Improvements in engineering properties of soils through microbial-induced calcite precipitation.pdf

KSCE Journal of Civil Engineering (2013) 17(4):718-728 DOI 10.1007/s12205-013-0149-8

Geotechnical Engineering

www.springer.com/12205

Improvements in Engineering Properties of Soils through Microbial-Induced Calcite Precipitation Ng Wei Soon*, Lee Min Lee**, Tan Chew Khun***, and Hii Siew Ling**** Received March 22, 2012/Revised May 25, 2012/Accepted July 12, 2012

···································································································································································································································

Abstract Microbial-Induced Calcite Precipitation (MICP) has recently emerged as a sustainable technique for soil improvement. This paper aims to study the effectiveness of MICP in improving the shear strength and reducing the hydraulic conductivity of soils. A species of Bacillus group, B. megaterium was used to trigger the calcite precipitation. The experimental variables included soil types (tropical residual soil and sand), soil densities (85%, 90%, and 95% of their respective maximum densities), and treatment conditions (untreated, treated with cementation reagents only, treated with B. megaterium only, and treated with B. megaterium and cementation reagents). The results showed that MICP could effectively improve shear strength and reduce hydraulic conductivity for both residual soil and sand. The improvements, however, varied with soil densities, soil types, and treatment conditions. With MICP treatment, the improvement ratios in shear strength of the residual soil specimens were significantly higher (1.41-2.64) than those of the sand specimens (1.14-1.25). On the contrary, the sand specimens resulted in greater hydraulic conductivity reduction ratios (0.09-0.15) than those of the residual soil specimens (0.26-0.45). These observations can be explained by the particle-particle contacts per unit volume and pore spaces in the soil specimens. Both soil specimens when treated with cementation reagents only exhibited slight alterations in the shear strength (ranging from 1.06-1.33) and hydraulic conductivity (ranging from 0.69-0.95). The results implied that natural calcite forming microorganisms only exist for insignificant amount. The amount of calcite precipitated in the treated residual soil specimens ranged from 1.080% to 1.889%. The increments of calcite content in the treated sand specimens were comparatively higher, ranging from 2.661% to 6.102%. The results from Scanning Electron Microscope (SEM) analysis confirmed the experimental findings. Keywords: microbial-induced calcite precipitation, soil improvement, shear strength, hydraulic conductivity, B. megaterium ···································································································································································································································

1. Introduction Nowadays, construction on problematic soils is inevitable owing to the growing scarcity of land worldwide. Problematic soils are commonly characterized by low strength and high compressibility (Ho and Chan, 2011; Huat, 2006; Kazemian et al., 2011). In tropical regions like Malaysia, soils are subject to further softening due to infiltration of intense and prolonged downpours. Consequently, development on the problematic soils is highly susceptible to severe geohazards including excessive settlement of embankment or foundation, debris flow, and catastrophic landslide. Studies on the soil improvement techniques can be found in abundance. The important features of soil improvement include: improving the shear strength of soil, reducing the potential for total and differential settlement, reducing the time during which

the settlement takes place, reducing the potential for liquefaction in saturated fine sand or hydraulic fills, reducing the hydraulic conductivity of soil, removing or excluding water from the soil etc (Kazemian and Huat, 2009; Krebs and Walker, 1971; Leonards, 1962). Conventionally, the norm is to replace low strength soil deposits with engineering fill. Presently, the use of chemical grouting is becoming increasingly popular owing to its economical benefit. Chemical grouting can be achieved with a variety of additives including Portland cement, lime, asphalt, sodium silicate, acrylate, lignin, urethane, and resins. While many of these additives have proven successful (Anagnostopoulos and Hadjispyrou, 2004; Basha et al., 2005; Karol, 2003; Peethamparan et al., 2009; Xanthakos et al., 1994), the additives often modify the pH of soils, and may contaminate the soils and groundwater (Dejong et al., 2006; Karol, 2003). In recent years, with increasing awareness of environmental

*Postgraduate Candidate, Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Kuala Lumpur 53300, Malaysia (E-mail: [email protected]) **Assistant Professor, Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Kuala Lumpur 53300, Malaysia (Corresponding Author, E-mail: [email protected]) ***Assistant Professor, Faculty of Engineering and Green Technology, Universiti Tunku Abdul Rahman (Perak Campus) 31900, Malaysia (E-mail: [email protected]) ****Associate Professor, Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, Kuala Lumpur 53300, Malaysia (E-mail: [email protected]) − 718 −

