Biogrouting with microbial-induced carbonate precipitation (MICP) for improving the physical and mechanical properties of granular soils potential liquefaction

Abstract

Biogrouting, a method to enhance soil properties using microorganisms and mechanical techniques, has shown great potential for soil improvement. Most studies focus on small sand columns in labs, but recent tests used 0.5 m plastic boxes filled with sand stabilized with microorganisms and fly ash. The experiments, conducted over 30 days, applied injection and infusion methods with microbial fluids, maintaining groundwater levels to simulate field conditions. Mechanical properties were analyzed through unconfined compressive strength (UCS) tests on extracted samples. Researchers also assessed calcium carbonate distribution and shear strength. Results showed water saturation significantly influenced vertical stress (qu), while UCS correlated with the permeability of sand containing varying calcium carbonate levels. Bacillus safensis, a resilient bacterium used in this process, can withstand extreme conditions. After completing its task, it enters a dormant state and reactivates when needed. The bacteria produce calcium carbonate by binding calcium with enzymes, which cements soil particles, enhancing strength and stability.

• •Testing enzymes on microbes and natural soil
• •Installation settings for drip tools using infusion
• •Soil resistance testing after stabilization using UCS

Graphical abstract

Specifications table

Subject area:
More specific subject area:	Microbial bioremediation, Eco-friendly Soil Improvement
Name of your method:	Microbial bioremediation
Name and reference of original method:	Not applicable
Resource availability:	Not applicable

Background

Referring to the liquefaction potential distribution map of Yogyakarta [1], it shows that the Yogyakarta International Airport (YIA) research location has the highest liquefaction potential. Which is due to the high groundwater table and mainly sandy soil. Under such conditions, the soil at the YIA area becomes the most critical to investigate in order to minimize the effects of liquefaction that could occur during an earthquake. Preserved soil conditions using the undisturbed method at 1 meter depth below the ground surface using Shelby tubes for natural soil.This study does differ from previous work in that it includes the influence of soil salinity. This is because YIA is located in the coastal area, and the salinity of the groundwater in the coastal area is not high. Similar studies have indicated that salinity highly impacts the strength of soil and structural buildings due to higher corrosion rates in structures. Thus, the bioremediation process in this study involves the effect of salinity on the conditions of the groundwater table, which were utilized in the formulation of the soil samples [2].

In recent years, a novel biogrouting technique utilizing biocement as the primary grouting material has been introduced. This process effectively binds sand particles by facilitating cementation within the sand. The deposition of calcium carbonate occurs in proximity to particle contact points, which serves to reduce pore throat sizes and thereby impede the flow of water. In comparison to traditional cement, the biogrout exhibits a markedly reduced viscosity, enabling it to flow in a manner analogous to water [3] .

Micrometer-sized particles aid in penetrating small pores or fissures in rock formations. Moreover, the grout's ability to flow freely in soil reduces the energy needed for widespread distribution in comparison to cement. Furthermore, the physical properties of the grout can be altered by environmental factors like pH or chemical reagent concentration, enabling precise and targeted application. Moreover, the utilization of industrial by-products or waste materials as resources for biogrouting [[4], [5], [6], [7], [8]] offers an economical and sustainable solution to a range of geotechnical challenges. The introduction of biogrouting represents a novel approach to construction, whereby biogrout is employed as a material for ground improvement or soil stabilization. In contrast to traditional grout materials, such as cement or chemical compounds (e.g., silicates, acrylates, acrylamides, and polyurethanes), biogrout utilizes gel-forming biopolymers or biomineralization as binding or sealing agents across a range of applications [9]. These applications include seepage control, erosion remediation, the formation of subsurface barriers, the manipulation of subsurface flow paths, the treatment or remediation of fractures, and the sealing of ponds or reservoirs, among others. Biomineralization is the process in which minerals are precipitated through various microbiological activities by different biological organisms [[10], [11], [12]].

