Effect of natural carbonates on microbially induced calcite precipitation process

Abstract

Microbially induced calcite precipitation (MICP) is an emerging ground improvement technique that uses microbes to induce cementation between soil particles. To date, the majority of research has focused on exploring MICP with silica-rich sands; however, the present study investigates the process and efficacy of MICP in a carbonate-rich natural soil, and a comparison is made with benchmark silica-rich sands. MICP column experiments were performed with a range of treatment formulations to optimize and understand the MICP process in carbonate-rich soil. Performance was quantified using chemical (pH, urea, and ammonium concentrations) and physical measurements (TGA and LOI tests). Micro-scale characterization of the cemented soils was performed with XRD, SEM, and EDS, while shear-wave velocity (V_s) and unconfined compressive strength tests were performed to evaluate the effect of precipitated calcite on macroscopic engineering properties. Natural carbonates were found to have a significant impact on the MICP process, resulting in an increase in MICP efficiency of 23% and increases in precipitated calcite contents by as much as 82% when compared to benchmark silica-rich soils receiving similar treatments. These results suggest that the presence of natural carbonate minerals within soils may lower the energy barrier and act as preferential sites for calcite precipitation during the MICP process. Furthermore, SEM images highlighted the association of bacterial cells with precipitated calcite crystals, differences in calcite morphologies and more widespread cementation bonds in carbonate-rich soil when compared to silica sand. Generated cementation also resulted in a linear increase in V_s with increases in precipitated calcite contents for MICP treated carbonate-rich soil, consistent with past results for silica sands. Lastly, differences in yeast extract concentrations applied in treatment solutions were also found to significantly impact the development of ureolytic microbial capacity and the efficiency of the MICP process in the considered soils.

Keywords: Chemical measurements; Microbially induced calcite precipitation; Microscale characterization; Natural carbonates; Shear-wave velocity.

Fig. 1

Schematic of the MICP column setup.

Fig. 2

Comparison of precipitated calcite profiles between carbonate-rich Blessington soil and silica-rich concrete and Ottawa F-110 sands.

Fig. 3

Comparison of shear-wave velocity profiles between carbonate-rich Blessington soil and silica-rich concrete and Ottawa F-110 sands.

Fig. 4

Comparison of precipitated calcite profiles for different MICP treatment formulations for Blessington, Ottawa, and concrete sands.

Fig. 5

pH measurements before and after pumping of each treatment for different MICP column experiments: (A) BS_0.02_50_1:1 and BS_0.02_350_1.4:1, (B) BS_0.2_50_1:1, BS_0.2_350_1.4:1, BS_0.2_350_1:1, and BS_2_350_1:1, (C) BS_0.2_50_1:1 and BS_0.2_50_1:1_COM, and (D) CS_0.2_50_1:1, CS_0.2_50_1:1_COM, OS_0.2_350_1:1, and OS_2_350_1:1.

Fig. 6

Aqueous urea concentrations before and after pumping for each treatment for different MICP column experiments: (A) BS_0.02_50_1:1 and BS_0.02_350_1.4:1, (B) BS_0.2_50_1:1 and BS_0.2_350_1.4:1, (C) BS_0.2_50_1:1 and BS_0.2_50_1:1_COM, and (D) CS_0.2_50_1:1 and CS_0.2_50_1:1_COM.

Fig. 7

Concentrations of aqueous urea during time course sampling at different stimulation and cementation treatment cycles for different MICP treatment formulations: (A) 2nd stimulation treatment, (B) 4th stimulation treatment, (C) 6th stimulation treatment, (D) 2nd cementation treatment – YE 0.2 g/L, (E) 6th cementation treatment – YE 0.2 g/L, (F) 10th cementation treatment – YE 0.2 g/L, (G) 2nd cementation treatment – YE 0.02 g/L, (H) 6th cementation treatment – YE 0.02 g/L, and (I) 10th cementation treatment – YE 0.02 g/L.

Fig. 8

Changes in aqueous ammonium concentrations from the pore fluid of BS_0.2_50_1:1_COM and CS_0.2_50_1:1_COM versus (A) rinse treatment and (B) number of pore volumes. Ammonium removal versus rinse treatment for rinsed columns (BS_0.2_50_1:1_COM and CS_0.2_50_1:1_COM) in (C) cumulative ammonium concentration and (D) percent ammonium removal with respect to sorption concentrations obtained from unrinsed columns. Ammonium concentrations in unrinsed columns (BS_0.2_50_1:1 and CS_0.2_50_1:1) along the column length from (E) soil pore fluid and (F) sorbed ammonium in soil samples.

Fig. 9

Thermogravimetric measurements for pure calcium carbonate, untreated, and MICP treated Blessington, concrete, and Ottawa sands. (A) Normalized mass fraction curves for untreated and MICP treated Blessington soil samples, (B) Normalized mass fraction curves for untreated and MICP treated concrete and Ottawa F-110 sand samples, (C) DTG (first differential of mass loss) curves for untreated and MICP treated Blessington soil samples, and (D) DTG curves for untreated and MICP treated concrete and Ottawa F-110 sand samples.

Fig. 10

Comparison of X-ray diffraction patterns of pure calcium carbonate and MICP treated samples. (A) Pure calcium carbonate, (B) OS_0.2_350_1:1, (C) CS_0.2_50_1:1, and (D) BS_0.2_350_1:1.

Fig. 11

Scanning electron microscopy views of MICP treated Blessington sand: (A) Presence of microbes (shown in orange color) on the surface and within the calcite crystals, (B) Morphology of precipitated calcite crystals on the surface of a soil particle, (C) Muscovite particle cemented between two other soil particles, (D) Larger, blocky calcite crystals in the cemented samples (E) Aragonite polymorph and cemented soil particles, and (F) Low magnification view showing a representative view of cemented soil unit: widespread precipitated calcite crystals, cementation bonds and matrix, and calcium carbonate polymorphs.

Fig. 12

Energy X-ray dispersive spectroscopy (SEM–EDS) of MICP treated Blessington and Ottawa F-110 sand for comparison of precipitated calcite morphology. (A,B) SEM view and SEM–EDS layered image, respectively, of MICP treated Blessington sand showing the detected elements. (C,D) Silicon and Calcium element map scan, respectively, of MICP treated Blessington sand show in (A,B). (E–F) SEM view and SEM–EDS layered image, respectively, of MICP treated Ottawa F-110 sand showing the detected elements. The difference in precipitated calcite morphology on natural carbonate and silica particles can be observed from SEM–EDS and elemental scans.

Fig. 13

Shear-wave velocity profiles with increasing cementation treatments for different MICP treatment formulations of Blessington and concrete sands.

Fig. 14

Unconfined compressive strength testing results for three MICP treated Blessington sand samples: BS_0.02_50_1:1, BS_0.02_350_1.4:1, and BS_0.2_350_1.4:1.