Main bronze index page
What can you expect on this page?
- Detailed information about:
- The start of the British bronze age.
- The developement of the techniques
- Moulds
- Copper alloys with arsenic
- Copper alloys with tin
- Copper alloys with lead
- Copper alloys with minor elements
- Hardness
- Some artifacts
- Copper alloys of the iron age
More info about 5300 BC - 4900 BC
Information about the appearence of copper in the prehistory
Information about the appearence of bronze in Great-Britain
The British bronze age - Metallurgy
The next information is a full version of the general site about the British bronze age
The introduction of copper.
The smelting of copper ores need a high pyro-technical level of knowledge,.so it usually first occurred after the introduction of pottery.
The developement stages are in general:
- the winning and use of hammered, "metallic" copper
- the smelting of metallic copper
- the mining and smelting of copper ores
- casting the copper ores in moulds
- the preference of the harder arsenic coppers
- the preference of tin-copper (bronze) alloys
Approximate calibrated C14 dates for the mining of copper ores are as follows:.
.Ai Bunar 4000 B.C. Rudna Glava (Serbia) 3800 B.C. Iberia 3000 B.C. Mitterberg (A) 2500 B.C.
These dates suggest an east to west movement of people seeking copper ores from the 5th millenium B.C.
It is unlikely that they were seeking copper ores merely for pigments since there were traces of smelting in the first and third areas.
These sites tend to confirm the old theory of Mediterranian diffusion of technology. We have as yet no evidence for the spread of copper mining across Europe from south-east to north-west, .apart from the Austrian Mitterberg evidence which is somewhat later. It is therefore possible to envisage diffusion of copper smeltng from the Near East taking place along the Mediterranian by the use of the minerals on the northern shores in the years around 4000-3000 B.C. and reaching Ireland by about 2500 B.C.
Meanwhile there was another, trans-European, current from south-east to north-west which arrived in mainland Britain a little later, about 2200-2000 B.C., which can be identified with Beaker .traditions of burial. By 2000 B.C. these two currents had merged and spread in many directions.
one inteeresting feature of early British metallurgy is the rapid introduction of the use if tin bronze.
Whereas in the Near East it took from 6000 B.C. to 3000 B.C. to introduce bronze, in Britain it was being used in 2000 B.C., only 400 years after the introduction of a copper-using technology. It has been suggested the the tin was panned from the stream beds of rivers in Ireland in sufficient quantities to make about half the tin needed (200 kg) for the bronze made in early bronze age Ireland.
It is certain that in Britain, Cornish deposits were exploited and tin bronze gradually ousted .arsenical copper throughout the British Isles in the years 2000-1800 B.C. The earliest date for the .production of metallic tin in Britain is given by the piece of tin smelting slag found in a rtual burial in Cornwall, dated about 1800 B.C. Before this, bronze bronze could have been made either by adding smelted tin metal or by adding high-grade tinstone pebbles under a reducing cover to a copper charge in a crucible.
It is very probable that in the early phases of the bronze age the metallurgists were a class.apart, and were not entitled to ceremonial burial. Thus detribalized, they were liberated from the .bonds of local custom and they therefore had the freedom to travel, and perhaps settle where they.could find markets for their products, even abroad. Since copper and, more particularly, tin ores .are restricted to certain well defined areas of the British Isles, the early metallurgist had to .have his own personal trade contacts. It is perhaps possible that he mined and smelted his own material and then carried the finished products round certein limited areas. Alternatively, the itinerant caster with his few stone moulds travelled the area calling at intervals on the smelters in permanent settlements in order to replenish his supplies.
The late bronze age.
The late bronze age altered much of this. The use of clay moulds, instead of a few stone moulds, widened the technique; bronze workers seem to have been more common and processes wre capable of greater and more rapid output. Much of the metal of this period occurs in the form of hoards. .
These are mainly of two types. Firstly, those which seem to have been deposited as part of some ritual requirement, often in marshy ground as part of some water cult; these are often bent or mutilated.