Ng Wei Soon, Lee Min Lee, Tan Chew Khun, and Hii Siew Ling

issues, there has been a remarkable shift toward “green” and sustainable technologies. As all chemical grouts except sodium silicate are toxic and hazardous, there are expressed concerns over their use for soil improvement (Dejong et al., 2010). A new soil improvement technique that utilizes a biological process, which is termed technically as MICP, has emerged recently. MICP has been enabled through interdisciplinary researches at the confluence of microbiology, geochemistry, and geotechnical engineering, to find natural treatments for soil improvement (Dejong et al., 2010). MICP is a biological process occurs in nature. It is intensified by introducing a large population of urease-producing microorganisms and cementation reagents into the soil matrix, whereby a cement compound is generated to improve the engineering properties of the soil. The environmental friendly characteristics of the ureaseproducing microorganisms will cause very little, if not no impairment to the soil, human health, and environment. Despite of the fact that the MICP soil improvement technique is still a relatively young technology, many studies pertaining to the topic have been reported including Baveye et al. (1998), Castainer et al. (1999), Ehrlich (1999), Mitchell and Santamarina (2005), Lian et al. (2006), Ivanov and Chu (2008), Dejong et al. (2010), Okwadha and Li (2010), Harkes et al. (2010) and, Lu et al. (2010). Dejong et al. (2006) performed laboratory experiments to investigate the MICP on loose, collapsible sand specimens using B. pasteurii. They found that MICP-treated specimens exhibited a noncollapsing strain softening shear behavior, with higher initial shear stiffness and ultimate shear capacity than untreated loose specimens. Baskar et al. (2006) isolated the calcite-forming microorganisms, B. thuringiensis and B. pumilis from stalactites, sampled from three caves in Sahastradhara, Dehradun, India. They concluded that microbial activity and temperature are the key factors promoting calcite precipitation. Harkes et al. (2010) researched on the methodology to distribute and fix B. pasteurii homogeneously in sand bed. They found that the homogeneous distribution can be achieved by performing twophase injection procedures: injection of a bacterial suspension into the sand body, immediately followed by injection of a fixation fluid of high salt content. Ivanov and Chu (2008) provided a detailed review on MICP applications for soil improvement. Although there are diverse potential applications of MICP, they concluded that, at present, promising applications are only focusing on bioclogging and biocementation. Biocementation is defined as the improvement of soil strength by the production of particle-binding materials through microbial means, while bioclogging is the reduction of hydraulic conductivity of soil or porous rock by pore-filling materials generated by microbial processes. Most Bacillus strains can precipitate calcite through the conversion of urea into ammonia and carbon dioxide (Hammes et al. 2003). The previous reported studies however, mainly adopted B. pasteurii as the urease-producing microorganism to consolidate loose sand columns. Studies which employed other Bacilli and different soil types are still very limited. The present experimental work investigates the effectiveness of a different Bacillus strain, B. megaterium, in improving the shear strength

(biocementation) and reducing the hydraulic conductivity (bioclogging) of two soil types, i.e., tropical residual soil and sand.

2. Testing Materials 2.1 Soil Specimens The two types of soil employed in this study were tropical residual soil and sand. The tropical residual soil specimen was obtained from a site in Universiti Tunku Abdul Rahman, Kuala Lumpur campus compound, while the sand specimen was of typical concrete sand. Table 1 tabulates the values of physical indices of the soil specimens obtained from the standard soil properties tests. The soil particle size distributions are presented in Fig. 1. Based on the British Standard soil classification system, the tropical residual soil specimen was classified as Sandy Silt with high plasticity, while the sand specimen was classified as Well Graded Sand. Standard proctor compaction test was performed to obtain the compaction curve of the tropical residual soil. The Maximum Dry Density (MDD) of the soil was found to be 1563 kg/m3. To investigate the effectiveness of the MICP treatment for soils of varying density, the residual soil specimens were compacted into Table 1. Physical Properties of the Soil Specimens

Composition Gravel (%) Sand (%) Silt (%) Clay (%) LL (%) PL (%) PI Soil Classification BSCS Maximum Dry Density (MDD) Maximum Index Density (ρmax) Minimum Index Density (ρmin)

− 719 −

Residual Soil (Sandy Silt)

Sand

0 29 55 16 58.0 44.3 13.7 MHS 1563 kg/m3 -

2 96 2 0 SW 1842 kg/m3 1439 kg/m3

Fig. 1. Particle Size Distributions of the Soil Specimens KSCE Journal of Civil Engineering

Improvements in Engineering Properties of Soils through Microbial-Induced Calcite Precipitation

three different densities, i.e., 85% of MDD, 90% of MDD, and 95% of MDD. For the sand specimen, the minimum (ρmin) and maximum (ρmax) index densities were determined in compliance with the procedures of ASTM D4254 (ASTM 2000) and ASTM D4253 (ASTM 2000), respectively. Three densities were compacted within the range of the minimum (ρmin = 1439 kg/m3) and maximum (ρmax = 1842 kg/m3) index densities, i.e., 85% of ρmax, 90% of ρmax, and 95% of ρmax.

residual soil specimens, the desired densities were achieved by compacting the soils into prefabricated moulds with the moisture content corresponding to the desired density determined from the compaction curve. For the sand specimens, the desired densities were achieved by tamping the predetermined amount of sand specimens in layers. Upon compaction, all the soil specimens were saturated to ensure steady flow of cementation reagents through the soil specimens during the MICP treatment.

2.2 Microorganism The urease-producing microorganism used in this study was B. megaterium ATCC 14581. It is a rod-shaped, Gram positive soil bacterium of diameter 2 µm to 5 µm. It can be found in diverse environment from rice paddies to dried food, seawater, sediments, fish, and even in bee honey (Wittmann et al., 2010). The role of B. megaterium is to produce enzyme urease through its metabolic activity under proper cultivation process. The enzyme urease triggers the MICP biochemical reaction by hydrolyzing urea (CO(NH2)2) through the following reaction:

3.2 MICP Treatments As explained in the previous sections, MICP can be achieved by introducing the urease-producing microorganisms into soil together with the external sources of urea and calcium chloride solution (Martinez et al., 2011; Stocks-Fischer et al., 1999; Whiffin et al., 2007). The technology has been applied successfully in diverse engineering fields including the consolidation of sand columns, improvement of concrete strength, repairing the concrete cracks etc (Achal et al., 2011; De Muynck et al., 2010; Van Tittelboom et al., 2010). The rate of calcite precipitation is an important process variable. The rate should not be too high to avoid rapid precipitation and consequent clogging near the inlet of the specimen mould. The precipitation rate also should not be too low as the specimen would take a long duration to cement (Van Paassen et al., 2010). To obtain a significant improvement in strength for loose granular sand, at least 60 kg calcite per meter cube of soil has to be precipitated. This corresponds to approximately 2 mol CaCO3 precipitate per liter of pore space (Van Paassen et al., 2010; Whiffin et al., 2007). Qabany et al. (2011) found that this amount of precipitated calcite can be achieved by injecting one pore volume of 0.25 M CaCl2 / urea solution into the soil specimen at intervals of 6 hours for a treatment period of 38-48 hours. In the present study, B. megaterium was cultivated in nutrient broth (pH 7.0 prior to autoclaving) at 130 rpm and incubation temperature of 37oC. 13.0 g of nutrient broth was dissolved in one liter of distilled water, which consisted of 5.0 g of peptone, 5.0 g of sodium chloride, 2.0 g of yeast extract, and 1.0 g of meat extract. The culture with concentration of 5 × 107 cfu/mL (optical density of 1.3) was then mixed with air-dried soil specimens. The concentration of B. megaterium was determined by the spread plate technique. Serial dilution was performed on the incubated culture. The diluted culture was then spread on the surface of agar plate where the colonies were counted after 16 hours of incubation. Subsequently, the soil mixture with the addition of appropriate moisture content was compacted to the desired densities. Figures 2(a) and (b) show the schematic diagram of laboratory setup and soil specimen mould used for the MICP treatment, respectively. Prior to commencing the MICP treatment process, the soil specimens were flushed through with a fixation fluid (50 mM CaCl2) to saturate the soil specimen and fix the B. megaterium homogenously in soil specimen. The MICP treatment was

