Note
Property of Bronze Knives: An Experimental Replication of a Bronze Knife
2014 Volume 54 Issue 5 Pages 1167-1171
Details
2014 Volume 54 Issue 5 Pages 1167-1171
This study aims to examine the property of bronze tools and weapons through experimental replication. The target of this replication is the bronze knives of the “Spring and Autumn” period from Xinjiang, China. The composition and the element ration of the alloy, the metal construction, along with other aspects of the bronze knives of Xinjiang have been analyzed in previous studies. The alloy used for a particular knife included 10% tin, and was forged after casting the bronze to increase the strength of the metal.
Referring to these previous researches, I replicated a knife casted with an alloy consisting of 10% tin and 90% copper, followed by repeated implementing of cold forging and annealing. I explored the property of the bronze alloy by measuring the strength of the replicated knife and comparing with iron.
Xinjiang is located in northwestern China. As an important part of the Silk Road, people, products, techniques, and cultures were exchanged there. This region has yielded iron artifacts dating to the first millennium BC.1) Although raw material for tools and weapons gradually changed from bronze to iron in Xinjiang as iron artifacts became popular, bronze continued to be used. Iron was almost exclusively used as a raw material for tools and weapons during the Han Dynasty (second century BC to third century AD). It took nearly 800 years from the first appearance of iron artifacts for this material to become fully established there.2)
This phenomenon occurred because bronze has superior properties as a material for tools and weapons, although it is problematic to obtain materials for iron products. Bronze fulfilled the role for material use in that time period. Compared to iron, which was quickly introduced, bronze was superior in regard to its hardness and tensile strength. In fact, iron performed better than bronze only after it was stabilized with steel.
Through this study, I aim to inspect the bronze tools and weapons used in the Early Iron Age. Referring to previous research that contained metallurgical analyses of unearthed artifacts, we reproduced bronze knives and conducted an experiment to measure bronze’s performance. This reproduction is then compared with both hard and soft steels, followed by an evaluation of the property ofbronze tools and weapons. I aim to study the technical background that changed human societies from the Bronze Age to the Iron Age on the Eurasian steppe.
The problems in the transition from the Bronze Age to the Iron Age in Xinjiang were mainly studied in regard to changes in specific cultural areas3,4,5,6) (Fig. 1). The area of Xinjiang where some iron artifacts appeared first time is Hami, placed in the east of Xinjiang. There was excavated the iron artifacts, e.g. an iron ring and an knife from Yanblack tomb in ninth century BC. In Turfan, placed in the central of Xinjiang, iron artifact was yielded from Yanhai tomb at eighth century BC. Ili, in the west of Xinjiang, iron knives were unearthed from tombs at sixth century BC. Though iron artifacts were found at the early time of first millennium BC in these specific cultural areas, materials of weapons and tools was continued to use bronze. The quantity of Iron artifact exceed bronze at second century BC, it was Han dynasty.
Specific cultural areas in Xinjiang.2)
In Central China, iron artifact appeared at the beginning of eighth century BC, then a fragment of cast iron was found at sixth century BC, then started using at fifth century BC. The iron was used as main material of farming implement at 4 century BC. Compare with Central China, it needed a long time that iron was used commonly in Xinjiang.
The metallurgical studies of bronze and iron artifacts from Xinjiang mainly examined archaeological artifacts using a microscope that facilitated analysis of the composition of alloys.7,8) More than 17 bronze knives were analyzed. The compositions of alloys were copper and tin, copper and arsenic, copper, and so on. Bronze knives in the first millennium BC mainly used tin and copper bronze. Microscopic observations revealed small, depressed grains and twin crystals, which resulted from casting, followed by cold forging and annealing (Fig. 2 left, center).
Several iron artifacts were also analyzed. An iron knife from the Yanghai No. 265 tomb was made from low carbon steel and produced by forging.8) Iron artifacts such as rings and fragments from the Yanbulake tombs were also analyzed. One ring was made from low carbon steel, while an unknown fragment included various steels.9,10)
Concerning the practicability of bronze knives used in the Early Iron Age, we use the method that compares numerical performance by testing the hardness and tensile strength. However, because hardness and tensile strength values are obtained by means of a breakage test, it is difficult to analyze the actual artifacts. This is because they are fragile due to rust damage and therefore unsuitable for measuring. In order to understand the conditions at that time, we created a replica from the same materials using comparable production methods as the prehistoric artifacts. My experimental replication of the bronze knife included analyzing the alloy compositions and the processing methods.
The sample fulfilling these conditions was a knife from the Chawufu tombs, located at the southern foot of the Tianshan mountain range in central Xinjiang (Fig. 2). Bronze knives and daggers unearthed from sites in the northern steppe area of Xinjiang were likely used to butcher livestock and cut up meat for distribution.11) More than 140 bronze knives were unearthed from the Chawufu tombs; both edges of these artifacts measure between 10 and 30 cm.12) Such items were common tools for cutting meat. The bronze knife from M256 of the Chawufu tombs dated to the middle of the first millennium BC. Part of this knife’s blade was analyzed; the composition of alloy was 89.16%Cu and 10.84%Sn. In other words, it was a tin bronze alloy produced through forging.7) It was first casted, then forged and annealed.
Therefore, I produced bronze knives by the following process and then measured its numerical value.
1. Cast the samples composed of 90%Cu and 10%Sn
2. After casting, cold forge and anneal, then form both knife edges by grinding
3. Cut part of the samples, implant in resin for observing with microscope, evaluate the metallic organization, and measure hardness
4. Compare with iron knives
The handle and blade of bronze knives unearthed from the steppe area were cast at the same time, though in replicating bronze knives, only the blade parts were cast due to the experimental nature of the study. The samples used 36 g of Cu and 4 g of Sn for a weight with the ratio of 9:1. For Cu, electrolytically refined copper was used; and the tin (Sn) had 99.99% purity for an experimental metal. Using excavation reports for reference, two plates were cast using a metal mold; they measured 70 mm in length, 15 mm in width, and 4 mm in thickness.
