ZINC AND RELATED ALLOYS : THE PIONEERING TRADITIONS IN THE ANCIENT AND MEDIEVAL INDIA
Professor Arun Kumar Biswas
The Asiatic Society
1, Park Street
Calcutta � 16

INTRODUCTION

India achieved the distinction of being the only country in the ancient and the medieval world to produce pure zinc metal and high zinc-brass alloys. The saga of zinc in ancient India has been established only recently by a team of scholars from India (Hegde, Biswas) and England (Craddock, Willies). The present author has recorded the current status of our knowledge on the subject in a paper and in a book. This provides a brief summary and also includes a hitherto unpublished material on the Bidri alloy.

The earlier occurrence of zinc in man - made artifacts is in the form of the copper alloy known as brass. Ever since the discovery of copper and the alloying elements of tin, arsenic, lead, etc., different materials, including zinc, were used to alloy and harden copper.

The earliest method of making brass was possibly the cementation process in which finely divided copper fragments were intimately mixed with roasted zinc ore (oxide) and reducing agent, such as charcoal, and heated to 1000oC in a sealed crucible. Zinc vapour formed, dissolved into the copper fragments yielding a poor quality brass, zinc percentage of which could not be easily controlled.

Fusion of zinc with copper increases the strength, hardness and toughness of the latter. When the alloy is composed of 10-18% zinc, it has a pleasing golden yellow colour. It can also take very high polish and glitter like gold. For this property, brass has been widely used for casting statuary, covering temple roofs, fabricating vessels, etc.

Reduction of zinc oxide around 1000oC is crucially important : below 950oC no zinc is produced. Zinc is obtained in the vapour form at this temperature, since its b.p. is 913oC. With trace of oxygen, the zinc vapour would be reoxidised and hence the successful operations in the past must have been done in closed crucibles. If the temperature were higher than 1083o C during brass-making, then copper would melt and flow down to the bottom of the crucible forming a puddle there, exposing a very small surface area of the metal for alloy formation.

Brasses containing upto 36% zinc are known as a- brasses, which undergo easy cold work. Brasses containing more than 46 % zinc are brittle. With zinc content between 36 and 46%, we have a+b brasses which are lighter, harder and more suitable for casting statuary.

Werner and Haedecke demonstrated experimentally that brass produced by the cementation process could not contain more than 28% zinc. Brass founders trying the cementation process have verified this observation.

The materials of antiquity containing more than 28% zinc in copper matrix must have been prepared by mixing the two metals, which could have been possible only after the discovery of zinc as a separate metal and its preparation by a process such as distillation. The antiquity of brass artifacts can, therefore, be divided into two eras, one preceding, and the other following the discovery of zinc as a separate metal.

BRASS BEFORE THE DISCOVERY OF ZINC

We claim that the earliest artifact noted so far containing an appreciable amount of zinc anywhere in the world is from India. Lothal (2200 � 1500 BC) showed one highly oxidised antiquity (No. 4189) which assayed : 70.7 % copper, 6.04 % zinc , 0.9% Fe and 6.04 % acid-soluble component (probably carbonate, a product of atmospheric corrosion). The material could have been prepared through smelting of zinc-bearing copper ore or the cementation route described earlier. The raw materials might have come from the Ahar-Zawar area. The Harappan site of Rosdi, also in Gujarat, has yielded a few samples of chisel, celt, rod and bangle, made of brass and assaying upto 1.54% zinc.

Similar materials might have been used for making the brass-bronze items of Atranjikhera during the PGW era (1200-600) BC. One copper-based item contained 11.68% Sn, 9.0% Pb and 6.28% Zinc, while another item assayed 20.72% Sn and 16.20% Zn. Both the samples contained traces of iron and sulphur, indicating the possibility of chalcopyrite and sphalerite-galena having been the source materials, which could easily come from the Ahar-Zawar area. Most of the brass samples in ancient India contained variable proportions of Zn, Sn and Pb (Table 1).

Table 1 � Analysis of Some Brass Object in Ancient India

S.No Date & Site Object Analysis % Remarks
Cu Zn Sn Pb
1. - 1500 BC Lothal No.4189 Copper Object 70.70 6.04 - - Ref.5
2. Harappan 

Rosdi

Chisel, Celt 

rod, bangle

95.5 � 98.5 Upto

1.54

Upto

1.23

Upto

1.20

Ref.6
3.
  • 1000 BC
  • Atranjikhera
No.1

No.2

70.99

52.42

6.28

16.20

11.68

20.72

9.00

9.77

Ref.7
4. 4th century BC

Taxila

Vase Bm

215-284

55.39 34.34 4.25 3.08 Ref.8
5. 2nd century BC Bangle 73.72 19.70 0.10 5.84 Ref.8
6. 2nd Century AD

Gujarat

Female figure carrying flower container Indo Parthian 88.60 7.60 0.13 2.49 Ref.6 pp.56-57
7. 5th century AD Gandhara Buddha 68.50 20.25 3.86 3.62 Ref.13
8. 6th century AD Akota Ambika 76.70 16.32 1.61 4.04 Ref.6, pp.104-105
9. 7th Century AD Mahudi Rishabhanatha 66.00 12.80 5.90 1.50 Ref.6, p.66
10. 8th century AD Kashmir Shiva 82.00 17.00 1.00 - Ref.14
11. 9th Century AD Nalanda Buddha 78.95 15.15 0.74 3.03 Ref.15
12. 11th century AD, W.Tibet Manjusri 65.50 30.40 0.30 1.70 Ref.16

We have drawn attention to the brass items of Lothal and Atranjikhera and their possible link with the 1260 + 160, 1136 + 160 BC and 1050 + 150 C-14 dates of the timber samples in the Rajpura � Dariba silver-lead-zinc mine near Udaipur.