CO (NH2)2 + 3H2O → 2NH4+ + HCO3− + OH−

(1)

The ammonium (NH4+) increases the pH and causes the bicarbonate (HCO3−) to precipitate with calcium ion (Ca2+) from the calcium chloride supplied in order to form the calcium calcite (CaCO3): Ca2+ + HCO3− + OH− → CaCO3 + H2O

(2)

The calcite generated is responsible for cementing and clogging the soil specimens. Biocementation is achieved when the calcite crystals precipitate on the surface or form bridges between the existing soil grains. These calcite crystals bond the soil particles and forbid movement of the grains, and hence enhance the strength and stiffness properties of the soil (Harkes et al., 2010). Bioclogging, on the other hand, is the formation of pore-filling materials (calcite) through microbial process and subsequently results in a reduction in soil porosity and hydraulic conductivity (Ivanov and Chu, 2008). 2.3 Cementation Reagents Cementation reagents serve as the raw materials for calcite formation in the MICP process. The cementation reagents employed in this study comprised 0.25 M solution of urea (CO (NH2)2) and calcium chloride (CaCl2). The cementation reagents also contained 3 g nutrient broth, 10 g ammonium chloride (NH4Cl), and 2.12 g sodium bicarbonate (NaHCO3) per litre of deionized water (Dejong et al., 2006; Qabany et al., 2011; Stocks-Fischer et al., 1999; Stoner et al., 2005). All chemicals used were of Analytical Reagent (AR) grade.

3. Methodology 3.1 Preparation of Soil Specimens Prior to the MICP treatment, the soil specimens were air-dried in the laboratory at room temperature for 1 week. For the Vol. 17, No. 4 / May 2013

− 720 −


hours. The flow velocity was controlled at approximately 1.7 × 10−5 m/s. This was done by regulating the pressure of the air compressor. The pressurized air compressed the reagent solution in the pressure tank, and eventually created a hydraulic gradient between the inlet and outlet of the soil specimen mould. This treatment configuration was applied for all soil specimens. 3.3 Shear Strength and Hydraulic Conductivity Tests Shear strength and hydraulic conductivity of the treated soil specimens were tested in the laboratory. For the residual soil specimens, the shear strength was determined by performing the unconfined compression test on 50 mm diameter saturated specimen. The hydraulic conductivity was determined by the falling head permeability test. For the sand specimens, the shear strength was determined by the standard direct shear box test on 60 mm × 60 mm dry specimens. The hydraulic conductivity was determined by the constant head permeability test. All test procedures were in compliance with the British Standard 1377 (BSI, 1990). The same mould was used for the MICP treatment and subsequent testing in order to ensure minimal disturbance on the tested soil specimens. Fig. 2. (a) Schematic Diagram of Laboratory Setup: (b) Soil Specimen Mould

performed by injecting one pore volume of 0.25 M CaCl2/urea solution into the soil specimens at intervals of 6 hours over 48

3.4 Experimental Design The experimental design focused mainly on the effects of soil types (residual soil and sand), soil densities (85%, 90%, and 95% of maximum density), and treatment conditions (untreated, treated with B. megaterium only, treated with cementation reagents only,

Table 2 Experimental Design No. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Experiment abbreviation 0.95RU 0.95RM 0.95RR 0.95RT 0.90RU 0.90RM 0.90RR 0.90RT 0.85 RU 0.85RM 0.85RR 0.85RT 0.95SU 0.95SM 0.95SR 0.95ST 0.90SU 0.90SM 0.90SR 0.90ST 0.85SU 0.85SM 0.85SR 0.85ST

Soil type Residual soil Residual soil Residual soil Residual soil Residual soil Residual soil Residual soil Residual soil Residual soil Residual soil Residual soil Residual soil Sand Residual soil Sand Sand Sand Residual soil Sand Sand Sand Residual soil Sand Sand

Soil density 0.95 MDD 0.95 MDD 0.95 MDD 0.95 MDD 0.90 MDD 0.90 MDD 0.90 MDD 0.90 MDD 0.85 MDD 0.85 MDD 0.85 MDD 0.85 MDD 0.95ρmax 0.95ρmax 0.95ρmax 0.95ρmax 0.90ρmax 0.90ρmax 0.90ρmax 0.90ρmax 0.85ρmax 0.85ρmax 0.85ρmax 0.85ρmax − 721 −

Treatment method Untreated B. megaterium only Reagents only B. megaterium & Reagents Untreated B. megaterium only Reagents only B. megaterium & Reagents Untreated B. megaterium only Reagents only B. megaterium & Reagents Untreated B. megaterium only Reagents only B. megaterium & Reagents Untreated B. megaterium only Reagents only B. megaterium & Reagents Untreated B. megaterium only Reagents only B. megaterium & Reagents KSCE Journal of Civil Engineering


and treated with B. megaterium and cementation reagents) on the shear strength and hydraulic conductivity of soils. Each experiment was repeated for three times, and the average of these three values was considered for the shear strength and hydraulic conductivity determinations. The untreated soil specimens served as controls. The treatment with cementation reagents only was used to investigate the existence of naturally inhibited calcite forming microorganisms in the soil specimens. The specimens treated with microorganism only were used to monitor the effect of biomass on the shear strength and hydraulic conductivity of the soils. The details of the experimental design are tabulated in Table 2. A total of 144 laboratory experiments were performed to cater for the 24 combinations of soil specimens and treatment conditions.

