For the axe-making project we chose to stick to refining the iron by folding and forging, but in doing so, we realised just how labour and resource-intensive this method is. Re-smelting would have provided a considerably more efficient processing method.
The process
A large forge or small furnace is filled with charcoal and preheated to a high temperature. Small pieces, or preferably bars, of bloom iron are introduced to the charcoal through the top of the furnace. The bellows are pumped continuously, and as the charcoal burns, more is added. Once the back end of the first iron bar has sunk below the surface, another is introduced. The furnace heats the tip of the iron bar so that it melts; the following bars are also liquefied, forming a sticky pool of iron in the bottom of the furnace. The end result is a continuous, relatively compact and pure iron bloom. While slag is mostly extracted from the iron, some of it still remain.
In order to make iron forgeable by re-smelting it, yet avoiding carburizing the iron into steel in the process, it is necessary to control the supply of air to the furnace so as to keep the temperature at a moderate level. Also, to prevent the bloom’s environment from becoming reducing and too rich in carbon gases, it is important that the furnace below the tuyere, or air inlet, is not too deep. It is also important to keep introducing crushed slag, as this will form a molten screen that protects the iron surface from oxidation (burning).
Saving resources
The re-smelting process makes it possible to refine one kilogram of iron by using only 2-3 kilograms of charcoal; the burning time is approximately 30 minutes. This is considerably less wasteful than processing by folding. Also, the iron’s weight loss is considerably lower compared to fold-processing. The re-smelting method involves a loss of material which is considerably lower than 50%, and the level of loss is most probably related to the amount of slag in the sponge iron. This has been shown by various tests in the so-called Aristotle furnace (Sauder 2010). It seems that Mads Jylov has had similar experiences (Jylov 2009:123,127).
Ole Evenstad
In 1782 Ole Evenstad wrote a textbook on the extraction of iron and steel from bog ore; this was published in 1790. Chapters 8 and 9 describe the process of re-smelting for the purpose of producing iron and steel (Evenstad 1790):
Chapter Eight. About the re-smelting of bloomery- or crude iron, in the forge, to make finer iron.
Iron which has only been transformed from ore to crude iron in the bloomery, is ill suited for any practical purpose whatsoever, even if it is of the very best kind; it may well be wrought and worked into crude tools, but when it breaks, it is impossible to weld back together. It retains a number of impurities, which have to be expelled before the iron is perfected, and this can be achieved by re-smelting in the forge.
Read more ...
Evenstad describes his experience of yield and loss of material when re-smelting sponge iron as follows (Evenstad 1790:439):
In the first re-smelting process [of the day], the iron often loses some of its weight [... but provided ample quantities of crushed slag are introduced ...] and the smelting is continued, then only very little, or even nothing, is lost.
In chapter 12 of his textbook Evenstad calculates the cost of production, quoting a weight loss of only 6–11% when bloomery iron is re-smelted into finer iron (Evenstad 1790:447). This means his sponge iron must have had a very low slag content. It has been suggested that Evenstad may have re-smelted sponge iron for other reasons than merely to reduce the slag content and consolidate the iron. Lee Sauder has considered the removal of phosphorous.
Was the method of re-smelting used in the Viking Age?
The best written source available to us that describe the re-smelting is Ole Evenstad’s textbook from the late 18th century. Sadly, this is far more recent than the Viking Age, but we do have a few less detailed sources implying use of the method further back in time. One source is from the pre-roman Iron Age, the Meterologica by Aristotle. Another is from c. 1120, De diversis artibus by the monk Theophilus. You can read more about these older sources in the section Making steel by re-smelting iron.
It is challenging to establish with certainty whether archaeological finds show use of the re-smelting method. In Scandinavian archaeology the terms used for bloomery iron are somewhat undefined and interchangeable (Martens 1979, Hauge 1946, Mortensen 1940). We have terms like 'luppe' (bloom), 'blesterjern' (bloomery iron) and 'fellujern' (NO: fellujárn). The term 'blesterjern' must refere to the actual raw bloom; an uneven lump, with almost the same height and width. We also find much flatter blooms, probably what the medieval sources call 'fellujern'. These flatter blooms are reminiscent of the end result from the re-smelting, as described by Evenstad:
When the lump of iron and the slag have melted and fallen down in the forge the iron takes the form of a flat lump; the flatter it is, the better the iron. This lump is taken up immediately with tongs; on a large stone, which must be ready by the forge, it is held at the edge and cut into pieces as desired. Then it is ready to be worked up as one wishes.
Written sources, like the icelandic law book Grágás from the 13th century, are the basis for the use of the terms 'blesterjern' (NO: blástrjárn) and 'fellujern' (NO: fellujárn):
Vétt blástrjárns fyrir 5 aura. Vétt fellujárns fyrir 6 aura.
One vett bloomery iron costs 5 øre. One vett 'fellujern' costs 6 øre. [vett = a unit of weight]
'Fellujern' is thus priced somewhat higher than the same weight of bloomery iron. This points to the two terms representing different qualities of iron, and that the 'fellujern' is more refined. The term 'blesterjern' probably referes to the raw iron, but the big question is if 'fellujern' referes to re-smelted iron. One could make that assumption, but a more definitive answer should be sought by comparing archaeological finds with experimental results, assessed by analysis. This lies beyond the scope of this project.
Literature
Evenstad, Ole (1790): Afhandling om Jern-Malm, som findes i Myrer og Moradser i Norge, og Omgangsmaaden med at forvandle den til Jern og Staal. Det Kgl. Danske Landhusholdningsselskab, København.
Jylov, Mads Rohde (2009): Fra malm til stål. Jernudvindingsteknologi i perioderne vikingetid og tidlig middelalder belyst ved eksperimentelarkæologiske forsøg og metallurgiske analyser. Afdelingen for Middelalder- og Rænæssancearkæologi, Aarhus Universitet Moesgård.
Hauge, T. Dannevig (1946): Blesterbruk og myrjern. Universitetets oldsaksamlings skrifter bind III. Oslo
Martens, Irmelin (1979): Blåsterjern og fellujern. Noen synspunkter på en lite påaktet funngruppe. Universitetets Oldsaksamlings Årbok 1979, page 190-197. Oslo.
Mortensen, Rasmus (1940): Jysk jærn. Undersøgelser vedrørende den gamle Jærnudvinding af Myremalm.
Sauder, Lee (2010): Making Steel in the "Aristotle Furnace" (20.04.2016).
Sauder, Lee: Remelting Bloom Iron for Phosphorus Reduction (27.05.2016).