Maastritchian Limestones of Feriana Mountain used in White Cement Production (Central West Tunisia) Tahar Aloui, w Anouar Ounis, and Fredj Chaabani Laboratory of Mineral Resources and Environment, Department of Geology, Faculty of Mathematical, Physical and Natural Sciences, University Campus, 1060 Tunis, Tunisia The Campanian to early Maastrichtian period is marked by important deposits of limestones in most parts of Tunisia and Algeria. These limestones form the so-called ‘‘Abiod forma- tion’’. The present paper focuses on the ability of limestones of Feriana Mountain (Jebel Feriana), Central West part of Tunisia, to produce high-quality white cement clinker. From detailed experimental and analytical techniques’ data, the lime- stones of Abiod formation are particularly characterized by a high purity degree and whiteness index. They have a bright white color with a slight visible tendency toward yellow-red. Based on Bogue calculation, X-ray diffraction, and microscopical point- count methods, these particular limestones can conveniently form a white Portland cement when mixed with B12.5% of Neogene sands of Maamoura syncline and 7–10% by weight of common Mediterranean kaolins. I. Introduction T HE Feriana Mountain is situated in the Central Tunisian Atlas, close to the Algerian border (Fig. 1). Its outcrops emerged during the Santonian and Coniacian ages, which are (from oldest to youngest): (1) The shale of Aleg formation (Santonian to Coniacian), which is poorly exposed along the Choubet Ali and Choubet Ettaref; (2) the chalk of Abiod formation (Campanian to early Maastrichtian), which has typically 150–200 m thickness; it is subdivided into three members: lower and upper layers of chalks separated by a middle shale, which is missing at Feriana Mountain 1–7 ; and (3) the transgressive siliciclastic series, which are present mainly in the west–east direction, in Maamoura and Bouhaya synclines (Fig. 1). They are composed of coarse whitish quartz sand, clay, gray to brown siltstone, and gypsum. According to Robinson and Wiman, 8 these series include the Messiouta formation (Aquitanian), which is underlain by the Abiod formation, the Mahmoud continental formation (Burdigalian-Langhian), the Beglia formation (Serravallian-Tortonian), and the Segui formation (Tortonian-Quaternary). Over these siliciclastic sediments, there are Quaternary alluvium and recent soils associated with coarse conglomerates, encrustations, yellow- colored sands, and brown clays. 2,5,9 From the beginning of the 1980s, many investigations and valorization of these geological series were undertaken by Trabelssi, 10 Aloui, 5 and Aloui and Chaabani. 6,7,9 Only Maast- richtian limestones are valuable because of their high-purity degree and whiteness index and can be used in several industrial products, such as paper, white cement, tiling, dimension stone, glass manufacture, painting materials, and pharmaceutical products. The present work uses a multidisciplinary approach to evaluate the ability of Maastrichtian limestones of Feriana Mountain to supply a second line of production of white cement, implanted at B4 km west from Feriana town (Fig. 1). The alumina should be provided by a clay extremely poor in coloring oxides, particularly, Fe 2 O 3 , Cr 2 O 3 , Mn 2 O 3 , and TiO 2 . 10 The literature on clay rocks from various scientific surveys highlights the occurrence of kaolin deposits in Aı¨n Khmouda village located at about 40 km North East of Feriana Mountain. Unfortunately, this material cannot be used to produce larger quantities and this is the why several mineral industries need to import kaolin from European providers. The deficiency in silica is compensated by controlled addition of Neogene sands from Maamoura syncline (Fig. 1). II. Experimental Procedure Seven cores (SC01–SC07) were drilled to a depth of approxi- mately 70 m at Feriana Mountain (Jebel Feriana) to valorize the Maastrichtian limestones and to evaluate the exploitable reserves (Fig. 2). The step of the sampling through the whole lithology depends on homogeneity and thickness. If the sample thickness is o2 m, one sample is taken from its middle. If the sample thickness exceeds this limit, one sample is taken every 2 m. The contents of alumina oxide (Al 2 O 3 ), calcium oxide (CaO), chrome oxide (Cr 2 O 3 ), phosphorus pentoxide (P 2 O 5 ), ferric ox- ide (Fe 2 O 3 ), magnesium oxide (MgO), dimanganese trioxide (Mn 2 O 3 ), titan dioxide (TiO 2 ), and silicon dioxide (SiO 2 ) from the raw material were determined by X-ray fluorescence (XRF) analysis on a Philips PW 1606 spectrometer (France) with automated sample feed, reverse potential end window with Rhodium anode and operated at 50 kV, 40 mA. Beads were prepared by fusing mixtures of 0.7 g of powdered sample with 6 g of lithium tetraborate (LiB 4 O 7 ). 11–13 This step of pre-preparation of the sample leads to a more homogeneous material and, consequently, has the advantage of obtaining a more accurate XRF analysis. The contents of potassium oxide (K 2 O) and sodium oxide (Na 2 O) were obtained by atomic absorption spectra. The determination of the content of sulfur trioxide (SO 3 ) was carried out by a gravimetric technique. The methods used in the analysis of composition of phase include the theoretical approach via Bogue calculation, the microscopical point-count procedure, and X-ray diffraction (XRD). XRD analyses were carried out using a Philips PANalytical X’Pert PRO X-ray diffractometer with an automatic divergence slit, a spinner, an X’celator, and CuKa radiation at a scan speed of 0.011 2y/s. The acceleration power applied was 40 kV, with a current of 40 mA. The difference between the experimental and the theoretical peaks of Si (111) was within 2y 5 0.0021 and the diffraction data were evaluated by X’pert HightScoreplus s program. The percentages of the different mineral phases were calculated using data obtained by XRD and Rietveld refine- ment. The global parameters of refinement control were based on the scale factor and unit cell of each phase (alite, belite, aluminate, ferrite, free lime, and periclase). P. Brown—contributing editor w Author to whom correspondence should be addressed. e-mail: tahar.aloui@edunet.tn Manuscript No. 23900. Received October 25, 2007; approved August 18, 2008. J ournal J. Am. Ceram. Soc., 91 [11] 3704–3713 (2008) DOI: 10.1111/j.1551-2916.2008.02725.x r 2008 The American Ceramic Society 3704