4. Experimental Results

Fig. 4. Shear Strength Improvements of Residual Soil Specimens

4.1 Shear Strength of Residual Soil Figure 3 compares the stress-strain curves of the untreated residual soil specimens (RU) and those treated with B. megaterium and cementation reagents (RT). The unconfined compressive strength (qu) of the soil is defined as the peak stress or the stress that yields 20% of axial strain, whichever is lower. The shear strengths of the residual soils in this study were characterized by the undrained shear strength parameter (cu), which was taken as half of the unconfined compressive strength (qu), i.e., cu = ½ qu. Fig. 4 summarizes the shear strength results of the residual soil specimens. The shear strength of MICPtreated residual soils was improved for all densities (0.85RT, 0.90RT, 0.95RT). The shear strength improvement ratio increased with increased density, i.e., 1.41 (41%), 2.59 (159%) and 2.64 (164%) for specimens of 0.85RT, 0.90RT, and 0.95RT, respectively. The soil specimens treated with cementation reagents only (0.95RR, 0.90RR, and 0.85RR), also exhibited increased shear strength. The undrained shear strength parameter improved by 1.11, 1.25 and 1.33 for the specimens of 0.85RR, 0.90RR, and 0.95RR, respectively. The results implied that MICP was triggered by the microorganisms inhibiting naturally in the soil deposits. The improvements (ranging from 1.11-1.33), however, were

Fig. 3. Stress-strain Curve for Residual Soil Specimens Vol. 17, No. 4 / May 2013

lower compared to the specimens treated with B. megaterium and cementation reagents (ranging from 1.41-2.64). This was because the introduction of B. megaterium resulted in a higher production of urease enzyme. The enzyme triggered more calcite precipitation and led to greater enhancement in shear strength. The results for the specimens treated with microorganism only, i.e., 0.85RM, 0.90RM and 0.95RM were not shown in Fig. 4 because no visible improvement in shear strength was observed in these specimens. The results implied that biomass was ineffective in improving the shear strength of the residual soil. 4.2 Hydraulic Conductivity of Residual Soil Figure 5 shows the results of hydraulic conductivity for the residual soil specimens. The saturated hydraulic conductivity (ksat) of untreated residual soil specimens ranged between 1.0 × 10−7 m/s and 9.3×10−7 m/s, in which the values were directly proportional to the soil density. The saturated hydraulic conductivity

Fig. 5. Saturated Hydraulic Conductivity Reductions of Residual Soil Specimens − 722 −


of tropical residual soil typically lies in the range of 1×10−6 to 1×10−8 m/s (Tan et al., 2008). The saturated hydraulic conductivity of MICP-treated soil was markedly reduced for all densities. The reduction in hydraulic conductivity inflicted by the calcite can be explicitly seen by observing the margin between the saturated hydraulic conductivities of the untreated specimens (0.85RU, 0.90RU and 0.95RU) and those treated with B. megaterium and cementation reagents (0.85RT, 0.90RT and 0.95RT). The greatest reduction in hydraulic conductivity occurred in the densest specimen (0.95RT) where the ratio of saturated hydraulic conductivity between treated and untreated specimens was 0.26 (a reduction of 74%). The reduction ratios for 0.90RT and 0.85RT specimens were 0.40 (60%) and 0.45 (55%), respectively. Similar to the observations in shear strength, the residual soil specimens treated with cementation reagents only (0.85RR, 0.90RR and 0.95RR) exhibited slight alteration in saturated hydraulic conductivity (decreased by not more than 30% of untreated). These observations confirmed the earlier finding that a relatively small amount of calcite precipitating microorganism exists naturally in the residual soil. Furthermore, the residual soil specimens treated with microorganism only, without the supply of cementation reagents (0.85RM, 0.90RM, and 0.95RM) experiences no significant diminution in hydraulic conductivity. In the nutshell, diminution in hydraulic conductivity of soil was mainly inflicted by calcite and the effect was proportional to the soil density. The formation of calcite clogged most of the pores and reduced the saturated hydraulic conductivity of the residual soil effectively. 4.3 Shear Strength of Sand Figure 6 shows a sample of shear stress versus horizontal displacement results obtained from the direct shear tests for the untreated sand specimens (SU), and those treated with B. megaterium and cementation reagents (ST). The shear strength of the dry sand specimens used in this study was characterized by the effective internal friction angle (φ ') only. Fig. 7 summarizes the improvement in the effective internal friction angle of the sand specimens. The effective internal friction angles of the untreated sand

