The title is Non-destructive dating of ancient flax textiles by means of vibrational spectroscopy. It is published in Vibrational Spectroscopy, an Elsevier journal.
The authors are, in order listed, 1) Giulio Fanti, Department of Industrial Engineering, Padua University, via Venezia 1, 35131, Padova, Italy; 2) Pietro Baraldi, Department of Chemical and Geological Sciences, Modena University, via G. Campi, 183, 41000 Modena, Italy; 3) Roberto Basso, Department of Industrial Engineering, Padua University, via Venezia 1, 35131, Padova, Italy; 4) Anna Tinti, DIBINEM, via Belmeloro 8/2, Bologna University, 40126 Bologna, Italy
It was received 15 November 2012, revised 3 April 2013 and accepted 4 April 2013. It is now available online 18 April 2013 to subscribers and for purchase for $27.95
The Abstract:
The possibility to define a two-way relationship between age and a spectral property of ancient flax textiles has been investigated in the present paper employing both FT-IR and Raman analyses on selected samples dated from about 3250 B.C. to 2000 A.D.
After a first selection to eliminate polluted samples, based on visual inspection, on proper mechanical tests and on a first glance at the resulting spectra, eleven samples of the original 14 have been used for Raman analysis and eight for FT-IR analysis.
For the first time, the possibility to define a correlation among spectral properties and age of flax samples, by using calibration curves, has been proved.
In agreement with the kinetics theoretical model, the experimental relationships are of an exponential type, giving correlation coefficients higher than 0.9. The better results were obtained using FT-IR because Raman analysis needs to consider an additional variable due to the non negligible influence of fluorescence.
Presently, the method allows to assign an uncertainty of centuries to the measured data, but future calibrations based on a greater number of samples (though it is not easy to find ancient samples adequate for the test) and coupled with ad-hoc cleaning procedures could significantly improve its accuracy.
It is available at ScienceDirect.com – Vibrational Spectroscopy – Non-destructive dating of ancient flax textiles by means of vibrational spectroscopy
Hat tip to Blog reader Cazab.
An excellent paper for a number of slightly off-field reasons.
Firstly, although the ‘Shroud of Turin’ was apparently used as a way of attracting funding, it has no other mention in the paper at all.
Secondly, there is no mention of any of the controversial mechanical testing that formed the basis of Fanti’s ‘third method’ of dating the shroud.
And thirdly, anyone who thought Andre Marion’s description of his methods (for clarifying putative inscriptions) was at all adequate need only look at this paper to understand why I called it ‘sketchy.’ Fanti’s paper is a paragon of clarity by comparison.
What I would like to see now is his actual results from the Shroud, in the same format as he presents the results for his other textiles, so that we can derive our own opinion from them.
Hugh raises excellent questions. I think it safe to say for all the justified skepticism of Fanti, a new door to study of the Shroud has been opened. I hope it’s not just an illusion.
After reading the paper, it is fair to admit that this paper is published in a JCR journal which in the ranking of “Chemistry ANalysis” journals is in position 42 out of 73. Perhaps, it does not have a high impact factor (1.6) but it can be thought that has gone through a peer-review process. In this sense, my apologies to Fanti for some comments I wrote on this point.
However, as Hugh says, there is not a single mention to the Shroud.
My concern is that results are given at only a 68% confidence level, while tipically they are given at a 95% confidence level in most fields of science for practical reasons. If applied here, that would mean that the error bars would be twice those provided by Fanti. That means that uncertainty would move from 200 years for the FT IR test to 400, from 400 years to 800 years in the low fluoresecence test and from 600 to 1200 years in the high fluorescence test.
Due to very low amount of samples, values of the correlation coefficient and calibration curves should also be provided at a 95% confidence level