The dating of cellulosic ancient objects is of great importance as paper and textiles allow us to transfer cultural and social information from generation to generation. Many experimental methods are traditionally applied to this problem such as C14, dendrochronology, NMR, ir, racemisation. For printed paper dating is favoured by the contained text and by the structure of the paper sheets.
All these methods present some doubts about their results, even if toward some of them, specifically C14, there is a general confidence from the whole scientific community.
Our experience in the field of sensoristic sciences has advised us to look for some methods based on these devices able to be applied to the dating of cellulosic materials. The first approach we adopted is based on the progress of the carboxylation process of cellulose molecule with the passage of time due to hydrolysis and oxidation by radicalic reactions and ageing. The presence of carboxylic groups allows cellulose to behave as an immobilising agent of an enzyme, catalyst of a specific reaction. The catalytic action ensures a multiplying effect on the consumption of the reagents and consequently on the concentration of the final products. The number of immobilised enzyme molecules can be so sensitively evaluated and related to the age of the sample.
A second approach is based on another process suffered by cellulose that is the increase with the time of the alkylation degree. So by measuring the alkyl content by a demethylating enzyme is possible to evaluate the age of the samples. The enzyme for this aim is a transmethylase active in presence of a cofactor and what is measured by HPLC is the concentration of the product of the demethylation. This approach results to fail in the case of coloured samples.
On comparing the sensitivities of the two methods the former results more sensitive (about 100,000 times) than the latter: for the same age 6,8 . 10-3 moles determined against 81,6 nanomoles. The former one is also more rapid and needs less weight of samples, that is a precious quality in this specific case.
We successfully applied both the approach to some samples of papers, textiles, woods supplied to us by private and public institutions, in some cases previously dated in some other blind. In two cases we revealed two falses and a trap sample (modern qualified as old).
The environmental conditions of ageing can affect the results but, according to what we found, at not so determining level. This method does not want to represent a revolutionary one but only a contribution able to bring information and evaluations starting from a point of view largely different from the other ones applied till now.

References
[1] Campanella L., Favero G., Tomassetti M., Vecchio S.: Metodi termoanalitici ed enzimatici per l'analisi e la datazione di reparti cellulosici (lignei, tessili, cartacei). Workshop "Il legno nella storia e nell'arte", Pisa, 30 Settembre - 1 Ottobre 2002. (Testo in http://www.clubarke.it/atti.pdf).
[2] Campanella L., Favero G., Rodante F., Tomassetti M., Vecchio S.: Thermogravimetric and kinetic methods to date wood finds. First results. Annali di Chimica, 93, 2003, 897-907.
[3] Campanella L., Chicco F., Favero G., Gatta T., Tomassetti M.: Further Applications of a new biosensor method for dating cellulosic finds. Annali di Chimica, 95, 2005, 133-141.
[4] Campanella L., Battilotti M., Costanza C.,: Studies on simulated ageing of paper by photochemical degradation. Annali di Chimica, 95, 2005, 727-740.

