Minerals and Metals in Ancient India v1.c22.s6: Copper and its Alloys in Taxila
Minerals and Metals in Ancient India v1.c22.s6
- Language
- English
Copper and its Alloys in Taxila
v1.c22.s6.1
Hardly any copper or bronze objects have been unearthed at Taxila of a date earlier than the fourth century BC, and very few until after the establishment of the Indo-Pārthian empire in the first century AD. Pliny stated (N.H., XXXIV.17) that India possessed neither bronze nor lead, and the *Periplus* recorded (49.56) that copper, tin and lead were imported to the Indian ports of Barygaza (Broach), Muziris and Nelcynda (near Cochin) etc. On the other hand, the widespread use of copper and its alloys in ancient India has been well-established, as recorded in our earlier chapters.
v1.c22.s6.2
The occasional scarcity of copper in Taxila was a local and temporary phenomenon. There was no dearth of copper coins in the third century BC Mauryan era. During the period 200 BC-AD 25, the struggles between the Yavana (Greek), Śaka (Scythian) and Śuṅga dynasties, specially over the mines of copper and non-ferrous metals in Rajasthan etc., caused the shortage, not only in Taxila but also in the entire western India. While some of these metals were imported to the Western ports, the eastern states around Pāṭaliputra, Vārāṇasī etc. did not encounter any shortage of copper. Marshall rightly observed (chapter 28, p. 566 fn.) that “it is quite impossible that the large quantities of copper used under the Pārthian rule at Taxila (after AD 25) could have been imported by sea; the chief sources of supply must have been the copper mines of Afghanistan and Kashmir”. We might add that the copper mines of Rajasthan must have been productive during peace time.
v1.c22.s6.3
‘Pārthians continually expelling each other’, as reported in *Periplus*, alludes to the fight between the Pārthians and the Kuṣāṇas during the first century AD, over not only the seats of power but also the mines of copper, non-ferrous metals and gems. Eventually, Trajan of Rome vanquished the Pārthians, destroyed their silk route, and indirectly helped Kuṣāṇas who used to import gold through the friendly Romans. When Shapur II established the Sassanid (Persian) empire at Taxila, defeating the Kuṣāṇas in AD 350, he emulated the first century AD Pārthians (also from Persia) in re-introducing copper money (vide Table 14.1).
v1.c22.s6.4
The tin ore or cassiterite, necessary to produce bronze, might have been obtained from Pālanpur state of western India, or Dhārwār district or Hazārībāgh in Bihar, but the most likely source for Taxila could have been the Khorassān and Kārādāgh hills in the Hindukush region where the close proximity of copper and tin ore reserves could have led to the earliest discovery of bronze in the Pre-Harappan era. In India, tin has been known as trapu or the ‘quick melting’ metal since the Vedic times, but around third century BC the other word for tin that entered the Sanskrit vocabulary was kastira, a modification of the Greek kassiteros or cassiterite (SnO₂). Similarly, the Greek oreichalcos for brass became ārakūṭa in Sanskrit which normally used the term riti for the said alloy. These are the metallurgical evidences for the early Indo-Greek contact in the north-west.
v1.c22.s6.5
Lead in Taxila was probably obtained from the same mines as silver at Fāranjāl and neighbouring localities in Southern Afghanistan or from the lead-zinc ore mines of Rajasthan. When Pliny said that lead and bronze, being imported in the west Indian ports were not available in India, he had little knowledge about the rest of India, or about the fact that the Andhras and Śakas had earlier made the metals from indigenous sources and used them.
v1.c22.s6.6
We reproduce the analytical data regarding copper and alloys of Taxila reported by Sana Ullah and M.A. Hamid to Marshall (chapter 28, pp. 567-69). The data correspond to (a) soft copper objects (Table 14.4), (b) objects of bronze unalloyed with lead (Table 14.5), (c) bronze alloyed with lead (Table 14.6), (d) brass objects (Table 14.7), (e) objects of copper-nickel alloy (Table 14.8), and (f) solder (Table 14.9).
