Non-invasive imaging of the effect of injection strategy on the spatial and temporal development of enzymatically-induced calcite precipitation

Induced calcite precipitation, where CaCO₃ closes voids inside porous media and unconsolidated samples are solidified, is an important technique in geotechnical engineering. To optimize these applications, it is crucial to understand how the dynamics of mineral precipitation affect flow and transport in porous media. The aim of this study is to investigate how different injection strategies affect the spatial and temporal development of calcite precipitation using time-lapse non-invasive imaging with magnetic resonance imaging (MRI) and X-ray microcomputed tomography (µXRCT). These two imaging methods are complementary because µXRCT aims to detect structural changes of the solid matrix, whereas MRI focuses on the liquid phase in the pore space. Together, these methods enable time-resolved observations of the three-dimensional development of porosity, and thus have the potential to offer valuable insights into the spatial and temporal dynamics of the precipitation process.

 

We performed two distinct types of experiments to induce precipitation by simultaneous injection of a cementing solution consisting of 0.5 M CaCl₂ and 0.5 M urea and an enzyme solution containing 5.0 g/l of Jack Bean meal into homogeneous sand packings prepared in 30 mm long sample cuvettes with a diameter of 15 mm. Two injection strategies were realized. In a first experiment, a constant flow rate of 0.01 mL/s was maintained during six injection cycles. Pressure development was monitored in parallel. In a second experiment, the solutions were injected  at a constant pressure that was increased stepwise during six cycles from initially 50 mbar to 300 mbar to maintain moderate flow rates. Following each cycle, both samples were imaged using XRCT and MRI and the intrinsic permeability was determined.

 

Imaging results indicate that calcite preciptation occured more strongly close to the inlet, as manifested by water content and relaxation maps from MRI and density maps from XRCT. Only during the last two injection cycles, zones with increased precipitation became visible in the center of the column. The MRI relaxation maps suggest a reduction in pore size due to precipitation, which agreed with increased surface-to-volume ratio of the pores. Vertical porosity profiles derived from XRCT showed an average change of 12 and 11 vol.% for the constant flow and constant pressure inection strategies, respectively, and confirmed the non-uniform distribution observed with MRI. The permeability decreased by two orders of magnitude for both injection strategies. However, this decrease was achieved already after 90 injected pore volumes in case of the constant pressure injection strategy, whereas the constant flow strategy required 165 pore volumes for a comparable decrease. This is attributed to the increased tendency for preferential flow in case of the constant-rate injection strategy, but this needs to be confirmed through a detailed analysis of the variability of calcite precipation within the sample cross-section. Overall, this study showed the feasibility of monitoring induced calcite precipitation using both MRI and XRCT.