Comparison between XRF, TXRF, and PXRF analyses for provenance classification of archaeological bricks. (24th April 2013)

• Record Type: Journal Article
• Authors: Bonizzoni, Letizia; Galli, Anna; Gondola, Marco; Martini, Marco; Streli, Christina; Markowicz, Andrzej
• Is Part Of: X-ray spectrometry. Volume 42:Number 4(2013:Jul./Aug.)
• Page Start: 262
• Page End: 267
• Publication Date: 2013-04-24
• Subjects: X-ray spectroscopy -- Periodicals; 543.62
• DOI: 10.1002/xrs.2465
• Languages: English
• ISSNs: 0049-8246

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <p>Ceramics provenance studies based on minor and trace elements are widely used to help in gathering knowledge of local furnace presence and commercial trades. Most of the times, destructive analyses are allowed because of the high amount of material, but sometimes (integer vessels, architectural decorations, etc.), a nondestructive examination is required. In both cases, X‐ray fluorescence (XRF) analysis can be a useful tool.</p> <p>In this work, we compare the classification results obtained on a set of bricks from the medieval Chartusian monastery <italic>Certosa di Pavia</italic> in the Po valley, northern Italy, exploiting XRF, portable XRF, and total reflection XRF (TXRF) spectrometers.</p> <p>The XRF nondestructive analyses were performed both <italic>in situ</italic> before sampling and in the laboratory without preparing the shards. Grounded samples were instead used to obtain semi‐quantitative TXRF analysis without dissolving the samples but using a suspension in high‐purified water. The project aimed at verifying if a correct classification was possible even if a quantitative determination of minor and trace elements is abandoned, allowing a faster response when a large amount of data is concerned. For this reason, weight concentration was determined from laboratory XRF measurements only, whereas semi‐quantitative results and entire spectra were considered for TXRF and the <italic>in situ</italic> campaign, respectively. Data were then submitted to different statistical data treatments to obtain provenance classification. Copyright © 2013 John Wiley & Sons, Ltd.</p> </abstract>