Arjuna Thantilage¹,Senake R. Ratnayake²

1 Lecturer, Post Graduate Institute of Archaeology

2 Chemical Conservation Officer, Department of Archaeology

Abstract

It is a well known fact that there was no way of producing liquid iron from wrought iron in historical period and hence, casting of iron implements was impossible. An ancient bowl made of iron was subjected to microscopic investigations with a view to find its manufacturing technology. By means of these microscopic images, it was possible to understand some important technological aspects of ancient Sri Lankan iron implements manufacturing process.

Introduction

In 2004, remains of an iron bowl and many other iron implements were discovered from an accidental digging by a villager at Akurugoda, in the suburb of Tissamaharama (maritime city of southeast Sri Lanka) about six feet below the earth surface. Akurugoda and its suburb areas were subjected to extensive systematic archaeological explorations and excavations under the KAVA project. According to the results of the KAVA project excavations, archaeological evidences belonging to 1^st Century BC to 8^th Century AD were discovered from eight different phases¹ from the Tissamaharama, Akurugoda area. It is, therefore, reasonable to think that this bowl also belongs to the above mentioned period of time. Authors of this article were able to compromise with villager to acquire findings for scientific investigations.

During the historical period, Sri Lanka had possessed a rich iron production tradition. In recent times, a wide range of evidence has been collected for its technological superiority². In the course of this historical period, iron had been produced by bloomery process where smelting involved creating a bed of red-hot charcoal to which iron ore was mixed in order to produce metallic iron by reduction of the ores. Normally, the bloomary furnaces produced iron not in a molten state but reduced to a spongy aggregate of iron and slag formed at a temperature well below the melting point of pure (1,535 ⁰C).

Thus iron is formed at the solid state. Since these furnaces were incapable of reaching temperatures (average value) higher than 1200 ⁰C, the normal product was a solid lump of metal known as a “bloom”. The forced – draught (bellow operated) furnaces known from archaeological studies is usually regarded as the pinnacle of this early smelting technology³. Subsequently, those iron blooms were subjected to forged, annealed or cold work by the blacksmiths to produce various iron implements. Even though significant attention has been paid to understand ancient Sri Lankan iron production processes and furnace technology^2,4 little is known about the technological aspects of manufacturing goods and implements using the iron produce by the bloomary process. Pure iron melts at 1535 ⁰C⁵. It is a well known fact that, until the nineteenth century, in any where in the world, it was not possible to extract the iron in liquid state or melt the iron blooms  and made the iron implements by casting with molds, simply because remained metal smelting technologies were not sufficient to reach the temperature of melting point of iron. However, casting of iron implements had been done in antiquity by lowering the melting point of iron by making cast iron where carbon content is 3%-4%. Since malleability is extremely low in cast iron, it is impossible to make iron implements by hammering cast iron.  Hence, we were astonished by coming across this rare iron bowl. This paper is an attempt to understand its manufacturing process utilized by the ancient Sri Lankan blacksmiths and ascertain their technological aspects in the production of iron implements.

Cross sections, inside and outside surfaces of the bowl were examined with both naked eye and with the Microscope. The photomicrographs were taken. Depending on the situation, magnification values were varied and these values are indicated with the figures.

Dimensions of bowl and visual observations

Following dimensions were taken after reconstruction of the bowl (Fig- 1 and 2).

Average Radius of the rim = 17.1 cm

Average Maximum Radius of the bowl = 21.2 cm

Owing to the unauthorized excavation, unearthed bowl had been broken into six pieces. At a glance, bowl seems to be made by casting. It is clear that from reaching to rim areas to the bottom areas its thickness is gradually decreasing (Fig-3). The measured minimum thickness of the bottom (by measuring all six pieces of the bowl) is 2 mm. Because of very thin nature of the bowl bottom, it is reasonable to think that the bottom had not been preserved.  Investigations of the cross sections. showed that the walls of the bowl consisted of thin plate like structures and both inside and outside surfaces covered with a thick patina. As a result of the patina layer, internal metallic structure has been preserved acquiring extreme stability.

