Of iron and fire: the metals of Notre-Dame through the centuries

Research carried out over the past twenty years has shown that the Gothic cathedrals of the XNUMXth^e century like Bourges, Chartres, Rouen or Troyes used iron as a building material. The great churches of the twelfth^e century remain relatively unknown.

The fire of Notre-Dame de Paris revealed that the structure of the cathedral, erected from the 1160s, was also reinforced with numerous iron frames. Some of them, hidden in the stone or by the framework, were hitherto unknown.

The methodologies developed over more than twenty years and already tested on dozens of medieval buildings make it possible to question the iron constructions discovered at Notre-Dame in order to renew our knowledge of this building and ancient construction techniques, but also that of the economy of iron in medieval times.

Several dozen iron fittings, damaged by the fire or removed during restorations, can now be scrutinized under the eye of optical and electronic microscopes. These are both "staples" implemented in the masonry, in the stands, in the monolithic columns, and at the top of the walls under the burned frame, but also reinforcements linked to the frame itself. (nails of various sizes, keyed and bolted rods) which testify to restorations over the centuries.

The microstructures of these metallic elements contain part of the history of Notre-Dame de Paris and its site, which today are explored by archaeologists, chemists and archaeometers - specialists in the physico-chemical analysis of ancient materials, gathered within the Metal working group du Notre-Dame scientific site.

What metals for cathedral builders?

The first question is that of the quality of the metal used by the builders. What choices did they make both in medieval times and during the restorations? Has the fire altered the properties of these construction irons? Observation of polished surfaces under an optical microscope, after etching with specific chemical reagents, reveals the microstructure of these ferrous alloys, their shaping (folds, welds, etc.) and assesses their degree of heterogeneity.

Old irons are indeed often heterogeneous, in connection with the production processes used, in particular low furnaces, which do not allow production of iron in the liquid phase before the middle of the XNUMXth century.^e century. In particular, they contain varying levels of carbon or phosphorus (the two main elements that combine with iron), forming composite materials between iron, mild steel and phosphorous iron.

Hardness tests can be carried out to find out their respective hardnesses. These old irons also contain many non-metallic impurities, called "slag inclusions", which are all weak points in the structure of the material.

From samples of whole objects, it is possible to machine tensile specimens, ie parts of standard manufacture and dimensions, to subject them to mechanical tests and determine the physical properties of these materials (elastic modulus, tensile strength, elongation at break, etc.). At Notre-Dame, analyzes have shown that medieval iron, like that used in the XNUMX^e century, has lower mechanical properties than those of contemporary alloys, like what has already been established on other medieval and modern monuments.

However, the fire does not seem to have altered the iron's microstructure or its already limited mechanical properties. This information can be used for restoration and fuel reflection on the reuse of these materials.

Better understand the construction timeline

The second question concerns the chronology of the reinforcements implemented.

The small amounts of carbon present in the steely zones of these ferrous alloys (steel is an iron-carbon alloy) provide access to crucial data in terms of dating. We know today extract that carbon, which comes from the coal used in the reduction furnace that produced the iron, to carry out a carbon-14 dating.

The steely zones revealed by the metallographic attack are removed with a drill, then the carbon is extracted by combustion and recovered in gaseous form (CO₂) and finally reduced to graphite. Mass spectrometry with particle accelerator (AMS) carried out at the LMC14 then makes it possible to determine the quantity of carbon 14 remaining and to compare it with the reference curves to determine the age of the material and deduce the period of iron production.

This technique, developed less than ten years ago to date old ferrous alloys, makes it possible to Distinguish reinforcements contemporary with the construction from those added a posteriori as consolidations.

For example, the comparison of the dating of the staples placed in the galleries of the choir (built around 1160-1170), those of the nave, later, and those at the top of the high walls at the beginning of the XNUMXth^e century, will also shed light on the gradual insertion of this material into buildings and reconstruct the thoughts of the builders who used it.

Similarly, in the frame, which has been reworked many times and where multiple frames are used, the dating will make it possible to highlight the restoration phases prior to the XNUMXth century.^e century, still totally unknown, by crossing this information with the study of wood carried out in parallel.

Where does the Notre-Dame iron come from?

The third question concerns the provenance of these materials. Where were these tons of iron produced? How was the construction site supplied with metal in the twelfth^e century, in the XNUMXth^e century and for subsequent campaigns; in what forms circulated these irons ?

We know that iron circulated in the Middle Ages in both directions on the Seine and could also be supplied by land; it could have been produced by domains of the bishopric or the cathedral chapter (the college of canons in charge of the management of the building), and particularly in the many domains located throughout the current Île-de-France.

It is also not uncommon for several sources, local and more distant, are mixed up during construction campaigns. The welds observed in the iron matrix of Notre-Dame suggest the assembly of several parts to make each staple, whose origins may be different. The impurities contained in these ferrous alloys make it possible to approach this question of the circulation of these materials.

In the old production processes in low furnaces, mainly used until the end of the Middle Ages, the iron, produced in the solid state, carries in its matrix small fragments of slag (the waste of the reduction), called inclusions slag, the composition of which depends on the unreduced fraction of the ore used.

The chemical analysis of these inclusions, performed by mass spectrometry coupled with a laser ablation module, which facilitates the analysis of volumes with a diameter between 30 and 100 micrometers, provides access to their trace element composition.

This composition can be compared using statistical tools to that of the macroslag rejected by the metallurgists on the production sites.

Thus, a vast archaeological investigation must be carried out in parallel, in building on existing findings in the region, in order to identify the slag sites that may have provided this iron and to carry out the analyzes necessary to establish these comparisons of chemical signatures.

The investigation has only just begun.

Maxime L'Héritier, Lecturer in medieval history, Paris 8 University – Vincennes Saint-Denis

This article is republished from The Conversation under Creative Commons license. Read theoriginal article.

Image credit: Shutterstock.com / godongphoto / Notre-Dame de Paris cathedral after the fire, France, June 30, 2019