Discussion of “Swelling characteristics of bentonite in artificial seawater”Appears in Canadian Geotechnical Journal, 46(2): 177–189.

Abstract

The authors have to be congratulated for presenting such interesting research. The results are important not only in terms of buffer material to fill disposal facilities, but also in terms of understanding the effect of rising seawater level and regression of seawater inland due to excessive fresh water withdrawal at coastal areas (Yukselen et al. 2008). Thus, the discussers are very interested in the results presented in this paper, and would like to offer the following comments on the authors’ interpretations. The experimental results reveal that the maximum swelling pressures of bentonites B–E are affected by artificial seawater to varying degrees. However, results are obscure for several reasons. For example, it is known that calciumtype bentonite has lower swelling potential than sodiumtype bentonite under the same testing conditions (Azam et al. 2000). However, the comparison of maximum swelling pressures shown in Fig. 6 indicates that the maximum swelling potential of calcium-type bentonite yields a larger swelling pressure than sodium-type bentonite. Furthermore, even though the other sodium-type bentonites (B, D, E) are affected by artificial seawater, bentonite A is not affected. These results from Fig. 6 are also in apparent contradiction with Figs. 9 and 12. For example, the maximum swelling strain difference between distilled water and artificial seawater is negligible for bentonite C. On the other hand, there are appreciable strain differences in bentonite A. We wonder if the presented maximum swelling pressures of the bentonites are merely a function of the water content of the samples. To this end, we plotted the maximum swelling pressure of bentonites as a function of their average water content (Fig. D1). We also plotted the maximum swelling pressure as a function of the montmorillonite content of bentonites (Fig. D2). As seen from Fig. D1, the swelling pressure of bentonites increases linearly with average water content, implying that the reported maximum swelling pressures are also a function of moisture content. In conclusion, the results show that the maximum swelling pressure of bentonites appears to increase concomitant with montmorillonite and water contents (Fig. D2). This should be expected as the listed average water content is a function of the montmorillonite content (Fig. D3). We believe that the reported swelling pressures of bentonites are a coupled function of their water and the montmorillonite content. Thus, from the presented data it is hard to reach the conclusion that the ‘‘influence of artificial seawater on the swelling characteristics of sodium-type and artificial sodium-type bentonites is low’’ and ‘‘the influence of artificial seawater on swelling characteristics of calcium-type bentonite is more slight than sodium-type and artificial sodium-type bentonites.’’ Further evidence of the above statements is the variation of maximum swelling pressures with liquid limit. As seen from Fig. D4, the swelling pressure decreases with an increase in the liquid limit of bentonites. This appears to contradict previous research that documents swelling pressures that increase (not decrease) with an increase in liquid limit of soils (Vijayvergiya and Ghazzaly 1973; Chen 1988; Issa 1997). Additional evidence of the above stated statement comes from the authors’ own data. For example, if we take the maximum swelling strain differences between distilled water and artificial seawater at 1.4 Mg/m3 of initial density under 10 kPa vertical pressure and plot them as a function of the liquid limit of bentonites, the maximum swelling differences increase with an increase in liquid limit (Fig. D5). We chose the initial dry density of 1.4 Mg/m3 as this density is the only initial density in all tested samples. We validate our observations by plotting the observed maximum swelling strains presented in Fig. 9 as a function of the liquid limit of samples at 1.65 Mg/m3. As seen from Fig. D6, the observed maximum swelling strain is a function of liquid limit. We would appreciate the authors’ comments on our interpretation of the reported results.