During the Harappan era, copper used to be alloyed with tin and arsenic; since these were scarce commodities, alternative alloying elements had to be looked for. Artisans in the Rajasthan-Gujarat region might have stumbled on to zinc ore deposit as a new source of alloying element.

Craddock et al surveyed the evidences of early brass artifacts in the West. The earliest brass artifacts known in the West come from excavations at the Gordion Tomb in Phrygia, dating from the 8th and 7th centuries BC onwards. These came after the Lothal and Atranjikhera traditions. From the 7th Century BC, the Greeks commented upon brass or oreichalkos, but always as an expensive, exotic metal not produced in Greece. There was no zinc in the early Greek bronzes, Etruscan bronze of the 5th century BC contained 11% zinc.

ZINC METAL AND HIGH ZINC BRASS

The earliest brass containing more than 28% zinc, which could be made only after the isolation of pure zinc metal came from Taxila. Craddock pointed out the overriding importance of the vase (BM 215-284) excavated from the Bhir Mound at Taxila and dated to the 4th century BC. This brass sample contains 34.34% zinc, 4.25 % Sn, 3.0 % Pb, 1.77% Fe and 0.4% Nickel. This is very strong evidence for the availability of metallic zinc in the 4th century BC. Possibly India was the first to make this metal zinc (rasaka) by the distillation process, as practised for the other metal mercury (rasa).

There are references to zinc and brass in the lost (4th century BC) text Philippica or Theopompus, quoted by Strabo in his Geography :

"There is a stone near Andreida (north west Anatolia) which yields iron when burnt. After being treated in a furnace with a certain earth it yields droplets of false silver. This added to copper, forms the so-called mixture, which some call oreichalkos" (Strabco, Geogrraphy, Book XIII, Sec 56).

The above reference pertains probably to the process of downward distillation of zinc (�droplets of false silver�) and its subsequent mixing with copper to make brass oreichalkos (arakuta in Kautilya�s Arthasastra) described in detail in the post-Christian era Sanskrit texts.

It is quite possible that the zinc making technology travelled west from India during 6th-5th centuries BC, as it did later again in the 18th century AD. The pseudo-Aristotelian work �On Marvellous Things Heard� mentioned :

"They also say that amongst the Indians the bronze is so bright, clean and free from corrosion that it is indistinguishable from gold, but that amongst the cups of Darius there is considerable number that could not be distinguished from gold or bronze except by colour" (quoted by Craddock)

The Indian emphasis was on the �gold-like� brass and not on the zinc metal. The Greeks, however, used zinc metal as such in a few cases. In the course of the excavation of the Agora in Athens, a roll of sheet zinc was found in a sealed deposit dating from the 3rd or 2nd century BC. Analysis showed it to be nearly pure zinc with 1.3% lead, 0.06 % Fe and 0.005% Cu with traces of Mn, Mg, Sn, Ag and Sb (quotated by Craddock). Although Needham and Forbes doubted the above evidence on the ground that �the pieces were beyond the contemporary technology�. Craddock certifies this to be genuine sample. It is quite possible that the Greeks had carried the material or the technology which had existed in Taxila as early as 4th century BC and possibly much earlier in Rajasthan.

MINING ARCHAEOLOGY AND SMELTING RELATED TO INDIAN ZINC ORE

The recent pioneering work on the zinc-lead-silver mining archaelogy in the southern part of Rajasthan by Willies et al and the relevant C-14 dates have firmly established India�s primacy in non-ferrous ore mining in the ancient world.

The ancient workings in the South Lode (100 m depth) of Rajpura-Dariba mine (80 km north-east of Udaipur) have been C-14 dated as 1260 BC, 1050 BC and the East Lode workings as 375 BC, 360 BC, 250 BC, 120 BC, 150 AD etc. Thus, it is clear that the tradition of underground mining in India goes back to the thirteenth century BC, if not earlier. The earliest emphasis was possibly on copper ore; at Rajpura-Dariba, the other targets were lead, silver and possibly zinc ore, which is strongly suggested by the brass artifacts of Lothal and Atranjikhera.

The art of smelting zinc ore and recovery of zinc metal by distillation must have been discovered before 4th century BC when Taxila produced the brass vase containing 34.34 % zinc. This possibility is reinforced by the facts of mining archaelogy. Starting from the 5th century BC, we have many C-14 datings in Rajpura-Dariba, Rampura-Agucha (40 km south of Ajmer) and most crucially, in the Zawar mine systems.

Zawar (24o21' N, 73 o41'E) is about 30 km south-west from Udaipur, where the ancient mines (earliest C-14 date obtained so far is 430 BC) are found, both opencast and underground. Zawar Mala, Mochia and Balaria are some of the specific mines in this area.

The host rock of the Zawar mines is sheared dolomite, the result of metamorphism of sedimentary dolomites. The ore was geologically deposited syngenetically as disseminated lenses within the dolomite beds. Zinc occurs as sphalerite or as marmatite in which zinc sulphide is in solid solution with iron sulphide. Also associated with the rock are galena, hydrozincite, pyrite, silver, etc.

Ancient workings are found at outcrops on the ridge of Zawar Mala, and as deep as the 470 m (above sea) levels of the modern Zawar Mala mine, some 120 m below surface. The upper parts accessible from there were a few tens of metres below surface, isolated by a roof fall. At the surface, mine openings occur at intervals of 50 m or so. Willies investigated one mine of more commodious proportion � Pratapkhan or Pratap�s mine � in which Rana Pratap Singh, rival of Akbar, took refuge during 1595 � 1600 A.D. A flat �room� floored with phyllite slabs is inferred as Pratap�s refuge. The quarried materials used to feed a zinc smelter just below the narrow valley.

