H. Ben Boubaker
LabTEN – Water Technologies Research Center, Technopole of Borj-Cedria, Tunisia
Abstract – This study focuses on calcium sulfate (gypsum) and calcium carbonate (CaCO3) simple and mixed precipitations. These forms of scaling are still an issue in several industrial applications such as cooling towers and water desalination, either by thermal-based or membrane-based processes. CaCO3 precipitation has been studied by the degassing method while the gypsum simple and mixed precipitations have been studied by the double decomposition method, and by following 4 parameters simultaneously, namely: pH and total alkalinity serve to detect the CaCO3 precipitation, calcium ion concentration allows following the gypsum germination, and the quartz crystal microbalance (QCM) response gives information about the salts seeding, especially gypsum. A kinetic study of CaCO3 and CaSO4∙2H2O mixed precipitation resulted in a different form to that of the simple precipitation. Indeed, even though the pH remains the key parameter to detect CaCO3 precipitation either under simple or mixed precipitation forms, as the QCM response was found not to be a good parameter to detect the CaCO3 germination in a mixed system, especially when it precipitates after calcium sulfate, the QCM response remains a specific parameter to gypsum crystals because even after the appearance of the first CaCO3 crystals, gypsum germination continues with faster kinetics. Gypsum germination time in co-precipitation is higher than that corresponding to the simple precipitation (CaSO4-H2O system) because of the competition between the different co-precipitating ions.
Key words: Scaling / Simple and mixed precipitations / Germination / Degassing / Double decomposition.
Scaling, a frequent phenomenon occurring in water distribution pipes and some industrial applications such as cooling towers and water desalination, either by thermal-based or membrane-based processes (Waly et al. 2012; Ghaffour et al. 2013), is characterized by the appearance of an adhering crystalline deposit constituted essentially by calcium carbonate (CaCO3) and gypsum, amongst other scaling forms, on the pipes, distillers’ walls or on membranes surfaces (Bernard and Pascal 1960; Powel et al. 1964; Amathieu 1985; Ghorbel et al. 1991; Rola 1994; Gal et al. 1996; Sheikholeslami 2000; Sheikholeslami 2003; Arras et al. 2009; Rahmawati et al. 2012).
Crystallization, mainly kinetic and thermodynamic aspects, has been widely studied for many years (Sohnel and Mullin 1988; Mullin 1993; Elfil and Roques 2001; Alimi and Elfil 2003; Elfil and Roques 2004; Elfil 2006; Elfil et al. 2007) and exhaustive information is available in literature for pure salts, mainly CaCO3 and calcium sulfate dehydrate (CaSO4·2H2O) which are the major scaling contributors (Marshall and Slusher 1966; Elfil and Roques 2001). In fact, recent studies have been interested to the possibility of controlling or reducing this problem by several methods, mainly chemical treatment (Xyla et al. 1992; Gal et al. 1996; Barrett and Parsons 1998; Elfil and Nawel 2004; Szczes 2008; Wu et al. 2010). There are also a number of alternative non-chemical treatment options available for scaling control. Amongst these is the use of magnetic, electronic and electrolytic treatment devices (MacAdam and Parsons 2004; Al Nasser et al. 2011). Each of these scale controlling methods has its advantages and a number of factors need to be considered before choosing the right option. However, only few studies have been interested in mixed precipitation phenomenon, due mainly to its complexity (Sheikholeslami and Sudmalis 2000; Sheikholeslami 2003; Zarga et al. 2013). Indeed, the study of the interactive effect of salts in mixed precipitation with or without common ions is not widely discussed. Solubility effect, ionic strength, crystal structure and inhibitory effect are still not well defined in the co-precipitation process.
Studies that have been interested on CaCO3 and gypsum mixed precipitation (MacAdam and Parsons 2004) using temperature-controlled batch tests showed that the co-precipitation germination time and kinetics follow that of pure CaCO3, while the thermodynamic calcium concentrations on co-precipitation follow that of pure gypsum. In addition, the structure of precipitates, compared with that observed by simple precipitation is influenced by the coexistence of the two salts.
Other studies have shown that in the case of mixed precipitation, three possibilities may occur: 1) the first salt can act as a foreign body and reduce the energetic barrier of the second precipitating salt, therefore, the germination time of the second compound is reduced (Sohnel and Mullin 1988; Zarga et al. 2013), 2) precipitates may have morphologies that are different from those obtained with the pure body formation reactions, and 3) the presence of a co-precipitation salt can affect the thermodynamic of the solution and subsequently influence the germination kinetic of other salts (Lu and Czanderna 1984; Sohnel and Mullin 1988).
This paper looks at qualitative analysis of gypsum and CaCO3 simple and mixed precipitations from Kinetic and structural point of view. It also aims to find the appropriate method to follow the germination parameters of the two precipitating salts and to compare the simple and mixed precipitation kinetics.
Materials and methods
Double decomposition method and experimental unit
Solutions were prepared with double decomposition method using two solutions of sodium sulfate (Na2SO4), sodium bicarbonate (NaHCO3) and calcium chloride (CaCl2). This method is mainly used to reach high supersaturations. In contrast to the dynamic method for which supersaturation is gradually decreased up to precipitation, the double decomposition method allows only for a specified supersaturation to observe or not the precipitation (Elfil and Roques 2001).
Solutions may contain calcium, bicarbonate, sulfate, sodium and chloride ions in significant quantities. Na + and Cl- ions are involved in the precipitation process by means of ionic strength which is adjusted by the addition of sodium chloride (Sohnel and Mullin 1988; Elfil and Roques 2001; Wu et al. 2010). A schematic of the experimental setup used in this investigation is presented in Figure 1.