DOI: 10.1148/rg.251045175
RadioGraphics 2005;25:209-213
© RSNA, 2005
Scenes from the Past
Carthaginian Pair of Tweezers1
Roel J. Jansen, RT,
Hans F. W. Koens, MSc and
Jaap Stoker, MD, PhD
1 From the Department of Radiology, Academic Medical Centre (R.J.J., J.S.) and the Department of Mediterranean Archaeology, Allard Pierson Museum (H.F.W.K.), University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands. Received August 27, 2004; revision requested September 10 and received October 8; accepted October 11. All authors have no financial relationships to disclose. Address correspondence to R.J.J. (e-mail: r.j.jansen@amc.uva.nl).
 |
Introduction
|
|---|
The oldest known pair of tweezers dates back to the 3rd millennium BC. The tweezers were cast from molten bronze (1,080°C) in a double mold made of baked clay. After cooling down, the mold was opened and the tweezers removed. Some cold working, consisting of forging the cooled metal to improve hardness and flexibility, was needed before a bright, shiny "gold" object was ready for use. Bronze, an alloy of copper and tin, was the most suitable metal in this period. The melting point of bronze could be reached in the forge, and the metal was then easy to work up. Changing the tin content of the alloy made the metal more or less hard, flexible, and elastic. The right combination of copper and tin resulted in a high-quality object.
In several campaigns, a group of archaeologists from the University of Hamburg, Germany, led by Professor H. G. Niemeyer, excavated three houses built in the 7th century BC. The houses were located under the Decumanus Maximus, a main Roman road dating from the 1st century BC and an important part of the road system of ancient Carthage. Carthage was founded around the end of the 9th century BC by the Punicians and for hundreds of years exerted a powerful influence on the political and cultural life of the western Mediterranean basin. The Romans destroyed the city in 146 BC. The excavation and the corresponding part of the Decumanus Maximus are now located in the center of a suburb called New Carthage in Tunis. The oldest remains could be dated to the Archaic period (700–600 BC). During the excavations, a large number of metal objects were found. One of the objects, a pair of tweezers (Fig 1), will be discussed in this article. Tweezers were used for medical purposes and for the removal of superfluous hair. In the Punic period, hairless skin was a sign of taking good care of oneself, being both a mark of beauty and a hygienic measure.
View larger version (48K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Photograph shows Carthaginian tweezers dating from the 7th century BC. The first step in conserving the tweezers was to clean the iron using an electrolytic reduction process to disjoin the surface corrosion layer from the iron parts. After being restored, the tweezers were impregnated with a two-component solvent-free epoxy compound to protect them from active corrosion.
|
|
The tweezers date from the 7th century BC. Although the shape of the tweezers is not unusual, the fact that they are made of iron is highly unusual. Excavation showed the tweezers to be in fragments; however, the fragments were in reasonably good condition for two reasons: (a) The iron contained a high level of calcium, which neutralized the corrosion that is normally caused by combinations of sulfur and phosphorus in iron ore. (b) The tweezers were found in a sandy, dry, warm environment, and the absence of large amounts of water delayed the corrosion process. However, although depositional conditions were ideal, the passage of 2,700 years during which the tweezers were buried caused alterations due to corrosion from water and chemical reactions, and restoration and conservation were time consuming. This pair of iron tweezers is unique in that iron would have been suitable for tweezers only if the blacksmith were able to create steel to guarantee the right amount of flexibility. Deliberate steeling was not a controlled process before Roman times (1,2).
Until now, metallurgic examination has been the standard method of revealing the techniques used by a blacksmith to create such an object. However, this type of examination has two disadvantages: (a) it is destructive because samples need to be taken, and (b) only small parts of the object of interest can be examined. Radiography offers the advantages of being nondestructive and allowing examination of the entire object. Because the use of iron was highly unusual in making this type of object, we decided to first examine the tweezers radiographically to identify regions of interest for the metallurgic examination.
The purpose of our study was twofold: (a) to establish which technical skills a Carthaginian blacksmith possessed in the 7th century BC, and (b) to study the potential of radiologic examination with radiography and computed tomography (CT) as a nondestructive alternative to metallurgic examination.
|
Materials and Methods
|
|---|
The tweezers had a length of approximately 11 cm and a thickness of approximately 0.7 cm. One leg was 0.8 cm longer than the other. The width of each leg was approximately 0.5 cm.
Radiographic Examination
Radiographs were obtained with a Polyphos generator and P125/30,50CR x-ray tube (Siemens Medical Systems, Erlangen, Germany) with a 0.8-mm focal spot. The radiographs were obtained in the anteroposterior, lateral, and 45° oblique planes using extremity film (Agfa Gevaert, Antwerp, Belgium) at 70 kV and 16 mAs. An x-ray grid was not used. CT was performed on an MX 8000 scanner (Philips, Eindhoven, the Netherlands) with different settings for section thickness, milliamperes, and kilovolts. However, because the tweezers were a small metal object, they produced too many artifacts for CT to provide any information.