v1.c22.s6.7
The analysis given in Table 14.4 shows that soft ductile copper of great purity was freely used at Taxila for hammering out domestic utensils, bars, sheet and wire. In one of the specimens (No. 7), the impurity amounts only to 0.32 per cent. Marshall thought that this suggest the use of natural copper such as that exist in the Zanskar Valley, a specimen of which was assayed by Sana Ullah: 99.48 Cu, 0.08 Fe and 0.34 insoluble (SiO₂, etc.).
v1.c22.s6.8
Other samples show the presence of sulphur indicating that sulphide ore of copper was the source. High purity of copper (97-98%) in these cases, showing 0.5 per cent sulphur suggest a high technological efficiency in copper metallurgy. 0.99, 1.13 and 1.23 percentages of antimony in copper samples No. 3, 13 and 14 respectively are too low to be considered as the results of deliberate alloying; on the other hand such appreciable values indicate that the antimony-rich copper ore from Badakshan, Balkh and Kabul might have been tapped to make the stated alloy. Such an ore might have been also used to make the earlier stated Taxilian white opaque glass (5.08% Sb₂O₃) and turquoise blue powder (2.42 Sb₂O₃ and 3.6 CuO). During the medieval period, antimony sulphide of this region was indeed used to make khārsini alloy and cast into ‘hairclips and amulets for the women of Hindustan’ (Al-Bīrūnī), and antimony mirrors. This trend might have started with the first century AD Taxilian mirrors (see No. 14). The sample No. 8 contains a high percentage of lead which was added probably to facilitate casting. Unalloyed copper was never cast at Taxila.
v1.c22.s6.9
Bronze containing 21-25 per cent tin, with little lead but some antimony, was in general use at Taxila for casting domestic utensils, bells, ornamental pieces (vide Table 14.5). This alloy is very hard and easy to cast on account of its low fusibility but has the disadvantage of being brittle. Nearchus, Alexander’s admiral, remarked that the Indians at that time employed only cast bronze, not hammered, so that ‘their vessels broke like earthenware if they fell’.
v1.c22.s6.10
It is surprising that Taxilians did not make much of the modern gun-metal type of bronze containing 8-12 per cent tin, which possesses great strength and elasticity. The ram-headed handle No. 16 containing 11.66 per cent tin, was evidently hollow cast by the cire perdue process. The Taxilians preferred high tin bronze presumably on account of its easy melting and casting properties. (Their preference also indicates easy availability of the metal from Afghanistan).
v1.c22.s6.11
The bonze samples alloyed with lead (Table 14.6) were also amenable to easy casting but resulted in weak structure. Lead was added to partially substitute tin. The analytical results show the relative proportions of the elements varying erratically; this has been explained by Marshall suggesting that scrap copper and bronze might have been added to molten lead in a haphazard manner before casting.
v1.c22.s6.12
An unguent flask (No. 21 in Table 14.6) obatined from the Greek stratum of Sirkap deserves particular notice because its shape, material and technique all combine to indicate its Hellenistic character. With its flared base, long neck and wide-projecting brim, it is obviously quite foreign to India. It contains large percentages of tin and lead (7.34 and 7.78 % respectively) — an alloy which appears to have come into vogue during the Greek period, the advantage of adding the lead being that it rendered the metal less brittle and easier to cast than the ordinary high-tin Indian bronze of that period. The unguent flask of Hellenistic design might have been fabricated by the cire perdue process.
