Limits...
Direct Analysis in Real Time Mass Spectrometry for the Nondestructive Investigation of Conservation Treatments of Cultural Heritage

View Article: PubMed Central - PubMed

ABSTRACT

Today the long-term conservation of cultural heritage is a big challenge: often the artworks were subjected to unknown interventions, which eventually were found to be harmful. The noninvasive investigation of the conservation treatments to which they were subjected to is a crucial step in order to undertake the best conservation strategies. We describe here the preliminary results on a quick and direct method for the nondestructive identification of the various interventions of parchment by means of direct analysis in real time (DART) ionization and high-resolution time-of-flight mass spectrometry and chemometrics. The method has been developed for the noninvasive analysis of the Dead Sea Scrolls, one of the most important archaeological discoveries of the 20th century. In this study castor oil and glycerol parchment treatments, prepared on new parchment specimens, were investigated in order to evaluate two different types of operations. The method was able to identify both treatments. In order to investigate the effect of the ion source temperature on the mass spectra, the DART-MS analysis was also carried out at several temperatures. Due to the high sensitivity, simplicity, and no sample preparation requirement, the proposed analytical methodology could help conservators in the challenging analysis of unknown treatments in cultural heritage.

No MeSH data available.


Related in: MedlinePlus

Results of canonical analysis: canonical scores of Root 1 and Root 2 (a). The samples are separated along the two roots in the three classes: untreated parchment sample (red), the parchment sample treated with castor oil (blue), and the parchment sample treated with glycerol (green). Circles represent samples from the training set while crosses represent samples from the test set. (b) and (c) represent the weights of the original variables on Root 1 and Root 2, respectively: original variables are reported on the x-axis and the weights on the y-axis. Only the most significant weights are reported (identified by normal probability plots).
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC5121580&req=5

fig7: Results of canonical analysis: canonical scores of Root 1 and Root 2 (a). The samples are separated along the two roots in the three classes: untreated parchment sample (red), the parchment sample treated with castor oil (blue), and the parchment sample treated with glycerol (green). Circles represent samples from the training set while crosses represent samples from the test set. (b) and (c) represent the weights of the original variables on Root 1 and Root 2, respectively: original variables are reported on the x-axis and the weights on the y-axis. Only the most significant weights are reported (identified by normal probability plots).

Mentions: The overall best model provided very good results in calibration and prediction: all samples from the training and test sets were correctly assigned (NER and specificity of 100% for all classes) and no overlap was detected (selectivity of 100% for all classes). The results are slightly worse in cross-validation, with a NER% equal to 98.68%. The model includes 6 PCs: PC1, PC2, PC9, PC10, PC12, and PC15. Table 2 reports the coefficients of each PC included in the LDA classification model on the first two canonical roots calculated by canonical analysis. This result is graphically represented in Figure 7(a), reporting the samples along the two discriminant roots calculated by canonical analysis: circles correspond to samples belonging to the training set and crosses represent samples from the test set, while color indicates the class. The first root separates glycerol treated samples (at negative values) from the other two classes (at positive values), while the second root separates untreated samples (at positive values) from castor oil treated samples (at negative values). The figure clearly shows the perfect classification of the samples from both training and test sets, since the three groups of samples are well separated.


Direct Analysis in Real Time Mass Spectrometry for the Nondestructive Investigation of Conservation Treatments of Cultural Heritage
Results of canonical analysis: canonical scores of Root 1 and Root 2 (a). The samples are separated along the two roots in the three classes: untreated parchment sample (red), the parchment sample treated with castor oil (blue), and the parchment sample treated with glycerol (green). Circles represent samples from the training set while crosses represent samples from the test set. (b) and (c) represent the weights of the original variables on Root 1 and Root 2, respectively: original variables are reported on the x-axis and the weights on the y-axis. Only the most significant weights are reported (identified by normal probability plots).
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC5121580&req=5

fig7: Results of canonical analysis: canonical scores of Root 1 and Root 2 (a). The samples are separated along the two roots in the three classes: untreated parchment sample (red), the parchment sample treated with castor oil (blue), and the parchment sample treated with glycerol (green). Circles represent samples from the training set while crosses represent samples from the test set. (b) and (c) represent the weights of the original variables on Root 1 and Root 2, respectively: original variables are reported on the x-axis and the weights on the y-axis. Only the most significant weights are reported (identified by normal probability plots).
Mentions: The overall best model provided very good results in calibration and prediction: all samples from the training and test sets were correctly assigned (NER and specificity of 100% for all classes) and no overlap was detected (selectivity of 100% for all classes). The results are slightly worse in cross-validation, with a NER% equal to 98.68%. The model includes 6 PCs: PC1, PC2, PC9, PC10, PC12, and PC15. Table 2 reports the coefficients of each PC included in the LDA classification model on the first two canonical roots calculated by canonical analysis. This result is graphically represented in Figure 7(a), reporting the samples along the two discriminant roots calculated by canonical analysis: circles correspond to samples belonging to the training set and crosses represent samples from the test set, while color indicates the class. The first root separates glycerol treated samples (at negative values) from the other two classes (at positive values), while the second root separates untreated samples (at positive values) from castor oil treated samples (at negative values). The figure clearly shows the perfect classification of the samples from both training and test sets, since the three groups of samples are well separated.

View Article: PubMed Central - PubMed

ABSTRACT

Today the long-term conservation of cultural heritage is a big challenge: often the artworks were subjected to unknown interventions, which eventually were found to be harmful. The noninvasive investigation of the conservation treatments to which they were subjected to is a crucial step in order to undertake the best conservation strategies. We describe here the preliminary results on a quick and direct method for the nondestructive identification of the various interventions of parchment by means of direct analysis in real time (DART) ionization and high-resolution time-of-flight mass spectrometry and chemometrics. The method has been developed for the noninvasive analysis of the Dead Sea Scrolls, one of the most important archaeological discoveries of the 20th century. In this study castor oil and glycerol parchment treatments, prepared on new parchment specimens, were investigated in order to evaluate two different types of operations. The method was able to identify both treatments. In order to investigate the effect of the ion source temperature on the mass spectra, the DART-MS analysis was also carried out at several temperatures. Due to the high sensitivity, simplicity, and no sample preparation requirement, the proposed analytical methodology could help conservators in the challenging analysis of unknown treatments in cultural heritage.

No MeSH data available.


Related in: MedlinePlus