Fig. 6. Stress-strain Curve for Sand Specimens

Fig. 7. Shear Strength Improvements of Sand Specimens

specimens were between 39.9o and 48.8o. The internal friction angles of the MICP-treated sand specimens (0.85ST, 0.90ST and 0.95ST) were generally higher than the untreated specimens (0.85SU, 0.90SU and 0.95SU). The 0.85ST specimen had the greatest improvement ratio (1.25), followed by 0.90ST specimen (1.17) and 0.95ST specimen (1.14). The improvement ratio decreased with increased density. The trend was opposite to the results observed in the residual soil specimens. Furthermore, the sand specimens (1.14-1.25) exhibited significantly lower improvement ratios than the residual soil specimens (1.41-2.64). For the specimens treated with cementation reagents only (0.85SR, 0.90SR and 0.95SR), the shear strength improvement ratio was markedly lower (1.06-1.15) compared to the specimens treated with B. megaterium and cementation reagents (1.141.25). The specimens treated with microorganism only (0.85SM, 0.90SM, 0.95SM) barely had any effect on the shear strength alterations. 4.4 Hydraulic Conductivity of Sand The saturated hydraulic conductivity (ksat) of the sand specimens is illustrated in Fig. 8. The saturated hydraulic conductivities of the untreated sand specimens were in the orders of 10−4 to 10−3 m/s. These values are in close agreement with typical saturated hydraulic conductivity of fine to medium sand specimens (Brassington, 1988). Similar to the trend observed in shear strength of sand specimens, the reduction in hydraulic conductivity becomes less effective with the increased density. The greatest reduction in hydraulic conductivity occurred in the 0.85ST specimen in which the hydraulic conductivity decreased by approximately one order of magnitude from 3.5 × 10−3 m/s to 3.2 × 10−4 m/s (a reduction ratio of 0.09). As the density of the specimen increased, the reduction ratios of hydraulic conductivity were marginally lesser as observed in 0.90ST (0.14) and 0.95ST specimens (0.15). For the sand specimens treated with cementation reagents only,

− 723 −

KSCE Journal of Civil Engineering


4.5 Quantitative Analysis of Calcite Precipitated The quantitative measurements of the carbonate content in the soil specimens were carried out by gravimetric analysis using the acid-treatment weight loss technique. Dry soil samples of 20 g were prepared for the tests. The carbon dioxide deliberated from the reaction between diluted hydrochloric acid (2 molar) and carbonate in soil was indicated by the effervescence. The residue was collected using a filter paper and oven-dried at temperature of 105°C. The measured weight loss of the soil sample was used to estimate the percentage of carbonate content in the soil specimens. It was assumed that the increment of carbonate content in the soil specimens after MICP treatment was caused by the formation of calcium carbonate.

The carbonate contents of the untreated residual soil and sand were 0.670% and 0.274%, respectively. Figure 9 correlates the carbonate content of the MICP-treated residual soil specimens (0.85RT, 0.90RT and 0.95RT) with their shear strength and hydraulic conductivity alterations. The carbonate contents of the 0.85RT, 0.90RT and 0.95RT soil specimens were found to be 1.750%, 2.559%, and 2.381%, respectively. By subtracting their initial content in untreated specimens, the increments of calcite contents after the MICP treatment were 1.080%, 1.889%, and 1.711%, respectively. The 0.85RT specimen recorded the lowest carbonate content, and hence contributed to the lowest alterations in both shear strength and hydraulic conductivity. The highest carbonate content was measured in the specimen of 0.90RT, followed by the specimen of 0.95RT with a slightly lower carbonate content. The overall trend of the carbonate content showed reasonably good comparisons with the trends of the shear strength improvement and hydraulic conductivity reduction. Dense soil has a close arrangement of soil particles, and this contributes to more inter-particle contact points per unit volume. The precipitation of calcite at these inter-particles contact points reduces the pore size (reduces hydraulic conductivity) and improves the bonding between soil particles (improves shear strength). The correlations between the carbonate content of the MICPtreated sand specimens (0.85ST, 0.90ST and 0.95ST) and their shear strength and hydraulic conductivity alterations are demonstrated in Fig. 10. The calcite contents of the 0.85ST and 0.90ST specimens were almost identical to each other, i.e., 6.376% and 5.943%, respectively. However, a significant drop of carbonate content (2.935%) was observed in the 0.95ST specimen. The dense sand (0.95ST) experiences low calcite precipitation after the MICP treatment because it has a relatively smaller pore space. By comparing the amounts of calcite precipitated in both the sand and residual soil specimens, it can be found that the soil particle size has a significant effect on the calcite precipitation. The amounts of calcite in the treated sand specimens (ranging from 2.661% to 6.102%) were generally higher than those of the

Fig. 9. Correlations between Carbonate Content and Alterations in Shear Strength and Hydraulic Conductivity of Residual Soil Specimens

Fig. 10. Correlations between Carbonate Content and Alterations in Shear Strength and Hydraulic Conductivity of Sand Specimens

Fig. 8. Saturated Hydraulic Conductivity Reductions of Sand Specimens

the reductions in hydraulic conductivity were negligible. The hydraulic conductivities of the 0.85SR, 0.90SR and 0.95SR were only reduced by an average of 7%. Similarly, the reductions in hydraulic conductivity of sand specimens treated with microorganism only (0.85SM, 0.90SM, and 0.95SM) were also negligible.

Vol. 17, No. 4 / May 2013

− 724 −


residual soil specimens (ranging from 1.750% to 2.559%). This observation can be explained by the pore spaces available in the soils. For the residual soil, some portions of the soil particles were smaller than 2 µm. As the bacteria used in this study has a diameter of 2 µm to 5 µm, the limited pore spaces controlled the passage of bacteria through the soil, and hence lower calcite amount was precipitated. The large pore space in sand can accommodate for more bacteria and reagent solutions, and allow for more calcite precipitation. From the Fig. 10, it is apparent that the overall trend of the calcite content showed good agreement with the trends of improvement in shear strength and reduction in hydraulic conductivity of the treated sand specimens. It should be noted that despite of significantly higher calcite content precipitated in sand as compared to the residual soil, the improvements in shear strength of sand specimens were still much lower than those of residual soil specimens. The reason could be the calcite was not formed at the inter-particle contact points in view of large pore size in sand. Whiffin et al. (2007) found that calcite content of above 3.5% was effective in improving the shear strength of sand. However, the present experimental results showed that the shear strength was not improved considerably despite calcite content of 6% had been produced. The possible explanation of this discrepancy is that the sand specimen used in the study of Whiffin et al. (2007) was finer than the sand adopted in the present study. For example, the median grain size (D50) of the sand adopted in the study of Whiffin et al. (2007) was 165 µm, which was much finer than the D50 of 520 µm adopted in the present study. The results suggested that the shear strength of soils with fine grain size could be effectively improved at lower calcite content due to the dominant role of inter-particle contact points per unit volume. Nevertheless, there was still a great reduction in hydraulic conductivity due to the high calcite contents in treated sand. The precipitated calcite caused clogging at the sand pore spaces and pore throats. 4.6 Scanning Electron Microscopy Analysis Scanning Electron Microscope (SEM) imagery was performed on the selected soil specimens to visualize the calcite precipitation in the soil specimens. The SEM images were captured using the