The earliest C-14 datings in the Zawar mines are 430 + 100 BC of the PRL 932 sample from the Zawar Mala mine and 380 + 50 BC of the BM 2381 sample from the Mochia mine. Similar datings from Rajpura � Dariba (e.g. 375 BC), Rampura Agucha (370 BC), etc. confirm widespread underground mining of lead-zinc ores in the southern Rajasthan during the fifth-fourth centuries BC onwards.

Subsequent C-14 datings in the said mining area are : 250 BC, 200, 170, 140, 120 BC, 60 AD, 110AD, 150 AD.

As regards the recovery of zinc from the ore, the crucible reduction/distillation method was put to large scale commercial practice in the 13th century AD; this will be described later. Indirect and circumstantial evidences suggest that distillation method was in vogue much earlier, probably from the 4th century BC onwards, although not on a large scale, as we find in the 13th century AD context.

In this connection, we recall the evidence from Rampura Agucha. The zinc-lead-silver ore at the site was selectively mined at least as early as 370 and 250 BC. An appreciable amount of zinc must have been separated from the zinc-rich ore (present-day ore in the site contains 13.5% zinc), as revealed from the low-zinc content slag. One sample of slag assayed as low as 0.01% zinc. Near the slag dump area several retort-like pieces were reported. When assembled, their appearance suggested a cylinder approximately 20 cm long with walls 4-5 cm thick and an innermost pipe-like feature with a coating dirty white material, mainly zinc sulphate. They could be mistaken for tuyeres but for their closed pointed ends. This is highly suggestive of a used retort. Along with this, some thin walled tube-like object containing a thin coating of blister type material was also found.

It is conceivable that the retorts were being used in the said context for roasting zinc ore to obtain the light, white, smoky zinc oxide, which the ancient Greeks called pompholyx or philosopher�s wool. In the modern zinc plant at Udaipur, roasting of zinc sulphide concentrate produces not only zinc oxide (and sulphur dioxide gas) but also some zinc sulphate, which was detected in the 4th-3rd century BC retort in Rampura Agucha.

The said retorts, already found sealed at one end, must have been closed or sealed at the other and also to prevent the escape of smoky zinc oxide into the atmosphere. The retorts were possibly modified to serve as reduction distillation chambers (to produce metallic zinc), the final version of which, notified in the 13th century AD context, would be described later. Very significantly, Tiwari et al. noted that the slag sample from Agucha containing only 0.01% zinc but as high as 9.30% lead, was �attached to baked earthen materials which could be part of the earthen appliance used for smelting�. We suggest the possibility that the earthen appliance was a zinc distillation retort. Remains of zinc furnaces have been found at Sojat in Jodhpur also.

BRASS AND ZINC IN ANCIENT INDIA

We may now turn our attention to the antiquity of brass in ancient India. Before the discovery of zinc metal in India (made by the distillation route) sometime during the fifth-fourth century BC, brass could be made, as in Lothal and Atranjikhera, only by the cementation route in which one of the following was smelted along with copper ore : zinc ore, sphalerite concentrate or the roasted product, philosopher�s wool or zinc oxide. The traditions of making philosophers� wool and cementation brass could have persisted even after the discovery of the distillation process of making zinc. We invite attention of the readers to the analysis of some Indian brass objects made before 4th century BC (Table I)

As we have indicated earlier, the distillation route of making zinc and alloying this with molten copper was the only way of making high-zinc (more than 28%) brass, such as the 4th century BC Taxila vase (34.34% zinc). The said vase (BM 215-284), excavated from the Bhir Mound site, was made before the Greek settlement in Sirkap.

One bangle from the second century BC Sirkap settlement assayed 19.70% zinc. The Dharmarajika settlement of the post-Christian era produced brass objects like bangle and pot with controlled compositions 77-79% Cu, 12.88-13.07% Zn, 2.5-3.5% Sn and 3-6% Pb.

Some of the other early brass samples from ancient India have been reviewed by Neogi and Ray, an extract of which is presented below :

Brass articles of 1st century BC or AD have been found on excavation of some ancient stupas. General ventura executed operations for the examination of the stupas at Manikyalaya in 1830. Three deposits were obtained, of which the third, at a depth of 64 ft. consisted of a copper box enclosing a brass cylindrical box cast and a beautifully turned on the lathe. The lid of brass casket was found on cleansing to be inscribed. From the inscriptions on the various articles of this deposit and the accompanying Indo Scythian coins, the great tope at Manikyalaya has been identified to be a mausoleum of the Indo-Scythian King Kanishka (1st century BC or AD).

Another inscribed brass urn of the same date as the former has been discovered in a tope about 30 miles west of Kabul in Wardak district. This urn, which in shape and size approaches closely the ordinary water-vessels in use in India to this day, was originally thickly gilt and its surface has in consequence remained well preserved.

As regards coins, both brass and bronze were used in ancient India for coinage. Circular punch-marked brass coins of Dhanadeva and Aryavarma of Ayodhya (c, 1st century BC) have been found. Brass coins of kings of several other dynasties living at that time have also been collected. From these archaeological and numismatic evidences it is clear that brass was in common use in ancient India during the first century BC. A small number of die-struck coins of the Pre-Gupta and Gupta periods, including a piece attributed to Chandragupta II, are considered to be made of brass.

Table I features some of the typical brass objects in ancient India up to 11th century AD, before the advent of Muslims in the country.

DISCOVERY OF THE MEDIEVAL ZINC SMELTING OUTFIT AT ZAWAR, NEAR UDAIPUR, RAJASTHAN

In April 1980, the Hindustan Zinc Limited (HZL) sponsored a three-year research project on recovery of zinc from the ancient slags which was successfully conducted at the Indian Institute of Technology (IIT) Kanpur by the present author. In 1982, HZL collaborated with British Museum Research Laboratory (P.T.Craddock, Lyon Willies, etc.) and the Department of Archaeology, M.S.University of Baroda (K.T.M.Hegde) on archaeological investigations, and this led to the spectacular discovery of the zinc distillation outfit, including furnaces and retorts showing the production strategy, Craddock described the exciting discovery.

