عوامل موثر بر کارایی رسوبات کلسیت میکروبی Factors Affecting Efficiency of Microbially Induced Calcite Precipitation

These findings are based on the test results using 0.25 and 0.5 M liquid media. During the termination of the experiment, samples that were treated with a higher chemical concentration and a smaller number of injections seemed to produce more hardened specimens in which it was harder to dissociate the calcite by acid, and they had a greater tendency to clog. This indicates that the input chemical concentration potentially has an effect on the precipitation pattern. The authors investigated this by conducting SEM imaging on samples that had undergone treatments of different chemical concentrations. Scanning electron microscopy was conducted on two series of samples; the first series was of samples injected with a 0.5 M solution every 12 h, whereas the second series was of samples injected with a 0.25 M solution every 6 h. The total mass injections were varied. Figs. 7 and 8 show the development of CaCO3 crystals in the two series. To ensure that images taken were representative of the entire specimen, for each specimen, more than one sample was taken and several images were taken for each of these samples. In Fig. 9 an image of a sample treated with a 1 M urea-CaCl2 solution is shown at a precipitation value of 70 kg∕m3. Despite the apparent exterior resemblance between samples treated with different chemical concentrations, it was clear from the images that the cementation distributions were different at the microscale. A low-concentration treatment (0.25 M) was found to generally result in a uniform distribution of calcite precipitation at different levels of cementation, as shown in Fig. 7. At low precipitation values [Figs. 7(a) and 7(b)], crystals were distributed all over the sand grains where no areas of concentrated precipitation could be found, because precipitation seemed to take place over the surface of the sand grains rather than accumulating over the crystals. This was even clearer as precipitation increased [Figs. 7(d) and 7(e)], where a larger number of crystals covered the sand grains uniformly and the crystals size did not seem to be any larger as precipitation increased. When an intermediate concentration treatment (0.5 M) was performed, a more random distribution of cementation was observed, as shown in Fig. 8. This nonuniformity was clearer for lower precipitation values [up to 67 kg∕m3, as shown in Figs. 8(a) and 8(b)]. As more calcite accumulated, the precipitation pattern was more random. At a given amount of precipitation, some samples showed a nonuniform pattern [Fig. 8(d)], whereas others showed relatively more even spreading of CaCO3 crystals all over the samples [Fig. 8(e)]. Figs. 9 and 10 highlight the differences obtained between the 0.25 and the 0.5 M treatments at the particle contact points. For the 0.25 M treatment (Fig. 9), the calcite crystals all had a similar size, were very well distributed spatially, and covered the contact area uniformly. For the 0.5 M treatment (Fig. 10), the crystals were not very well distributed spatially and had different sizes (where at some locations precipitation accumulated over the calcite crystals, resulting in larger crystals rather than being uniformly distributed over the surface of the sand grains). When a high-concentration treatment (1 M) was performed, the precipitation pattern was less uniform with larger crystal sizes, as shown in Fig. 11. These observed patterns indicate that—in samples tested in this study—the use of lower chemical concentrations over a larger number of injections resulted in a more homogeneous cementation. Two mechanisms are typically reported for precipitation in MICP (Stocks-Fischer et al. 1999; DeJong et al. 2006; RebataLanda 2007). The first mechanism is that bacterial cells act as nucleation sites for CaCO3 precipitation (Ca2þ is bound to the cell, which is active precipitation). The second mechanism is the urea hydrolysis, which raises the pH around the cells and produces the conditions favoring precipitation. In this study, bacterial distribution was not expected to have an effect on precipitation patterns for different concentrations because bacteria were applied in the same way to all samples. An explanation of the noted variation in precipitation distribution could be the distribution of the urea molecules with respect to the bacterial cells. In the case of higher urea concentrations, a higher and more localized rise in pH takes place around some bacterial cells as more urea molecules are available, and thus a thick layer of precipitation takes place (i. e., the second mechanism is much more dominant than the first). In contrast, in lower concentrations, smaller amounts of calcite precipitate at every injection and the higher number of injections could allow the urea molecules to be distributed over more bacterial cells, resulting in an overall better distribution of precipitation. Gandhi et al. (1995) reported that nucleation of new crystals would compete with the process of crystal growth for the available supersaturation, such that the formation of fine particles depends on circumstances in which nucleation of new particles prevails over the growth of those that exist. Taking this finding into account, Somani et al. (2006) reported that the higher the carbonate concentration is, the larger the average particle size of precipitates becomes. At lower carbonate concentration, on the other hand, the carbonate ions may be consumed mainly by the nucleation of calcite rather than the growth of calcite crystals. Such results suggest that, at higher supersaturation resulting from bacterial activity (i. e., when enough urea for hydrolysis is available), there could be a greater tendency for precipitation over existing crystals (i. e., growth of crystals) rather than nucleation in new sites. This is also further supported by Snoeyink and Jenkins (1980), who stated that the necessary degree of supersaturation for precipitation tends to be larger for homogeneous nucleation (i. e., growth of calcite crystals) than for heterogeneous nucleation (i. e., nucleation over sand grains). Such high supersaturation may also be a result of organics produced by bacteria [such as extracellular polymeric substances (EPS)] acting as crystallization inhibitors (Rodriguez-Navarro et al. 2007). The observations made by Somani et al. (2006) could also defend that at high calcium concentrations precipitation will start at a relatively low carbonate concentration, which could result in smaller crystals precipitating. However, this could be the case only initially until carbonate levels increase as a result of high urea concentration (despite ongoing precipitation). Thus, increasing supersaturation in the solution would result in precipitation accumulating over these initial small crystals and formation of larger crystal sizes, or possibly a mixture of different sizes, as shown in the 0.5 M input cases (Figs. 8 and 10). For a low-input CaCl2-urea concentration (0.25 M), the level of supersaturation would not be expected to increase to such a condition, which results in continuous heterogeneous precipitation over the sand grains. On the other hand, some studies such as Sondi and SalopekSondi (2005) showed that, in addition to general precipitation rules, the presence of organic macromolecules, such as enzymes, directly affect the precipitation process either through nucleation, crystal growth, or even morphology obtained. Furthermore, StocksFischer et al. (1999) reported that the availability of microbial cells and the extracellular urease enzyme produced around these cells have a significant impact on the rate of ammonia production and, consequently, precipitation. Van Paassen (2009) reported how reaction (hydrolysis) and diffusion rates could also have a large impact on crystal properties at different stages of precipitation and should be taken into account when discussing distribution of solutes (urea molecules) with respect to crystals, because they could affect the supply of carbonate ions toward the crystal surface. These different findings and observations may make it difficult to predict a precipitation pattern for different chemical concentrations in the presence of bacteria and organic substances. However, the precipitation patterns observed in this study suggest that the use of higher concentrations not only results in thicker calcite matrices, but possibly also gives a faster decline in the bacterial activity, because the urea becomes less available to the encapsulated microbial cells to hydrolyze. A more detailed discussion of these differences in precipitation pattern and crystal size, along with the quantification of these differences, could be the subject of future study. The CaCO3 precipitation pattern could have a significant impact on applications targeted for biocementation. It will influence the amount of contact between soil grains by forming bridges between these grains, and thus the way in which load is transferred between them and the strength and stiffness of the material (Harkes et al. 2010). It also influences the pore space shape/structure through local accumulation of crystals, which could have an effect in the flow properties of porous media. Further work is needed to examine the effect of precipitation pattern on the change in engineering properties, such as permeability and stiffness/strength.