Fig. 11. SEM Images for the Residual Soil Specimens: (a) Untreated, (b) Treated with Cementation Reagents Only, (c) Treated with B. megaterium and Cementation Reagents

Hitachi S-3400N Scanning Electron Microscope. The tests were of particular interest on the formation of calcite crystals upon MICP treatment. Fig. 11 shows the SEM images of the residual soil specimens treated under three different conditions, i.e. untreated, treated with cementation reagents only, and treated with B. megaterium and cementation reagents. Relatively smooth particle surfaces were observed in the untreated specimen (Fig. 11a). For the treatment with cementation reagents only, some calcite crystals were observed on the soil particles (Fig. 11b).

Fig. 12. SEM Images for the Sand Specimens: (a) Untreated, (b) Treated with Cementation Reagents Only, (c) Treated with B. megaterium and Cementation Reagents − 725 −



Abundance of calcite crystals were observed in the specimens treated with B. megaterium and cementation reagents (Fig. 11c). On closer observations, rod-shaped B. megaterium was found in intimate contact with the calcite crystals. Similar patterns were observed for the SEM images of sand specimens (Fig. 12). Comparatively, the quantity of the precipitated calcite crystal for the sand specimen treated with cementation reagents only (Fig. 12b) was less than that observed in residual soil specimen (Fig. 11b). This observation confirmed the results from the shear strength and hydraulic conductivity tests that the natural microorganisms only exist for an insignificant amount in the sand specimens, and hence induced slight alterations to their properties. Furthermore, Energy Dispersion Spectroscopy (EDS) was performed on the MICP-treated specimens using EDAX (AMETEK Materials Analysis Division) to analyze the concentrations of calcium (Ca), carbon (C), oxygen (O), and silica (Si) elements. It is an analytical technique used for elemental analysis or chemical characterization of samples. Fig. 13(a) illustrates the EDS performed on a soil particle. The intensity of Si element was the highest among all elements concerned since it is the major element in soil. As for the Fig. 13(b), Ca and O recorded the

Fig. 13. EDS on (a) Soil Particles, (b) Crystal Calcite Vol. 17, No. 4 / May 2013

highest intensity. An increased intensity was also encountered for C. As these three elements form the main components of calcite (CaCO3), the EDS was probably performed on a calcite crystal.

5. Discussion Microbial induced calcite precipitation has been shown to be an effective method to enhance the shear strength and reduce the hydraulic conductivity of soil. The soil with enhanced strength can contribute to a greater ground bearing capacity, while reduced hydraulic conductivity can minimize settlement, shrinkswell tendency, seepage, and infiltration of rainfall into soils. The experimental results indicated that MICP was more effective in improving shear strength for residual soil (1.40-2.64 increment ratios) than for sand (1.14-1.25 increment ratios). With respect to the reduction in hydraulic conductivity, the sand specimen (0.09-0.15 reduction ratios) was found to be more effective than the residual soil (0.26-0.46 reduction ratios). The improvement in shear strength and diminution in hydraulic conductivity of residual soil increased with increased density. However, the sand specimens exhibited a reverse trend. These different observations between residual soil and sand need clarification through an insight into the behaviour of the B. megaterium in soil. According to Achal et al. (2009), the effectiveness of MICP treatment on a soil specimen can be attributed to both the ability of the microorganism to move freely throughout the pore space and the sufficient particle-particle contacts per unit volume at which cementation occurs. These two attributes, however, are contradicting each other as the soil with large pore space tends to have less particle-particle contacts per unit volume, and vice versa. These conditions require a balance relationship between the microorganism size and the pore structure characteristics, namely the pore throats. The relatively low improvements in shear strength of sand compared to residual soil can be explained by the insufficient concentrations of particle-particle contacts per unit volume. This is because the sand specimen contains coarser granular particles. The contacts between the coarse particles are lesser compared to the residual soil specimen that consists of wide range of particle sizes (ranging from smaller than 1 mm to 2 mm). The pores between the coarse particles in residual soil are filled with the smaller grains, thus results in greater particle-particle contacts. The improved shear strength and reduced hydraulic conductivity with increased density of residual soil can also be explained by the particle-particle contacts. As the residual soil compacted to a higher density, the particle-particle contacts increase. This facilitates greater calcite bonding at particle-particle contacts. Pore spaces also decrease with increased compaction. However, the long treatment duration (2 days) and intermittent cementation reagents injection method (flush the cementation reagents through the sample every 6 hours at a moderately high velocity of 1.7 × 10−5 m/s) employed in this study are believed to have minimized the inverse impact of small pore throat on the MICP