Secondly, those which are clearly founder's hoards, containing scrap metal and pieces of copper ingot.
The pieces are often broken or bent to make them more suitable for remelting.
Many of these hoards can be dated to the later phases of the late bronze age and it is possible that the smith never returned to collect his hoard because the demand for this metal had ceased. Hoards of this period contain tens, or even hundreds of items.
Secondary (scrap) metal seems to have constituted a large proportion of the total amount of metal used.
Moulds.
The bronze founder no longer needed to seek out a piece of suitable stone and carve it, but could now make a clay mould or cast a bronze mould whenever it was necessary. Most of the fragments of clay moulds found are from two-piece moulds and are bot single-piece 'investments', which tends to show that a straight-forward moulding process was used in which the two pieces of moulding material were pressed around an oiled wooden or metal pattern to obtain the impression. The pattern was removed by stripping off one of the half-moulds from the pattern, whereupon the pattern could be removed from the other half-mould and re-used. After drying, the two half-moulds could be reassembled like the stone moulds but could be held together by an outer wrapping of mould material. It works even better when the half-wet moulds dry upon eachother.
The alternative technique is the "lost wax" process in which a wax model is invested with clay or loam (with a charcoal layer!), then dried, and the wax melted out. This needs a new pattern every time, or some technique for the rapid reproduction of wax models (Bronze moulds couldn't have been used for this). There is no evidence for the use of this process in the British Isles before the early iron age.
By the late bronze age there was a good deal of international trade and an enormous increase in output.
It seems that bronze workers travelled considerable distances on pack-horses, so large are the hoards now found. The number of late bronze age objects surviving is large; the hiards exceed those of all previous phases put together by ten times, and as a rule contain a larger number of castings. Previously, copper or bronze was reserved mainly for weapons and smith's tools, but now the community could afford to use cauldrons, buckets, carpenter's tools and cermonial objects, after the reduction in price ensuing from increased output.
Naturally, the smelting branch had to expand its activities, and the supply of metal from Ireland was supplemented by the exploitation of sources in Wales, Scotland and from the Continent.
It was probably not long before peoples entering the country by way of the Irish Sea found plentiful supplies of material exposed in outcrops, whereas those entering the country from the east would be disappointed, and would take time to develop and exploit the ores of the western seaboard and Ireland.
It is prabable that south Ireland was the main source of copper in the west.
Primitive tools such as stone hammers and wooden shovels have been reported from many workings. .
In the absence of other more rliable dating evidence these finds cannot be accepted as proof of prehistoric working, for primitive tools were used in mining until recent times. But the Mount Gabriel mine has yielded giving a date in the 14 th century B.C.
To get a better view over the life of a bronze smith and the cultures during the British bronze.age it is good to research the alloys and forms of metal objects in that era.
We than discover the great metallurgic knowledge that was available by prehistoric man. .
Copper alloys with arsenic.
The following is an example of alloy developement in the beginning of the British bronze.age. Irish halberds were invented in the copper-using period and are similar to the Portuguese.Beaker type. They started as impure copper (<1 % arsenic) , became copper (Cu) or copper -.arsenic (CuAs) alloys, copper - tin (CuSn) alloys in Britain, and later copper - arsenic alloys with.high As (under international influence). The As present in the early Irish-made artifacts was.natural, while the later, high As levels were due to added As minerals under continental.influence. But the easy availability of tin on the British mainland favoured the use of tin bronzes.rather than the alien Cu-As tradition.
Still, high As-contents of up to 7,0 % are encountered.
Effect of arsenic and tin on the hardness (HV) of copper
composition (%) as cast work-hardened
=============================================================
A.Pure copper 100 40 120
=============================================================
B.Arsenic 2 140
4 195
6 210
8 87 215
=============================================================
C.Tin 2 50 140
4 165
6 185
8 210
10 230
15 140 300
100 5*)
*)pure metallic tin: 5 hv
Whereas the effect of arsenic on the strength of copper in the cast or annealed condition is.relatively slight, as little as 1,04 % As (arsenic) will raise the maximum strength of hammered .copper from 124 to 177 HV. (watch table) Thus the enormous effect of arsenic on the hardness of a hammered .cutting edge would not fail to have been noticed, and efforts would have been made to reproduce.the conditions which gave this result. Amounts greater than 1,25 % have a special significance. .