Based upon the Shroud's 1988 radiocarbon date of approximately 1325 AD [1], the hypotheses of Phillips [2], the comments of Hedges [4], the experiments of Rinaudo [3] and Moroni [5] and the robust statistical analysis of Riani et al. [6], we conducted additional experiments to determine a way to detect in-situ production of carbon-14 from a neutron-producing event or other ambient contamination that would decrease its measured age.
Unbleached modern flax linen whose nitrogen content was measured to be 720 ppm was used in these experiments. At three different times, two linen pieces were irradiated in air and a third piece in carbon dioxide gas. Thermal neutrons having a fluence of 1.055 (±0.021) x 1014 n/cm2, were used.
A theoretical model, based on the known reaction between neutrons and nitrogen to produce carbon-14, predicted the radiocarbon date would be in the future, 3972 AD. The radiocarbon date of the piece irradiated in carbon dioxide was in good agreement; it dated to 4043.
A sample irradiated in air dated to 8451, but another dated to 4165 after it was heated to 245 Celsius for 75 minutes; this date differed by only 5% from the calculated date. This result demonstrates that the heat treatment removed the extra carbon-14 that came from the nitrogen in the air surrounding the linen that then diffused into the linen. Additional heating experiments showed that 99% of this extra carbon-14 would diffuse out again at room temperature in about 6 years.
A variety of experiments also showed that the carbon-14 produced by neutron collisions with the indigenous nitrogen in the linen was not removed by various combinations of heating up to 175 Celsius for 75 minutes followed by all seven pretreatment cleaning procedures used in the 1988 radiocarbon dating.
Experimental results from the second piece of irradiated linen taken at a location 0.5 meter distant from the first piece were inconsistent with results from the first piece. Measurements of the nitrogen content at 18 different locations in the linen cloth showed that these inconsistencies were caused by varying nitrogen content at different locations, which ranged between 453 and 764 ppm and had gradients ranging between 3 and 96 ppm/cm, with and average gradient of 40 ppm/cm.
Using these measured nitrogen variations and the verified theoretical model, calculations were made to predict the variation in radiocarbon dates if the linen cloth used in these experiments had been made in 33 AD and then received a neutron fluence of 8 x 1013 n/cm2. Its average radiocarbon date, as measured in 1988, would have been 1331 and would have ranged between 1018 and 1632 at different locations spanning a distance of 1.5 m. The gradients of radiocarbon age at different locations would have ranged between 190 yr/cm and 5.9 yr/cm, with an average gradient of 78 yr/cm, which is comparable with the gradient observed for the 1988 radiocarbon results of 41 yr/cm.
It is concluded that if the Shroud received a neutron flux, its carbon-14 content would increase, causing its radiocarbon age to be too young despite receiving any of the above-mentioned heating treatments and seven pretreatment cleaning procedures. In addition, in agreement with Riani et al. [6], the measured radiocarbon dates varied from place to place because of variations in nitrogen content that occurred in the Shroud.

References
[1] Damon, P.E., et al, Nature 337, 594 (1989)
[2] Phillips, T.S., Nature 337, 594 (1989)
[3] Hedges, R. E.M., Nature 337, 594 (1989)
[4] Rinaudo, J., 3rd Int. Congress on the Shroud of Turin (1998)
[5] Moroni, M., Barbesino, F. and Bettinelli, M., 3rd Int. Congress on the Shroud of Turin (1998)
[6] Riani M, Aktinson A., Crosilla F., Fanti G.: Int Workshop on the Scientific Approach to the Acheiropoietos Images, Frascati (2010).

Ever since the publication of the Damon et al. paper [1] about the radiocarbon dating of the Shroud, the validity of the claimed 95 percent confidence has been questioned.
ANOVA (Analysis of Variance), a novel method to judge multiple measurements and to evaluate random and systematic errors has been used and promoted by the National Institute for Standards and Technology (NIST) in the United States [2].
ANOVA is also referred to as Inverted Beta [3].

The Nature article [1] noted that this type of analysis had already been known to Dr. Leese of the British Museum [4], who used an analysis of variance to estimate the td value, which lies between the inter- and intra -laboratory degrees of freedom.

In this paper we will use ANOVA to analyze for verification the 1986 radiocarbon dating of the Shroud, using a theoretical model, based on counted ¹⁴C. By this method one can compare the internal (or within) variability with the external (or between) variability of radiocarbon measurements.

To pass the F test, the calculated Ratio F = Σ (EČ)/Σ (IČ) must be equal or lower than the tabulated values, given in statistical tables. In theory, one should use only ¹⁴C particles counted by a computer operated AMS facility. In practice, radiocarbon years are much easier, and provide a valid way to evaluate scientifically AMS measurements.

The theory of the method and all calculations made by a computer program will made available to the conferees.


1. Damon, P.E. et al. "Radiocarbon Dating of the Shroud of Turin" Nature 337 p. 594 (1989).
2. Taylor, C. and Kuayt, C. "Guidelines for Evaluating and Expressing Uncertainties" (National Institute for Standards and Technology, 1994).
3. Perry, C. Chemical Engineer's Handbook, fourth edition, p. 273 (McGraw Hill).
4. Leese, M.N. and Tite, M.S. (private correspondence) .