The purpose of the radiographic investigations was (a) to obtain information about homogeneity and structural differences in the metal (corrosion, linear patterns), (b) to investigate whether there were differences between one leg and the other, and (c) to establish a correlation between the metallurgic and radiographic findings.
Metallurgic Examination
A sample was taken from each leg (Figs 2, 3) and cut with a liquid-cooled cutoff wheel. Axial and longitudinal sections were made from each sample. The sections were mounted in a two-component epoxy compound consisting of epoxide 20–8130-032 and hardener 20–8132-008 (Buchler, Lake Bluff, Ill). This cold-mounting epoxy resin system adheres well to samples and has a very low rate of shrinkage during curing. The epoxy compound sets in about 8 hours at room temperature. The samples were mounted in such a way that both cross sections could be examined in detail. After the epoxy compound had completely set, the surface was polished with graded grit paper and polish paste according to standard metallurgic procedures. The metal was etched with 3% concentrated nitric acid in ethyl alcohol. The sections were examined with an Olympus type 404028 optical microscope (Olympus, Tokyo, Japan) at a magnification of x125.
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Photograph shows a cross section of the shorter leg with a homogeneous appearance (the sample was obtained at the location indicated by the arrow in Fig 4). The color variation and the linear ridges and grooves are due to the cutting of the leg with a liquid-cooled cutoff wheel to harvest metal for the metallurgic examination. The empty holes contained slag inclusions (ie, remains of the mineral parts of the iron ore—in many cases, silicon oxide) that were lost during cutting.
|
|
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Photograph shows a cross section of the longer leg with corrosion in the core (double-headed arrow) and a steeled rim (black arrow). White arrow indicates surface corrosion (the sample was obtained at the location indicated by the arrow in Fig 4).
|
|
|
Results
|
|---|
Radiographic Examination
The radiographs were of good quality and revealed the inner structure of the tweezers (Fig 4). There were multiple fractures in both legs and structural differences between one leg and the other. The shorter leg appeared homogeneous and opaque in the proximal part (ie, toward the place where it joins the longer leg) and somewhat inhomogeneous in the distal part (ie, toward the tip of the leg). There were seven fractures in the shorter leg (Fig 4). Fracture a had a small loose fragment. Fracture c fit badly because of loss of material. At e, three fractures were present, resulting in two small loose fragments near the distal end of the tweezers. Fractures b and d were normal fractures with no fragments. The shorter leg had retained its original shape, although a part of the top of the leg was missing.
View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Anteroposterior (a) and lateral (b) radiographs show multiple fractures in both legs. The shorter leg appears homogeneous, whereas the longer leg appears inhomogeneous, with both radiolucent and opaque areas. Arrow indicates the location from which material was taken for metallurgic examination. a-h indicate the location of the fractures.
|
|
View larger version (37K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Anteroposterior (a) and lateral (b) radiographs show multiple fractures in both legs. The shorter leg appears homogeneous, whereas the longer leg appears inhomogeneous, with both radiolucent and opaque areas. Arrow indicates the location from which material was taken for metallurgic examination. a-h indicate the location of the fractures.
|
|
The longer leg appeared inhomogeneous, with both opaque and radiolucent areas. The opaque areas were located in the proximal part of the leg and at the exterior part of the leg, where a small (0.5–1-mm-wide) band of increased opacity was seen. Some focal broadening was present, especially in the distal part of the leg. In these locations, the leg had lost its original shape. There were three fractures in the longer leg; fractures g and h had loose fragments.
Metallurgic Examination
The sections were of good quality and revealed the characteristics of a process known as case-hardening. The longitudinal section of the longer leg revealed a needlelike structure at the boundary of the sample consisting of martensite (Fig 5), a very hard constituent of steel that is formed when carbon-rich iron is quenched in liquid that is at or near room temperature. The blacksmith would reduce the extreme hardness and brittleness of martensite by tempering the metal. The core was made of impure iron (wrought iron) and showed slag inclusions. These slag inclusions were scattered and mostly lay lengthwise along the longitudinal section. It was not possible to melt iron at the time the tweezers were made because the fuel used (charcoal) could not generate enough heat. The maximum temperature that could be reached (1,200°C) caused melting of only the mineral parts of the ore, and a porous lump of wrought iron formed on the bottom of the furnace. The core had been heavily attacked by corrosion (Fig 5). Color differences in the sections indicated that the carbon content was high near the exterior and decreased as one moved toward the center. Corrosion was most severe in areas with a low carbon content. It could also could be established that, after the tweezers had been removed from the forge, a gradual cooling down to approximately 300°C had taken place. The tweezers had then been cooled rapidly, thereby optimizing the steeling process. Metallurgic examination revealed complete steeling of the shorter leg.
