters tools from Naxos, circa 2700 to 2200 BC". British Museum.
- "Results page, with some information on arsenical bronze". Sican archaeological project.