v1.c22.s6.13
The brass items in Taxila (Table 14.7) are of outstanding importance insofar as the evidence proves the earliest use of zinc in the world. There are many Soft Copper Objects in Taxila
v1.c22.s6.14
| Object | Reference number | Age (century) | Copper | Tin | Arsenic | Antimony | Iron | Nickel | Zinc | Lead | Sulphur | Total |
| 1 Flat bar | Bm.'30-1,087 | 5th BC | 97.78 | — | — | — | 1.79 | — | — | — | — | 99.57 |
| 2. Rod | Bm.'20-302 | 4th BC | 98.33 | 0.19 | 0.19 | 0.35 | 0.45 | 0.51 | — | Trace | — | 100.02 |
| 3. Ear-cleaner | Bm.'19-275 | 3rd BC | 97.11 | — | 0.25 | 0.99 | 0.70 | 0.29 | — | 0.20 | 0.46 | 100.00 |
| 4. Plate | Sk.'14-1,779 | 1st BC | 98.93 | — | Trace | 0.16 | 0.17 | 0.52 | — | — | 0.22 | 100.00 |
| 5. Sheet | Sk.'26-2,194 | 1st AD | 98.13 | — | 0.23 | 0.59 | 0.24 | 0.30 | — | — | 0.51 | 100.00 |
| 6. Ladle | Sk.'30-337 | " | 99.10 | — | 0.02 | 0.16 | 0.44 | 0.12 | — | — | — | 99.84 |
| 7. Pan | Sk.'28-1307/c | " | 99.68 | 0.11 | Trace | — | 0.09 | 0.11 | — | — | — | 99.99 |
| 8. Spout of pot | Sk.'16-252 | " | 77.45 | 0.74 | 0.24 | — | 0.56 | 0.43 | — | Trace | 18.65 | 98.07 |
v1.c22.s6.15
Objects of Bronze Unalloyed with Lead at Taxila
v1.c22.s6.16
| Object | Reference number | Age (century) | Copper | Tin | Arsenic | Antimony | Iron | Nickel | Zinc | Lead | Sulphur | Total |
| 9. Rod | Bm. '19-558 | 4th BC | 87.24 | 8.28 | 0.55 | 0.46 | 1.23 | 0.91 | — | 0.94 | 0.39 | 100.00 |
| 10. Round bowl | Bm. '30-916 | 3rd BC | 76.76 | 21.55 | 0.16 | — | 0.95 | 0.48 | — | — | — | 99.90 |
| 11. Bell | Sk. '30-216 | 2nd BC | 74.80 | 24.14 | 0.10 | 0.20 | 0.55 | 0.21 | — | — | — | 100.00 |
| 12. Mirror | Sk. '14-1,385 | 1st BC | 74.28 | 24.85 | 0.04 | 0.12 | 0.71 | — | — | — | — | 100.00 silver trace |
| 13. Dish with central boss | Sk. '17-109 | 1st AD | 75.69 | 22.40 | 0.06 | 1.13 | 0.30 | 0.34 | 0.07 | 0.01 | — | 100.00 |
| 14. Mirror | Sk. '33-143/10 | " | 74.61 | 22.00 | 0.11 | 1.23 | 0.71 | 0.47 | — | — | — | 100.00 |
| 15. Bowl | Sk. '15-1, 115 | " | 75.54 | 23.20 | 0.28 | 0.30 | Trace | 0.49 | — | Trace | — | silver = 0.87% 99.81 |
| 16. Ram-headed handle of saucepan | Sk. '33-143 | " | 87.15 | 11.66 | — | — | 0.88 | — | — | Trace | — | 100.00 |
| 17. Bowl | Sk. '22-645/16 | " | 75.69 | 23.69 | — | — | 0.37 | 0.25 | — | Trace | — | 100.00 |
| 18. Goblet | Sk. '27-1, 021 | " | 73.39 | 25.59 | — | 0.12 | 0.71 | 0.15 | — | 0.05 | — | 100.00 |
| 19. Mixing pan | Sk. '14-111 | " | 74.24 | 24.58 | 0.18 | 0.02 | 0.72 | 0.26 | — | — | — | 100.00 |
v1.c22.s6.17
Table 14.6
v1.c22.s6.18
Objects of Bronze Alloyed with Lead at Taxila
v1.c22.s6.19
| Object | Reference | Age (century) | Copper | Tin | Arsenic | Antimony | Iron | Nickel | Zinc | Lead | Sulphur | Total |
| 20. Handle of bell | Sk.'29-2,653 | 2nd BC | 85.24 | 9.79 | 0.13 | 0.27 | 0.48 | 0.71 | — | 3.38 | — | 100.00 |
