Behavior of blended cement mortars exposed to sulfate solutions cycling in relative humidity M. Nehdi * , M. Hayek Department of Civil and Environmental Engineering, University of Western Ontario, 1151 Richmond Street, London, Ontario, Canada N6A 5B9 Received 12 June 2003; accepted 20 May 2004 Abstract In recent years, several cases of damage to concrete structures due to sulfate exposure have occurred essentially in the above ground parts of structures. Such distress, often characterized by white efflorescence and surface scaling, is driven by salt crystallization in pores and/or repeated reconversions of certain sulfates between their anhydrous and hydrated forms under cycling temperature and relative humidity (RH). However, the effect of the water/cementitious materials ratio (w/cm), pozzolanic additions, and other parameters on the durability of cement- based materials under such exposure conditions is still misunderstood. In this study, 12 cement mortars having different w/cm (0.30, 0.45, and 0.60) and made with ordinary portland cement (OPC) or OPC incorporating 8% silica fume, 25% class F fly ash, or 25% blast furnace slag were made. Standard bars from each of these mortars were submerged in both 10% magnesium sulfate (MgSO 4 ) and 10% sodium sulfate (Na 2 SO 4 ) solutions; their expansion and surface degradation was monitored for up to 9 months. In addition, cylinders made from these 12 mortars were partially submerged in 50-mm-deep 10% MgSO 4 and 10% Na 2 SO 4 solutions. Half of the cylinders were maintained under constant temperature and RH, whereas the others were subjected to cycling RH. The effect of the w/cm and mineral additions on the classic chemical sulfate attack and development of efflorescence was investigated, and the results are discussed in this article. D 2004 Elsevier Ltd. All rights reserved. Keywords: Durability; Sulfate; Expansion; Salt hydration; Efflorescence 1. Introduction Research on sulfate attack of concrete has flourished in recent years owing to related lawsuits in the United States and Canada. Consequently, the traditional view of sulfate attack has been challenged. For instance, it was suggested that the current testing procedures are not indicative of field condi- tions and that there is a need for the development of failure criteria and parameters that can enable the predicting of sulfate attack in field structures [1]. A critical analysis of sulfate attack mechanisms, related test techniques, field monitoring methods, and development of pertinent modeling criteria was recently presented [2]. The controversies and confusion caused by existing literature lead Mehta [3] to present a critical examination of the state-of-the-art research information on this subject. Skalny et al. [4] recently pub- lished a comprehensive review on sulfate attack in concrete. There has been growing emphasis of recent research in this area on the distinction between chemical and physical types of sulfate attack. In the so-called chemical attack, the sulfate ions (SO 4 2  ) enter in reactions with cement hydra- tion products leading primarily to the formation of ettringite and gypsum. However, ongoing controversy exists as to what is the effect of each of these products on the deteri- oration of concrete. Researchers have often focused on the effect of the SO 4 2  without equal attention to the cations associated with them (Ca, Na, Mg, and Fe). It is now known that low tricalcium aluminate (C 3 A) cements, which are considered resistant to sodium sulfate (Na 2 SO 4 ) attack, can be detrimental in exposure conditions involving magnesium sulfate (MgSO 4 ) or sulfuric acid [2], because the prevailing low pH conditions are conducive to direct attack on the calcium silicate hydrates (CSH). Formation of ettringite and subsequent expansion and cracking have often been invoked as the driving mecha- nisms of chemical sulfate attack [5–7]. At high SO 4 2  concentrations and/or with gradual decrease of pH, ettringite is not stable and decomposes to form gypsum. Yet, the 0008-8846/$ – see front matter D 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.cemconres.2004.05.032 * Corresponding author. Tel.: +1-519-661-2111x88308; fax: +1-519-661- 3779. E-mail address: mnehdi@eng.uwo.ca (M. Nehdi). Cement and Concrete Research 35 (2005) 731 – 742