Examination of Bronze Implements
T HE results of modern methods of research on the constitution of metallic alloys and on the effects of varying mechanical and thermal treatment on their structure and mechanical properties have increased our knowledge of these subjects enormously. Of these methods, the use of the microscope for the examination of the micro-constitution and crystalline condition of suitably polished and etched specimens is one of the most important. This method of examination is capable of indicating not only
... dicating not only how the component metals exist in the alloy, but also what mechanical or heat treatment the alloy has undergone. As regards constitution, the microscopical ·examination is used in order to ascertain whether the component metals are present in the free state, in chemical combination, or mutually dissolved and retained in solution in the solid state. As regards treatment, the structure observed is used to decide whether the metal object was obtained by a casting process or was shaped by mechanical work, and is also capable of indicating whether any work has been applied to the object and also whether this work has been followed by an annealing process. Below are given some results of an examination of a chisel and a palstave of the Bronze Age found about 1898 in a brickfield near the Hanwood-Shrewsbury road and supplied by Mr. F. Drinkwater of West Kirby through Miss L. F. Chitty of Yockleton, near Shrewsbury. With regard to the constitution of bronzes, these consist of alloys of copper and tin in varying proportions, and if the molten metals be allowed to solidify and cool very slowly, or if cooled normally and then reheated for the necessary length of time, the tin is soluble in the copper in the solid state up to 13 per cent. of tin. After such treatment, the bronze would show one constituent only under the microscope. On cooling a liquid alloy containing 8-13 per cent. tin, the first portions to solidify are richer in copper than the portions solidifying last. On polishing and etching such an alloy, these copper-rich portions are found to have a fern-like structure, and the presence of these dendrites, as they are called, is indicative of cast metal. The alloy not being homogeneous throughout its mass has areas with a greater tin percentage than 13, and in these areas a pale blue constituent, which is a very intimate mixture of two solid solutions, makes its appearance. The more of this second constituent that is present the harder and more brittle does the metal become, and with any definite percentage of tin the more quickly the metal is cooled, within limits, the more of this constituent is present in the casting. If the cast bronze be now heated for some hours at a temperature below its melting point, diffusion takes place, the metal becomes more or less homogeneous in composition, and the dendrites and the second constituent more or less disappear, the metal becoming more homogeneous the longer the annealing or the higher the temperature. With cast bronze of less than 8 per cent. tin, the dendrites would appear as above, but little or no blue constituent would be present because the tin rich areas would contain less than 13 per cent. of tin. NO. 2926, VOL. I 16] If the cast metal were cooled in the mould very slowly, this would have, as regards the homogeneity of the metal, a similar effect to annealing. The rate of cooling of a casting will depend on three factors : first, the temperature of the metal when it is cast, it being obvious that the higher this temperature the longer will the solidifying and cooling take. Secondly, the material of which the mould is made. If the material conducts heat rapidly from the metal, the rate of cooling will be greater than in the case of a mould from which the heat was less rapidly removed. The stone mould commonly used for casting antique bronzes would be a bad conductor of heat, and the cooling would take place very slowly. A bronze mould would chill the casting, and cooling would be very rapid. A sand mould would have an effect intermediate between these two. The third factor would be the size of the casting, one of large bulk cooling more slowly than a smaller casting, as there would be more heat to disperse . Another important characteristic in the structure of bronze is the outline of the crystals revealed by the etching. In the cast metal the crystals are very irregular, some crystals being of large size compared with others, and having interlocked boundaries. If the metal be now worked in some way by hammering or rolling, an examination will show the results of this work in the distortion of the original crystals, which are often elongated to a considerable extent at right angles to the direction of the work when this has been excessive. The effects of less drastic amounts of work are seen in the appearance of parallel lines on the crystals, these lines being known as slip bands. If the metal be annealed after working, new crystals will make their appearance, appearing quite different from the casting crystals and characteristic of the annealed metal. These crystals are much more uniform in size, and have sharp lines as their boundaries. They invariably show a great deal of twinning, which is recognised by the crystals having two or more parallel lines traversing them. As a result of this working the dendritic structure will be somewhat destroyed. Instead of being fern-like, it generally appears as thick, dark, irregular parallel bands over the specimen in a direction at right angles to the force applied. A long annealing entirely eliminates this structure, but the new crystals grow before this is effected, so that in annealed metal, traces of the dendrites may still be seen. It will be gathered from these notes that much can be deduced from the micro-examination of antique bronzes. From the fern-like, dendritic, or core structure, as it is variously called, and the type of crystal boundaries, the cast structure is at once recognised, and with an 8-13 per cent. bronze Sl(me sort of deduction can be made as to the mould used, because with very little of the blue constituent present the metal would have cooled slowly, and it is probable that a stone mould would have been used. The twinned equiaxed crystals and the partial or total absence of dendrites indicate that the metal had been hammered and annealed.