Introduction
Optical microscopy study of exterior surfaces can find evidence of alteration, such as over-painting and modern tool marks, as well as evidence of antiquity, such as fossilized organic matter. It is often one of the first examinations performed on an object, and is commonly used to determine the necessity and order of further examination. The non-invasive technique is frequently performed in conjunction with a microscopic examination of the internal structure of a removed sample.
One of the primary analytical techniques performed on site at TK is optical microscopy. TK regularly uses two high-power metallurgical microscopes. The equipment permits study of small samples, removed from objects, and study of the unaltered surfaces of whole objects, even those of considerable size. Metal is the principal material of study by microscopy at TK. Most of the metal objects examined are bronze, some with inlays of precious metal, glass or semiprecious stone. Pigments and objects of copper, iron, bone, ivory, jade, and textile are also examined, though less frequently. Bone and ivory, being organic, are often suited to other forms of testing, and microscopic examination is primarily used to detect alterations, repairs and tampering on such materials. Textiles are suited to radiocarbon testing, and microscopy is generally performed for fiber identification. Both the exterior surface of a whole object and the internal structure of a removed sample may be examined microscopically. Internal structural study is generally reserved for metals and this specific application will be discussed in chapter five. Chapter four pertains to microscopy in general, and deals largely with the exterior surfaces of metallic objects.
Operated in conjunction with the microscopes is a digital imaging system, designed by Syncroscopy.1 The system utilizes a high-resolution digital camera, motorized microscope focus control, and image capture and processing hardware and software, which dramatically increases the information that can be obtained through microscopy. At high magnifications the depth of field that is in focus is usually far less than the overall depth of the site being examined. The imaging system is used to capture up to 255 sequential source images, which are then processed by Syncroscopy's Auto-Montage™ software, creating a single in-focus photomicrograph, which has a greater depth-of-focus than any single image. The increased depth-of-focus allows for more accurate interpretation of the site details. Auto-Montage™ can also be used to create depth maps, models, color relief images, line profiles and anaglyphs. The system is used for all micrograph image capturing at TK, including metallographic examination of polished samples, but is especially useful on uneven surfaces, such as corrosion layers on metal.
Tool Signatures
Tools, such as files, leave distinctive marks, generally referred to as tool signatures. Under magnification, there are visible differences between the marks left by hand tools and modern high-rpm tools. On a purported artifact the presence of modern signatures is cause for additional study, but may not indicate a serious problem. It is not uncommon to find evidence of the use of modern tools on truly archaic metals. Thick corrosion is often removed from exterior surfaces, in order to reveal decorative patterns, inlays or inscriptions. Such cleaning carries the risk of damaging the artifact. However, the risk is often outweighed by the benefit of exposing the underlying detail.
Aggressive cleaning should be avoided, not only due to the damage it causes, but the possible loss of information resulting from significant alteration of the surface. Corrosion crusts and original tool marks can provide tremendous data about the history of an object. In the case of stone, tool signatures and surface encrustations may provide the only strong evidence that the material was carved in antiquity. Many authentic artifacts have been cleaned to the point that they are devalued, as the visible surface is no longer representative of the original.
Aggressive cleaning of metals is frequently followed by repatination, in an attempt to give exposed metal a more aesthetically pleasing appearance.
Patination
When discussing metallic objects, the term patina is often applied to the appearance of corroded metal, but the connotation of the term varies. Technically, it refers specifically to the green corrosion products that form on copper and its alloys. In common usage, it is applied to any alteration in color, on a wide range of materials, as a product of age. For the purpose of this publication, patina refers to any visual characteristic other than that of new unaltered material, whether an encrustation or simply a discoloration. It should be noted that there is evidence of the deliberate patination of metallic objects, for artistic purposes, in ancient China (Chase 1994). As such, the presence of a patina on metal is not necessarily a function of age or an attempt to create the appearance of age.
The corrosion of metal can result in the formation of a mineral crust on the surface. Obviously, chemicals can be applied to hasten this process. Several substances may be applied to the surface of a metal to cover damage or a repair, or to disguise a modern metal. Patination chemicals can turn copper and its alloys most any color, including those commonly encountered on antiquities.