The formation of minerals occurs as a direct result of cellular activity or as a consequence of chemical alterations in the environment due to microbiological processes [13,14]. The most prevalent technique for ground improvement is microbial-induced carbonate precipitation (MICP). It has been established that MICP can occur through a range of metabolic processes, including urea hydrolysis, photosynthesis, sulfate reduction, and denitrification [[15], [16], [17]]. In essence, these processes serve to enhance the saturation conditions for calcium carbonate precipitation. Among these processes, urea hydrolysis is considered the most efficient method for producing a carbonate source, thus aiding in calcium carbonate precipitation when a soluble calcium source is present. This method has attracted considerable research interest on a global scale [3,18]. Unless otherwise stated, the term "biogrouting" in this study is used to refer to carbonate precipitation induced by ureolytic bacteria.

Method details

Biogrouting

Biogrouting represents an innovative construction technique that employs a novel material, designated as biogrout, with the objective of enhancing ground stability and reinforcing soil. In contrast to traditional grouting materials, such as cement or chemical compounds (including silicates, acrylates, acrylamides, and polyurethanes), biogrout utilizes gel-forming biopolymers or biomineralization as binding or sealing agents for a range of applications. Such applications include the control of seepage, the mitigation of erosion, the creation of underground barriers, the alteration of subsurface flow paths, the treatment of fractures, and the sealing of ponds or reservoirs. The term ‘biomineralization’ is used to describe the precipitation of minerals through microbiological processes carried out by various organisms [3,17,19]. Mineral formation can be attributed to either direct cellular activity or chemical changes in the environment induced by microbial processes. [18,20]. The most prevalent method for ground improvement is microbial-induced carbonate precipitation (MICP).

MICP may be facilitated by a number of metabolic processes, including urea hydrolysis, photosynthesis, sulfate reduction, and denitrification. These processes serve to enhance the saturation state of calcium carbonate, as outlined in the studies conducted by [15,21]. Additionally, [[22], [23], [24]] have highlighted the potential role of these processes in promoting MICP. Among these processes, urea hydrolysis is considered the most effective in producing a carbonate source, thereby promoting calcium carbonate precipitation in the presence of a soluble calcium source. Unless specified otherwise, the biogrouting discussed in this study primarily focuses on the precipitation of carbonates induced by ureolytic bacteria.

Microorganisms for biogrouting

The urease-producing bacterial strain employed in this study was Bacillus safensis, as illustrated in Fig. 1. The observations indicated the ability of the bacterial isolate to degrade urea, substantiated by positive results from both urease and bacterial protease assays. Hence, it can be inferred that the sample has the ability to synthesize the urease enzyme, which hydrolyzes urea to produce ammonium.

Fig. 1.

Bacteria test result, a) SEM 2500x, b) SEM 10000x, c) enzyme test bacteria.

Moreover, in Fig. 1c, the bacteria can hydrolyze casein, as evidenced by a clear zone around the bacteria. Thus, it can be suggested that the bacterial sample can produce the caseinase enzyme. Caseinase is a protease enzyme that can hydrolyze casein into its constituent amino acids [2]. The formation of a clear zone (b) around the colony (a) is a reliable indicator of protease enzyme activity (see Fig. 1c).

Model preparation

A controlled environment was set up to replicate the expected conditions and techniques used in real-world engineering applications of grouting. This allowed for the simulation of in-situ biogrouting processes. To this end, three model tests were conducted, with dimensions of 0.270 × 0.190 × 0.210 m, 0.362 × 0.24 × 0.156 m, and 0.557 × 0.384 × 0.344 m, respectively (see Fig. 2).

Fig. 2.

Sample box size.

The tests were conducted in a controlled environment similar to the conditions and grouting techniques expected in actual engineering practice. The first model uses fully stabilized soil, while the second utilizes soil that has undergone initial stabilization on the surface. The third model includes soil stabilization in the specified area, with the addition of a 7mm-thick retaining wall serving as a boundary and retainer for both the initial and stabilized soil (see Fig. 3).