− 726 −


treatment. The sand specimen has a higher hydraulic conductivity reduction ratio than the residual soil. This can be explained by the greater porosity of sand. Greater porosity means more pore space available for calcite deposition by B. megaterium, and hence results in greater reduction in hydraulic conductivity. As mentioned earlier, the pore throat size can also affect the effectiveness of MICP. The improvement in shear strength and reduction in hydraulic conductivity of the sand specimens decreased with increased density. This is because denser sand contributed to a smaller pore throat size. Consequently, the movements of B. megaterium and reagent solutions within the sand specimen were restrained, and hence retarded the MICP process slightly. Although denser specimen may have greater particle-particle contacts leading to enhanced improvement in the soil properties, however, particle-particle contacts of sand have not been improved by greater compaction due to lack of finer particles that act as filler to the voids between the large particles. The smaller pore throat in the sand specimen plays a more dominant role in controlling the effectiveness of MICP. The residual soil and sand specimens treated with cementation reagents only exhibited slight increment in the shear strength and diminution in the hydraulic conductivity. The results imply that the amount of natural calcite forming microorganisms is insufficient to trigger effective MICP. Comparatively, the residual soil specimens show marginally greater improvements than the sand specimens. This is because the residual soil specimen was taken in-situ from a site which could be rich of natural microorganisms, while the sand specimen was of typical concrete sand which was left exposed under the extreme tropical climate. As the result, the natural inhibited microorganisms in the sand specimens are lesser than the residual soil specimens.

with the increased density. The results implied that the particle-particle contacts of sand have not been improved markedly by the higher degree of compaction. 4. The saturated hydraulic conductivities of the MICP-treated residual soils exhibited reduction ratios of 0.26-0.45. The reduction is less significant than the reduction ratios of the sand specimens (0.09-0.15). This can be explained by the greater porosity and pore spaces in sand that are available for bioclogging. 5. The amounts of calcite precipitated in the treated residual soil specimens ranged from 1.080% to 1.889%. The calcite content increased with the increased soil density. Opposing trend was observed for the sand specimens, whereby the calcite content decreased with the increased sand density. Nevertheless, the amounts of calcite in the treated sand specimens (ranging from 2.661% to 6.102%) were generally higher than those of the residual soil specimens. 6. For both residual soil and sand specimens, treatment with cementation reagents only exhibited slight alterations in shear strength and hydraulic conductivity. The results indicated the presence of natural calcite forming microorganisms which existed in insignificant amount. The results from SEM analysis confirmed this finding. 7. The effects of microorganism only (biomass) on the shear strength and hydraulic conductivity of both residual soil and sand specimens were negligible.

Acknowledgements This project is funded by the Ministry of Higher Education, Malaysia under Fundamental Research Grant Scheme (FRGS). Besides, the authors acknowledge the useful comments from the reviewers.

6. Conclusions Twenty four configurations of experimental variables were designed to investigate the effectiveness of MICP in improving the shear strength and reducing the hydraulic conductivity of two types of soil selected for the study. The following findings are drawn from the study: 1. B. megaterium was demonstrated to enhance the shear strength and reduce the hydraulic conductivity of both residual soil and sand specimens. The improvement in the engineering soil properties varied with soil densities, soil types, and treatment conditions. 2. The MICP-treated residual soils exhibited significant increment ratios in shear strength, i.e., 1.41-2.64. The rate of improvement increased with increased density. This can be explained by high particle-particle contacts in residual soil particles. 3. The MICP-treated sands improved in shear strength by ratios of 1.14-1.25. The lower improvement compared to residual soils can be attributed to the lesser contacts between sand particles. The rate of improvement decreased

References Achal, V., Mukherjee, A., Basu, P. C., and Reddy, M. S. (2009). “Strain improvement of Sporosarcina Pasteurii for enhanced urease and calcite production.” J. Ind. Microbiol. Biotechnol., Vol. 36, No. 7, pp. 981-988. Achal, V., Pan, X., and Özyurt, N. (2011). “Improved strength and durability of fly ash-amended concrete by microbial calcite precipitation.” Ecol. Eng., Vol. 37, No. 4, pp. 554-559. Anagnostopoulos, C. and Hadjispyrou, S. (2004). “Laboratory study of an epoxy resin grouted sand.” Ground Improv., Vol. 8, No. 1, pp. 3945. ASTM (2000). ASTM D 4253-00, Standard test methods for maximum index density and unit weight of soils using a vibratory table, American Society for Testing and Materials. ASTM (2000). ASTM D 4254-00, Standard test methods for minimum index density and unit weight of soils and calculation of relative density, American Society for Testing and Materials. Basha, E. A., Hashim, R., Mahmud, H. B., and Muntohar, A. S. (2005). “Stabilization of residual soil with rice husk ash and cement.” Constr. Build. Mater., Vol. 19, No. 6, pp. 448-453. Baskar, S., Baskar, R., Mauclaire, L., and Mckenzie, J. A. (2006).