They're called "arsenical coppers".
They either stem from specially selected ores or represent coppers to which additions of arsenical minerals have been made.
It is unlikely that such additions would have been pure. .
One wonders how much early smelters knew about the hardening effect of impurities, and whether .the arsenic and antimony contents were maintained at a certain level intentionally. (resp. 1,5 and 0,5 % average) .
Where arsenic occurs in an otherwise pure metal, it was added deliberately perhaps as a high arsenic mineral. .
It has been noted that halberd rivets were often softer and had lower arsenic contents than the blades.
This mow seems to have a quite simple explanation. When arsenical copper is heated to a hot forging temperature (800' C) .some of the arsenic is lost as white fume owing to oxidation on the surface. .Naturally, the smaller the artifact, the more arsenic is likely to be lost as the surface area / .volume ratio increases and also the time required to make the smaller object. The volatilization .of arsenic under these conditions shows itself as a white fume, and by observing the extend of this, .one can get some idea of the arsenic content and the amount being lost..
Copper alloys with tin.
Examinations of 450 British middle and late bronze age artifacts contained 0,7 - 37 % Sn (tin), but .the majority were near to the normal figure of 10 %.
The 2 traditions, the Irish impure copper and the British tin bronze, were carried on side by.side until the Arreton Down phase (Wessex II). By about 1500 B.C. the British Isles were.converted to tin bronzes throughout.
The odd case of a Beaker arsenical copper knife associated with a bronze rivet is put down.to re-use and re-riveting in a later period when the normal stock-in-trade of the smith was bronze.But the high nickel (Ni) content of both rivet and knife suggests a continental origine.
A Ni content higher than 0,1 - 0,35 % is considered representing a probably Central European.origine.
As the reduction in Ni (nickel) content suggests a different sorce of ore, the change in As and.Sb (antimon) could be due to differences in smelting technique. .
Copper alloys with lead.
In Wales, the Acton Park phase saw the introduction of 7 % or more lead to the bronzes, undoubtedly added as lead metal. .This stage did not last and the lead content was progressively diluted, probably with lead-free scrap, until its .revival in the south-east of England in the Wilburton phase (950 - 750 B.C.). Although about 2 % Pb (lead) has a pronounced .beneficial effect on the casting properties of a tin bronze, greater amounts than this are probably added as a diluent. .
In southern Britain the first phase of the late bronze age is represented by the Wilburton.hoard which, as it has high levels of As, Sb, Ni, and Ag (silver), is probably heavily contaminated with scrap .from the continent. It also contains large amounts of additional lead. The material from northern .Britain is free of lead, but the ingot material often associated with this is of high purity.
On the whole, by the end of the late bronze age the lead content had fallen to 2 - 3 %.
Having established that lead was an intentional addition in the middle and late bronze ages, two .things follow: the metal was added either to facilitate casting, i.e. to increase the fluidility .of the bronze, or as a cheap diluent to an expensive metal. But it also establishes the important .fact that lead was readily smelted by the early middle bronze age, which, considering its apparent .scarctiy as a metal in its own right at this time in Britain, is of some importance..
Minor elements, impurities.
Some elements (like silicium, iron and mangan) are mainly present in the form of slag. The quantity .of slag is mainly between 1 and 3 % of the metal.
Antimony and silver will also contribute something to the hardness.And so by obtaining a purer copper by improving their technique, early smelters had reduced its strength. .
Zinc doesn't play a part in this research. It is volatile but no more so than arsenic, but while the arsenic like .sulphur has a strong affinity for the copper, the zinc tends to find its way into the slag. ..