A 
B 
C
D 
E 
F
Figures A-F, above, are of tin-bronze wafers that were deliberately patinated with readily available materials and techniques.2 Wafers A and B were corroded using common household chemicals and an exposure time of only 42 hours. Wafer C was heated in a wood-burning fireplace for three and one-half hours. D and E were suspended half in and half out of solutions made from baking ingredients. D was exposed for eleven days, and E only four days. Wafer F was patinated using a solution available in most hobby stores. The liquid was applied several times over a period of four days. Corrosion on all of the wafers is insoluble in water. These images may provide some idea of how quickly and easily bronze can be patinated. Some forgeries are treated for months, or even years, to achieve the desired appearance, which is often remarkably similar to that of natural long-term corrosion on archaic metals.
Induced patination can produce anything from a discoloration to an encrustation. Most of the recipes used to patinate metal produce a relatively stable surface layer. Copper-alloy forgeries and some repairs on archaic metals may be disguised through an inexpensive process using common table salt. The proper mixture can cause a relatively thick corrosion layer to form in a very short time. Unfortunately, chloride contamination can cause aggressive corrosion on copper and its alloys, resulting in disfiguration of the metal.
Most often, naturally formed corrosion products have a different morphology than those rapidly produced by artificial aging techniques. The natural corrosion tends to be relatively hard and non-porous, while induced, rapidly grown corrosion products tend to be porous and relatively soft. It should be noted that some natural corrosion on artifacts is also soft and porous. Careful study is necessary to determine the probable origin of any corrosion layer. For this reason, examination of the external corrosion layer is not always sufficient to support the classification of a metal object as an antiquity.
Applied Coatings
In cases where repatination of a surface is insufficient to mask a repair or other feature surfaces may be coated with various substances. In one common technique, used on metals, natural corrosion products are removed from a metallic object, crushed, mixed with adhesive or patination chemicals and applied to a surface in imitation of a natural corrosion layer (Fig. 4·2). The scavenged material may have come from the same item, or an entirely different one.
Figure 4·2 Applied coating with coarsely ground minerals. Scale bar = 0.5mm
Within applied coatings, especially those covering repairs, it is not uncommon to find very coarse mineral fragments, such as these. In the image at left, corrosion products have been scavenged and poorly ground, leaving large fragments with sharp edges. Some pieces display the malachite over cuprite layering common in naturally formed corrosion on bronze.
Figure 4·3 Applied coating with pigment grains. Scale bar = 0.5mm
Applied coatings often contain pigments, which may be applied with a simple binder, or they may be added to mixtures of adhesives, ground minerals, dirt and other substances. Tiny clumps of pigment are readily detected under magnification. In this image, the arrows point to clumps and individual grains of pigment in an applied coating on a bronze ding. Not all colorants have a visible granular structure. Stains and very finely ground pigments are often used, making detection more difficult.
Figure 4·4A Spring and Autumn period hu
The vessel at left is a large Spring and Autumn period bronze hu with basic (alkaline) copper carbonate corrosion products. Visual inspection detected slight differences between the color and texture of material at sites 1 and 2, prompting microscopic examination.
Figure 4·4B Portion of inlaid decoration on hu
Figure 4·5A is a photomicrograph of the naturally formed corrosion products at site 1 on the above hu. The blue and green minerals are consistent with azurite and malachite, respectively. Both are basic copper carbonates and are often found together on ancient bronze and copper. By comparison, Figure 4·5B, of site 2, shows an area of touch-up, consisting of pigment, texturing agent and adhesive. The fiber, visible at the lower right, is a modern brush fiber, caught within the applied substance. Modern, applied surface coatings are often textured by adding sand or a man-made substance, such as that used to add texture to acrylic paint. During the examination of an artifact, areas of touch-up are noted and inspected to determine the nature of the underlying surface. Among collectors, over-painting of repairs is a commonly accepted practice, provided that no attempt is made to deny the existence of such repairs. It is proper to use only those materials that can be removed without causing additional damage to the artifact. A related practice, which is not encouraged, is the application of material to the natural un-repaired surface of an artifact. When detected, such altered areas are inspected proceeding on the assumption that the site is repaired. When it is determined that no repair exists a decision must be made concerning removal of the applied substance.