After casting 40 g of metal, the sample became 39.1 g, with a prestress density of 7.82 g/cm3. The density of electrical copper is 8.92 g/cm3, while tin’s is 7.3 g/cm3. The ratio of alloy will be 8.73 g/cm3, rendering these samples too light. It was possible to discern bubbles in the alloy.
After cutting the casting mold, the blade was forged to a thickness of 0.5 mm. The method involved forging 300 times at a fixed temperature, then annealing just once. After continuing this process 10 times, forging a total of 3000 times, the sample blades were produced.
As a result of extensive forging, the blade became 75.6 mm long, 17.7 mm wide, and 0.5–3.0 mm thick. After being fixed by a grinder, the edges were made with a file. It obtained sharp edges in the end, easily cutting beef in experimental attempts (Fig. 3).
A replica knife.
For comparison to archaeological artifacts, the metal organization of our replica knives was observed by the following method. The casting sample was named No. 1, while the knife that was cast, forged, and annealed was No. 2. First, samples No. 1 and No. 2 were cut by diamond saw, and implanted in resin (Struersco. N105) for microscopic observation. After setting, it was delicately ground using several grades of waterproof sandpapers (#340, 500, 800, 1000, 1200, 1500, 2000); #8000 oxidation alumina powder; #12000 cerium powder; 3 μm diamond paste, and finished with a mirror surface (Fig. 4). After etching with 5% iron chloride liquor, I observed the metallograph. A Meiji ML8100 microscope was used.
Implanted samples.
The results of microscopic observation are as follows.
Sample No. 1, casting (Figs. 5, 6)
Sample No. 1.
A dendrite crystal of sample No. 1.
The crystal was relatively large, observable with the naked eye. When the view was expanded, a dendrite crystal was visible. Sample No. 1 was cast with a metal mold and considered a segregation of tin, forming α-phase dendrite.
Sample No. 2, forged knife (Figs. 7, 8)
The oval-shaped bubbles of sample No. 2.
Magnified sample No. 2.
Fine crystals were observed compared to sample No. 1. The oval-shaped bubbles formed by pressure from above and below during forging (Fig. 7). When it was magnified, a straight linear division of recrystallization could be seen (Fig. 8).
5.2. Measurement of HardnessHardness of the alloy was measured using a micro Vickers hardness tester (Akashi HM-101) at the Tochigi prefecture industry technology center. The test force was 98.07 mN (10 gf). The measure points on sample No. 1 were placed horizontally at 1 mm intervals (Fig. 5), while those on sample No. 2 were near the surface and center parts of the knife (Figs. 9, 10). Sample No. 1 (casting) was 95.8 – 114.3 Hv, 106.5 Hv average (Fig. 11). Sample No. 2 (forged knife) was 119.7–169.2 Hv, 144.9 Hv average (Fig. 12). The center section near the edge and the knife’s backside were the hardest parts. Continued cold forging and annealing would reach a hardness strength 1.5 times greater than casting alone.
The sample No. 2.
The measure points of sample No. 2.
The sample No. 1.
The sample No. 2.
The result of this experiment showed that casting after forging created a stronger blade. Based on our measurements, I then compared the bronze replica samples with iron artifacts. Iron transforms the metal’s character by including carbon. A 0–0.2% carbon content is known as soft steel, 0.2–2%C is hard steel, and over 2%C is considered casting iron in this paper. I shall now refer to the carbon steel used in today’s tools.
Carbon steels (JIS)
SK65 (0.6–0.7%C) cold press; 190–280 Hv
SK140 (1.3–1.5%C) cold press; 230–320 Hv
S10CS09CK (0.05–0.15%C)
No-quenching; 109–149, quenching; 121–149 HB
S15CS15CK (0.10–0.2%C)
No-quenching; 111–149, quenching; 143–235 HB
Compared with the above JIS values, the following strengths can be considered: bronze (10% Sn) = soft steel < quenching soft steel < hard steel. Figure 13 shows that bronze increases in tin content as hardness increases, but there is a decrease in malleability. For knives, cold forging using 10%Sn was to obtain a soft and fragile phase and hardening of the malleable α-phase (Fig. 14), which would be practical for use.
Hardness and tensile strength of Cu–Sn alloy.13)
Cu–Sn phase diagram.14)
Based on this study, we see that a bronze knife subjected to continuous cold forging and annealing is 1.5 times stronger than one created by casting. This was the same strength of a no-quenching soft steel. Therefore, when bronze knives were produced in the first millennium BC in Xinjian, they were hardened by using the forging production method. Conversely, for iron to become a more practical material than bronze, it would require heat treatment that increased and decreased the carbon concentration. Alternatively, annealing could be used to obtain soft or hard steel to fill the object. If the Iron Age was a time when the raw material of tools and weapons changed from bronze to iron, and the constant use of iron established this cultural phase, heat treatment technology would be required.
After the appearance of iron, bronze knives were still used for hundreds of years in Xinjiang. It was thought that iron was too difficult to process and that bronze strong enough to use as knives was worth continued use.
The bronze knives that pastoral people used for butchering animals were also practical as ordinary tools for cutting meat and tendons or breaking small bones. This required keen flexibility and balance to avoid nicking the blade. Sufficient processing in bronze knife production would result in superior tool performance. For future evaluation of ancient bronze tools and weapons, I plan to accumulate additional data on tensile strength using the Charpy impact test.
※ This study was done in collaboration with the late Mr. Hiroyuki Ito (the institute of Wako Metal Technology).