"On the third day of the excavation (December 1983), one of the Baroda team (Hegde) spotted the corner of a refractory plate sticking out from a heap of spent retorts beside a goat track in a valley on Zawar Mala �With mounting excitement we cleared a small area above and around it to reveal first, the edges of furnace walls, and then the tops of retorts still in situ."

Extensive archaeological and archaeo-material investigations followed : we now present the summary of the results of the historical experiments performed in two continents.

ANCIENT RETORTS AND FURNACES AT ZAWAR

During the 1983 excavations, two groups of furnaces were uncovered. A single bank of seven furnaces upon Zawar Mala contained small retorts 20 cm long and 10 cm in diameter. In old Zawar, there was a more extensive arrangement of furnaces using a larger retorts (30-35 cm long and 10-15 cm diameter). In both groups, 36 retorts in a 6 x 6 arrangement were contained within the truncated pyramid of each furnace. Thus, no less than 252 retorts were fired simultaneously in a single bank. The retorts were supported vertically on perforated bricks through which the condenser tubes passed into the cooler zinc collectors beneath.(Figs. 1-3).

The furnaces are in two parts consisting of a zinc vapour condensation chamber at the bottom and a furnace chamber at the top. The two chambers are separated by a perforated terracotta plate. The condensation chamber measures 65 x 65 cm and 20 cm in height. The perforated terracotta plate that separates the two chambers is a composite unit made up of four equal segments of 35 cm2. It is 4 cm thick, well-baked, and sturdy. Its perforations include circular holes of two sizes : larger ones of 4 cm diameter each of which is surrounded by a number of smaller holes of 2.5 cm diameter. Within the furnace, the composite terracotta plate was found to be supported on a ledge in the furnace walls on all four sides and a single solid terracotta pillar placed below the junction of its four segments.

Up above the perforated terracotta plate is the furnace chamber, in which 36 charged retorts (Figs.1-2) were arranged, inverted vertically; it may be presumed that 36 vessels were placed, one underneath each retort, to collect the condensed zinc vapour. This arrangement of downward distillation retort with the condensing unit underneath or �distillation per descensum�, is precisely what had been described in Rasaratnasamuccaya text (2,157-166, 9.48-50). The brinjal-like retorts in Zawar (Fig.3) are also similar to the vrntakamusa described in Rasaratnasamaccaya (10.22-23).

It appears that a cylindrical reed of 1.5 cm diameter was inserted into the retort after it was charged and the funnel part was luted on it (Fig.3); this is evident from the central hole which is preserved in many retort residues. The reed helped to keep the charge within the retort when it was inverted and placed in the furnace. When the furnace was fired, the reed burnt away leaving behind a cylindrical flow channel for the zinc vapour to flow freely out of the retort. The ore must have been roasted before smelting.

The smelting charge must have included a small quantity of common salt (as surmised from the chlorine and sodium contents in the retort residue) and an adequately large quantity of carbonaceous matter, apart from the calcined ore, and then rolled into pellets of 1 cm3 volume

( Rasa Ratna Samuccaya or RRS 2.163-164 refers to the gutikakrti pellets containing sodium bicarbonate and borax). The charge (about 1.5 kg per retort) was loaded into clay retorts fitted with funnel like condenser tubes, as described before (Fig.3). These were indeed the brinjal-shaped crucible or vrntakamusa, as described in RRS(2.157, 2.163, 10.23-24, etc.). On heating in the furnace, zinc oxide was reduced by a carbonaceous matter to zinc vapour. The reducing blue flame of carbon monoxide was observed to be substituted by white flame of zinc vapour, indicating that reduction had taken place (bhavet bila sita yadi � RRS 2.159-160).

Using a scanning electron microscope and observing the vitrification textures of the Zawar retort and clay materials, Freestone estimated that the temperature reached in the Zawar zinc distillation furnace was of the order of 1150-1200oC, and that this temperature was maintained for 5 hours ore more. The highly endothermic reduction of zinc oxide must have been achieved at a very low partial pressure of oxygen (less than 10-20 atm) to prevent re-oxidation of the metal. Zinc vapour condensed in the tube, the temperature being around 500oC, and collected in the vessels placed below. This kind of downward distillation or tiryakpatana of zinc vapour, produced under a highly reducing atmosphere, has been described in RRS (2.163-168, 10.48-50).

A part of the zinc oxide was converted to well-identified silicate phases and thus could not be recovered as reduced metal.

Freestone et al. estimated that 200-500 g zinc was extracted per retort, or 7-18 kg per smelt of 36 retorts. Each retort weighs about 3 kg. Thus, the debris of around 6 lakh tons of spent retorts corresponds to about 1 lakh ton of zinc, according to Freestone et al. which might have been produced at Zawar during 13th-18th centuries AD. This has been indeed one of the most outstanding levels of industrial production in the medieval world.

PHASE STUDIES ON ZAWAR SLAGS AT IIT, KANPUR

Six lakh tons of spent retorts contain two lakh tons of residues within, assaying about 3% zinc. Therefore, some 6000 tons of zinc metal remain within the retort residue, and probably another 1000 tons in the lead slag. Our research at the Indian Institute of Technology (IIT) Kanpur, sponsored by the Hindustan Zinc Limited, was directed towards the recovery of zinc from these two kinds of slags. The first step in our world was characterization of these residues, which turned out to be very useful and relevant to the archaeo-metallurgical problem.

The phases identified by X-ray and electron diffraction studies by Biswas et al. are summarized in Table 2.

Table 2 - Phases Identified in Zawar Retort (Content), Wall and (Lead) Slag by X-ray and Electron Diffraction Studies (Biswas et al.)