Fig. 3.

A model of the preparation sample.

The distinction in box size is designed to facilitate the observation of the MICP's performance on biogrouting by focusing on the development of bacteria. Box 1 (small box) is intended to assess this aspect, while Box 2 (medium box) is designed to evaluate the capillary process of groundwater level simulation. Box 3 (large box) is intended to assess the performance of simulated field conditions, with a portion of the area stabilized using the retaining wall method and a controlled groundwater level, thereby approximating the conditions encountered in the field.

Soil preparation

The initial soil testing was carried out to determine the characteristics of the loose sandy soil through the use of physical, mechanical, and chemical tests in conjunction with soil stabilization with fly ash. The physical properties were investigated using soil from the Yogyakarta International Airport area. The results of the sieve test showed that the soil was identified as loose sand, and a summary of the soil physical tests is presented in Table 1.

Table 1.

Soil physical properties test YIA soil.

Gravimetri dan Volumetri
1. Specific Gravity	3.12
2. Water Content, w (%)	26.47
3. Soil Volume Weight, γt (gr/cm3)	1.60
4. Volume Dry Weight, γd (gr/cm3)	1.20
5. Void ratio, e	0.42
6. Degree of saturation, Sr (%)	100
7. Cohesion, c (kg/cm2)	1 × 10−16
8. Friction angle, ø (°)	21.81
9. Gravel (%)	0.11
10. Sand (%)	96.86
11. Silt (%)	3.04
12. Clay (%)	0

The sieve analysis graph presented in Fig. 4 indicates that the soil conditions at the research site are predominantly sandy, with a percentage exceeding 96 %. The prevalence of sandy soil types, in particular, increases the potential for liquefaction. The objective of the design mix formula is to determine the optimal mix content for improving potentially liquefiable soils with coarse-grained, homogeneous soil types or sandy soils [1].

Fig. 4.

Sieve analysis graph.

A microscopic model experiment to investigate the sealing of fractures by MICP, while simultaneously examining the impact of hydraulic environmental factors. In the context of flow conditions, the spatial distribution of calcite carbonate precipitation on fracture surfaces is found to be contingent upon fluid velocity [7,25]. The precipitation forms on the surfaces of bacteria, while the settling mechanism regulates the transport of the aggregated to the fracture surfaces [26,27]. The settling velocity interacts with the fluid shear velocity, resulting in the suppression of calcium carbonate deposition due to entrainment [[27], [28], [29]]. Furthermore, the introduction of calcium carbonate was observed to result in a more channelized flow within the fractures, with increased deposition occurring in areas of low flow and decreased deposition in high-velocity channels. Consequently, injection flow rates that are insufficient may result in the obstruction of calcium carbonate precipitation at the injection site, preventing uniform distribution across the fracture. Conversely, excessively high injection flow rates may dissolve the calcium carbonate precipitation, leading to the formation of seepage channels within the fractures.

The soil that underwent remolding was initially a sandy granular soil, with an initial void ratio of approximately 0.927 and a relative density of Dr = 10 %. Two apertures were filled with geotextile membranes, which were installed for the purposes of injection and extraction. The installation of the microbial fluid injection apparatus also illustrates the geotextile membrane in Fig. 5a. The objective of the membrane is to filter the soil, thereby preventing the clogging of the injection holes. Throughout the soil modelling process, all three boxes were treated identically. Additionally, an aperture was created and connected to a transparent plastic hose. This setup facilitates the monitoring of groundwater levels in the field, with a 2 cm depth below the surface and a groundwater table depth of 2 m from the ground surface. Groundwater levels are monitored using transparent pipes and the connected vessel method, allowing for the visualization of the groundwater level in the sample, as shown in Fig 5b.

Fig. 5.

Equipment preparation includes: a) infusion installation, b) drip hole, c) communicating vessels.

