Analytical Method

 

Neutron activation analysis is used to determine the concentrations of elements present in each sample of stone, thus identifying its compositional profile or "fingerprint." This method is the best presently available for limestone analysis for several reasons:

• its precision and accuracy allows the testing of very small samples;
• it determines many elemental concentrations simultaneously with good precision and accuracy;
• its sensitivity permits the determination of constituents present in very small concentrations (parts-per-million or parts-per-billion) in even very small samples. 

 

     Samples of limestone are collected, weighed and sealed in high-purity quartz for irradiation in a nuclear reactor such as the University of Missouri Research Reactor (MURR).   Samples of known standards are included for calibration. 

 

 

The encapsulated limestone is bombarded with neutrons that convert the nuclei of the constituent elements to radioactive isotopes. These radioactive isotopes eventually achieve stability by emitting their extra energy in the form of gamma-ray photons. The process is summarized in the schematic diagram.

 

(Click here for diagram)

 

Each radioactive isotope of a given element has a characteristic energy and mode of decay, and a characteristic decay life time. 

 

The energies and the intensities of the gamma rays emitted by the radioactive isotopes of the elements in each irradiated sample are measured using a germanium detector. The resulting spectrum shows not only which elements are present but their relative amounts as well.

 

(Click here for expanded diagram)

 

 

 

LIMITATIONS of compositional analysis in determining provenance:

 

When attempting to relate one sculpture to another or to a limestone formation on the basis of composition, several limitations intrinsic to the process must be kept in mind:  

• The stone source that provided the sculptor's raw material is assumed to be homogeneous.
• Variation in composition of the stone within a given source is assumed to be small compared to variations among geographically and geologically different sources.
• We can characterize a stone source and evaluate the natural variations in concentration to be expected from that source only if a statistically significant number of samples from the source is available for analysis and comparison.
• The sample taken from a sculpture for analysis is assumed to be representative of  the sculpture's stone. This may not be the case because the sampling location was chosen to minimize visible damage, or because stone from only one location was tested. In such cases sampling error may be important.
• The analytic procedure can introduce an error in concentrations of the order of 5-10%, depending on the compositional variable determined.  This error is usually small compared with sampling errors.
• The results can be expressed only in terms of statistical probabilities, not certainties