Sample Phases identified
X-ray diffraction  Electron diffraction
Retort content
Zn2P2O7.5H2O
Goslarite ZnSO4.7H2O
Hemimorphite,Zn4(OH)2Si2O7H2O
Esperite Ca2Pb (ZnSiO4)4
Sphalerite Zns, Chalcopyrite
CuFeS2, larsenite, PbZnSiO4
Zn2P2O7.5H2O, goslarite,
hemimorphite, esperite,quartz. 
Mg2Al2Si4O18
Ca2ZnSi2O7, hardistonite
or aurichalcite Zn5(CO3)2(OH)6
 
Retort Wall
Goslarite, hardistone, hemi-
morphite, Zn2P2O7.5H2O
aurichalcite, quartz,sphalerite
Goslarite, hemimorphite,
 willemite Zn2SiO4
  hardistone
Slag
Quartz, goslarite,
Zn2P2O7.5H2O, hemimorphite
Sphalerite, chalcopyrite, hardistone
Hemimorphite, goslarite,
hydrozincite, quartz.
sphalerite

           The results show that in the roasting operation prior to retort distillation, a small part of sphalerite was not converted into oxide and remained in the retort as ZnS and ZnSO4. The presence of goslarite or ZnSO4.7H2O (hydrated at a later stage) was confirmed by the endothermic DTA peaks apart from X-ray and electron diffraction studies. Some ZnO might have remained unconverted in the retort to undergo atmospheric conversion to basic carbonate at a later stage. While most of the ZnO was reduced to metal, a part of it must have been converted at a high temperature to phosphate and silicates from which the metal could not be recovered. Biswas et al. detected a number of zinc, calcium-zinc, lead-zinc, calcium-lead-zinc and magnesium-aluminium silicates. Later, Freestone et al. reported that lead slag contained iron, calcium-iron, calcium-zinc, magnesium and calcium-magnesium silicates. The only silicate phase that Freestone et al. could report in the retort residue was CaMgSi2O6 or diopside.

Biswas et al obtained secondary electron images of retort residue samples and found that the particles have a variety of morphology : plates, needles and spheres. Scanning electron microscopy and X-ray microanalysis showed characteristic peaks of many elements, the most prominent being those of silicon and calcium. Scanning was done a fine electron probe on the zinc-containing particles in the size range of 0.5 � 40 m m.

Several of the small particles (0.5-8.0 m m) show approximately constant values for the ratio of intensities corresponding to the elements Si, Ca, Mg, Fe, Al, Zn, Pb, Mn, Na, K and S (in decreasing order of occurrence), indicating that these particles containing a few of the above elements are homogenous in nature. The larger particles show wide variation in intensity ratios and hence variation in the chemical composition of the grains from point to point. Thus, the approximate size of the zinc containing and other homogenous grains in the retort residue is in the range 0.5-8.0 m m.

The size of the particles in the lead slag sample was found to be lower than that in the retort residue. The size of the zinc-containing grains was estimated to be in the range 0.5-6.0 m m. m m, by carrying out point to point analysis. The X-ray spectrum showed the presence of titanium apart from the other elements noted in the retort residue.

The work at IIT Kanpur was directed primarily towards the recovery of zinc from the siliceous retort residue and slag at Zawar. We found that the non-silicate phase, such as hydrozincite Zn5(CO3)2(OH)6 could be easily leached by acid. A special �fast leaching� technique, making use of the water-starved nature of the silicate-sulphuric acid system, could recover more than 80% of the zinc contained in the silicate phases, such as hemimorphite Zn4(OH)2Si2O7H2O, willemite Zn2SiO4 etc.

The characterization work done at IIT, Kanpur, has established the existence of complex zinc phosphate-silicate phases in the slag which could have been produced only by the above 1000oC pyrometallurgical smelting process. This corroborates the conclusions reached by Craddock et al.

The large scale manufacture and widespread use of zinc and brass in medieval India need to be fully chronicled. Table 3 records a few brass icons of India for the period 1350-1752 AD; the high zinc content, sometimes in the range 35-40%, in these icons is particularly not worthy. Item No.9 in Table 3 is dated 1752 AD � five years before the War of Plassey, and eight years before the Zawar production was slowed down on account of the Maratha invasions. During the medieval period, the Moghuls had used brass (as well as bronze) for manufacturing guns and artisans of Bidar (83 km from Hyderabad) used high zinc (84%) brass or bidri alloy for ornamentation over it by gold or silver ware.

Table 3 - Brass Icons in India (1350-1752 AD)


S.No. Site  Item Elemental percentages Impurities
Cu Sn Zn Pb Fe
1. Medieval
Gujarat
1350 AD
Period
Ambika
68.4 1.6 18.5 9.5 - Fe, Ag, Bi
2. Gujarat, 1480 AD Model Temple with four doors 10 x 24.5 cm 68.6 0.2 28.9 1.6 - Fe, Mg, Bi
3. Gujarat 1485 AD Vishnu-Narayana 58.9 0.5 36.8 2.1 - Fe, Al, Ag, Si, Mg, Bi
4. Rajasthan
15-16 AD
Rajput Prince on Horse 72.9 0.4 21.8 2.7 1.0 Al ).3, Ag, Mg
5. Gujarat 1554 AD Kal Bhairava 76.7 1.2 13.8 6.3 1.5 Al, Mg
6. Gujarat
17 AD
Chauri-Bearer 58.3 1.5 35.5 2.3 - Fe, Ni, Al, Ag, Cd, Si, Mg
7. Rajasthan
17 AD
Dipalakshmi Rajput Girl 58.8 0.9 33.2 4.5 0.8 Al 1.6, Mg
8. Gujarat
18 AD
Dipalakshmi 52.8 1.1 39.9 2.9 - Fe, Al
9. Gujarat
1752 AD
Tirthankara-Seated 62.3 - 36.0 0.5 - Fe, Ag, Bi
10. Nepal Sadakasari Lokesvara Form of Avalokitesvara 60.5 2.75 35.3 2.37 - Fe, Ni, As, Au

The etymology of the words denoting zinc became clearer. Madanapala � Nirghantu of 1374 AD mentioned yasadam vangasadrsam or the �zinc metal (yasada) like tin� (Table 4). Yasada means that which gives yasa or fame; the connection was clear in so far as zinc was known to produce the famous gold-like yellow alloy of brass (vide 2nd century AD text Rasaratnakara 1.3). The European word �zinc� was probably derived from yasada; the Sanskrit word became jast (Abdul Fazl) and dasta in several Indian languages.

