Quantitative mineralogical analysis of hydraulic limes by X-ray diffraction G. Mertens a, ⁎ , P. Madau b , D. Durinck c , B. Blanpain c , J. Elsen a a Fysico-chemische Geologie, Katholieke Universiteit Leuven, Celestijnenlaan 200E, 3001 Heverlee, Belgium b Chimie Industrielle, Université Libre de Bruxelles, avenue F.D. Roosevelt 50, 1050 Bruxelles, Belgium c MTM, Katholieke Universiteit Leuven, Kasteelpark 44, 3001 Heverlee, Belgium Received 9 February 2007; accepted 13 August 2007 Abstract A combined selective dissolution/quantitative X-ray diffraction (QXRD) approach is proposed for the quantitative mineralogical phase analysis of hydraulic limes. The proposed methodology is validated by the analysis of two model mixtures. Afterwards two commercial hydraulic binders and one self-burned hydraulic quicklime were analysed. Chemical, thermal and microprobe analyses were performed to check the results. It is shown that the proposed selective dissolution/QXRD approach yields reliable quantitative mineralogical information for hydraulic limes in spite of their complex phase composition and the presence of amorphous material. © 2007 Elsevier Ltd. All rights reserved. Keywords: Amorphous material; Characterisation; Thermal analysis; X-ray diffraction; Hydraulic limes 1. Introduction Quantitative X-ray diffraction has been mainly directed towards applications concerning Portland cement and other common cement types [1–6]. However, there is a renewed interest in hydraulic limes and natural cements because of their use in the restoration of historic buildings [7]. The hydraulic properties of these limes cannot be accurately estimated from the chemical analyses alone [8]. Moreover it seems that the mineralogy cannot be simply predicted from the overall chemistry of the limestone and the associated raw materials [9]. Even more than for cement production, the mineralogy of the limes depends on the nature of the limestone, the burning conditions, the type of kiln used, etc. The amount and nature of the constituent phases of a material is of utmost importance when trying to understand its properties. For this reason and because of the need for quality control in modern production [1], the importance of X-ray diffraction for quantitative phase analysis of these hydraulic binders is apparent [10]. Moreover, minor phases present in some limes may indicate a unique source-rock or source-area. Such information is invaluable in provenance studies where the origin of the binder must be retraced. In the cement industry, the usual way of quantifying clinker compositions is by using the Bogue calculation [11] or by means of microscopic methods [12]. Since a fundamental assumption of the first method is the requirement of thermodynamic equilibrium conditions—not usually met dur- ing clinker production [5,13], the results are not always reliable. In the production of hydraulic limes, sintering/equilibrium conditions are only attained occasionally. The Bogue calcula- tion cannot therefore be applied. Microscopic methods are also used for the determination of clinker/cement composition. Both optical and electron microscopy, often combined with Image Analysis techniques, have been shown to give good results [14,15]. However, microscopy is probably more operator/equip- ment dependent [16]. Early critical results concerning the quantification of mineral phases showed that ‘neither the Bogue, nor the X-ray diffraction methods gave accurate analyses of the cement’ [17]. However, efforts have since been made to make X-ray diffraction a suitable technique for the quantification of cement phases [18,19]. Since cement mineralogy is to some extent similar to that of other calcareous hydraulic binders, the Rietveld method has been chosen for quantification in this study. However, most Available online at www.sciencedirect.com Cement and Concrete Research 37 (2007) 1524 – 1530 ⁎ Corresponding author. Tel.: +32 16 327586; fax: +32 16 327981. E-mail address: Gilles.Mertens@geo.kuleuven.be (G. Mertens). 0008-8846/$ - see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.cemconres.2007.08.002