EARLY BYZANTINE METALLURGICAL OBJECT AT THE SITEGAMZIGRAD – ROMULIANA IN EASTERN SERBIA

M. Živić #

National Museum, Zaječar, Serbia

(Received 01 April 2009; accepted 28 April 2009)

Abstract

The processing of waste powders from steel production in electric arc furnaces is a world problem. Thepresence of Fe, Zn, Cd, Pb, etc. in the powders makes them a valuable source for these metals. Zinc is the metalthat is most often utilized, which leads to reduction of environmental pollution.

The greatest problem in this connection is the presence of chlorine and fluorine in the powders, whichinfluences the electroextraction of zinc in a significantly negative way.

The presented paper shows the results from the study of the possibilities for zinc leaching from powdersobtained in the steel production in Bulgaria. A detailed characterization of the powders was made usingchemical analysis, XRD, DTA and TGA and Mössbauer spectroscopy. The results from the powder leachingwith different solvents give the reason to recommend a technological scheme for the complete and totalprocessing of the waste powders from steel production.

Keywords: Gamzigrad - Romuliana - Romulianum; Early Byzantine period; Mettalurgy; Iron ore;Smelting furnace; Workshop

#Corresponding author: majazivicnika@yahoo.com

DOI:10.2298/JMMB0902197Z

J o u r n a l o f

M i n i n g a n d

M e t a l l u r g y

Journal of Mining and Metallurgy 45 (2) B (2009) 197 - 206

The most important find from thearchaeological excavations conducted in2004 – 2005 was large metallurgicalworkshop, situated in the southeastern cornerof the fortified Galerius’ Palace. Theworkshop was discovered in the area definedon the north by Galerius’ baths, from the eastand west by ramparts of younger fortification(fig. 1). Metallurgical object spread to the

west to the western wall of square tower ofolder fortifications, tower V. The facilities

was held area of about 250 m2 and was a truemetallurgical plant in which the melted iron,as well as iron objects were made by castingand minting (fig. 2). Based on thearchaeological findings, one could say thatthe building dates from the end of V and thefirst half of VI century.

M. Živić / JMM 45 (2) B (2009) 197 - 206198

Fig. 1. General plan of Romuliana

Fig. 2. Metallurgical workshop in SE corner of Romuliana (sector of Galerius’ baths)

Furnace in the metallurgical object, markedas furnace 7/2004, was used for melting ironore (fig. 3). It’s basis was of horseshoe-plan,

5. 00 x 2. 50 m in size, on the foundationmade of broken stone bounded with clay.The interior of the furnace was lined withclay, which was due to high temperatureburnt. According to this, furnace liningvaries in color from white through orangeand light red to red. Upper furnaceconstruction was most probably domed, builtwith bricks bounded with clay, by system ofdropping square stones (fig. 4). Calotte

(dome) collapsed into the interior of thefurnace, while the opening for dischargingslag was in its north-eastern corner (fig. 5).

In the central part, the basis of furnace wasentrenched, and the ellipsoidal trench wasfilled with ash, soot and charred wood (fig.6). In order to determine layers in the interior

of the furnace, a cross-section of it wasmade, in the east - west direction (fig. 7). On

M. Živić / JMM 45 (2) B (2009) 197 - 206

Fig. 3. Furnace 7/2004

Fig. 5. Opening for discharging slag in north-eastern corner of furnace 7/2004

Fig. 4. Upper construction of furnace 7/2004 Fig. 6. The ellipsoidal trench in the central partof furnace 7/2004

199

the cross-section, three layers was clearlydefined, of which the first consisted of slag,the second of iron slag mixed with coal, ashand soot, while the third layer consisted ofash and soot mixed with minute sand andgravel.

It is interesting that in the furnace a lot ofdifferent parts of deer horns with traces ofworking were found. The most probably isthat the cuttings of deer horn, ensue duringthe process of making of various items fromthe raw materials in the workshop onRomuliana, were put into the furnace withcoal and iron ore made for melting. Deerhorn contains a high percentage of oxygen,so it is, in the process of melting iron,probably used as a means of binding carbon.In this way, the iron was made more elasticand much more convenient for furtherprocessing by coinage.

On the floor level of metallurgical object,in the vicinity of the furnace, two pits wereentrenched, about 2, 50 m in diameter, usedfor depositing slag (fig. 8). Since the intensemelting of the iron, these pits had been filledup and the slag discharged from the furnacewas running further to the north-east, whereits solified remains, in a length of about 12

m, were found. Slag layer thickness wasperiodically up to 1 m. In the cross-section ofthis layer, various lenses were established(fig. 9): 1 – slag; 2 - iron slag; 3 - charcoal,soot and ash mixed with sand, and 4 - sandand gravel with ash and soot.