View larger version (184K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Photomicrograph (original magnification, x125) shows the longitudinal metallurgic section of the longer leg. Double-headed arrow indicates a needlelike structure (martensite), arrow indicates a heavily corroded area in the core, arrowhead indicates a slag inclusion.
|
|
Comparison of Radiographic with Metallurgic Examination
Radiography demonstrated the results of case-hardening very well. The homogeneous opaque areas corresponded with areas of steeling, which, as mentioned earlier, was complete in the shorter leg. However, this was not an ideal situation because, although the steeled parts would be resistant to corrosion, the steeled iron would also become too brittle.
The inhomogeneous radiographic pattern and the radiolucent areas in the longer leg indicated areas where corrosion had occurred, since corrosion products absorb fewer x rays than does metal iron, making the areas of corrosion darker at radiography. Corrosion is also the reason that some parts of the tweezers had lost their original shape. The volume of the corrosion products was larger than the volume of the iron of which the tweezers were made and caused the areas of broadening. The small, opaque band at the exterior part of the longer leg indicated areas of steeling.
|
Discussion
|
|---|
During the production of iron in the 7th century BC, the melting point of this metal (1,537°C) could not be reached. Even with the highest quality charcoal, a maximum temperature of only 1,200°C could be achieved. Until the Middle Ages, when fossil fuels were discovered, the iron produced was actually wrought iron, which is very low in carbon and therefore very soft. The metallurgic process used to harden wrought iron is complicated. One way to harden an iron object is to cool it in water during forging, but this has no effect if the carbon content of the object is not raised first. The process used to raise the carbon content is called carbonizing and yields a piece of iron containing steel structures. Carbonizing can take place spontaneously during forging: Carbon from the charcoal of the forge fire will penetrate the iron, and, if the object is cooled in water, some carbonizing will occur. Tholander (3) performed metallurgic examination of Cypriot knives from the 11th century BC and found microscopic steel structures (cementite, martensite). These structures were found only inside the knife blades. However, steel structures inside a blade will not increase the hardness of the edge of the blade and therefore could not have been made on purpose, being in fact useless.
Tweezers will function only when they are flexible, and flexibility can be achieved only by steeling the iron. Because iron was chosen for making this pair of tweezers, the blacksmith must have had knowledge of carbonizing and hardening iron. The tweezers were formed from a single small rod of wrought iron and were deliberately carbonized and steeled by means of case-hardening. The object was put into the forge under reducing conditions (ie, an oxygen-free environment), surrounded with carbon (eg, animal bones), and left there for some time. The carbon penetrated the iron from the outside. The amount and depth of carbon penetration would depend on how long the object was left in the forge and on the temperature that was reached. Cooling an object rapidly once it reaches about 300°C will start the process of steeling. Case-hardening is a well-controlled process if forge conditions, temperature, and time are controlled (Fig 6), but such control was not possible during the period in question, and the exact amount of carbon penetrating the iron would have been unpredictable (4). This was the reason why there was a difference in steeling between one leg and the other: The shorter leg had absorbed a large amount of carbon, so that steeling was complete in that leg.
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Graph illustrates the diffusion of carbon into iron in a forge at a temperature of 920°C. The depth of carbonizing is given in millimeters. The x axis represents the amount of time elapsed, and the y axis represents the carbon content in percent carbon by weight.
|
|
Fracture c in the shorter leg (Fig 4) was ancient. The other fractures were due to postdepositional circumstances. The ancient fracture could be distinguished from the others because of its poor fit and the loss of material. It is tempting to assume that this fracture occurred during usage in the past and was the reason the tweezers were thrown away. The shorter leg was of poor quality because it became too brittle during the process of case-hardening.
|
Conclusions
|
|---|
A pair of iron tweezers dating from the 7th century BC were found in an excavation in Carthage. The tweezers were made with a process known as case-hardening. They are the oldest product known to have been produced with this technique and are, therefore, unique. Their discovery demonstrates that case-hardening was used to steel iron long before Roman times by a Carthaginian blacksmith. Radiography revealed the characteristics of case-hardening very well. The small, opaque band seen at radiography at the exterior of the longer leg and the homogeneous and opaque areas seen in the shorter leg corresponded with the areas of steeling seen at metallurgic examination. The inhomogeneous pattern and radiolucent areas seen at radiography corresponded with areas of corrosion.
|
References
|
|---|
- Tylecote RF. A history of metallurgy London, England: The Metals Society, 1976; 164-165.
- Forbes RJ. Studies in ancient technology Vol VIII. Leiden, the Netherlands: Brill, 1962; 111-114.
- Tholander E. Experimental studies on early iron-making. Hist Metallurgy 1987; 124-128.
- Maddin R, Muhily JD, Wheeler TS. How the Iron Age began. Sci Am 1977; 237:122-131.
Top
Introduction
Materials and Methods