| 21. Flask | Sk.'29-2,010/1 | " | 80.38 | 7.34 | — | — | 1.33 | 0.26 | 2.49 | 7.78 | — | 99.58 |
| 22. Stand of goblet | Sk.'29-1,855 | 1st BC | 87.42 | 2.12 | 0.15 | 0.06 | 2.49 | 0.65 | — | 7.06 | — | 99.95 |
| 23. " | Sk.'13-1,015 | 1st AD | 80.98 | 15.03 | 0.10 | 0.13 | 0.09 | 0.27 | — | 2.59 | 0.81 | 100.00 |
| 24. Lion | Sk.'20-318 | " | 83.75 | 7.34 | 0.20 | — | 1.94 | 0.78 | — | 5.42 | — | 99.43 |
| 25. Cock | Sk.'30-159 | " | 83.35 | 5.62 | — | — | 0.56 | 1.00 | — | 9.61 | — | 100.14 |
| 26. Flask | Sk.'13-A 492 | " | 83.57 | 4.50 | — | 0.30 | 1.48 | 1.51 | 0.52 | 7.92 | — | 100.00 |
| 27. Rod | Jn.'16-F 556 | 5th AD | 76.50 | 9.39 | 0.25 | — | 0.21 | 0.05 | — | 12.71 | 0.89 | 100.00 |
v1.c22.s6.20
Table 14.7
v1.c22.s6.21
Objects of Brass at Taxila
v1.c22.s6.22
| Object | Reference number | Age (century) | Copper | Tin | Arsenic | Antimony | Iron | Nickel | Zinc | Lead | Sulphur | Total |
| 28. Vase | Bm.'15-284 | 3rd to 2nd BC | 55.39 | 4.25 | — | 0.26 | 1.77 | 0.40 | 34.34 | 3.08 | — | 99.49 |
| 29. Bangle | Sk.'29-2,681 | 2nd BC | 73.72 | 0.10 | — | 0.18 | 0.42 | — | 19.70 | 5.84 | — | 99.96 |
| 30. Bangle | Dh.'30-20 | 1st to 5th AD | 76.75 | 2.58 | — | — | — | 0.11 | 13.07 | 6.33 | — | 98.84 |
| 31. Pot | Dh.'33-273 | " | 79.19 | 3.58 | Trace | 0.10 | 0.94 | 0.09 | 12.88 | 3.22 | — | 100.00 |
v1.c22.s6.23
Objects of Copper-Nickel Aloy at Taxila
v1.c22.s6.24
| Object | Reference | Age (century) | Copper | Tin | Arsenic | Antimony | Iron | Nickel | Cobalt | Lead | Sulphur | Total |
| 32. Antimony rod | Bm.'21-432 | 3rd to 2nd BC | 87.05 | 2.01 | — | — | 0.98 | 9.02 | 0.86 | — | — | 99.92 |
| 33. Bangle | Bm.'14-290 | " | 78.09 | 0.64 | — | — | 1.53 | 18.88 | ? | — | — | 99.46 |
| 34. Coin of Euthydemus II | — | 2nd BC | 77.58 | 0.04 | — | — | 1.05 | 20.04 | 0.55 | — | 0.09 | 99.35 |
| 35. Plate | Sk.'14-1,746 | 1st BC | 77.45 | — | — | 0.30 | 0.68 | 21.35 | 0.52 | — | — | 100.30 |
v1.c22.s6.25
Table 14.9 Solder Objects at Taxila
v1.c22.s6.26
| Object | Reference number | Age (century) | Copper | Tin | Arsenic | Antimony | Iron | Nickel | Cobalt | Lead | Sulphur | Total |
| 36. Filling from bottom of Kohl flask | Sk.'29-2,010/1 | 2nd BC | 0.62 | 0.20 | — | 0.85 | 0.35 | — | — | 97.98 | — | 100.00 |
| 37. Solder | Sk.'28-1,307/c | 1st AD | 3.11 | 46.13 | — | 0.37 | 0.72 | — | — | 49.67 | — | 100.00 Zinc trace |
v1.c22.s6.27
claims of brass objects having been made in the ancient world (Turkey, Greece, Iran) during first millennium BC by the accidental cementation process viz., reducing a zinc-containing copper ore at a temperature above 917°C (b.p. of zinc) but not far above 1000°C. By this process zinc vapour gets absorbed in copper. It has been proven by Haedecke (1973) that this cementation process cannot give brass containing more than 28 per cent zinc.