− 727 −



“Microbially induced calcite precipitation in culture experiments: Possible origin for stalactites in Sahastradhara caves, Dehradun, India.” Curr. Sci., Vol. 90, No. 1, pp. 58-64. Baveye, P., Vandevivere, P., Hoyle, B., DeLeo, P., and de Lozada, D. S. (1998). “Environmental impact and mechanisms of the biological clogging of saturated soils and aquifer materials.” Crit. Rev. Environ. Sci. Technol., Vol. 28, No. 2, pp. 123-191. Brassington, R. (1988). Field hydrogeology, Halsted Press, New York. BSI (1990). British standard 1377: 1990, Methods of test for soils for civil engineering purposes, BSI Group, United Kingdom. Castainer, S., Le Métayer-Levrel, G., and Perthuisot, J.-P. (1999). “Cacarbonates precipitation and limestone genesis -- The microbiogeologist point of view.” Sediment. Geol., Vol. 126, Nos. 1-4, pp. 9-23. Dejong, J. T., Fritzges, M. B., and Nüsslein, K. (2006). “Microbially induced cementation to control sand response to undrained shear.” J. Geotech. Geoenviron. Eng., Vol. 132, No. 11, pp. 1381-1392. Dejong, J. T., Mortensen, B. M., Martinez, B. C., and Nelson, D. C. (2010). “Bio-mediated soil improvement.” Ecol. Eng., Vol. 36, No. 2, pp. 197-210. De Muynck, W., De Belie, N., and Verstraete, W. (2010). “Microbial carbonate precipitation in construction materials: A review.” Ecol. Eng., Vol. 36, No. 2, pp. 118-136. Ehrlich, H. L. (1999). “Past, present and future of biohydrometallurgy.” Process Metallurgy, R. Amils, and A. Ballester, eds., Elsevier, pp. 312. Hammes, F., Boon, N., de Villiers, J., Verstraete, W., and Siciliano, S. D. (2003). “Strain-specific ureolytic microbial calcium carbonate precipitation.” Appl. Environ. Microbiol., Vol. 69, No. 8, pp. 49014909. Harkes, M. P., Van Paassen, L. A., Booster, J. L., Whiffin, V. S., and Van Loosdrecht, M. C. M. (2010). “Fixation and distribution of bacterial activity in sand to induce carbonate precipitation for ground reinforcement.” Ecol. Eng., Vol. 36, No. 2, pp. 112-117. Ho, M. H., and Chan, C. M. (2011). “Some mechanical properties of cement stabilized malaysian soft clay.” World Acad. Sci. Eng. Technol., Vol. 74, No., pp. 24-31. Huat, B. B. K. (2006). “Deformation and shear strength characteristics of some tropical peat and organic soils.” Pertanika J. Sci. Technol., Vol. 14, Nos. 1-2, pp. 61-74. Ivanov, V. and Chu, J. (2008). “Applications of microorganisms to geotechnical engineering for bioclogging and biocementation of soil in situ.” Rev. Environ. Sci. Biotechnol., Vol. 7, No. 2, pp. 139-153. Karol, R. H. (2003). Chemical grouting and soil stabilization, M. Dekker, New York. Kazemian, S. and Huat, B. B. K. (2009). “Assessment and comparison of grouting and injection methods in geotechnical engineering.” Eur. J. Sci. Res., Vol. 27, No. 2, pp. 234-247. Kazemian, S., Huat, B. B. K., Prasad, A., and Barghchi, M. (2011). “A state of art review of peat geotechnical engineering perspective.” Int. J. Phys. Sci., Vol. 6, No. 8, pp. 1974-1981. Krebs, R. D. and Walker, R. D. (1971). Highway materials, McGrawHill, New York.

Vol. 17, No. 4 / May 2013

Leonards, G. A. (1962). Foundation engineering, McGraw-Hill, New York. Lian, B., Hu, Q., Chen, J., Ji, J., and Teng, H. (2006). “Carbonate biomineralization induced by soil bacterium Bacillus megaterium.” Geochim. Cosmochim. Acta, Vol. 70, No. 22, pp. 5522-5535. Lu, W., Qian, C., and Wang, R. (2010). “Study on soil solidification based on microbiological precipitation of CaCO3.” Sci. China Technol. Sci., Vol. 53, No. 9, pp. 2372-2377. Martinez, B. C., Barkouki, T. H., Dejong, J. D., and Ginn, T. R. (2011). “Upscaling of microbial induced calcite precipitation in 0.5 m columns experimental and modeling results.” Geo-Frontiers 2011, pp. 4049-4059. Mitchell, J. K. and Santamarina, J. C. (2005). “Biological considerations in geotechnical engineering.” Journal Geotech. Geoenviron. Eng., Vol. 131, No. 10, pp. 1222-1233. Okwadha, G. D. and Li, J. (2010). “Optimum conditions for microbial carbonate precipitation.” Chemosphere, Vol. 81, No. 9, pp. 11431148. Peethamparan, S., Olek, J., and Diamond, S. (2009). “Mechanism of stabilization of Na-montmorillonite clay with cement kiln dust.” Cem. Concr. Res., Vol. 39, No. 7, pp. 580-589. Qabany, A. A., Mortensen, B., Martinez, B., Soga, K., and Dejong, J. (2011). “Microbial carbonate precipitation correlation of s-wave velocity with calcite precipitation.” Geo-Frontiers 2011, pp. 39934001. Stocks-Fischer, S., Galinat, J. K., and Bang, S. S. (1999). “Microbiological precipitation of CaCO3.” Soil Biol. Biochem., Vol. 31, No. 11, pp. 1563-1571. Stoner, D. L., Watson, S. M., Stedtfeld, R. D., Meakin, P., Griffel, L. K., Tyler, T. L., Pegram, L. M., Barnes, J. M., and Deason, V. A. (2005). “Application of stereolithographic custom models for studying the impact of biofilms and mineral precipitation on fluid flow.” Appl. Environ. Microbiol., Vol. 71, No. 12, pp. 8721-8728. Tan, L. P., Lee, C. Y., and Sivadass, T. (2008). “Parametric study of residual soil slope stability.” International Conference on Construction and Building Technology 2008, 16-20 Jun 2008, UniTen, Malaysia, pp. 33-42. Van Paassen, L. A., Daza, C. M., Staal, M., Sorokin, D. Y., van der Zon, W., and van Loosdrecht, M. C. M. (2010). “Potential soil reinforcement by biological denitrification.” Ecol. Eng., Vol. 36, No. 2, pp. 168175. Van Tittelboom, K., De Belie, N., De Muynck, W., and Verstraete, W. (2010). “Use of bacteria to repair cracks in concrete.” Cem. Concr. Res., Vol. 40, No. 1, pp. 157-166. Whiffin, V. S., van Paassen, L. A., and Harkes, M. P. (2007). “Microbial carbonate precipitation as a soil improvement technique.” Geomicrobiol. Journal, Vol. 24, No. 5, pp. 417-423. Wittmann, C. and Krull, R. (2010). Biosystems engineering I: Creating superior biocatalysts, Advances in Biochemical Engineering / Biotechnology, Springer-Verlag Berlin Heidelberg, Berlin, Heidelberg. Xanthakos, P. P., Abramson, L. W., and Bruce, D. A. (1994). Ground control and improvement, J. Wiley, New York.

− 728 −

Improvements in engineering properties of soils through microbial-induced calcite precipitation.pdf

Recommend Documents