Microscopic observations

Microscopic investigations of the cross sections clearly show that the body of the bowl consists of several layers of “metal sheets” which had been bonded together forming a single wall. Number of layers varies approximately from 6 to 3 from top to bottom. See fig. 3 and fig. 4. Rim of the bowl has been made up by several layers and outermost “metal sheet” seems to be bent inwardly (See fig. 5).

Discussion

Joining of pieces of metal can be achieved by riveting, soldering or by welding. Soldering should be used only in the case of the use of any metal or alloy whose melting point is lower than that of the metal or alloy to be soldered⁶. Since this bowl is made out of iron, soldering process can be ruled out. In the case of welding it can be divided into three types.

(1). Pressure welding cold or hot

(2). Sweating together or surface welding without pressure

(3). Fusion welding (fusion together, modern welding like electric welding or acetylene welding)

The bonding of “iron sheets” together to form the wall of the bowl can be attributed to the pressure welding cold or hot without fusion. The microscopic photographs of cross sections of the bowl wall clearly show pile of welded “sheets”. In this particular case, a metallographic investigation would only reveal that it could be either hot or cold. Most probably it could be a hot pressure welding process owing to the practical reasons. Because copper, bronze or brass never join properly with pressure welding and only hot pressure welding worked mostly with iron⁶.

Thickness of the bowl wall (and also the hardness of the metal) has been increased by hammering several thin “sheets” (which were made out by hammering of blooms) to form a composite wall structure.  There are places where “metal sheets” were not bonded together exactly leaving tiny spaces among sheets which are only observable through microscope Fig. 6 and fig.7.

Availability of triangular shaped endings of “metal sheets” within the bowl structure (fig.8) confirms the usage of overlapping sheets of different sizes. It is reasonable to consider that these triangular shaped endings were caused by hammering on superimposed (heated ?) “iron sheets”. See fig.9.1 and fig. 9.2. Close to the bottom there are fewer metal “sheets” than at the other areas of the bowl. Microscopic images clearly displayed that sheets close to the bottom, had been more successfully pressure welded together leaving minimum spaces among them. In order to obtain shape of the bowl, there is a very high possibility to use an anvil having a spherical shaped head. The bonding of iron sheets together to form wall of the bowl can be attributed to the pressure welding cold or hot without fusion. But most probably it may be a hot pressure welding process for practical reasons.

This type of welding, piling of many sheets into a composite laminate was the method of making massive structures using iron such as the Delhi pillar⁶ (310 A.D)  weighing about six tons. So this bowl is a valuable example of the achievement of technology that the ancient Sri Lankan metal smiths possessed. And hence, there is no doubt about that the availability of highly organized, complex technological and social structure behind this object. In future, it is possible to expect unearth larger iron structures produced by the ancient Sri Lankan metallurgists using this technique.

References

1. Jochen Gorsdorf; 14C Dating of Tissamaharama(Akurugoda) Excavation. pp-279-282,Ancient Ruhuna Sri Lankan – German Ar ; Vol 1; Verlag Philipp Von Zabern . Mainz Am Rhein; 2001 Ed. H.J.Weisshaar,H.Roth.W.Wijepala, Somasiri Devendra.
2. Gill Juleff, An ancient wind-powered iron smelting technology in    Sri Lanka, Nature, Vol. 379, Issued No. 6560, 1996, pp 60-63
3. Bamberger M., In furnaces and smelting technology in antiquity, Ed. by Craddock P.T., and Hughes M.J., British Museum London, 1985, pp 151-158
4. Dehigala-ala-Kanda (KO.14) at Alakolavava:An Early Iron Production Sites with a Highly Developed Teachnology; Svaute Forenius and Rose Solangaarachchi; Further Studies In The Settlment Archaeology Of The Segiriya – Dambulla Region; Ed, Senake Bandaranayake, Mats Mogren, PGIAR Publication
5. Isaacs A., Daintith J., Martin E., Concise science dictionary, 2^nd Edition, 1991. Oxford university press, p-359
6. Forbes R.J; Studies in Ancient Technology, Vol VIII (2^nd Edition),   pp 138-141, E.J. Brill, Leiden 1971

Posted in Articles