1

Archaeometallurgy of Iron and Steel.

Vincent Serneels
Department of Geosciences, Mineralogy and Petrography
University of Fribourg, Switzerland

Introduction

Iron based alloys are a very important class of materials. Carbon is the most important alloying element (iron < 0.02 % C ;
steels 0.02 – 2 % C and cast iron 2 – 6 % C). The presence of carbon increases the hardness and brittelness of the alloy. It
decreases the melting point (from 1530° to 1130°). Quick cooling (quenching) can also harden the steels. Other elements like
phosphorus, sulfur, manganese, chromium, nickel and so on, influence in various ways the physical properties of the material.
Iron is one of the most important elements in the Earth crust (7 %) and is present in all kinds of rocks. The grade of an iron ore
must be high to be economically interesting (40 to 80 % Fe2O3). Many different metallogenic processes result in the formation
of iron ores (magmatic, sedimentary, metamorphic and alteration). Iron ores are abundant and widespread at the Earth surface.
Actually, only very large deposits are runned but during the Past, small occurences were mined with profit.
Only in meteorites, the iron occurs in the Nature as a metal. In general the iron is combined with oxygen atoms to form oxides
(magnetite: Fe3O4, hematite: Fe2O3, “ limonite ª: FeO.OH). There are also carbonates (siderite: FeCO3), sulfides (pyrite: FeS2),
silicates and so on.
The metallurgical treatment of an iron ore is basically a chemical reduction. The oxygen atoms bounded to the iron atoms must
be separated. This reaction requires a high temperature and the presence of a reducing agent. In most case, carbon plays this
role and the heath is provided by the combustion of a carbon-rich fuel (coal or charcoal). The temperature is one of the factors
controling the process. The time and the fugacity of oxygen are also very important.
At low temperatures (800°-1300°), the reduction process is possible in the solid state. It produces a piece of solid but spongy
iron poor in carbon. This piece must be consolidated by hammering but can be immediately used. This is the “ direct method”.
In Europe, it has been the normal way to produce iron until the medieval period. It was still practised at the end of the 19

th

century in some areas.
At higher temperatures (1300°-1600°), carbon diffuses in the iron and the melting occurs. The product is cast iron that can be
moulded but is too brittle to be hammered. It is necessary to remelt the alloy in an oxygen-rich atmosphere to reduce the
carbon content (finery process). Then, the metal can be shaped by hammering. This is the “ undirect method ”. The first blast
furnaces appear during the 12

th

century AD. It becomes the predominant method only during the 16

th

century.

To work the iron into objects is a process that can be very simple or very complex, depending on the quality of the raw metal
used (spongy iron, wel consolidated iron, recycled scrabs) and of the final product expected (from simple iron nail to
elaborated composite blade, heath treated and decorated). The basic actions are the hammering (plastic deformation) at high or
low temperatures and the procedures of heathing and cooling. The hammering gives a shape to the object and increases
hardness by reducing the scale of the grains. The heathing allows the grains to recristallise and to grow up, reducing the
brittleness. During the heathing, the carbon can be partly released from the alloy. The speed of cooling controls the formation
of special components increasing hardness. The joining of pieces of metal is a very important aspect of the work of the smith.
Iron has the capacity to be hot welded.

From the ore to the object, the process can be describe as a production line (“ chaîne opératoire ”).

Archaeological approach

The archaeological evidences related to iron working are numerous. The remains related to mining (pits, galeries, and dumps of
refuses and so on) are in general not very spectacular due to the fact that most iron ores are superficial, used as a bulk and
easily worked.
Furnaces are used for the reduction of the iron ores and they can be found as ruins during archaeological excavations. Several
types of furnace were used to produce iron by the direct method during the Past. A major difference is the way the iron is
separated from the slag during the operation. The slag can be tapped outside of the furnace or allowed to flow in the lower
part of the furnace. Some furnaces can be used only once and others are reusable several times. The air supply can be
provided by natural draught or using belows. The size and shape of the furnace can vary a wide.
During the reduction of iron ore, a slag, containing the impurities of the ore and the unreduced iron oxides, is formed. Very
large accumulations of slags are found in the areas of intensive production.

To work iron, the smith needs a fireplace and an anvil. This can be a very simple device but large workshops with elaborated
working structures are also known. Smithing requires high temperatures and for this reason the smithing hearth is always
equiped with belows.
Typical residues are formed during this activity. In the hearth, fused materials accumulate in the lower part. In general, those
slags have a hemispheric shape (PCB or plano-convex bottom). During hammering, the superficial layer of oxides formed on
the surface of the hot metal is broken into fine flat shaped particles, the hammerscales. Frequently, small bits of iron are lost

2

during the smithing.

To understand the production of iron at a given time, in a given area, one needs to know for each steps of the production line
(mining, reduction, smithing), the techniques used, the location of the activities and the size of the operations. Moreover, one
needs to know how and in what quantity iron has been used. The study of iron production must not be restricted to
technological questions but must be included in a general approach of the ancient societies, including economical, ecological
and social aspects.