In 1597, Libavius (AD 1545-1616) received Indian zinc, which he called �Indian or Malabar lead�. He was uncertain what it was. Although Paracelsus (1616 AD) is generally credited to have given the name �zinc� to the metal, large scale export of the metal from India to the West started later in 17th century, and according to Roscoe, the identification of zinc as the metal from blende or calamine was accomplished by Homberg in 1695.

Table 4 � Some Literature on Zinc
 
Period Topics
1374 AD Madanapala � Nighantu refers to Yasam vangasadrsam (zinc � tin like) : Yasada means that which gives yasa or fame, converts copper into yellow gold-like brass
1597 AD Libavius receives a sample and calls it Indian or Malabar lead
1616 AD Paracelsus calls it �zinc� from Yasada, in several Indian languages : dasta
1695 AD Homberg identifies Indian zinc as the same metal from European calamine
Before 1730 AD An Englishman transmitted Zawar technology to the West � identity not known
1730 AD William Champion�s experiment at Warmley near Bristol; patent in 1738. Cost of metal £ 260 a ton, whereas calamine cost only £ 6 a ton
1743 AD Champion starts manufacturing zinc by distillation per descensum � process �notoriously close to the Zawar process� (Morgan and Craddock)
1751 AD Postlewayt�s Disctionary of Trade and Commerce admits ignorance about zinc technology. India continues making high Zn brass statues
1800-1820 AD Zawar zinc industry devastated by famine and Marhatta invasion
1886 AD V.Ball quotes Beckmann�s History of Inventions (Bohn�s Edition, ii.p.32) : "An Englishman went to India in the 17th century to discover the process used there in the manufacture of zinc, and returned with an account of distillation per descensum. I have not yet been able to identify this Englishman".

 

BIDRI AND OTHER ART WORKS IN PRE-MODERN INDIA

In pre-modern India several traditions of art works based on metals, alloys, gems and stones flourished and became internationally famous. Many of these traditions started in ancient India and continue vigorously in modern India.

Bidri ware, the sleek and smooth dark coloured metalwork with intricate eye-catching designs on its glossy surface, is famous all over the world. This metalwork as well as the technique to produce it are found in India alone.

Bidri is an alloy which contains 76 to 98 % (normally in the neighbourhood of 95 %) zinc, 2 to 10 % copper, upto 8 %. lead, 1 to 5% tin and trace of iron (vide Table 5). Occasionally high percentage (19.9) of lead and (11.4) tin have been noted. However mostly it is high zinc low copper alloy. Up to 1 % copper in the zinc forms the terminal solid solution h ; above 1 % copper the Î phase precipitates at the grain boundaries in this phase field. The usual yellow brass may contain not more than 40-50 % zinc, often less; copper constitutes the remainder or the predominant phase. Thus brass and bidri represent the two opposite ends of the zinc- copper phase diagram.

The Bidri ware�s surface is first made smooth and a solution of copper sulphate applied to it to darken it temporarily for engraving. The engraving tools cut the intricate but delicate tapestry of design into the metal which is then lighter in colour than the darkened surface and enables the pattern to be seen clearly.
 
 

Table 5 � Results of Atomic Absorption Spectrometry Analysis of Birdi Ware (Taken from La Nilece et al)
 
Victoria & Albert Museum

Acquisition nos.

Description Percentage by weight content
Zinc Copper  Lead  Tin Iron
2539-1883 I.S. Huqqa 91.2 3.2 2.9 1.2 0.2
1578-1904 Huqqa 98.0 3.3 1.4 0.1 0.2
45-1905 Huqqa 88.2 3.0 2.9 0.7 0.5
I.S. 46 � 1977 Weight 92.6 4.7 1.4 5.7 0.1
1479 � 1904 Ewert 89.3 4.8 0.8 < 0.1 0.8
I.S. 10 � 1973 Bowl 92.1 3.6 1.4 0.3 0.1
02942 (I.S.) Huqqa 84.6 3.3 1.0 0.1 0.5
I.S.131 � 1958 Pan box 95.0 3.1 1.5 0.1 0.1
I.S. 181 � 1965 Huqqa 76.1 10.1 8.2 11.4 0.6
I.S. 31 � 1976 Bottle 98.6 2.6 1.0 0.4 1.2
857 � 1874 Huqqa 92.1 3.6 0.8 0.6 0.4
I.S. 17 �1 1970 Pan box 81.3 2.0 2.0 < 0.1 0.5
02949 Bottle 83.8 2.5 6.6 < 0.1 0.2
02949 Bottle 93.2 2.4 1.1 0.1 0.2
855 � 1874 Huqqa 99.2 2.6 0.4 0.2 0.6
02941 (I.S.) Basin 91.2 2.8 2.1 0.1 0.2
I.S. 4 �1977 Huqqa 83.6 3.7 0.8 < 0.1 0.6
I.S. 19 � 1978 Huqqa 85.3 4.6 3.0 1.3 < 0.1
120 � 1886 Bottle 86.7 5.3 1.9 0.2 0.1
I.M. 224 � 1921 Bottle 91.7 3.6 0.7 0.7 0.7
2066 � 1883 Box 79.6 2.7 19.9 0.1 0.5
1402 � 1903 Huqqa 95.3 2.7 1.1 3.0 0.5
I.S. 11 � 1973 Vessel 97.6 3.4 1.8 < 0.1 0.1
I.S. 39 � 1976 Huqqa 89.2 3.4 0.5 0.7 0.7
I.S. 19 - 1980 Huqqa 80.6 3.6 7.3 6.4 0.9
856 - 1874 Huqqa 77.4 4.4 0.9 0.4 0.1