Hardness.
The best way of obtaining some idea of the mechanical properties and therefore of the value of an implement as.a cutting tool is to measure its hardness. This is normally done by an indentation test where a pyramidal diamond.is pressed into the prepared metal under a controlled load and the size of the indentation measured with a microscope .(Vickers test). The result is normally expressed in kg/mm2 ("HV").
Since the artifacts can be hammered on different portions, the hardnesses can also vary on the same subject.E.g., the cold-worked flanges can reach over HV200, thick parts can remain on HV 140..
Some bronze age artifacts.
An Irish halberd (2,62 % As) had a hardness of 80 (in the cast portion) and 132 in the cold-worked portion.
This should be compared with the hardness of the cast tin-bronze of 120-160, and the pure copper of 40...
The high arsenic contents contribute appreciably to the hardness of these weapons.
A flat axe (one of a hoard of 7) contained 8 % Sn and 0,5 % As. It had not retained its original cast structure .as it had been considerably worked. It would appear that some of this work had been done cold, during ritual fracturing.along a line of porosity and inclusions, before being buried.
Four early bronze age halberds and a dagger appeared to have been cast from the pointed end in two-part moulds.and slowly cooled. Three of these objects have been made from arsenical copper (between 0,62 and 2,6 %).
They have predominantly cast structures but have been forged to their final shape and cold-hammered at the edges.
This has caused an increase in hardness from 73 to 153. Some oxidation was apparent, as shown by the presence of.As- and Sb-oxides. This may seem to be very difficult to avoid, especially in the high-arsenic alloys (>2,5 %) and may .lead to brittleness. It is one of the reasons why high arsenic alloys are not used today.
Two halberds were made from tin bronzes containing 8,1 and 5,8 Sn. These had received much the same .treatment as the arsenical coppers. This had had the effect of raising the hardness from about 65 to 133. .
The rivets in this case had not been made of a more easily workable alloy than the blades themselves. .
A rivet in an arsenical halberd had the same structure and hardness as the blade, and had been made of the same material.
A rivet in the bronze dagger contained 11,28 % Sn, and had been finally forged cold after previous working and.annealing. The hole in which it had been inserted was drilled from one side only.
Conclusions of the research from these items:.
Irrespective of material, the halberds had been.
- .
- made from blanks, cast in a two-part mould.
- the mould had been made of stone or clay.
- the mould had been pre-heated.
- the halberds then received a light forging to trim of the flash at the parting line.
- the halberds had then been forged and annealed, and finally.
- the cutting edge had been cold-hammered and ground.
The above examples show that it was common practice to increase the strength of a cast alloy by working.
However, this was not always done.
A cast copper, thick butted flat axe from Ireland appeared to have been slowly cooled after casting, possibly in .a pre-heated closed stone or clay mould. This is not a good casting by modern standards; the porosity and inclusions may.be due to the volatilization and oxidation of arsenic (1 %).
An ogival bronze dagger belonging to the Wessex culture. The cutting edge had been worked,.but the hardness so obtained had been removed by subsequent heating.
The sact that the cold-work resulting from decorating the surface with incised grooves still remained was proof that.it had not been softened by heating in a funeral pyre. It appears, therefore, that either its maker was unfamiliar with the fact.that cold-hammering hardens metal, or else had made a mistake.
An examination of an Irish flat axe made from a 1,5 % As copper showed that it had been cooled fairly slowly in a clay .or stone mould.
The melt was deoxidisedand poured from a temperature of about 1100'C, i.e. just above the melting point of pure copper.
After casting, the axe had been forged at the top (butt) and near the blade, but not in the centre.
For some reason the edge of this axe had also been left in the soft condition, and it was suggested it had been used as a.pattern for clay moulding and had been annealed during the heating of the mould to harden the clay.
While this is not impossible, it is rather surprising, since a two-part mould would be required and the pattern could be removed.before firing the clay. It seems more likely that the axe has been annealed inadvertently either by faulty workmanship or, later, .by falling into a fire.