In the western part of the building, two

M. Živić / JMM 45 (2) B (2009) 197 - 206

Fig. 7. A cross-section of furnace 7/2004

Fig. 8. Pits for depositing slag

Fig. 9. The cross-section of slag layer

200

rectangular furnace bases were discovered,marked as furnace 1 and furnace 2/2005 (fig.10). Furnace 1/2005, 1. 25 x 1. 00 m in size,

with dug out bottom, was most probablysmelting furnace. Its wall in the lower zonewas built of broken bricks and stone laid inclay. Firebox, as is confirmed by the zone ofburned earth and ash, was in the east. Upperstructure of this furnace was not preserved.Furnace 2/2005, 0. 90 x 0. 60 m in size, withfloor paved with tegulae, could have been theblacksmith’s furnace. This furnace had been

restored and between the floors wasencountered the lense of soot, ash and burnedearth. Its upper structure was not preserved,but it was most probably similar to upperconstructions of blacksmith’s furnacesdiscovered in the southwestern tower ofyounger fortification of Romuliana (tower19) in 2002, as well as structures of furnacesfound in the sector of Galerius’ baths inprevious campaigns. In both furnaces a largenumber of iron objects was found, as well asan iron anvil, and a large quantity of ironslag. These two furnaces were most probablypart of the blacksmith workshop, which was,in the most intense period of activity in themetallurgical furnace (furnace 7), itsindivisible part, forming together real plantfor the processing of iron ore.

Without incomplete chemical - physicalanalysis of slag, iron slag and iron objectsfrom the furnaces, one could not withcertainty discuss about their function.Malachite findings from the layer directlyabove the furnace 7, could perhaps indicatesthe processing of non-ferrous metals(copper) in the explored metallurgy facilities,along with the obvious black metallurgy.

At the end of V and the first half of VIcentury, in the time of the rule of Zeno,

Anastasius and Iustinus the Ist [1],performed a new epoch for Coastal Daciaand other northern provinces of Illyricum,marked by consolidating of Byzantinepower. In this very period Romuliana wasrestored. The first horizon of early-byzantinefortified settlement was characterized byintensive handicraft, the testimony of whichwas metallurgical object investigated from2004 to 2005. Melting and casting of iron inthis workshop were very intense and largerscaled. This could be primarily seen by the

M. Živić / JMM 45 (2) B (2009) 197 - 206

Fig. 10. Furnace 1 and furnace 2/2005

201

dimensions of metallurgical furnace (furnace7), as well as the amount of slag removed.Also, according to the dimensions of the pitsenvisaged for outpouring of slag, it seemsthat, at the beginning, small-scale work wasplanned, but melting, for some reason,continued, the pits were completely filledand that the slag from them was runningfurther to the North East. The workshop wasa time abandoned, its interior nivelated withsend above which the horizon of the secondhalf of VI century was formed, with thecorresponding layer, which was formed bythe beginning of VII century. The layer,illustrated, among other things, with poorarchitecture, clearly speeks about the processof penetration of the population from thevicinity into the city, established during theseventieths of the VI century.

Activities of metallurgical workshop,where smaller iron objects were castaccording to the findings of moulds (fig. 11),probably started at the time of EmperorAnastasius (491 - 518), when a restoration oflife was intense in Coastal Dacia, derelictedafter invasion of Huns and constantirruptions of Barbarians during the secondhalf of V century. Its peak the workshopexperienced in the time of Iustinianus’restoration, crowned by establishing a newArchbishopric in new-built city IustinianaPrima. Production of Gamzigrad’s workshopcould, therefore, been associated with therenewal of the border on the Danube, whenweapons for the purpose of Danubiancastella could be make and mend in it. Forsuch a function could indicate arrow-peaksfound in layer above the floor ofmetallurgical object, as well as moulds formilitary belt-buckles, previosly found in thearea of Galerius’ baths. But, that is to saywith certainty is that the production of the

M. Živić / JMM 45 (2) B (2009) 197 - 206

Fig. 11. The findings of moulds for casting ofiron objects

Fig. 12. Iron arrow-peaks

202

workshop was intense throughout the fifthand sixth century, which testify to differentobjects of various materials (fig. 13 - 22).

