Innovative ground enhancement by improved microbially induced CaCO3 precipitation technology

The possibility of using microbiological processes to improve the mechanical properties of soil by undisturbed in-situ application has gained attention over recent years. This study has contributed to the technology of biocement, based on microbially induced carbonate precipitation (MICP), for the purpose of soil reinforcement application. MICP involves both the hydrolysis of urea by bacterial urease enzyme and calcium carbonate precipitation in the presence of dissolved calcium ions. Other previously published approaches were based on saturated flow (submersed flow), which is accomplished by pumping solutions from an injection point to a recovery point which is limited exclusively to water saturated soil. This work describes a new variation of in-situ soil reinforcement technology by using surface percolation via – for example – spray irrigation onto dry, free draining ground, such as dunes or dykes. In order to accomplish bacterial immobilization by surface percolation, it was necessary to alternately percolate bacterial suspension and cementation solution (CaCl2 and urea) to form sequential solution layers within the sand columns. By allowing Ca2+ ions diffusion between each layer bacterial immobilization could be enhanced from 30% to 80%. For a limited number of about 3 to 4 treatments this novel application method of cementation allowed homogeneous strength over the depth of the entire 1 m sand column. Although the strength was homogenous, CaCO3 analysis showed that about 3 times less crystals were precipitated in the top layer compared to the bottom layers suggesting differences in efficiency of the calcite crystal to provide strength. This work demonstrated that this efficiency of calcite crystals was related to the pore water content of the continuously drained column with less water content enabling more efficient strength formation. The geotechnical properties of bio-cemented sand samples under different degrees of saturation confirmed that higher strength could be obtained at lower degrees of saturation. To our knowledge, this study was the first study to demonstrate that the calcite crystals formed under a lower degree of saturation had more crystals formed in the contact points, contributing to the strength of the cemented samples. These preferred crystal formation was caused by the retained cementation solution situated in the form of menisci between sand particles at low degree of saturation. Scanning electron microscopy supported the idea that lower water contents lead to selective positioning of crystals at the bridging points between sand grains. After biocementation treatment, fine sand samples exhibited significant increase in cohesion from 1.1 to 280 kPa and friction angle from 23o to 41o. Similar improvements were also obtained for coarse sand samples. Overall, fine sand sample indicated higher cohesion but lower friction angle than coarse sand samples having similar CaCO3 content. The performance of cementation in large (2 m) laboratory scale trials indicated that subsequent treatments of more 4 times in fine sand caused clogging close to the injection end, resulting in limited cementation depth less than 1 m. This clogging problem was not observed in the 2 m treated coarse sand column, which had strength varying between 850 to 2067 kPa. This showed that the surface percolation technology was more applicable for coarse sand soil. The laboratory large scale application (80 L) of fine sand cementation indicated that relatively homogenous cementation in the horizontal direction could be achieved with 80% of cemented sand having strength between 2 to 2.5 MPa. This suggested that although the liquid infiltration flow paths could not be controlled in the surface percolation method, self-adjusting flow paths were triggered by the changed internal flow resistance caused by the precipitated crystals, favoring the homogeneous cementation. A simple mathematical model demonstrated that the cementation depth is dependent on the infiltration rate of cementation solution and the immobilized urease activity. Higher infiltration rate and lower urease activity will enable in deeper cementation. The model also predicted that repeated treatments will enhance sand clogging close to the injection point. The traditional production of ureolytic bacteria used for biocementation is very expensive, because of strictly sterile processing. This study described the sustainable, non-sterile production of urease enzyme using activated sludge as inoculum. By using selective conditions (high pH and high ammonia concentration) for the target ureolytic bacteria plus the presence of urea as the enzyme substrate, highly active ureolytic bacteria, physiologically resembling Bacillus pasteurii were enriched and continuously produced from chemostat operation of the bioreactor. When using a pH of 10, and about 0.17 M urea in a yeast extract based medium ureolytic bacteria developed under aerobic chemostat operation at hydraulic retention times of about 10 h with urease levels of about 60 U/ml culture. This activity is six times higher than required for successful biocementation. The protein rich yeast extract medium could be replaced by commercial milk powder or by lysed activated sludge, which could make the industrial production less costly. A method of in-situ production of urease activity was developed. This method involved providing selective growth medium to allow ureolytic bacteria to proliferate and produce urease activity in-situ of sand column. The aerobic ureolytic bacteria inoculum could only be enriched in unsaturated coarse sand column, where sufficient oxygen was available. However, high urease activities of 20 and 10 U/mL were obtained by growing soil bacteria under aerobic and anaerobic conditions respectively. The successful enrichment of highly urease active bacteria under anaerobic conditions could allow the in-situ production of urease activity at water logged soils. The in-situ produced urease activities by the enriched soil ureolytic bacteria were sufficient to allow successful cementation of fine (>500 kPa) and coarse (>1000 kPa) sand columns. The strength and CaCO3 analysis indicated that the common obstacle of surface clogging in deeper fine sand column was avoided, explained by avoiding bacterial accumulation at the top of the column. In combination, all findings of the present study imply that the cost of MICP technology can be reduced by optimizing the conditions for effective crystals precipitation by providing low saturation conditions when the cementation is operated. The cost reduction can also be achieved by producing urease activity more economically by omitting the requirement of sterilization (non-sterile cultivation) and bioreactor (in-situ growth). These are expected to make this technology more readily acceptable for field applications.