Multiariate characterization, using the SIMCA method, of mortars from two frescoes in Chiaravalle Abbey

Abstract The percentage of acid-soluble components (ASC) and the aggregate granulometric distribution in mortars from two frescoes (Gothic and Flemish wall paintings) from Chiaravalle Abbey were determined. Principal component analysis (PCA) of these data provided a “scores” plot where Gothic plaster samples were clearly grouped, and separate from Flemish plaster samples. Classification into the above two classes was achieved by the residual standard deviation values of all samples fitted into disjoint principal components (PC) models according to the SIMCA method. Furthermore the PCA scores for the “Flemish” class pointed out differences between painted and internal plasters, indicating the use of layers with different composition and function, while those for the “Gothic” class showed no grouping indicating either the use of a single mortar or inhomogeneous layers. The above findings confirm and extend previous results on the classification of mortars into groups with similar architectural function, pointing out the generality of the SIMCA classification for both historical attribution and conservation studies.