The softening temperature of 1,5 %As copper would not be much above 300'C which could easily be reached by a domestic fire.
A palstave was found to contain 12-15 % Sn. It had been cast in a two-part mould, and had been poured from a temperature of 1200'C.
This is unnecessarily high, since a 12 % tin bronze melts at about 950'C.
Insufficient feedinghad been employed so that the top was choked with dross.
After casting, it had been reheated and the flanges lightly forged until the temperature dropped to below 550'C.
The palstave was then reheated and a herringbone pattern engraved on the flanges.
A Breton-type socketed axe containing 43% Pb and 1,46 % Sn, hasn't been forged at any stage in its manufacture. .
It contained no slag or dross, but about 1 % oxygen. The lead was found to be evenly distributed.
It had been cast in a two-part mould, the untrimmed flash still being visible along a plane through.the loop of the axe. The core had been made of clay and had itself been made in a two-part mould as shown by the.grooves on the inside of the socket, which are not quite down the centreline.
Some clay and sand were found at the bottom of the socket, these being the remains of the core.
It is now believed that these "Breton" axes are some sort of currency or are, perhaps, ritualistic.
A wing-flanged axe consisted of a high purity bronze containing 17,3 % Sn. Here again the cast structure has entirely .disappeared, and it was clear that a considerable amount of work, some of it cold, had been done on the .original casting.
After cold-working it had been annealed at a comparatively low temperature and again cold-worked in places such as the .flanges. In the flanges the cold-working has increase the hardness to 200 and had caused some cracks.
It is clear that the flange had been bent over by cold hammering to make it grip the haft.
Experiments on pieces of metal of the same composition proved that the cast structure had been removed by annealing at 700'C .followed by repeated hammering and annealing at a somewhat lower temperature.
the actual hardness (HV) results were as follows: .
.Axe. Thick parts 140. Edges and sides 178. Flange 200. Made-up alloy . As cast 127. Worked and annealed 116-127. Annealed and cold-worked 182..
This table summarises the hardnesses actually achieved by bronze age smiths.
Cast tin bronzes should have hardnesses of the order of 100 HV and the table here shows that most of the .bronze objects have been well forged and homogenized and start at about this figure. When the lead is low,.edge hardnesses rach 170 owing to cold-working but drop to about 120 when appreciable lead is present. .Clearly, bronze age smiths rarely achieved the figure of 224 which is known to be possible on 13 % tin.bronzes azst in stone moulds and hammer hardened..
Copper alloys of the iron age.
By the end of the late bronze age there must have been a lot of tin bronze in circulation.
At the beginneng of the iron age much of it was lost or merely dumped by smiths who either had little use.for it or who went out of business altogether.
An increasing number of iron age hillforts are now found to have had a bronze age occupation so one can assume that .there was some continuity of metal working. It is possible that the realization that there was a connection between copper,.at least on the smelting side, had drawned on a large number of metal workers. For the smelters had been adding ferruginous.fluxes such as hematite to siliceous copper ores for centuries and the fact that there was more than one metal in raw copper must have been noticed by many bronze workers.
However, the idea that this 'impurity' was the new metal recently introduced by Hallstatt emigrants probably came as a surprise, and certainly iron working is an example of a diffused technique.
Many smiths would be converted to the new metal. Others would continue to make copper-base alloys on a reduced scale, although it is possible that specialization did not occur at this time on settlement sites but only on the smelting sites.
One important site is that of Gussage all Saints, a hillfort which shows detailed evidence of an intensive .industry for the manifacture of horse fittings. This is the first evidence in the U.K. for the use of the lost-wax process.
Small bronze fittings were made, together with bronze-plated steel bridlebits.
This seems to be evidence for an integrated ferrous and non-ferrous industry.
Unfortunatelyit is not known whether there were separate specialists doing the steel working and bronze working. Both aspects are represented on the site. The bronze was standard tin bronze with a low zinc content.
March, 5th 1998.