YOU ARE HERE:>>Real or Fake>>Fake Roman bronzes, section 2, page 3.



23rd December 2009



June '10

This is an update about the pair of Minerva figurines on this page.


Some  specialist investigation.

X-ray Fluorescence (XRF)

Some information here  .

XRF is an non destructive elemental analysis technique capable of highly accurate determinations for major elements.

A material is exposed to X-rays of high energy, and as the X-ray (or photon) strikes an atom (or a molecule) in the sample, energy is absorbed by the atom.

If the energy is high enough, a core electron is ejected out of its atomic orbital.

An electron from an outer shell then drops into the unoccupied orbital, to fill the hole left behind. This transition gives off an X-ray of fixed, characteristic energy that can be detected by a fluorescence detector.

The energy needed to eject a core electron is characteristic of each element, and so is the energy emitted by the transition.


There are basically two types of XRF.

WDXRF (wavelength dispersive X-ray fluorescence) separation is achieved by diffraction, using an analyzer crystal that acts as a grid.

The specific lattice of the crystal selects the correct wavelengths according to Bragg's Law.

A WDXRF spectrometer provides:



These are results we have obtained.


  • Although it's not strictly metallurgically correct the word bronze is used for all copper alloys.

  • Brass is any alloy of copper and  zinc while bronze is principally an alloy of copper and tin

  • The presence of zinc in anything but trace amounts  in a copper alloy is a reliable indication of relatively recent manufacture; almost certainly manufactured within the last 2000 years. Although copper alloying with zinc wasn't generally used in the west in antiquity, in other areas its use was known from mid-first millennium BC.

  •  The higher the zinc content in an alloy the more likely it is of recent  date of manufacture. Anything above 28% is considered reasonable  proof of manufacture after 1500 AD.

  • Low brass is a copper-zinc alloy containing 20% zinc with a light golden colour and excellent ductility; it is used for flexible metal hoses and metal bellow.

  • Although forms of brass have been in use since prehistoric times its  nature as a copper-zinc alloy was not understood until the post medieval period because the zinc vapour which reacted with copper to make brass was not recognised as a metal at all .

  • There are very early brass artefacts but they are small objects. The earliest brasses may have been natural alloys made by smelting zinc rich copper ores. The majority of the zinc however, being a very volatile metal would have been lost as a vapour . The boiling point of zinc is only  917C whereas the smelting of copper ores requires temperatures of over 100 C.

  • In the Roman Republic, all "copper alloy" items were real bronze, but around the beginning of the Empire brass came into use.  Brass was being deliberately produced from metallic copper and zinc minerals using the cementation process in which metallic copper was heated with calamine, a zinc ore   and variations on this method continued until the mid 19th century. The Romans used brass which has a much lighter much yellower colour than bronze for a while in  their Sestertii and Dupondii coinage after the reforms of 23 AD. This was a brass  alloy consisting of 80 percent copper and 20 percent zinc which is called orichalcum.

  • Orichalcum  contains  zinc, usually  in the range of 15 to 20 percent zinc for things like helmets, lorica fittings, etc.  But the percentage of zinc varied, from 5 to 10 percent for rivets up to a known maximum of about 26 percent, and for cast items a little lead was usually added. (Mixing lead into the copper-tin alloy produces "lead bronze," which may contain as much as 10% lead. The lead in the alloy does not become part of its crystalline structure, increasing the fluidity of compound when it is in its molten state. This facilitates casting, particularly the casting of finely detailed objects.)

  • Brass was also used in some fibulae and brooches , mirrors and much of the military used this copper alloy. Modern yellow brass is usually 30 percent zinc, so its colour will be slightly more yellow than many Roman items.

  • In Orichalcum and Related Alloys by Earle R. Caley (Numismatic Notes and Monographs No. 151, The American Numismatic Society, 1964)

Orichalcum coins listed as tested show a zinc content of 6.43% to 26.71%, the lowest being a dupondius of Commodus, the highest, a dupondius of Caligula.

  • Without a metallurgical analysis it's hard to tell the difference between ancient bronze and brass.  Although with a reasonably highish zinc content the bare metal tends to a yellowish golden  colour.

  • Copper-tin alloys are true bronzes, copper-zinc alloys are brasses. They can contain a small % of lead and other minerals to give fluidity during the working process and some impurities.


Lowering the % of Tin and Increasing the % of Zinc you have this sequence.

Bronze > Bronze/Gunmetal > Gunmetal > Brass/Gunmetal >Brass.

On  3425 Roman brooches studied from the Richborough Collection we have;

  • Brass: 1150
  • Brass/Gunmetal: 127
  • Gunmetal: 376
  • Bronze/Gunmetal: 195
  • Bronze: 1507
  • Silver/Other alloys: 70


Roughly you have these ratios.


Copper with < 2% Zinc and 5-15% Tin = 1507 Bronze Brooches.

Copper with < 4% Tin and 12- 24% Zinc = 1150 Brass Brooches.

with all the gunmetal alloys in the middle.




More "Roman"  bronzes which wouldn't fool a reasonably experienced collector or any genuine dealer for a moment>>>>


  • The fact that the "original" Minerva here is made in an alloy of copper-tin-lead* which is just like the known constituents of Roman leaded bronze figurines  of this period and that unlike the purported copy it appears so have lost it's hands antiquity, and given it's known provenance, suggests very strongly, that it is a genuine Roman bronze of the period roughly 1st -3rd century.

  • Though the other piece whose alloy contains zinc appears to have been cast from mould  taken of another which did not have hands (see the notes about the break points above) this metal analysis is not absolute proof of  it's modernity. However, taken in context, the fact of where it came from and the existence of all the others, it is almost impossible to consider it other than a modern fake. The additional tin and lead in the alloy and the iron  content especially, ( note that the other contains no iron at all) makes it even more likely that it is a modern alloy.

Sample preparation.

Procedures for sample preparation for XRF analysis vary considerably in the cases of in situ or intrusive measurements. Solid sample must be polished to assure surface homogeneity, while powders are usually pressed into pellets.

The XRF tests carried out on these two figurines by a  commercial organization were not by any means as sophisticated or careful as such tests normally  undertaken by archaeologists and the surfaces were tested without any polishing.

The result are subject to error by virtue of "surface enrichment" which is essentially an enrichment of some of the metal elements at the surface due to corrosion processes and hence the surface does not always directly reflect the internal composition.


Some information about the use of this technique with faience objects.


That all  said, the conclusions we can come to about these two pieces are rather likely to be correct.

Ideally I'd like to have another one or two of these Minerva pieces analysed.

Let me know if you see one for sale anywhere! Or if you would like to lend me yours......?



E. R. Caley: Orichalcum and Related Ancient Alloys, American Numismatic Society Publication, no. 151 (New York, 1964)

P. T. Craddock, ed.: 2000 Years of Zinc and Brass, British Museum Occasional Papers, no. 50 (London, 1990, rev. New Castle, DE and London, 3/1998)

PT Craddock: : "Zinc in classical antiquity

Discusses the evidence for the existence of metallic zinc in the classical world. Particularly deals with the well-known passage in Strabo's Geography concerning the mines at Andreida. Suggests this is a reference to a silver smelting process that produced speiss as a waste product, and small objects of metallic zinc as at the 17th-century German silver mines at Goslar.




23rd December 2009


Something appearing on the market  more recently.



Variously described, usually as "Roman" and often with the tag "erotic".


Neither Roman nor erotic!