In experimental models, administering infusions requires initially calculating the appropriate dose, considering various factors. These factors include the number of infusion droplets per minute, which depends on both the infusion volume and the administration rate. The formula mentioned is affected by both macro and micro drip factors, as shown in Eq. (1) and Eq. (2).

$$Dropletsperminute\left(Tpm\right)=\frac{VxFT}{tx60minutes}$$ (1)

$$Durationofinfusion\left(t\right)=\frac{VxFT}{Tpmx60minutes}$$ (2)

Where:

Tpm: droplets per minute,

t: duration of infusion (hours),

V: fluid requirement (ml),

FT: drip factor or number of drops per 1 ml.

The drip factor is the number of infusion drops per 1 ml.

The drip factor is categorized as macro and micro. Macro drip is commonly used for samples over 7 kg in volume. In Indonesia, two types of macro drip are available, differentiated by the infusion set brand:

• a.The Otsuka brand has a drip factor of 15 drops per milliliter, where one drop equals 0.067 ml.
• b.The Terumo brand has a drip factor of 20 drops per milliliter, equivalent to 0.05 ml per drop

Moreover, macro drip infusion sets with a drip factor of 10 drops per minute are rarely found in Indonesia. Micro drip is usually used for samples under 7 kg. The micro drip factor is 60 drops/mL, meaning one drop equals 0.01665 mL (refer to Fig.5a).

Method validation

After the biogrouting treatment and a curing period of 28 days, the sand models within the small, medium, and large boxes were dismantled to conduct tests assessing their engineering properties. Upon disassembly, it was noted that the stabilized soil conditions showed more solidity than the original soil. The groundwater table conditions in the samples remained consistent with the initial model design. The initial soil conditions were not affected by the stabilized soil (refer to Fig. 6).

Fig. 6.

Sample condition, a) sample condition in box, b) stabilized soil.

The stabilized soil model was divided into three sections based on location points (A, B, and C), calculated from the end of the sand sample. Each section consists of three samples molded using the unconfined compressive strength (UCS) testing method. The samples have a height of 80 mm and a diameter of 36 mm and are cylindrical, allowing for the assessment of engineering properties, as illustrated in Fig. 7.

Fig. 7.

UCS sample test preparation, a) test object parts, b) UCS sample soil.

The results from the Unconfined Compressive Strength (UCS) testing conducted on the three variations of the box model show a clear trend in stress values across the tested conditions at points A, B, and C. Significantly, point B consistently shows higher UCS stress values than points A and C, suggesting a localized strength increase under the specified conditions. This pattern is clearly depicted and further substantiated by the data outlined in Table 2.

Table 2.

UCS Sampel Test.

Sampel UCS test	Box 1 (Small box) (kg/cm2)	Box 2 (medium box) (kg/cm2)	Box 3 (large box) (kg/cm2)
A	205.832	138.624	85.764
B	268.726	213.720	179.971
C	198.062	90.543	105.764

The data presented in the table above demonstrates that the sample testing conditions at point B exhibit the highest compressive strength results when compared to conditions A and C. The graph of the test results can be seen in Fig. 8. This is because the conditions at point B are more stable, as there is no direct contact with the surrounding soil. Therefore, the stabilization and cementing processes can occur effectively and without interference.

Fig. 8.

The results of the compressive strength tests conducted on various box sample types.

The grain analysis graph illustrates an increase in grain dimensions of the sieve analysis results, attributed to the cementing reaction between fly ash and microbes. This reaction produces CaCO₃, serving as a bond between the granules of the sandy soil particles. The sieve analysis results demonstrate that the rise in soil grain size percentage is a consequence of the cementing reaction, as illustrated in Fig. 9.

Fig. 9.

Sieve analysis graph before and after.