Figure 4·6 Money tree branch with natural corrosion and applied coating.
This money tree branch, provides a good example of such alteration. Part of the branch retains its naturally formed corrosion (1), and part has recently been coated with a green material (2). The applied substance is not covering a repair, but the naturally corroded surface. The composition of the alloy of many money trees often results in the formation of corrosion with a smooth, warm reddish-brown to near black surface. This appearance is quite different from what many collectors expect on bronze. Certain individuals may coat the metal with greenish materials to make it more consistent with people's expectations.
Figures 4·7A and 4·7B are shown at the same magnification, for comparison. Figure 4·7A is of the natural corrosion at site 1 on the branch. The red is the copper oxide cuprite and the green is the basic copper carbonate malachite. Some of the light-colored material is the lead carbonate cerussite, while some is simply the granular dirt from the excavation site. Figure 4·7B is of the applied substance at site 2. As can be seen in the photograph of the branch (Fig. 4·6), the two sites do not appear significantly different to the naked eye. However, site 2 contains mineral fragments, patination chemicals, pigment and a binder. The minerals have been relatively well ground, and the patination chemicals have produced a coating on the pieces, giving them a rounded shape.
Though the above image pertains more to metallographic examination, it is included in this chapter to stress the importance of understanding the condition of an object as a whole, during both examination and restoration. Seen in cross-section, this corroded money tree branch retains a silvery metallic core. The outermost greenish layer is the exterior surface of the naturally formed corrosion. To the naked eye it appears a slightly shiny dark greenish-gray. As mentioned previously, it is this surface that is so often covered with applied substances, in an attempt to create an appearance more consistent with what people expect of ancient bronze. Unfortunately, removal of the applied material can result in significant damage to the artifact. As can be seen in fig. 4·8, though the metal core remains, corrosion has penetrated to a significant depth, making the branch quite fragile.
Mineralized Organic Matter
One circumstance where microscopic examination of a surface may, in and of itself, provide sufficient evidence of antiquity concerns organic matter. The corrosion crusts of some artifacts, especially metals, may contain traces of petrified organic matter. Mineralization of organic matter can occur in several forms. In the process of permineralization, the hollow pores of the material are filled with minerals, but the organic matter itself is not replaced. True mineralization is also referred to as replacement. In such cases, minerals replace the cell walls themselves. Organic matter in this condition is said to be petrified (Strömberg c1998). Several reactionary web sites exist which claim that organic matter can be petrified in a very short time, on the order of a few years. These sites erroneously identify permineralized material as true mineralized or petrified. Visible differences between the two can be detected during examination with sufficient magnification. Figure 4·9 Mineralized plant stem. Scale bar = 0.2mm Figure 4·9 Mineralized plant stem. Scale bar = 0.2mm
When mineralized organic matter is discovered on an object the area is carefully examined for any indication that fragments of authentic mineralized matter have been artificially attached to the surface. The process of true mineralization requires considerable time and the presence of mineralized organic matter on an object is an important indicator of antiquity. It should be noted that mineralized organic matter cannot be radiocarbon dated. This is because no trace of the original matter remains, only the mineral pseudomorph.
This image is of a sample of partially mineralized fabric, viewed in cross section. Some of the fibers themselves are petrified, and the whole sample is also permineralized.
Within a given sample of mineralized wood or fabric, individual fibers may have a different appearance and composition than those immediately adjacent. Differences in the chemical composition of the fiber, including the presence of fabric dye, may account for this phenomenon (Jakes and Sibley 1984). In Fig. 4·11, the minerals are a mix of oxides and carbonates.
Surface examination can provide tremendous information about the history of an object, though sufficient evidence to classify an object as an antiquity is not often found. In the case of metallic objects, the information gathered is generally used in conjunction with a metallographic examination.