The piece is then handed over to the inlayer. The inlay may be of silver, brass or occasionally gold. The final stage after the inlay has been burnished, is to blacken the surface of the piece so that the inlay stands out. This is done by applying a paste of ammonium chloride, potassium nitrate, sodium chloride, copper sulphate and mud which darkens the body by producing a characteristic black patina while having no effect on the inlay. The paste is later washed off and finally oil is rubbed into the piece to deepen the blackness of the patina. The result is a lustrous dense black body to contrast with the shining lining � white (silver) or yellow (brass or gold).

REPLICATION EXPERIMENTS AND CHEMICAL HYPOTHESIS

La Niece and Martin performed some replication experiments to show that the black colour of the patina was due to copper. The recipe for blackening the bidri metal has the main constituents in a warm solution of potassium nitrate (one part) often substituted by well-urinated soil, and ammonium chloride (four parts).

In a replication experiment a clean pure zinc sheet was immersed in the above solution, when a pale grey patina of zinc oxide and chloride was produced. When the experiment was repeated with the addition of copper sulphate, a reasonably good but superficial black patina formed almost instantaneously. Again only zinc oxide and chloride were the main crystalline products. XRD analysis identified Zn5(OH)8 Cl12H2O,ZnO and Cu2O (cuprite) as the crystalline phases. None of these explains the blackness of the patina which was amorphous and contained copper.

Similar experiments were performed with a specially prepared alloy of zinc containing 3 % copper. When this alloy was dipped in a warm solution of potassium nitrate and ammonium chloride, it turned as deep black instantly and the patina was even and adhered well to the metal surface. This was found to be non-crystalline. Scanning electron microscopy showed the patina to be about 10 m m thick assaying 30 % copper as a contrast to 3% in the bulk. Zinc and chlorine were the other elements detected.

La niece and Martin have postulated that ammonium chloride preferentially dissolves the zinc from bidri and the resulting copper-enriched surface or the copper-rich Î phase precipitate gets oxidised by potassium nitrate producing the black colour. The use of clay does not seem to be crucially important. It could merely serve as a source of alkali nitrate and as a poultice to absorb the unwanted zinc chloride formed during the process. The matt black patina is easily damaged by the standard cleaning techniques designed to remove the white decomposition products from the zinc.

HISTORY OF THE ART TRADITION

The mystery of the black patina is not yet fully solved. The inventors of the tradition did not have any ideal about the underlying scientific principles; they merely hit upon the arts and crafts aspect of the process.

The craft of bidriware is a kind of damascene work which has defined by sir George Birdwood as �the art of encrusting one metal on another, not in crustae , which are soldered or wedged, but in the form of wire, which by undercutting and hammering, is thoroughly incorporated into the metal which it is intended to ornament. The original tradition at Damascus was to encrust gold wire, and sometimes silver wire, on the surface of iron, steel or bronze.

A group of the damascene craftsmen moved from Syria or Iraq to india. Some of them were at Ajmer in Rajastan and hit upon the idea that damascening could be done on the base of high zinc low copper alloy. Zawar in Rajastan was the major zinc production center in the medieval world. the said craftsmen moved down south during the 15th century A.D. and settled at Bidar (17 o55�N and 77o3� E) near Hyderabad. when the art flourished in that place for centuries, it became known as Bidriware crafts.

The earliest known craftsmen like Abdullah-bin-khaiser and his pupil Sivanna worked at Bijapur. Historical evidences indicate that the beautiful articles presented to Alauddin Bahamani II (1434-1457 A.D.) on the occasion of his coronation impressed him very much, and he invited the craftsmen of Bijapur to settle at Bidar itself. The workers used to prepare beautiful huqqa base, ewer bowl, pan box, basin, bottle, slabchis, cot legs etc. The varieties of workmanship of the design consisted of tarakashi (inlay of wire), taihnishan (inlay of sheet) zaranishan (low relief, inlay levels with surrounding area), zarabuland (high relief, for examples silver over a lead pad, aftabi (design in overlaid sheet) etc.

The russian traveller Althanasins Nikitin, who visited Bidar during 1470-1474 A.D., took with him some of the early Bidriware specimens for presentation to the Russian Emperor. A large number of articles of Bidriware were made for presentation to the prince of Wales when he visited India in 1875. These now adorn the collection of the Victoria and Albert Museum which has published a comprehensive bibliography and illustrated catalogue on the subject. Bidar and Hyderabad museums also have beautiful collections of this kind of ware.

Mahmud has given some details about this craft under specific heading such as raw materials, tools, implements, process of production, preparation of alloy, mould making, casting (such as goblet ) etc. The manufacture of Bidriware has been carried on under a system of division of labour. The moulder prepares the alloyed metal, casts the vessel and turns it to its proper shape by his lathe. The carver engraves the patterns on the surface of the vessel, and the inlayer designs the patterns, inlays the ornament of gold, silver or brass, and finally polishes the article. In the pre-modern India there have been four seats of Bidriware manufacture: Bidar, Lucknow, Purnea (Bihar) and Murshidabad (Bengal ).