M. Živić / JMM 45 (2) B (2009) 197 - 206

Fig. 13. Golden tremisis (1/3 of solidus) ofEmperor Iustinus I (518 - 527)

Fig. 15. Comb case, lining and whorl of dearhorns

Fig. 16. Bronze fibulae

Fig. 14. Combs of dear horns

Fig. 17. Bronze linings from different objects andhandle of a bronze spoon

203

The fact is that the processing of iron inthe Byzantine Empire had a hugesignificance. Blacksmith was an integral partof economy, and very often was shown in theivory cases. As the main actors of thesescenes there were Adam and Eve: Eve wasusually shown kneeling or sitting, managingcylindrical bellows to fire, while Adam was

M. Živić / JMM 45 (2) B (2009) 197 - 206

Fig. 19. Different iron objects

Fig. 18. Belt vuckles of Suçidava type and bronzecresset holder

Fig. 20. Different glass vessels

Fig. 21. Ceramic jug

204

standing towards the firebox, dressed inpeasant suit, holding with pencers someobject in one hand, while with the other handwas hammering at anvil.

Expert archaeometallurgical and chemical- physical analysis of slag, iron slag and ironobjects from the furnaces, as well as of theobjects found on the same levels in tower 19and metallurgical objects built by southfaçade and southeast corner of Galerius’baths, will give more informations aboutcharacter of metallurgical activities in LateRoman Romuliana, as well as about theapplied technological process.

All data collected so far clearly indicatesthat iron ore was melted in the furnace7 / 2004 [2]. Most probably is thatintermediate reductional process wasapplied, and the iron, by openning thefurnace, cast in the fluid or liquid state. Sucha technological process, which requires avery high temperature, was first documentedin Germany and Scandinavia in the XIII -XIV century.[3] Before the discovery of

metallurgical workshop in Gamzigrad, thisway of melting iron ore was not confirmed inany Late Antique or Early Byzantinearchaeological site on our territory. The mostsimilar to metallurgical furnace inRomuliana are double furnaces, discoveredon the site Bellaire III, as well as the furnacefrom the site Boécourt in Jura, Switzerland,dated from the AD 500 to AD 650.[4]

Romuliana, with its many handicraftworkshops, is a perfect model for thereconstruction of the death of one earlybyzantine city. This unstoppable process ofruralisation of cities, started even in the Vcentury, Justinianus’ restoration was onlydelay, leading to the same time todisintegration of once urban tissue in lesseconomic groups. Perfect illustration of thisphenomenon is the metallurgical workshopin Gamzigrad, as well as numerous othertrade workshops discovered here. Ruralizedand impoverished Balkan cities sunken inisolation, resulted from an illusion ofpossible self-subsistence based on thecoexistence of its economic microregions.Thus made them easy prey for the Avars andjoined Slavs, who conquered last byzantinestrongholds in the interior of the Balkans in614/615. Romulianum and otherfortifications in the hinterland of Danubesucceeded to postponed their death until themiddle of VII century, when they finallysuccumbed barbaric attacks.

References

1. After the short and successful war withVandals in Africa 534, and before the beginning ofwar with Gots 535, Iustinianus fortified Danubiancoast with new fortifications in the district ofKatarakta – Đerdap as well as in Coastal Dacia. At thetime new ramparts of Aquis were built and all thebelonging castellan, among them Romulianum, has

M. Živić / JMM 45 (2) B (2009) 197 - 206

Fig. 22. Ceramic lids and ceramic pitos

205

been restored (about restoration of Danubian frontier:Procopii, De aedificiis, IV, 4). At the time, Gamzigradalready had church with baptistery (Basilica III) andcould be centre of church municipalities (ecclesia) forneighbouring fortifications.

2. D. Živković, N. Štrvac, M. Cocić, M.Šteharnik, Z. Stević, B. Marjanović, PreliminaryInvestigation of the Early Byzantine MetallurgicalActivities at Felix Romuliana (Zaječar, EasternSerbia), III International Symposium: Metallurgy inSoutheast Europe from Ancient Times till the End of19th Century, September 26th – 30th 2005, Sozopol,Bulgaria, Reports, Sofia, 125 – 129.

3. R. Pleiner, Iron in Archaeology. The EuropeanBloomery Smelters, Praha 2000, 179 sq

4. Ibidem, 182 – 183, fig. 20, fig. 46, Pl. XIII

M. Živić / JMM 45 (2) B (2009) 197 - 206206