YO ARE HERE:>>Collectors resources>>Scientific examinations>>XRF.



  • The first of these is the use of XRF analysis in the determination of the elemental composition of  faience artefacts which I began some time ago here
  • I previously  discussed the results with reference to only one fake shabti.
  • Here we look at the results of XRF anaylses of a group of eight shabtis.



An analysis of an analysis.



Egyptian faience is a non-clay based ceramic composed of crushed quartz or more usually, sand, with small amounts of calcite lime and a mixture of alkalis, displaying surface vitrification due to the soda lime silica glaze containing metal  pigments to create a bright coloured lustre. Plant ash, from halophitic plants typical of dry and sea areas, was the major source of alkali until the Ptolemaic Period, when natron - a naturally occurring alkaline mixture of sodium salts, carbonate, bicarbonate, chloride and sulphate, based alkalis almost completely replaced the previous source. 


The chemical composition of faience materials varies over time and depending upon the status of the workshop and  the varying accessibility of raw materials .There were also deliberate variations to achieve different colour and lustre effects in the surface and this brings a wide and varying spectrum to the elemental contents of faience glazes.

Although it is often written that crushed quartz pebbles were used to avoid contamination from such as iron in the sands, this is highly unlikely excepting for possibly the purposes of making a very fine white faience. (Kaczmarczyk, A., and R. E. M.Hedges. 1983. Ancient Egyptian faience: An analytical survey of Egyptian faience from predynastic to Roman times. Warminster, England: Aris & Phillips).


Prior to about 4000 BC only stones were glazed  but at about this time  stone cores were replaced by ceramic ones, made mainly from finely ground sand, but occasionally of comparatively coarser sand, which was modeled into shapes; cores also contain minor amounts of lime and either natron  or plant ashes.

A summary by Vandiver (1982: 167) of a composite range of chemical analyses of the body shows 92-99% silicon dioxide, 15% calcium oxide ,0.5-3% sodium oxide with small quantities of copper oxide,  aluminium oxide, titanium dioxide , magnesium oxide and  potassium oxide.


Often very friable, they are frequently white, or practically white in colour, but can be tinted brown, grey, yellow, sometimes very slightly blue or green (Lucas 1962: 157; Kaczmarczyk & Hedges 1983: 123; Vandiver & Kingery 1986: 20).  In the core, minute angular particles of quartz are bonded together by varying amounts of interstitial glass.

The cores are covered with an alkali-based glaze, a soda-lime-silica  generally 60-70% silica, 16-20% soda and 3-5% lime and various  metal oxides, principally copper to achieve a blue colour, and others used  to create  other colours.

Some companies offer XRF analysis http://en.wikipedia.org/wiki/X-ray_fluorescence in the determination of the authenticity of faience shabtis: this provides anelemental analysis of the surface glaze.


  • Is it reliable? Is it a helpful test?


Well, depends upon  how skilled they are at using the equipment and doing adequate surface preaparations when needed.

It then also depends upon their interpretation of the results.


The company used here utilises it's own database to interpret the results.


Take these two for example.



  • Number 1:  Careful and informed examination leaves one doubtful. On  a 1 to 5 scale it is exactly in the middle as equivocal at 3.
  • Number 5: is probably ok. Would be number 4 on such a scale.


XRF analysis results.




Note that the  percentages do not add up to 100.

Silica which will of course be  found as the  predominant  constituent has clearly been deliberately omitted here.



Both are deemed by this  laboratory to be consistent with their database for Egyptian shabtis of this period.


Note that their database can only be as useful as it is properly relevant; that is to say it is only valid if it is a database created by the XRF results of faience  pieces which are  by other means known to be certainly authentic.


Are there such valid and reliable databases which we can consult?

Yes, there certainly are.

I consider this a little later.



Both of these shabtis  were thought by several specialist shabti collectors rather uncertainly genuine: though number 5 thought  slightly more likely to be genuine.


To what extent does  the "positive " XRF results add any useful evidence  to our thoughts about them?


Both sides of the second one have been tested whereas only the front of the first one has. There are large absolute differences in elemental percentages between front and back which is interesting but not what we are considering here at the moment.


Both surfaces contain some calcium, copper and  iron albeit is  slightly different proportions. The first one contains no tin or manganese but the seconds does, albeit in trace amounts.


Let us compare this to a pair of what we assessed to be entirely genuine faience shabtis which the XFR tests also confirmed as genuine.


These are numbers 4 and 7. (Number 7 is a rare 19th dynasty shabti with a hieratic text)




Note that the percentages do not add up to 100.
Silica which will of course be found as the predomninant constituent has clearly been deliberately omitted here.


Here both front and back of each has been tested.

Again there are variations between fronts and backs.


Both surfaces contain some calcium, copper and  iron albeit in slightly different proportions.


Number 4 contains no tin but number 7 does, albeit in trace amount

Both contain traces of manganese on the front but none on the back

Number 7 contains a trace of titanium and tin whereas number 4 does not.

Number 7 also contains a trace of zinc on the  front but not the back.






Let's look at the these definitely fake shabtis submitted;only one of whch was said to be false by XRF determination.

Numbers 2, 3  and 8.




Has a lot of zinc. Only one other shabti here, a fake (number 8) had zinc: albeit only a trace amount. But the genuine 19th dynasty shabti had a trace of zinc.

Showed some titanium and so  did the fake number 8 albeit very little. But the genuine 19th dynasty shabti also showed a trace




FAKE  3: No zinc or  titanium , a lot of calcium, no copper  but  and both strontium and iron which fake 8 does not have.




FAKE 8: Has potassium and lead which neither other of the fakes appears to have .




So what can  we make of these results above?


Well, let us look in more detail at these results  and add in another shabti we think, but are not certain, is fake; though by this XRF analysis was said to be genuine.


The results are summarised  as follows.There were 3 definite fakes submitted. (numbers 2.3 and 8)

Only one of these, number 2,  was determined by XRF to not be genuine.

This was one with a very unusual and odd elemental content  You  can see it here>>> http://www.collector-antiquities.com/541/


It is however considered again below.

There was a further shabti  thought  to be "probably fake" (number 6)  (2 on the 1-5 scale) and this one too was determined by XRF to be genuine.There were 2 definitely genuine ones submitted. (numbers 4 and 7)

Both were then determined by XRF to be genuinely ancient.



These are the results for all.



Note that the  percentages do not add up to 100.

Silica whch will of course be  found as the  predomninent  constituent has clearly been deliberately omitted here.



A further analysis of these results and inferences drawn about the database which was used to interpret them.>>>>>>>