v1.c22.s6.28
We have reported earlier Lothal brass containing 6.04 per cent zinc (chapter 6) and Atrañjikhera bronze-brass containing 16.20 per cent zinc and 20.72 per cent tin (chapter 11) which must have been obtained by the cementation process. The vase sample No. 28 in Table 14.7 however assays much more that the critical 28 per cent in zinc. We may quote Craddock's (1981: 16) comment on this particular sample:
v1.c22.s6.29
This could have been produced only by mixing copper and zinc metals. Hence the overriding importance of the vase (BM 15-284) excavated from the Bhir Mound at Taxila, dated to the fourth century BC, and shown to contain 34.34 per cent zinc. (Marshall, 1951, 1975: 568) This is very strong evidence for the availability of metallic zinc in the fourth century BC.
v1.c22.s6.30
The Bhir Mound specimen was made long before the dates of the Greek settlement in Sirkap and Alexander's invasion of Taxila. Therefore the credit of discovering zinc as a separate metal and using it for making brass goes to the Indians and not to the Greeks. Marshall wrote that "brass was known to the Greeks at least as early as the 3rd century BC, but the knowledge of it may have come from India". (chapter 4, p. 129)
v1.c22.s6.31
Table 14.7 shows second century BC Sirkap bangle assaying 19.70 per cent zinc. In the Dharmarājika settlement, the post-Christian brass objects like bangle and pot were made of a remarkably controlled composition 12.88-13.07 per cent zinc. The brass specimens contain both lead and tin in appreciable proportions, so that the alloy was too hard for hammered work, and suitable only for casting.
v1.c22.s6.32
We would continue our general discussion on the antiquity of zinc and brass in India separately in chapter 18.
v1.c22.s6.33
Table 14.8 presents the composition of some 'white copper' objects consisting of copper and nickel. The alloy was evidently prepared by the smelting of mixed ores of copper and nickel which are known to occur naturally in the province of Yunan, China. This alloy is known as pai-tung meaning white copper in China, and as 'packtong' in Europe. The material was used in the coins of Euthydemus II, Agathocles and Pantaleon as well as in making antimony rod, bangle and plate etc. The use of this alloy proves the intimate trade relationship between the Greek dynasty in Taxila, Bactria and the Han dynasty of China during the second century BC.
v1.c22.s6.34
Articles of ‘white copper’ occurs at Taxila from the third century BC, but seem to have disappeared after the first century BC presumably because the trade link was snapped. The amount of nickel in the earliest specimen is only 9 per cent, the later ones containing a fairly regular proportion of 19-21 per cent. The alloy is very ductile and well-adapted for hammered work. It was valued for its durable silvery lustre, and employed for jewellery, coinage and small fancy goods. Lead and tin-lead solders have been reported in Table 14.9.
v1.c22.s6.35
The processes employed at Taxila in the manufacture of copper and bronze objects were as follows: (a) hammering, riveting and soldering, (b) solid casting, (c) hollow-casting by the *cire perdue* process, and (d) embossing or *repoussé* work.
v1.c22.s6.36
Of the thirty-two copper and bronze objects from the early Bhir Mound site, twenty-three are hammered and nine solid-cast. Compared to the Greek and early Śaka levels, the Pārthian rule of the first century AD yielded a larger number and greater variety of copper/bronze household utensils and ornaments such as bracelets, bangles, ear-rings, ear-pendants, brooches, neck-pendants and buckles. The hammering process is well-illustrated in some of the large cauldrons and cooking vessels of the first century AD, which were made in sections and riveted together.
v1.c22.s6.37
Stone or terracotta moulds were used for solid casting of high tin (21-25%) bronze objects. None of the statuettes unearthed is sufficiently large to have been hollow-cast, but from the second century BC onwards, the *cire perdue* process was frequently employed for the fabrication of household vessels and utensils. Copper, bronze and silver embossed reliefs were made by hammering the metal sheet on to a solid metal die. The techniques were known to the Indians but the Greeks, Śakas, Pārthians, Kuṣāṇas etc. introduced new patterns and styles for the art and household objects.
v1.c22.s6.38
Surgical instruments like spatula etc. made of copper and bronze were found in the Kuṣāṇa period layer at Taxila.
Linked texts (2)
- Naturalis historia — C. Plinius Secundus, Latin
- Bhir Mound