Petrochemical Identification and Insights on Chronological Employment of the Volcanic Aggregates Used in Ancient Roman Mortars

Abstract

Through the assistance of trace element and petrographic analyses on 14 samples of mortar aggregates from Roman monuments, including the Porticus Aemilia, the Temple of Concordia, the Temple of the Dioscuri, Temple B and other structures of the Area Sacra di Largo Argentina, and the Villa di Livia, we establish the source area and we investigate the chronological employment of the volcanic materials used in ancient Rome's masonry. In contrast to previous inferences, the petrochemical data presented here show that systematic exploitation of the local ‘Pozzolane Rosse’ pyroclastic deposit has occurred since the early development of concrete masonry, at the beginning of the second century bc, through the early Imperial age. Subsequently, exploitation was extended to the overlying Pozzolane Nere and Pozzolanelle deposits. Only during the early phase of development of the concrete masonry in Rome, volcaniclastic sediments outcropping near the construction sites were mixed with the sieved remains of the tuff employed as the coarse aggregate, to produce the fine aggregate. The results of the study on the investigated monuments suggest the possibility of establishing the chronological identification of three different types of mortars, as a function of the composition of the volcanic material employed in the fine aggregate, which, when implemented by future studies, may contribute to the dating of monuments and archaeological structures.

Provenance materials for Vitruvius' harenae fossiciae and pulvis puteolanis : Geochemical signature and historical–archaeological implications

“…The selective use of the local Pozzolane Rosse pyroclastic-flow deposit in the mortars and concretes of several monuments of ancient Rome spanning in age the II century BC through the IIIrd century CE has been documented by geochemical data presented in Marra et al (2015). Here we provide new petrographic evidence for the mortars of Temple of Castor and Pollux (Coarelli, 2008) and for Temple B of Largo Argentina (Caprioli, 2010) (II c. BC).…”

“…Subsequently, the clastic components employed as the aggregate were separated from the lime by bathing the samples in cloridric acid (HCl) for 4/5 days. As shown in Marra et al (2015) this treatment does not affect the Zr/Nb ratio of the samples, whereas it is able to produce more or less significant Y depletion. In contrast, a more stable behavior is observed for the TiO 2 content.…”

“…The aggregates of two mortar samples of the early Augustan and of the IInd-IIIrd constructive phases analyzed in Marra et al (2015) yielded geochemical compositions attesting their correlation with the deposit of the Grottarossa Pyroclastic Sequence (Karner et al, 2001), cropping out and exploited near the Villa di Livia in Prima Porta. In contrast, all the III-IV century mortars are characterized by the presence of large (2-3 cm) black scoriae (Fig.…”

“…For this reason results of the analyses, especially when very small amount of sampled material is concerned, must be discussed with care. To overcome this limitation, we have integrated the Zr/Y vs Nb/Y diagram with another diagram, where Y is replaced by TiO 2 for those cases in which we suspected the occurrence of selective depletion of the former element due to acid attack (Marra et al, 2015).…”

“…In contrast to the Augustan age mortars of the earliest Villa di Livia structures, in which the Roman builders employed the pyroclasticflow deposit of the Grottorossa Pyroclastic Sequence exploited locally (Marra et al, 2015), a pervasive use of the loose, pozzolanaceous facies of the Tufo Rosso a Scorie Nere Sabatino in the later mortars of the III and IV century AD is confirmed by the Zr/TiO 2 vs Nb/TiO 2 composition of the black scoria sample (LIV-B) analyzed in this work (Fig. 6).…”