engleski

Multi-analytical profiling of archaeometallurgical samples – A new insight into ancient materials

Analytical considerations on iron artefacts from ancient bloomery iron production are generally directed to analysis of main and trace elements content in raw materials and wastes. Elemental signature of artefacts usually refers to the carefully selected sample of final iron-making products, as well as to randomly collected samples of slag and pit deposit materials on excavation site. Archaeometric studies reveal the compositional relations between ore, smelting slags and bloom iron products, which is the utmost information for the provenance. Reconstructions of ancient iron making process have confirmed that different types of slag material are formed during smelting and smiting process such as tap slags, bloom slags and ceramic-rich slags.[1, 2] However, a significant chemical variability derived from furnace, fuel and fluxes in iron making process cause the provenancing attempts rather ambiguous. Another severe problem arises from representativeness and proper recognition of samples from iron making sites when they were collected from large clusters or dissipated piles of slag. A multi-sample and/or multi-method analytical approach has showed to be capable for collection of enough compositional data, which are adequate for characterization of iron-making sites.[3] Chemometric methods with univariate and multivariate statistical treatment of large set of spectrometric data are often used in experimental design and analytical method optimisation. In archaeometric studies, a multivariate statistics using principal component analysis (PCA) and hierarchical clustering (HC) are the most often applied modes of exploratory data analysis. It allows that the distinctions among objects with quite similar characteristics at macroscopically scale became visible. In this research study, the chemical composition of archaeological samples from bloomery iron-making site of NW Croatia was determined using multi-method approach.[4] Determination of elements content by XRF, ICP- AES and ICP-MS methods, as well as morphology study by SEM-EDS method is presented. The discrimination and classification of large group of archaeometallurgical objects was successfully achieved by exploiting of chemical signature as a ground level in PCA and cluster analysis. The main goal of this work was the confirmation of applicability and efficiency of combined multi- analytical and statistical tools in proper recognition of inhomogeneous bloom slags and mutually similar artefacts.

archaeometallurgy ; iron slags ; elemental profiling ; spectrometry methods

nije evidentirano