Doing so can make an iron disc as hard as steel.

Before cast iron began to be produced on a wider scale at the end of the 18th century, one of the most important iron–carbon alloys was steel. Humanity had been aware of this type of smelting since the 2nd millennium BCE, when it was produced in ancient Anatolia and the Middle East. Since steel is significantly more durable than carbon-free iron, it has throughout history been used to make various bladed tools—knives, axes, scythes, swords, and other weapons of both war and of peace — wrote University of Tartu junior research fellow in archaeology Kristo Oks and archaeology researcher Ragnar Saage, for ERR's Novaator portal.

Local steel production, in earlier times made in what is called a bloomery, has not yet been studied in Estonia, and it is not certain when the practice precisely began. The hope is that a doctoral dissertation recently launched at the University of Tartu will provide answers to this and many other questions.

Experimental archaeology is an excellent method of getting a better understanding of steel as a material, along with the details of its production.

There have been several methods for making steel reported through history and prior to the industrial revolution. At the beginning of the 12th century, a Benedictine monk called Theophilus recorded one method, which simply involved coating a piece of iron with pork fat, wrapping it in goatskin and heating it in the forge for an hour. Carbon originating from the animal additives bonds with the iron atoms and steel is formed, through the process.

Another method is remelting iron in a small furnace, where carbon released during the burning of charcoal bonds with the iron atoms. In 1790, Norwegian politician and writer Ole Evenstad described how people in his country smelted iron from bog ore and used a small remelting furnace. He noted that a furnace with a shallow bottom was most suited for working iron, whereas a furnace twice as deep (with the bottom about five cm below the tuyere – meaning the nozzle through which air is forced into furnace or forge) was more suited to producing iron.

This is likely a very ancient method: As early as the 4th century BCE, the Greek philosopher Aristotle described a similar process, whose aim was to purify iron and produce steel.

Iron to steel in several ways

In July of last year, history enthusiasts, craft aficionados, and University of Tartu students once again gathered at the Rõuge ancient farmstead in Võru County, to spend four summer days on the banks of the nearby Ööbiku river valley. The largest undertaking while at the camp was creating steel using both of the two methods described above. For this purpose, two similar steel furnaces were set up: One in the smithy above the forge hearth and the other next to the smithy. The furnaces had an internal diameter of 25–27 cm and a height of nearly 27 cm, considerably smaller than a typical smelting furnace.

Oks and Saage take up the story further: In the furnace built in the smithy, we tried to fuse smaller pieces of iron into one, but this proved unsuccessful — the same pieces stayed in the furnace after the experiment as had been the case before. The furnace proved to be too oval-shaped at the bottom, and the tuyere directed the airflow higher rather than into the center of the furnace, meaning the desired temperature was not achieved. Then the furnace built outdoors was about an inch deeper than the previous one, in order to create a carburizing zone at the bottom of the furnace, where iron gets enriched with carbon.

First, steel-making was attempted in this furnace, using the iron remelting method: The furnace was filled with charcoal, and after the charcoal was ignited, two iron blocks (of 2.5 × 3.5 × 5 cm size approx) were added, which gradually heated up, and sank downward.

Steelmaking was also tested using the method as described by Theophilus. To do this, five discs were cut from similar iron blocks; four were coated with pork fat, of which half were wrapped in wild boar skin, the others in an old leather strap. The fifth disc was set aside without these additions, as a control sample. After adding the animal organic material, the lumps were covered with clay, and the resulting balls were left to dry overnight.

The next day, they were placed on the furnace coals in the same way as in the previous method, until they sank downward and began to glow in the heat. At the commencement of the experiment, one of the four test bodies broke, causing the fat to flow out. As it turned out, the clay balls should have been dried longer, and care should have been taken to ensure their surface was not coated with fat, as this hinders moisture from escaping the clay. After an hour of maintaining the heat by using bellows, the test bodies were taken out of the hearth. The glowing clay layer was knocked off on an anvil stone, then the iron pieces were left to cool in the air.

Metallographic analysis carried out at the University of Tartu's archaeology laboratory showed that both tested-out steelmaking methods had proved successful. In a sample taken from the remelted iron blocks, the edge areas had been enriched with carbon and turned into steel with a carbon content of over 0.8 percent. In the central part of that sample, by contrast, softer iron predominated.

In the other case, carbon had penetrated into the center of the sample and soft iron was present in the edge areas. In the steel samples produced using Theophilus' method, the carbon-enriched areas were mainly at the edges, while the overall carbon content was lower than in the results of the first method. In places, carbon had also penetrated deeper, and smaller patches emerged where there was no carbon at all.

New insights from iron smelting experiments

After making the steel, we continued with iron smelting experiments in the summer of 2025. It can safely be said that these experiments were groundbreaking, as they shook our understanding of the role of slag in the iron production process. Slag is a by-product formed in the furnace during iron smelting, which helps small iron particles reduced from ore to unite into a single large bloom.

Whereas experiments in previous years were successful only when abundant slag flowed out of the furnace throughout the smelting process, two entirely new situations occurred in this summer's experiments. First, in June, in Araiši, in Latvia, we charged a furnace with iron ore collected near Palmse, in Lääne-Viru County.

The smelt proceeded as usual, and well-flowing slag had to be released from the furnace repeatedly. When the furnace was opened, however, the iron piece inside was not very impressive; during forging it quickly diminished, and in the end only about 200 grams of iron remained. Later analysis in the laboratory showed that the brittleness was caused by excessive carbon — in other words, we had produced cast iron

Then in July, Mart Paadik carried out an iron smelt at Metsakivi, where we witnessed the opposite result. He mainly used ore collected at Pedassaare in Jõgeva County, which we supplemented with ore from the Tallinn Botanical Gardens. Throughout the entire smelt, we released slag from the furnace only once, and we were already getting worried that something was wrong in the furnace.

Once the hearth was opened, however, we found inside one gigantic piece of iron, which had been forged for about three hours and resulted in two dense bars with a combined weight of 3.7 kilograms. This was the first smelt in which so little slag was produced.

So every new iron ore and every iron smelt teaches us something new about metallurgy. This summer's lesson was that slag abundance is no guarantee of good results, nor is it a prerequisite for the formation of iron.

--

Follow ERR News on Facebook and X and never miss an update!