Investigation on slags and related wastes

As furnace or workshop remains are scarce, frequently badly damadged and difficult to investigate without heavy
archaeological work on the field, slags are the most accessible material to study ancient metallurgy. It is of major importance
that slags were recorded by archaeologists as the other archaeolo gical remains and that they were investigated using
macroscopic observations and laboratory techniques. Before to select the samples for the analyses, it is essential to measure
the bulk quantities and to study the representativity of the samples. To measure the quantities is of great importance because,
at least up to a certain point, the amount of slags is related to the quantities of raw materials used and of iron products and
then reflects the size of the operations.
Different kinds of slags and associated wastes are produced during the reduction process and during the various stages of
the smithing process. They can then tell many different things.
Slags are very similar to natural rocks. They can be studied using basic or elaborated chemical and mineralogical techniques.
All kinds of information can be used to understand them. The chemical bulk compositions are giving many results about the
nature of the raw materials and the nature of the product. They can also be used to measure the efficiency of the processes
and to quantify the operations. The study of the mineralogical composition (bulk, phases or textural analyses) is giving more
informations on the conditions of formation.
The reduction slags are formed with the impurities and the unreduced iron oxides from the ore. From the chemical data, it is
possible to link the slags and the ore, to detect specific treatments of ore dressing and specific technologies (addition of
fluxes, etc). It is also possible to calculate a balance between ore, slag and iron that is a powerful tool to understand the
efficiency of the process and to measure its economic relevance. The interest and limitations of this calculation will be
discussed.
During smithing, the slag is formed by accumulation of different kinds of materials lost inside the hearth during the process.
Depending on what kind of work is going on, the sources of the materials will be different. The first source is the worked iron
itself. During heathing, the surfaces undergo oxidation and the crust of oxides will breake and fall down in the hearth. The
presence of trapped slag inclusions in the metal will also provide some material to the slag. To prevent oxidation and
decarburation or to clean the surfaces before welding, the smith can put sand, clay or some thing else on the metal. This is a
second imput to the slag. The hearth lining can provide some material if local melting occurs. Finally, the charcoal and the
other fuels give ashes that can be partly incorporated into the slag. The chemical composition will reflect those imputs and
allow describing them. On the other hand, the lost of iron during the process can be measured.
In this case, it is of particular importance to link the macroscopic observations and the analytical data.
Microscopic investigation of smithing slags is also a powerful tool. The slag rsulting of accumulation, the texture of cooling
can inform on the succesive steps of the work. A special attention is to be paid to the aspect, shape and nature of the
fragments of metal included in the slag. They reflect the nature of the worked metal.

General litterature on ancient iron

Benoit, P., Fluzin, P. (ed) 1995 : Paléométallurgie du fer et cultures. Symposium Belfort 1990, Belfort 1995.
Buchwald, V.F., Wivel, H., 1998 : Slag analysis as a method for the characterisation and provenancing of ancient iron objects,

Material Characterization 40 (1998), p.73-96.

Crew, P. 1991 : The experimental production of prehistoric bar iron, Historical Metallurgy 25/1 (1991), p.21-36.
Domergue, C., Leroy, M. (ed), 2000 : Mines et métallurgies en Gaule, recherches récentes, Gallia 57, p.1-158.
Dunikowski, C., Leroy, M., Merluzzo, P., Ploquin, A. 1996 : L’atelier de forge gallo-romain de Nailly (Yonne) : contribution à

l’étude des déchets de production, revue Arc héologique de l’Est 47 (1996), p.97-121

Eschenlohr, L., Serneels, V., 1991 : Les bas fourneaux mérovingiens de Boécourt – Les Boulies (JU/Suisse), Cahier

d’Archéologie Jurassienne 3, Porrentruy.

Feugère, M., Serneels, V. 1998 (dir) : Recherches sur l’économie du fer en Méditérranée nord-occidentale, Monographie

Instrumentum 4, Montagnac.

Fluzin, P., 1983 : Notions élémentaires de sidérurgie, p.13-44 in Echard, N. (ed) : Métallurgies africaines, nouvelles

contributions. Mémoires de la Société des Africanistes 9, Paris.

Leroy, M. 1997 : La sidérurgie en Lorraine avant le haut fourneau, l’utilisation du minerai oolithique en réduction directe,

Monographie du CRA, Paris.

Mangin, M. (ed), 1994 : La sidérurgie ancienne de l’Est de la France dans son contexte européen. Colloque Besançon 1993,

Annales littéraires de l’Université de Besançon 536, Paris 1994.

Magnusson, G. (ed), 1995 and 1996 : The importance of iron making. Technical innovation and social change, Conference

3

Norberg, Jernkontorets Berghistoriska Utskott (Stockholm) 58 and 62.

Pelet, P.-L. 1993 : Une industrie reconnue : Fer, Charbon, Acier dans le Pays de Vaud. Cahiers d’archéologie romande 60,

Lausanne.

Manning, W.H. 1985 : Catalogue of the Romano-British iron tools, fittings and weapons in the British Musuem, London.
Pleiner, R. (ed) 1989 : Archaeometallurgy of iron 1964-1987. Symposium Liblice 1987, Praha.
Pleiner, R. 2000 : Iron in archaeology, the european bloomery smelters, Praha.
Rostoker, W., Bronson, B. 1990 : Pre-industrial iron, its technology and ethnology, Archaeomaterials Monograph 1,

Philadelphia.

Scott, B., Cleere, H. (eds), 1986 : The craft of the blacksmith, Symposium Beéfast 1984, Belfast.
Serneels, V., 1993 : Archéométrie des scories de fer, Recherches sur la sidérurgie ancienne en Suisse occidentale, Cahiers

d’Archéologie Romande 61, Lausanne.

Tylecote, R.F., 1987 : The early history of metallurgy in Europe, London.