REFERENCE

  1. Biswas, Arun Kumar . The primacy of India in Ancient Brass and Zinc Metallurgy.Indian Journal of History of Science , 28,(4), 1993, pp. 309-330
  2. Biswas, Arun Kumar and Biswas, Sulekha; Minerals and Metals in Ancient India, 2 Volumes, U.K. Print world ( P ) Ltd. 1996. Chapter 18 in Vol .1 ( pp. 351-384 ) is entitled : Antiquity of Zinc and Brass in Ancient India
  3. Werner.O., Spektralanaivusche and Metallurgische Untersuchangen an Indischen Bronzen. Lerden.. Brill.,1972.
  4. Haedecke.K, Glerchgewichtsverhailtnisse bei der Messingherstellung nach de Galmerverfahren. Erzmetall. 26, 229-233.1973.
  5. Rao.S.R. Lothal - A Harappan Port-Town ( 1995-62 ) in 2 Volumes. Archaeological Survey of India, New Delhi. Vol.II. 1985 p.660. Chemicals analyses of the Harappan Level samples were provided by B.B.Lal
  6. Swarnakamal, Metallic Art and Technology of Gujarat, Museum and Picture Gallery,Baroda.1978. pp. 36-37. 56-57. 56.104-105.
  7. Gaur. R.C., Excavations at Atranjkhera - Earlv Civilisation of the Upper Ganga Basin.Aligarh Muslim University and Motilal Banarasidass. Delhi. 1983. p. 497
  8. Marshall.J., Taxila. 3 Volumes. Cambridge.1951
  9. Farnsworth. M., Smith. C.S., and Rodda. J.L., Metallographic Examination of a sample of Mettalic Zinc from Ancient Athens.Hesperia (Suppl. 3. ). 126-129. 1949
  10. Tiwari .R.K., and Kavdia N.K., Ancient Mining activity around Agucha Village.Bhilwara District, Rajastan. Man & Envir . 8, 81-87, 1984
  11. Neogi.P., Copper in Ancient India,1917 reprinted in 1979 by Janaki Prakashan,Patna.10-42.
  12. History of Chemistry in Ancient and Medieval India, edited by P.Ray, Indian Chemical Society, Calcutta, pp.94-95, 1956
  13. Lo Bue, E.Statuary Metals in Tibet, in Reference 21, p.38
  14. Werner, O., Spectro-chemical Analysis of Antique and Modern Bronzes, Material Prufung.7 (12). 463-470, 1965
  15. Lal, B.B., An Examination of some Metal Images from Nalanda, Anc. India, 12, 53-57,1956
  16. Schroeder, U. Von., Indo-Tibetan Bronzes. Visual Dharma Publications Ltd., Hong Kong 1981
  17. Biswas, A.K., Biswas, S. and Chakraborty, N.A., Archaeo-material studies in India and Literary Evidences in Historical Archaeology of India, edited by A.Ray and S.Mukherjee, Books and Books, New Delhi, 1990, pp.49-66.
  18. Willies, Lynn, Ancient Zinc-lead-silver Mining in Rajasthan, Interim Report, Bull.Peak Dist. Mines. Hist. Soc. Ltd. 10(2), Winter 1987, pp.81-123
  19. Craddock, P.T., The Composition of Copper alloys used by the Greek, Ertuscan and Roman Civilization � The Origins and Early use of brass, J.Archaeol. Sci., 5, 1978, pp. 1-16
  20. Craddock, P.T. et al, Hellenistic Copper-Base Coinage and the Origins of Brass, in Scientific Studies in Numismatics edited by W.A. Oddy, Occasional Paper No.18, British Museum, 1980, pp. 53-64
  21. Craddock P.T., The Copper Alloys of Tibet and their Background, in Aspects of Tibetan Metallurgy, Occasional Paper no.15, British Museum, 1981, pp.1-31
  22. Biswas, Arun Kumar, Annual Reports to Hindustan Zinc Limited, Udaipur. Studies on Recovery of Zinc form Floation Tailings and Ancient Slags. Indian Institute of Technology, Kanpur, 1981-84
  23. Biswas, Arun Kumar, Gangopadhyaa, A. and Ramachandran, T.R., Phase Studies on Zinc Residues of Ancient Indian Origin, Trans. Indian Inst. Metals, 37 (3), June 1984, pp. 234-241
  24. Craddock, P.T., Goat Path Discovery Unlocks Secrets of Medieval Zinc. IAMS Newletter, No.6, Institute of Archaelogy, London, 1984
  25. Freestone, I.C., Craddock,P.T., Gurjar, L.K.Hegde, K.T.M. and Paliwal, H.V., Analytical Approaches to the Interpretation of Medieval Zinc�Smelting Debris from Zawar, Rajasthan, J.Archaeol. Chem., 3, December 1985, pp.1-12
  26. Hegde, K.T.M., Craddock, P.T., and Sonavane, V.H., Zinc Distillation in Ancient India, Proceedings of the 24th International Archaeometry Symposium, Smithsonian Institution Press, Washington, 1986, pp. 249-258
  27. Hegde, K.T.M., An Introduction to Ancient Indian Metallurgy, Geological Society of India, Bangalore, 1991. A special chapter on �Zinc and Brass Production in Ancient India�, pp. 56-83
  28. Mahmud, Sayed Jafar, Metal Technology in Medieval India, Daya Publishing House, Delhi, 1988. Chapter 14 (pp.115-126) is devoted to �Bidri Metallurgy�
  29. Gairola, T.R., Bidri Ware, Ancient India, 12, 1956, pp. 116-123
  30. Agarwal ,O.P., Bidri, Museum XXVII, 1975, p.193
  31. Stronge, S., Bidri Ware, Inlaid Metal work from India, Victoria and Albert museum, London, 1985
  32. La Niece, S. and Martin, G., The technical Examination of Bidri Ware, Studies in Conservation, 32, 1987, pp.97-101