Subsequently, a bacterial counting test was conducted on the sample results to determine the extent of bacterial proliferation and viability in the cementing and MICP processes. The test demonstrated the successful execution of the MICP process, evidenced by the sustained viability of bacteria over time. This evaluation was done through measuring bacterial survival and proliferation at the testing intervals of 7, 14, and 28 days. The resulting bacterial counting test graph is shown in Table 3.

Table 3.

Bacteria Count.

Test time (days)	Sample code	Bacteria count CFU/gr
7	S1	191 × 101
14	S2	98 × 101
28	S3	570 × 102

The review of bacterial samples utilizes the Total Plate Count (TPC) analysis method. The TPC method is straightforward yet effective for quantifying the total number of viable bacteria in a sample. The process involves sampling, serial dilutions, plating on agar media, incubation, colony counting, and a final enumeration of the bacterial population. This method is crucial for quality assurance and monitoring environmental impact risks.

Table 3 illustrates that the bacterial count at seven days of age was 191 × 10¹ CFU/gr. At 14 days, the bacterial count decreased to 98 × 10¹, which can be attributed to the completion of the bacterial task of producing urease and protease enzymes and binding calcium from fly ash [1,2] thereby successfully forming CaCO₃. Upon completion of their function, the bacteria entered a dormant state until cracks in the soil permitted reactivation and the subsequent repetition of the process, during which soil particles were bound with the CaCO₃ produced in the reaction. This is corroborated by the bacterial count at 28 days, which demonstrates a notable increase to 570 × 10², presumably due to the reactivation of dormant bacteria prompted by soil fissures. Furthermore, the bacterial infusion applied during the biogrouting process facilitated the regeneration of bacteria that had perished.

in Fig. 10 explains that sample conditioning during testing has a great effect on bacteria number during counting tests. In combination 1 (C1) is a natural soil condition with normal air pH, combination 2 (C2) is a natural soil condition with high salinity air condition, combination 3 (C3) is soil sterilized by first being dried using an oven at normal pH, and combination 4 (C4) is soil sterilized with high salinity air. This test must be conducted to assess bacterial resistance to various conditions with the influence of native bacteria, as well as the influence of changes in soil and air characteristics. Results from the test at the age of 7, 14 and 21 days showed that in C1 and C2 at the incubation age of 7 and 14 days,the turning test results showed the highest number of bacteria were 137 × 10³ CFU/gr and 143 × 10³ CFU/gr, this is because the soil is not sterilized first, so the number of safensis bacteria coexists with the original bacteria. However, at the age of 21 days,the same because the bacteria through dormancy and died, which is 25.4 × 10³ CFU/gr. This decrease was possible because there was little reproduction at 7 and 14-day ages, where in C3 and C4 combination conditions, the results of the total count of bacteria was low, namely 1.91 × 10³ CFU /gr and 0.98 × 10³ CFU/gr possibly because safensis bacteria were still adapting to the environment. At the age of 21 days, there was a significant increase in the results of the total count of bacteria totaling 57 × 10³ CFU/gr.Curiouser and curious. in salinity conditions, the highest total number of bacteria was found in combination 4, indicating that safensis bacteria are resistant to extreme salinity 3.4 %.

Fig. 10.

Bacterial count.

The observed increase in grains can be attributed to the biocementing reaction, which involves the binding of soil particles to form larger aggregates. Chemical reactions between fly ash and microbes result in the formation of CaCO₃, acting as a binding agent, causing the soil grains to adhere to one another and form bonds between particles. The MICP process can be more readily discerned in the SEM test results, as illustrated in Fig. 11[2].

Fig. 11.

SEM scanning for stabilized soil, (a) 5000x zoom, (b) 10000x zoom.

Supplementary material and/or additional information [OPTIONAL]

Not applicable

Limitations

Not applicable

Ethics statements

Not applicable

Credit author statement

Diana: Conceptualization, Methodology, Writing original draft & editing. Soemitro: Review, Supervision. Ekaputri: Review, Supervision. Satrya: Review, Supervision. Warnana: Review, Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.